Content uploaded by Justus K. Musya
Author content
All content in this area was uploaded by Justus K. Musya on Jul 26, 2023
Content may be subject to copyright.
[Type the
document
title]
i
Ethnicity & Political
Violence
A Kenyan Perspective with Reference to
Mathare and Kibra Informal Settlements
Justus K. Musya, PhD
ii
COPYRIGHT@ 2023
JUSTUS K. MUSYA
ALL RIGHTS RESERVED
Contents and / or cover should not be reproduced in all or parts in
any form without the express authority of the author.
To contact the author:
P.O Box 40299 -00100 Nbi.
Tell No. 0721412562
ISBN No. 978-9914-49-439-6
Email: musyajustus@gmail.com
Published by
The Kairos Book Publishers
P.O. Box 1213 Nairobi , KENYA
Tel: +254 (0) 726244403 / +254 (0) 716302054
Email: cyrilchewhite@yahoo.com
iii
DEDICATION
I dedicate this book, "Ethnicity and Political Violence: A Kenyan
Perspective with Reference to Mathare and Kibra Informal
Settlements," to my late father, Musya Musyimi, and my mother,
Wanza Musya, whose love, support, and sacrifices have brought me
to where I am today. Your wisdom and guidance have been an
invaluable source of strength throughout my journey.
To my dear wife Elizabeth and our beloved children David and
Rose, thank you for your unwavering support and encouragement.
You have been my pillars of strength and have inspired me to work
harder and strive for excellence.
I also dedicate this book to the people of Mathare and Kibra informal
settlements, who have endured the devastating effects of political
violence and ethnicity. Your stories, experiences, and resilience have
been a driving force behind this work, and I hope this book will
contribute to shedding light on the challenges you face and the
solutions needed to address them.
To the community leaders, activists, and organizations who tirelessly
work towards promoting peace, social justice, and harmony, your
efforts are invaluable, and I dedicate this book to you.
Finally, to all the scholars, researchers, and practitioners who have
dedicated their lives to understanding and finding solutions to the
complex issue of political violence and ethnicity, I dedicate this
book to you. Your contributions and insights have been invaluable,
and I hope this book will add to the ongoing conversation and efforts
towards building a more just and peaceful society.
iv
ACKNOWLEDGEMENT
I want to express my heartfelt gratitude to all those who played a part in
helping me complete this work. Although it's impossible to thank each one
of you individually, I am deeply indebted to the following people for their
invaluable contributions.
First and foremost, I extend my profound thanks to Prof. Frank K. Matanga
of Masinde Muliro University, Prof. Maurice N. Amutabi of Technical
University of Kenya, and Prof. Nyamasyo formerly of Nairobi University.
Their generous sharing of insights, inspirations, and encouragement
propelled me to achieve my goals.
I would also like to acknowledge the late Mr. Thomas Ochieng, who
supported me in line editing the work and provided technical advice. Tom,
your guidance in collecting and analyzing data was invaluable, and I will
always treasure your friendship and support. Rest in peace, and may God
bless your family always.
I express my sincere thanks to Dr. Jacqueline Stinton for her financial and
moral support in conducting this research. Jackie, your kindness,
friendship, and unwavering support during times of need will always be
remembered.
I am also grateful to the members of AIC Jericho Church and my local
home church, AIC Mui, for their spiritual, financial, moral, psychological,
emotional, and physical support. Your assistance in financing my
education has helped me become who I am today, and for that, I will
forever be grateful.
Finally, I want to acknowledge the Lord Almighty, whose faithfulness has
seen me through all my academic pursuits, from nursery school to my
doctoral level. It's only by His grace that I have been able to come this far.
God, you alone are worthy of all the glory and honor.
Thank you all for being part of my journey.
v
Foreword by Julius Gathogo
Besides addressing the original and the deleterious effects of
negative ethnicity on Mathare and Kibra Informal Settlements,
within the city of Nairobi, this book also theorizes on the
broader philosophy behind ethnic tensions, and to an extent,
xenophobic behaviours troubling the post-colonial Africa. In
particular, the role of inequalities in causing ethnic grievances,
its role in fuelling ethnic mobilization, its global perspective, its
regional dimension, its national dimension, and its local
perspectives have all been surveyed so as to guide us in
addressing divisive citizenry. By the time we are done with
reading these dimensions of ethnic perspectives, we are driven
to start thinking deeper. And indeed, we are enabled to think
critically and creatively and brought to the understanding of the
problem: namely, the negative ethnicity (derogatorily called
tribalism) and we are simultaneously driven to start working
towards solutions and/or problem-solving.
Put differently: why shouldn‟t ethnic diversity help us
appreciate our great favours from God, as plurality is God‟s
economy for the world? Why should Africans utilise ethnic
cards to mobilize electoral politics? Why should we vote in a
person simply because he or she is from our local ethnic group?
Does it matter who messes up or builds the country; aren‟t they
the same practical terms? In the worst case scenario, we are
driven to ask: Why use ethnic mobilisation to instigate political
violence that will eventually lead to displacements, as in the
case of post December 2007 disputed elections where over 400,
000 were displaced from their farms; and Kenya became a home
of internally displaced persons (IDPs)?
Besides the above, this book theorizes the philosophy revolving
around ethnic conflicts. How can these theories help in
vi
understanding ethnic conflicts? The author has ably strived to
make sense out of this. In a nutshell, negative ethnicity is caused
by lack of vision among the leaders and/or elites of society. In
some cases, historical injustices, rooted in colonial hegemony,
have had their share of promoting ethnic divisions and
conflicts. Take for instance, the amalgamation of Nigeria of
1914, for instance. Yes, it created the North and South, can we
view it as a marriage of inconvenience, particularly when we
recall the terrorist activities in the north? Certainly, the
amalgamation of North and South Sudan since 1956 left a scar
within the rank-and-file of Southerners, the black Africans, who
felt that the Arabs in northern part were enslaving them. This
resulted in endless wars between the Arabs and the black
Africans. In 1983, the Sudanese People‟s Liberation Army (SPLA)
was formed so as to battle the then Arab-dominated
government of the northern Sudan. This was followed by the
formation of the Sudan People‟s Liberation Movement (SPLM),
as the political wing of SPLA. The formation of SPLM on 16th
May 1983 came after the Arab-dominated Sudan‟s government
reneged on the so-called Addis Ababa Agreement of 1972. This
agreement was previously signed by the then Sudanese
President Gaafar Nimeiry (1930-2009), who reigned from 1969 to
1985, and the then rebel movement, Anyanya leader, Joseph
Lagu, on 27th February 1972. It is in this Addis Ababa Agreement
where the Southern Sudanese were to be given economic,
social, political, religious, and educational rights. The signing of
the Addis Ababa Agreement of 1972 had thus brought down the
long ranging conflict.
Prior to this, Anglo-Egyptian Sudan had ceased after
independence in 1956, as it now became Sudan. This came
after both the Egyptian and the British governments recognised
Sudan as an independent country; and eventually terminated
vii
their respective shares of sovereignty over it. Reneging on the
1972 Addis Ababa Agreement of recognising the Southern
Sudanese religiosity by Field Marshal Jaafar Muhammad
Nimeiry (1930-2009), who served as President from 25 May 1969
to 6 April 1985, and his decision to begin Islamist rule, speaks for
the nature of African conflicts: a phenomenon where we
„suspend‟ our governing „constitutions‟ from family to national
levels without any iota of shame. Ripples are that we become
our own enemies who fuels conflicts that can be ethnic, racial,
social, political or xenophobic based. Thus the Sudanese case,
of reneging on a memorandum of understanding, ushered in a
conflict between the practitioners of Islam religion (northerners)
and the Christians and traditionalist of South Sudan, a
contestation that did not die till 2011 when South Sudan
became an independent country under the SPLA/SPLM in 2011.
Back to negative ethnicity, we appreciate that conflicts are also
caused by religious intolerance, as the Sudanese case has
demonstrated. Religious intolerance is manifested by blind
religiosity or blind denominationalism that doesn‟t see anything
good in others. Equally, blind ethnic loyalty doesn‟t see
anything good in other ethnic groups existing in the same
country. It is a psychological disorder that needs to be
diagnosed through public education. As a social construct, it
can be deconstructed through re-educating, de-education, and
be eventually reconstructed for the betterment of society. A
society that fails to educate people on the dangers of negative
ethnicity or racial-national prejudices faces extinction from the
map of the world. The height of madness in any nation reaches
its climax when a notoriously religious continent and her
countries allow religion to become the opium of, and for the
people, thereby eulogizing religio-denominational intolerance
and dehumanisations. When politicians use religion to confuse
viii
the vulnerable poor-weak-and-hungry citizenry, as they strive to
promote the Big Man‟s Syndrome and/or Messianic political
leaderships, the true rapture gets closer and closer. It could as
well mark the proverbial “last days.” Are we under borrowed
time? In Uganda, the contestations between the Anglicans who
were mainly members of Milton Obote‟s (1925-2005) Uganda
People‟s Congress versus the Democratic Party under
Paul
Kawanga
Ssemogerere
(1932-2022), whose supporters were
mainly Catholics, polarised Uganda for a long time; and has
remained a wound that takes long to heal. When a religious-
inclined society views some politicians as Mosaic-Messiah‟s,
while their contestants are effectively portrayed as Devil-
incarnates, it becomes a preparatory route for ending the life of
a nation. Besides this, all forms of marginalization, resource
control, and lack of vision among the emerging leaderships in
Africa, must be re-evaluated from time to time so as to ensure
that ethnic plurality does not become a curse. Rather, our rich
cultural diversity has to remain our strength rather than as our
weakness; and indeed, it has to remain the source of health and
wealth of a nation.
Having said this, I wish to recommend Justus K. Musya‟s book:
“Ethnicity & Political Violence: A Kenyan Perspective with
Reference to Mathare and Kibra Informal Settlements.” It brings
back the theme of Ethnicity that ought to be addressed from
time to time. With negative ethnicity, ethnic balkanisation,
banditry, refugee crisis, religious intolerance, narcotics,
dangerous liquor, terrorism, economic mismanagement,
suspicious and mysterious pandemics, and bad politics,
threatening the very existence of the African populace in the
twenty-first century, a sharp-shooter, of Musya‟s level, is needed
to outfox the enemy; and eventually call the society back to its
conscience. We thus need the likes of Musya to come and say:
“Look here! We are staring at danger; hence stop mishandling
ix
ethnic cards. Use it to enrich the society rather than impoverish
the very society that God so much loved.” The book is
recommendable to scholars of all nations under the sun!
Referring to the Africanist scholars in particular, the aspiring
scholars, and readers and leaders of all walks of life. This is your
book, and indeed it is a little encyclopaedia on how to handle
the ethnic card.
Julius Gathogo, PhD (Theology), PhD (History), PhD
(Educational Leadership)
Research Fellow, UNISA, South Africa, Senior Lecturer
Kenyatta University & a visiting Distinguished Professor of
Missiology and historiography, ANCCI University
c/o PO. Box 16778 Mombasa, Kenya
Email: juliusgathogo@gmail.com
4 April 2023
x
TABLE OF CONTENTS
DEDICATION iii
ACKNOWLEDGMENT iv
TABLE OF CONTENTS x
OPERATIONALIZATION OF TERMS AND CONCEPTS xvii
ABBREVIATIONS xx
Chapter One 1
INTRODUCTION 1
1.1 Background to the Study 1
1.2 Statement of the Problem 8
1.3 Research Objectives 9
1.4 Research Questions 10
1.5 Justification of the Study 10
1.6 Scope of the Study 14
1.7 Chapter Summary 16
Chapter Two 17
LITERATURE REVIEW 17
2.1 Role of Horizontal Inequalities in Causing Ethnic
Grievances 17
2.2 Role of Ethnic Grievances in FuellingEthnic
Mobilization 38
2.2.1 Global Level Perspectives 38
2.2.2 Regional Level Perspectives 41
2.2.3 National Level Perspectives 44
2.2.4 Local Level Perspectives 51
2.3 Role of Ethnic Mobilization in Instigating Political
violence 58
2.3.1 Global Level Perspectives 58
2.3.2 Regional Level Perspectives 63
2.3.3 National Level Perspectives 66
2.3.4 Local Level Perspectives 71
xi
2.4 Methodological Approaches in Relevant Studies 75
2.5 The Theoretical Framework 84
2.5.1 Theories of Ethnic Conflict 84
2.5.2 Application of Theories of Ethnic Conflict
to the Present Study 85
2.6 Summary and Areas of Focus of the Present Study 94
Chapter Three 99
RESEARCH METHODOLOGY 99
3.1 Research Design 99
3.2 Research Area 104
3.3 Study Population 108
3.4 Study Sample and Sampling Procedure 108
3.4.1 Probability Sampling for Quantitative Research 108
3.4.2 Non-Probability Sampling 112
3.5 Data Collection Instruments 113
3.5.1 Quantitative Instruments 113
3.5.2 Qualitative Instruments 115
3.6 Reliability of the Research Instruments 117
3.6.1 Reliability of the Scales 117
3.7 Exploratory Factor Analysis of Scales Developed 118
3.8 Confirmatory Factor Analysis 121
3.9 Validity of the Instruments 126
3.9.1 The Questionnaire 126
3.9.2 Trustworthiness of Qualitative Interviews 128
3.10 Data Analysis Procedures 129
3.10.1 The Quantitative Data 136
3.10.1 Descriptive Analysis 136
xii
3.10.2 The Qualitative Data 137
3.11 Limitations of the Study 138
3.12 Ethical Considerations 141
3.13 Chapter Summary 142
Chapter Four 143
QUALITY OF LIFE DIFFERENCES AND ETHNIC GRIEVANCES 143
4.1 Summary of the Validation Procedures for Quality of Life
Scale 143
4.2 A Comparative Analysis of Respondents in Relation to
Education, Livelihood, and Monthly Earnings 146
4.3 Descriptive Analysis of Quality of Life Construct 151
4.3.1 Study Area 151
4.3.2 Descriptive Analysis of Quality of Life for the
In Group 154
4.3.4 Descriptive Analysis of Quality of Life for
“Other group” 159
4.4 A Confirmatory Investigation into the Difference of
Sample Means of the In-group and the Out-group to
Quality of Life 161
4.4.1 Independent Means Test 161
4.5 The Validation Procedures of Ethnic Grievance Scale 163
4.6 A Comparative Analysis of the Ethnic Grievances
of Ethnic Groups 166
4.6.1 Descriptive Analysis of Ethnic Grievances in
Study Area 167
4.6.2 Descriptive Analysis of Ethnic Grievances in the
In-Group 169
xiii
4.6.3 Descriptive Analysis of Ethnic Grievances
in “Other group” 173
4.7 Confirmatory Investigation into Differences in Ethnic
Grievances of Ethnic Groups 174
4.7.1 Independent t-test for Ethnic Grievance
Scores per Ethnicity 174
4.7.2 Contingency Analysis of Ethnicity and Livelihood of
Respondents in Study Area 176
4.7.3 Contingency Analysis of Relationship between
Ethnicity and Livelihood among the In-group,
the Out-group, and “Other group” 182
4.8 A Confirmatory Investigation into Relationship between
Quality of Life Differences and Ethic Grievances 189
4.8.1 Correlation Analysis in Study Area 189
4.8.2 Regression Analysis for Link between Quality
of Life and Ethnic Grievances in Study Area 190
4.8.3 Correlation Analysis of Relationship between
Quality of Life and Ethnic Grievances in In-Group 192
4.8.4 Regression Analysis of Relationship between
Quality of Life and Ethnic Grievances in In-Group 193
4.8.5 Correlation Analysis of Relationship between
Quality of Life and Ethnic Grievances Out-Group 194
4.8.6 Regression Analysis of Relationship between
Quality of Life and Ethnic Grievances Out-Group 194
4.8.7 Correlation Analysis of Relationship between
Quality of Life and Ethnic Grievances Out-Group 195
xiv
4.8.8 Regression Analysis of Relationship between
Quality of Life and Ethnic Grievances Out-Group 196
4.8.9 Comparative Analysis of Correlation and
Regression Coefficients in the Link between Quality
of Life Differences and Ethnic Grievances by Ethnicity 196
Chapter Five
ETHNIC GRIEVANCES AND ETHNIC MOBILIZATION 209
5.1 Procedure for Scale Validation of the Ethnic
Mobilization Scale 209
5.1.1 Reliability and Validity of Ethnic Mobilization
Instrument 209
5.2 A Comparative Analysis of Respondents in Relation to
Political Developments of 2007 211
5.2.1 Descriptive Analysis for Ethnic Mobilization
for Study Area 216
5.2.3 Descriptive Analysis for Ethnic Mobilization of
Out-Group 221
5.2.4 Descriptive Analysis for Ethnic Mobilization of
Other-Group 224
5.3 A Confirmatory Investigation into the Differences
in Ethnic Mobilization in relation to In-Group and
Out-group, and Other group 226
5.3.1 Independent Means Test 226
5.4 A Confirmatory Analysis for the Link between Ethnic
Grievances and Ethnic Mobilization 235
5.4.1 Correlation Analysis of Link between
Ethnic Grievances and Political Mobilization 235
xv
5.4.3 Regression Analysis for Link between Ethnic
Grievance and Ethnic Mobilization 238
5.4.4 Correlation Analysis for Link between Ethnic
Grievance and Ethnic Mobilization for Out Group 239
5.4.5 Regression Analysis for Link between Ethnic
Grievance and Ethnic Mobilization for Out Group 240
5.4.6 Regression Analysis for Link between Ethnic
Grievance and Ethnic Mobilization 240
5.4.7 Comparison of Correlation and Regression
Coefficients of Link between Ethnic Grievances and
Ethnic Mobilization of In-groups, Out group,
and Other Group 241
5.5 Chapter Summary 242
Chapter Six 245
6.1 Summary of Procedures for Validating the
Political violence Scale 245
6.2 Exploratory Analysis for Political violenceScale 246
6.2.1 Descriptive Analysis for Study Area 246
6.2.2 Descriptive Analysis for Political violence
Instrument of In-group 249
6.2.4 Descriptive Analysis for Political violence
Instrument of the other group 254
6.3 A Confirmatory Investigation of Factors of Political
violence 257
6.3.1 Independent Means Test of the Differences
in Mean between the In-Group and Out-Group for Political
violence Mean Scores 257
xvi
6.3.2 A Confirmatory Investigation into the Association
between Ethnic Mobilization and Political violence in Study
Area 263
6.4 A Confirmatory Investigation into the Association
between Ethnic mobilization and Political violence
for In-Group 267
6.5 A Confirmatory Investigation into the Association
between Ethnic Mobilization and Political violence
for Outgroup 269
6.6 A Confirmatory Investigation into the Association
between Ethnic Mobilization and Political violence
of Other Group 271
6.7 Chapter Summary 273
Chapter Seven 275
SUMMARY, CONCLUSION, AND RECOMMENDATIONS 275
7.0 Introduction 275
7.1 Assessment of the Variations in Quality of Life of
Ethnic Groups in Relation to Ethnic 277
7.2 Analysis of Whether Ethnic Grievances in Mathare
And Kibra Have Been Used For Ethnic Mobilization In
Political Contests 281
7.3 The Role of Ethnic Mobilizationin Contributing to the
Escalation of Political Violence in Mathare and Kibra 285
7.4 Conclusion of the Study 289
7.5 Recommendations 289
8.0 REFERENCES 291
xvii
OPERATIONALIZATION OF TERMS AND CONCEPTS
Conflict Analysis: It is used to mean and to include
assessing the conflict mechanism
including context and structures,
plus mapping needs and fears of
parties in conflict.
Conflict Behaviour: It describes how in-groups and
out-groups viewed and treated
each other.
Conflict Escalation: It stands for the chain of conflict
triggered by ethnic grievances,
which led to ethnic mobilization,
and then to political violence.
Conflict Situation: It means the state of ethnic
relations due to the ethno-
political
conflict between in-group and
out-groups in the study area.
Conflict Transformation: It represents the moves
that could be taken to weaken the
ethno-political factors that
legitimise, validate, and animate
political violence in the study
area.
Ethnicity: It was used to label respondents
based on their identification with
an ethnic group, an identity that
framed how they behaved
politically.
xviii
Ethnic Grievances: It stands for group grievances
that arise out of discriminatory
policies and actions of ruling
elites creating spatial inequality
and/or horizontal inequalities
In-group: It defines political elite and ethnic
groups who benefit from state
patronage and resources for
human and community
development. In this proposal, the
in-group would be people who
voted for the incumbent
presidential candidate during the
2007 presidential election and
who the literature suggests
enjoys the traits of an in-group,
the Kikuyu.
Informal Settlements: It defines the areas of human
habitation, within the wider
Mathare and Kibra constituencies
in Nairobi county, in which land
title is irregular or illegal and
which is under invested in terms
of basic amenities, such as
schools, health facilities, and
other infrastructure.
Out-group: It refers to political elite and
ethnic groups who are deprived
of state resources, including
patronage and development
resources. In this proposal, the
xix
out-group is presumed to be
people who voted for the
opposition presidential candidate
during the 2007 presidential
election and who the literature
supposes bear the marks of an
out-group, the Luo and Luhya.
Political violence: It means the conflict attitude and
behaviour involving members of
the in-group and out-group
towards each other.
Ethnic Mobilization: It is a measure of how elites use
their ethnic groups in political
action including violence.
Quality of Life Differences: Refers to social, economic,
and cultural differences of ethnic
groups due to ethnic intolerance.
It is a measure of how in-groups
and out-groups interpreted their
quality of life based on ethnicity.
xx
ABBREVIATIONS
CBO: Community Based Organization
CDF: Community Development Fund
CFI: Comparative Fix Index
CFA: Confirmatory Factor Analysis
CIPEV: The Commission of Inquiry into Post-Election
Violence
CL: Community Leader
COD: Coefficient of Determination
DHS: Democratic and Health Survey
ECK: Electoral Commission of Kenya
EPR: Ethnic Power Relations Instrument
FBO: Faith Based Organizations
FGD: Focus Group Discussion
GFI: Goodness of Fit Index
GoK: The Government of Kenya
KNHRC: The Kenya National Human Rights
Commission
KMO: Kaiser-Meyer-Olkin
LDP: Liberal Democratic Party of Kenya
LISREL: Linear Structural Relations
LSK: The Law Society of Kenya
MAR: Minority at Risk
MFAF: Media Focus on Africa Foundation
NAK: National Alliance Party of Kenya
NFI: Normed fit Index
NGO: Non-governmental Organization
NICC: The National Integration and Cohesion
Commission
NSC: National Steering Committee on Peace
Building
xxi
NSIS: The National Security Intelligence Service
NNFI: Non-normed fit Index
NSCPB: The National Steering Committee on Peace
Building
ODM: The Orange Democratic Movement
ODM-K Orange Democratic Movement Kenya
PEV: Post-Election Violence
PNU: The Party of National Unity
PPP: Purchasing Power Parity
RMSEA: Root Mean Error Square of Approximation
SEM: Structural Equation Modelling
SRMR: Standardized Root Mean Square Residue
TJRC: The Truth, Justice, and Reconciliation
Commission
UNDP: The United Nations Development Programme
xxii
1
Chapter One
INTRODUCTION
In this section, the background from which the present
study ensued, defines the research problem, and
enumerates the objectives of the research is provided.
Thereafter, the merits of doing the study and outlines the
probable value of its findings. The scope of the study is
then specified.
1.1 Background to the Study
A growing body of evidence relates political violencewith
ethnicity.At the global level, (Østby 2007) argues that any
sound conflict analysis should chart or trace the course
and factors of conflicts that emanate from state-induced
inequalitiesthat triggers ethnic grievances, constituting
the onset of political violence. It is not the severity of
inequalities per se that contribute to ethnic grievance;
but rather, it is the comparisons that an ethnic group
make about their quality of life, vis a vis other groups.
Østby found that socio-economic inequalities are vital to
the onset of conflict and called on further research at the
sub-national level to assess how “national political
variables”, such as electoral systems,contribute to
political violences. Østby calls on further research at the
sub-national level to assess how “national political
2
variables”, such as regime type and electoral system,
influence conflict at this lower level.Likewise, Østby
(2006) called forresearch to explain the aetiology and
dynamics of ethno-political violences, in the context of
the inequalities and grievance dynamics, in theatres of
conflict, whether at national or sub-national level.
From regional studies, the major knowledge gap
identified was the need to developand test measures on
“multi-dimensional polarization” and test these estimates
of ethnicity on political violence. This research directive
invites analysts to develop scales to estimate dimensions
of ethnicity and assess their impact on political violence,
including at the sub-national level (Fjedle and Østby’s
(2012). Stewart (2008) investigate the role of politics on
social and economic inequalities in Kenya. Since its
independence, she argues, Kenya has witnessed conflict-
inducing inequalities, both social and economic as well as
political. Observed is the unbalanced distribution of
ethnic groups in senior executive positions, in the cabinet
and civil service especially. Violence broke out in Nyanza,
Western, and Coast provinces in 2007 presidential
electionsin regions that were underdeveloped, regions
populated by the out-group (Stewart, 2008). Implicitly,
the 2007 election can be used to study the in-group and
out-group phenomenon, the causal mechanism of
political violence in Mathare and Kibra. Rent disputes
assume an ethnic dimensionties, as it were Kinyanjui and
Mutsotso (2002) examined was the nature and
operations of the landlord-tenant relationship, a
relationship characterised by suspicion, hostility, and
3
hatred” (p.7), a relationship that is a dummy for
horizontal inequalities among ethnic groups. Further
research was needed to enlarge on Kinyanjui and
Mutsotso findings by providing answers on how and why
the seeming horizontal inequalities inn the study area
structured conflict behaviour during the PEV.
Theliterature treats the relationship between quality of
life differences and ethnic grievances as foundational to
the causal chain of political violence. Efforts to
understand the onset of political violences in Mathare
and Kibra would requires articulation of this
relationship.
On the role of ethnic grievances on conflict, Murshed and
Tadjoeddin (2007) surveyed developing countries
worldwide, with special attention given to 17 countries
with the highest incidence of political violence. With half
the countries surveyed in Africa, this study is rich in
insights and perspectives on the anatomy and
mechanism of political violence.Murshed and Tadjoeddin
(2007) developed a mechanism with which to measure,
not just social and economic inequalities, but also ethnic
grievances. Among the areas of research proposed by
Murshed and Tadjoeddin (2007) is the “need for data on
regionally based indices for horizontal inequalities
requiring information not only on income, assets, health,
and education, but also on subjective measures of well-
being”. Importantly, then, Murshed and Tadjoeddin
(2007) argue that a need existed to develop a mechanism
with which to measure, not just horizontal
4
inequalitiesbut also ethnic grievances, including at sub-
national levels.
At the national level, the Media Focus on Africa
Foundation MFAF examined the factors, actor, and
remedies for conflict in Kenya. This study grew out of the
2007 post-election violence and involved a nationally
representative sample. In regions of the country that
experienced political violence, during the 2007 post-
election violence, ethnicity was found to be the leading
cause of conflict. By far, the link between tribalism,
politics, and poverty is most to blame for political
violence. There was need to expand MFAF’s work by
developint scales that could estimate ethnic grievances of
ethnic groups and analyse their political implications in
an electoral setting.
Ethnic mobilization appears to be the proximate cause of
political violence, especially in a political and electoral
context.In a global study, Fearon (2004) examined what
explained the variations in political ethnicity in some
regions and countries of the world, by addressing two
important subjects: the factors of political ethnicity and
ethnic violence. Regarding the former subject, Fearon
(2004) argues that ethnic mobilization is easier to effect
than mobilizing other forms of identity, such as religion,
because political elites find it more convenient to
mobilize within such groups on socioeconomic grounds
or precisely due to ethnic grievances. Ethnic mobilization
morphs into violencewhen the elite of the out-group and
their ethnic groups feel frustrated that their status is
unlikely to change (Fearon, 2004).The motivations of key
5
actors who instigate violence then needs to be
understood in settings where violence occurs,
particularly the impact of disputed elections which the
out-group can interpret as hindering unfairly their
escape from marginalization.
In ethnically fragmented societies, both ruling and
opposition political elite would mobilize their ethnic
groups to retain or acquirepower (Cederman, Gleditsch,
and Hug 2013). This fight to control public resources,
economic and political, tends to be associated with
discriminative state systems and practices. Elections
reinforce ethnic cleavages and weaken national cohesion.
Ruling elite can impede efforts of elite from the out-
group to mobilize effectively, including through the
threat of exclusion from benefitting from state resources
and the use of violence in campaigns. Such actions only
serve to heighten ethnic tension and sharpen the in-
group and out-group divide.There is need to explore how
these elite dynamics at mobilization play out in theatres
of conflict and assess their impact on relationships
between ethnic groups who have a history of ethnic-
based conflict.
But some writers argue that the ethnic competition
perspective, which writers such as Cederman, et al,
2013) propose, only tells us who fights over what and
how they do the fighting. Even if national elections are
treated as a theatre of conflict, ethno-political violence is
a complex that cannot be adequately explained by ethnic
mobilization (Vermeesch, 2011). There is need for
6
careful accounts that explains why mobilization keels
over to violence. Ethnic violence is more likely to occur
when political leaders, from both the in-group and out-
group, have a motive to engage in violence, have the
means to sponsor such political manoeuvres, have the
means to sponsor such political manoeuvres, and could
do so (Vermeesch, 2011:11). He calls for research in
countries experiencing or susceptible to political violence
to establish “various interacting forces . . . responsible for
turning ethnic mobilization into violence.The precise
factors that cause the radicalization of ethnic
mobilization or political ethnicity leading to violence
needs to be identified. Central to understanding the
factors the cause of radicalization of ethnic mobilization
is knowing the factors that are associated with elevated
levels of ethnic mobilization and what the implications of
this is on political violence.
At the regional level, Eifert, Miquel, and Posner (2011)
examined the determinants of ethnic identification in ten
African countries, between 1999-2004. Such
identification grows appreciably during a competitive
presidential election. However, ethnic mobilization need
not involve violence. What remains unclear, nonetheless,
are the factors that cause such mobilization to tip over to
violence. Laakso (2007) examined the problem of
electoral violence in Africa, using Kenya, Tanzania, and
Zimbabwe. Some of the reasons these countries have
witnessed serious post-electoral violence is due to the
semi-democratic nature of their polities, with a
centralised executive. Countries with a culture of
7
election-related violence and those with elections that
are neither free nor fair or with electoral systems that
are not prepared to produce dishonest results.
Horizontal inequalities and identity-based political
parties only aggravated the problemin such political
settings (Laakso, 2007). In this vein, Omotola (2008:71)
calls for further research into “political economy of
electoral violence, including its democratic pay-offs. The
Kenya Human Rights Commission documented the
aetiology, dynamics, and consequences of the 2007 post-
election violence in the country, including in the study
area. The report suggests the post-election violence
(PEV)bore the traits of organization and direction. On
this score, there is need to identify agency in the
outbreak of violence, including what the goals of agents
were and how these goals relate with the goals of the
general population. Such evidence will contribute to the
body of knowledge on the factors associated with the
radicalization in the study area.
In the literature on the causal factors of political violence
in the study area are either qualitative in nature or
involve descriptive statistical analysis. Some of these
studies include (CIPEV, 2008; KHRC, 2008; Okombo and
Sana, 2010, and Murunga, 2011). As such, knowledge
produced so lacked the scientific rigour needed to make
firm conclusions about the causal factors and network of
the political violence in the study area. The general
literature on the factors of political has several major
methodological weaknesses. Many such works, including
notable studies by Collier, Hoeffler, and Rohner (2009),
8
conceptualise these conflict as insurgencies. This
definitional assumption leaves out non-armed rebellions
of the genre of the 2007 post-election violence. The
recurrent political violence in the study area are
consistent, not with an armed rebellion or insurgency,
but with inter-tribal clashes.It is against this substantive
and methodological weaknesses that the study emerged
and these deficiencies provide the context in which the
research problem, described below, is defined, justified,
and made urgent.
1.2 Statement of the Problem
Unattended, the conflict situation in Mathare and Kibra
would likely re-escalate. For the National Steering
Committee on Peace Building, renewed fighting in
Mathare and Kibra would likely poison the conflict
situation and leave it unyielding to conflict
transformation. Indeed, “after crossing the threshold of
violence, the conflict dynamic becomes not only more
destructive, but also very costly and difficult to
transform” (National Steering Committee on Peace
Building, NSC, 2010: 4). Renewed violence could entail
serious spill over effects. Re-escalation of political
violence, as (Miall, 2004: 7) rightly avers, tend to
“broaden (suck in new issues), widen (suck in new
actors), and intensify (suck in new victims)”. In Mathare
and Kibra, rent disputes have previously morphed into
ethno-political disputes, which have then diffused to
other slum areas (Dafne, 2009). The possible spiralling of
the conflict situation, in Mathare and Kibra, needs to be
9
stemmed. In short, unless such studies are done, efforts
to transform the conflict in Mathare and Kibra would be
“misguided and inappropriate” (Hendricks, 2009: 24)
and could even worsen the conflict situation.Despite
efforts to address the conflict situation, every time there
is an election, ethnic conflicts escalate.
Ignorance about the causal network and consequences of
the recurrent political violencebears steep costs. It is
detrimental to understanding how to deal with the
conflict situation there. It frustrates efforts to identify the
conflict’s structures and dynamics, including how
conflicting parties define, interpret, and respond to
conflict issues and controversies. It weakens the
understanding of the interplay and weight of factors that
trigger violence. It hinders efforts to understand the role
of agents of the conflict, whose actions are instrumental
to conflict escalation and de-escalation. Accordingly, this
study analysed the role of ethnicity in the political
violences in the study area.
1.3 Research Objectives
The general objective was to determine the role of
ethnicity in the perpetuation of political violence in
Mathare and Kibra. The specific objectives were to:
1. Assess the variations in quality of life of ethnic groups
in relation to ethnic grievances that are presumed to
be basic to the political violence in Mathare and Kibra;
2. Analyse whether ethnic grievances in Mathare and
Kibra have been used for ethnic mobilization
10
inpolitical contests;
3. Examine if ethnic mobilization has contributed to the
outbreak of political violence in Mathare and Kibra;
1.4 Research Questions
1. How do variations in the quality of life of ethnic
group's impact on ethnic grievances, grievances that
are presumed to be basic to the political violence in
Mathare and Kibra?
2. How have ethnic grievances influenced the nature and
level of ethnic mobilization used in political contests
inMathare and Kibra?
3. To what extent do the ethnic mobilization of ethnic
groups explain the outbreak of political violence in
Mathare and Kibra?
1.5 Justification of the Study
The burgeoning field of peace and conflict studies needs
growth especially in relation to the causal paths of
political violence. Highlighted in most of the commentary
on the subject, including the best known, the most
influential, and the most widely cited analyses on
political violence or violence from an international
perspective, is the need for articulation of causal
mechanism of ethno-political violences in national
settings. Involving as they do the examination of these
variables in different contexts and over considerable
time, such studies identify shared factors of political
violence and even explain differences in patterns or
11
trends in specific cases of conflict. These studies afford
analysts to make meaningful cross-national comparisons
of data, among other things.
Rubenstein (2008) called on researchers to explain the
undercurrents of conflict behaviour, particularly why
political economy controversies at the national level
trigger violence at the sub-national level. Shapiro (2006)
urged peace researchers to explain the role of
psychological factors in escalating conflict, specifically on
how conflicting parties think of themselves, their rivals,
and conflict issues. Sirin, Villabolos, and Nehemiah
(2011) wanted investigators to flesh out the actors,
values, and interests impeding conflict transformation in
theatres of conflict. Herera, Johnstone, and McDermott
(2006) wanted researchers to operationalise ethnic
identity and show how this identity shapes conflict
behaviour.These studies supply a strong ground for
justifying further research in national and even sub-
national levels.
Existing studies on the political violence in Mathare and
Kibra involve one-dimensional explanations of the
presumed effect of horizontal inequalities on ethnic
conflict. These studies do not specify the causal
mechanism of the conflict in the study area. For example,
Shilloh’s (2008) work was a descriptive analysis that
focussed mainly on land as an issue of conflict. The
findings have given us insights on the path of conflict
escalation and defined its moving parts. Some studies
(Republic of Kenya, 2009) portray a generous
12
understanding of the multi-layered dimensions of conflict
by describing the dynamics (the intensity) and triggers
(including the roles of actors) of the conflict. Still, these
studies do not reveal the conflict issues, including the
positions, interest, needs, and fears of conflicting parties.
Pointedly, research on the political violence in Mathare
and Kibra is undeveloped in two main areas. The causal
mechanism of the conflict lacks rigorous analyses. The
best way to explore the dynamics, structure, and agency
of the conflict would be by testing a theory, such as the
instrumental theory of ethno-political violence. Such a
studied exercise would mean applying a comprehensive
research methodology and using sophisticated statistical
procedures.
Systematic analyses of the conflict situation in Mathare
and Kibra were wanting. The transformation of conflict in
Mathare and Kibra would benefit immensely from
thorough research that produced evidence that allowed
policy makers and peace analysts to understand better
the causes and consequences of conflict situations there.
There was scarcity of studies that carefully examined the
conflict situation in the study area, using a sound
theoretical model and a rigorous, scientific methodology.
By gathering much-needed empirical data and
perspectives to support these efforts of peace building,
this study has fulfilled important evidence and
information shortfalls. To anticipate renewed ethnic
fighting in Mathare and Kibra, there was need to
understand better the mechanics and dynamics of the
conflict situation.
13
This research aimed to fill important knowledge gaps in
the burgeoning field of peace and conflict studies in
Kenya. We need to understand the motives of actors, who
were presumed to have been involved in escalating and
de-escalating the PEV. In the qualitative dimension of the
proposed work, the study determined the agents of
mobilization and demobilization, what are their goals;
what resources have they, and how do they relate with
the general population. Done in functional, rather than
personal, terms, this analysis will contribute to the body
of knowledge on the factors associated with the
radicalization in the study area, insights that will have
implications for ethnic conflicts occurring in the nation.
The findings will support effforts to build pece and
transform the recurrent political violence in Mathare and
Kibra. This meant producing evidence to understand the
structures of conflict in the study area through a
competent conflict analysis, with weakening of the
structures of conflict and conflict transformation in mind.
Policy makers are better armed with evidence with which
to interpret the ethnic conflict in Mathare and Kibra. Part
of the mandate of the National Integration and Cohesion
Commission (NICC) is to assess or evaluate political
violence in the country. This mandate would benefit from
the availability of diagnostic tools, analytical models, and
conceptual frameworks, as well as evidence and
information, all of which were in short supply. Evidence
would enrich programs launched to transform conflict in
Mathare and Kibra.
14
Inventory development and testing are substantial
analytical procedures. These analytical and statistical
operations could constitute the core forensic objective of
any substantial research, such as the one this study has
completed. The developed scales to describe the
dimensions (the conceptualizing) of ethnic conflict in the
study area, allowing analysts to make realistic
comparisons in other hot spots of violence about quality
of life, ethnic grievance, ethnic mobilization, political
violence, and readiness to forgive or reconcile.
Knowledge gleaned would enrich the tools that could be
deployed to monitor and evaluate programs of
transforming ethno-political violence in Kenya.
1.6 Scope of the Study
Mathare and Kibra are known flashpoints of conflict. For
example, of the five slums in
Nairobi County that experienced the post-election
violence, these slum areas were worst affected (The
CIPEV Report, 2008). Political violence in the study area
has tended to escalate periodically; as such, their episodic
nature means they need to be transformed. On this
premise, the conflict situation in Mathare and Kibra is
susceptible to re-escalation. Besides, the "informal
settlements in Nairobi’s remain the most inflammable in
Kenya’s cross-cultural political emotion” (Okombo and
Sana, 2010). Renewed conflict would probably
degenerate the conflict situation in the slum areas into an
intractable phase and unfurl the violence in other slum
areas. Cosmopolitan, Mathare and Kibra are, moreover, a
15
microcosm of ethnic conflict in Kenya, a convenient place
to study conflict formation, escalation, and
transformation.
The instrumentality theory was tested. Hypotheses
involved analysing the relationships between the
following variables: 1) quality of life differences and
ethnic grievances, 2) ethnic grievances and
ethnicmobilization, and 3) ethnic mobilization and
political violence. The context of analysis was the
national election of 2007. A representative sample of
respondents included people who took part in the 2007
national election, rather than adults living in the study
area in general. In line with the instrumentalist theory,
the study focused chiefly on an in-group and an out-
group. The out-group was only used for comparative
purposes. Nevertheless, the study explored the conflict
situation, its factors, and dynamics, before and after the
post-election violence of 2007.
Qualitative interviews were done to enrich our
understanding of the conflict situation. Through in-depth,
semi-structured, and focus group discussions,
information was obtained from community leaders on
the context, history, structures, dynamics, and
consequences of the political violence in Mathare and
Kibra over a ten-year period. These interviews produced
evidence on the respondents’ needs, fears, positions,
interests, and opinions about their readiness to engage in
peace building and conflict transformation. The
interviews looked at the conflict situation from a
16
historical point of view, considering the ethnic clashes or
political violence that occurred since 1992 the subject of
analytical attention. The study did not examine or assess
efforts to transform the conflict in the study area.
1.7 Chapter Summary
This chapter outlined the problem area and subsequently
defined the research problem and drew applicable
research objections and questions. The need and
importance of the research were subsequently outlined,
and the scope, limitations, and assumptions of the
research were enumerated and demarcated. The next
chapter will be the literature review.
17
Chapter Two
LITERATURE REVIEW
A thematic review of the literature on the role of ethnicity
in political violences was done. The review was organised
thematically as follows: 1) horizontal inequalities and
ethnic grievances, 2) ethnic grievances and ethnic
mobilization, and 3) ethnic mobilization and political
violence. In this chapter, discussions are held as well on
the conceptual framework, theoretical model, and
methodological stances related to this study.
2.1 Role of Horizontal Inequalities in Causing
Ethnic Grievances
2.1.1 Global Level Perspectives
Research into the relationship between quality of life
differences and collective or ethnic grievances places
partisan institutional norms structures at the centre of
ethno-political violence. Ethno-political violences are
likely to occur in quasi-democracies, with majoritatian
arrangements, which have been associated with
asymmetrical distribution of state resource (Østby,
2007). This practice of punishing elites and ethnic groups
who are deemed disloyal is characterised by
18
differentiation of quality of life between loyal and
disloyal ethnic groups. Wimmer Cederman, and Minn
(2007) found that ethnic exclusion is a powerful
explanatory factor in political violence. Elsewhere,
Cederman, Wimmer, and Minn (2010, p.2) rightly posit
“competing ethno-nationalist claims over the state
constitute the driving force behind many internal
conflicts”.
Global studies identify common antecedents and factors
of political violence and give insights on its causal
mechanism. In these studies, there is a growing
agreement that the state is an instrumental player in such
conflicts (Cederman, Weidman, and Gleditsch, 2010). The
state becomes an agent in the conflict by pursuing
policies and practices that create socioeconomic
inequalities between ethnic groups, inequalities that
form the onset of political violence. Therefore, political
violence conflicts are most likely to occur in settings in
semi-democracies, polities that blend features of a
democracy and autocracy. Østby 2007 found that socio-
economic inequalities are related to the inception of
political violence and are influenced significantly by
regime type: semi-democracies or a majoritarian polity,
with the winner takes it all phenomena in force. Only in
certain “political environments”, specifically countries
with majoritarian systems of government are horizontal
inequalities likely to be current. Wimmer, Cederman, and
Minn (2007) present an important framework that
defines the three-fold nature of how the ethnicization of
politics can entail exclusion and polarization.
19
The study was ken to establish the operation of ethnic
exclusion and polarization. The first one refers to the
extent to which some ethnic groups are excluded from
government resources, including situations where a
coalition exists and there is, as is likely to be, a “senior”
and “junior” partner. This kind of unequal power
relationship explains why even democracies could have
political violence and could shed light into the prime
mover of political violence-elite grievances.
Studies have examined the relationship between the
exclusion of some ethnic groups from power (ethnic
exclusion) and (2) unequal power sharing arrangements
in a coalition government, which give rise to “centre
exclusion”, with one partner in the coalition taking on a
junior role in government leading to elite grievances.
Wimmer Cederman, and Minn (2007) found that ethnic
exclusion is a powerful explanatory factor in political
violence. This finding is notable because it undermines, if
not belies, a huge and influential body of evidence that
had tended to disregard ethnicity as a factor in intra-state
conflict. Centre exclusion was found to be a significant
factor, though of less moment than ethnic exclusion, in
explaining political violence. Also observed is that only
the exclusion of “large numbers” of ethnic populations
can occasion ethnic mobilization and action, which can
include violence (p.39). Wimmer, Cederman, and Minn
(2007) have published an immensely important work on
the role of ethnicity in civil conflict at the global level. The
writers argue that the work is ground-breaking, mainly
because of its global scale. The writers did a cross-
20
country regression analysis of civil conflict in all
independent countries in the world since 1945. Before
this study was done, previous studies, such as the
insurgency, ethnic heterogeneity, and minority at risk
models, had failed to prove that ethnicity is a factor in
civil war; treatise is thus a significant movement in
knowledge, signalling as it does a departure from popular
but mistaken renditions on the cause of political violence.
This treatise rightly argues that knowledge about
political violence should be sought by assessing how any
state relates with ethnic groups, and how this
preferential or discriminatory treatment provokes
conflict behaviour among ethnic groups. The writers have
put forward a conceptual framework that comprised of
three elements - horizontal inequality, the exclusion of
ethnic groups from power; vertical inequality in the
segmentation of power, the proportion of power, political
elites control in power sharing regimes; and national
cohesion, the extent to which authority is exercised
effectively from the centre. The three-point
differentiation in the conceptualizing of ethnicity,
Wimmer, Cederman, and Minn (2007) argue, is important
because these three attributes of ethnicity, particularly
segmentation and exclusion, explain different kinds of
political violence and should be treated as such (p.3). In
this study, the trait of exclusion was used as it is most in
line with the notion of in-group and out-group.
(Østby 2007) argues that any sound conflict analysis
should chart or trace the course and factors of conflict,
including identifying the kind of polity that enables
21
asymmetrical distribution of resources. These
inequalities, her book held, are relevant to political
violence because they constitute the onset of the
phenomenon of political violence by triggering ethnic
grievances. For the researcher, however, the role of the
state in political violence conflicts is not limited to its
discriminative practices that lead to inequalities and thus
ethnic grievances in out-groups. In semi-democracies or
in polities where politics is organized about ethnicity,
elections can afford in-groups, who feel threatened by the
out-group, the opportunity to support political actions
aimed at keeping the elite of the group in power. This
bifurcated view of the impact of horizontal inequalities
entails the fear of the in-group in losing their privileges
and for the out-group, the question would be about
ending the experience of under privilege. As it happens,
Østby (2007) calls on further research at the sub-national
level to assess how “national political variables”, such as
regime type and electoral system, influence conflict, at
this lower level. Østby (2007) calls for urgent research to
inform interventions aimed at curbing the escalation of
political violence at national levels. Sub-national level
analysis on the role of horizontal inequalities in political
violence, such as the one done by the researcher, is a first
step in this direction. The spill over effect of this analysis
will help explain the etiology and dynamics of political
violence conflicts at the national level. Analysis of global
studies surface important knowledge gaps in the causal
path of political violence.
22
Cederman, Weidman, and Gleditsch (2010) examined
how the conflict behaviour of in-groups and out-groups is
shaped by horizontal inequalities. The work was set
against the background that global analysis of horizontal
inequalities (inter-group wealth comparisons) and
political violence conflicts, using the insurgency, ethnic
heterogeneity or minority at risk models, provide
mistaken and unconvincing accounts of inter-state
conflict. Indeed, these writers’ work is especially
significant, on this score, because it is, “the first truly
worldwide comparison of horizontal inequality and
ethno-nationalist wars” (p.2). Wide in scope, the study
involved 155 countries and examined political violence
wars between 1946-2005. To test the link between
horizontal inequalities and political violence, Cederman
Weidman, and Gleditsch (2010) found that horizontal
inequalities are the basis of other forms of social and
economic inequality. Social inequality refers chiefly to the
differences in access to health and education and the like.
Third, cultural inequalities refer to settings in which the
cultural lives of some ethnic groups, such as language or
customs, are denigrated, and the like.
Furthermore, Cederman, Weidman, and Gleditsch (2010)
call for further research to delve into the operation of the
“sore-loser logic” (p.11). This study sough to explain how
out-groups process an election loss, in which they
believed their political elites, stood a good chance of
winning. On their part, Wimmer, Cederman, and Minn
(2007) gave the following recommendations to push
forward conflict research. The first one was the need to
23
show how exactly ethnic exclusion, centre exclusion, and
ethnic polarization produce the adverse political violence
in country settings or in particular states. Only by using
this focus and approach, they add, can relevant peace
interventions be applied, interventions that would be
suited to a form of political violence. This study aimed to
uncover and explain the mediating variables of political
violence, especially about two variables: ethnic
grievances and political ethnicity, the elite mobilization of
these grievances, as distinct, yet related, variables. This
analytical effort was done in the context of the 2007
national election. It is the researcher’s contention that
the causal mechanism of political violence conflicts in a
local setting can be examined in the context of a dispute
about the outcome of a general election.
This study responded to Cederman, Weidman, and
Gleditsch (2010) call for investigators to understand the
impact of political and economic inequalities on ethnic
groups in specific country-settings.This study responded
to this area of knowledge deficiency by using Mathare
and Kibra as the unit of analysis. Cederman Weidman,
and Gleditsch (2010) want research undertaken to assess
the impact of social and cultural dimensions of group
inequality. The researcher’s primary interest was in a
setting in which elections was a precursor of inter-ethnic
fights in particular, rather than on the circumstances in
which rebellion or insurgency would break out, the area
of inquiry that the work of Wimmer, Cederman, and Minn
(2007) pursue. Differently put, this study contributed to
knowledge on election-related violence in which the state
24
was not the primary or exclusive source of violence. This
aspect of political violence is not discussed in Wimmer,
Cederman, and Minn’s (2007) important work. The
section below focuses on literature regarding horizontal
inequalities and conflict at the regional level.
2.1.2 Regional Level Perspectives
At the regional level, Østby (2008) has completed
perhaps the most insightful and current analysis of the
link between horizontal inequalities and political
violence. The work is current and broad in scope,
encompassing about 40 per cent of the countries in
Africa, including nations that have experienced the
severest forms of civil war: Rwanda, Uganda, Cote
Devoire, and Chad. A notable feature of this study is that
it examines whether inequalities entail group cohesion,
an outcome that other similar studies, including those
have been reviewed in this research, only assume, rather
than demonstrate. Central to the thesis in this study,
however, is that it is the effect of group, rather than
individual inequalities that need to be in focus and
explained or understood, a bias that introduces rightly
needed fresh thinking into the inequality-conflict debate
by exploring the effect, not just of income inequalities,
but also of ethnic/social and economic polarization.
Horizontal social inequality was found by Østby (2008)
to have the profoundest statistical effect on conflict
followed by socio-economic inequality. On its own,
economic inequality does not seem to generate the same
effects.
25
Cederman, Gleditsch, and Hug (2009)examined the
impact of elections on ethno-political violences. This
study forwards analysis on the connection between
elections and civil way by explaining its specific
mechanisms. In ethnically fragmented societies, two
things are likely to occur. The first one is that political
elite who are in power would likely to mobilize ethnic
groups to keep power and political elite who are in the
opposition are likely to mobilize ethnic groups to take
over power. This fight to control public resources,
economic and political, tends to be associated with
discriminative state practices. Elections would reinforce
ethnic cleavages and weaken national cohesion. More
particularly, the role of elites in making it difficult for the
other to mobilize ethnic support, including the threat of
exclusion from state resources and the use of violence in
campaigns, only serves to heighten ethnic tension. The
study, rightly observes, that such integrity questions
about the electoral process, when coupled with voting
irregularities, could result in violence.
The study under review involved examining a
comparative analysis of countries and ethnic groups
worldwide, along the aforementioned lines and
reasoning, and tested six hypotheses. The following
hypotheses merit attention. One, the prospects of
violence increases after elections and that elections,
whether competitive and non-competitive, increase the
prospect of ethnic conflict. The other hypothesis tested is
that marginalized groups, or out-groups, are more likely
to wage war against the state after an election. An
26
electoral process, the study argues, can serve to humiliate
out-groups further, deepening their grievances;
consequently, these groups could eventually revolt
against the state. Of note in the study’s analysis is the role
of the mediating variable, that is, the impact of the
threats to or changes in power structure and when in-
groups feel that their hold on power is under serious
threat and what happens when out-groups fail capture
power.
Nevertheless, this study considered these findings only
illustrative or facilitative to his primary interest in
elections as a precursor of inter-ethnic fights in
particular, rather than on the circumstances in which
rebellion or insurgency would break out. Differently put,
This study aimed to analyse election-related violence in
which the supports of an incumbent, rather than the state
per se, are the primary target of violence. This aspect of
ethnic conflict is not addressed in Cederman et al (2009).
Nevertheless, the study calls for further research to delve
into the operation of the “sore-loser logic” (p.11) in
ethnic conflict. This study sought to explain how out-
groups processed the announcement of the election loss
on the context of their quality of life.
Large in scope, Fjedle and Otsby’s (2012) study involved
studying communal conflicts, those that did not have the
state as a combatant, in sub-Saharan Africa, between
1990-2008. Two hypotheses were tested: intra-regional
inequality increases likelihood of communal violence;
inequality between groups increases probability of
27
violence. It is inequalities between groups, as a whole,
that led to a higher prospect of conflict. Fjedle and
Otsby’s (2012) main contribution to the inequality-
conflict debate is their focus on inter-group conflict
specifically, rather than insurgency or other forms of
rebellion in which marginalised or oppressed groups take
up arms against the state. Unlike most studies on
inequality, this paper does more than just describe the
impact of inequality on group grievance and mobilization
in broad terms; instead, it trains on intra-ethnic political
violence (such as the post-election violence in Kenya).
This is a major and important shift in the inequality-
conflict discourse - the taking of a sub-national approach
to political violence. Basic to the book in this study is the
need to measure horizontal inequalities among groups in
regions. There is needfor analysts to observe the
operation of the “we” versus “them” phenomenon and
mentality in specific theatres of conflict. It is this form of
categorization and identification, and how they arise, that
is really the core of political violence. As such, its
antecedents and dynamics need fleshing out.
Grudun Østby (2006) probes the link between socio-
economic inequalities and ethnic conflict. These
inequalities, her thesis holds, are factors in conflict
because they generate group grievances, the formative
stage of conflicts of this sort. Østby tests three
hypotheses, of which two are relevant to the rhetorical
objectives in the researcher’s proposal. The first one is
that severe socio-economic inequalities are likely to
increase prospects of ethnic conflict. This study follows
28
Østby’s argument, a received wisdom in the discourse
anyway, that such inequalities could occasion group
grievances. But perhaps the specifying the level of
inequalities that would induce conflict is not neccessary.
These inequalities might have been operationalised in a
way that one could only associate the severity of
inequality with conflict, based on an inferential analysis,
rather than by pre-specifying the conditions in which
such an association might prevail.
Significantly, too, Østby seeks to clarify the causal
mechanism implicit in her first hypotheses by arguing
that only in certain “political environments”, specifically
countries with majoritarian systems of government, are
horizontal inequalities likely to be current. The use of this
mediating variable was helpful because it focusses
attention on the polity that enables asymmetrical
distribution of resources in the first place Indeed, the
winner takes it all presidential elections in Kenya need to
be understood as a matrix of ethnic conflict in this sense.
Østby’s second hypothesis holds that socio-economic
inequalities are brought into sharp relief in semi-
democracies, those that have polities that blend features
of a democracy and autocracy. For the researcher,
however, the role of the state in ethno-political violences
is not limited to its discriminative practices that lead to
inequalities. In such polities, elections can motivate in-
groups, who feel threatened by the out-group, the
opportunity to support political actions designed to keep
them in power, actions that can include the use of
violence against the out-group or the taking of political
29
measures that spark violent reactions from the out-
group. This study followed this bifurcated view of the
impact of horizontal inequalities, existing or threatened,
in understanding ethnic conflict.
Regarding the findings, Ostby found that socio-economic
inequalities are related to the onset of conflict. This
outcome is influenced significantly by regime type, a
winner-takes it all polity. As it happens, Østby calls on
further research at the sub-national level to assess how
“national political variables”, such as regime type and
electoral system, influence conflict at this lower level. She
cautions that such efforts are susceptible to
measurement difficulties, specifically the variable of
quality of life differences.
This study will contribute to the analysis of ethnic
conflict at the national level by developing an inventory
of quality of life differences. This scale will be useful in
testing probable variations in living standards of in-
groups and out-groups at a sub-national level. Østby also
calls for urgent research to inform interventions aimed at
curbing the escalation of such ethnic conflicts at the
national level. Lessons gleaned on the role of horizontal
inequalities in ethno-political violences, at sub-national
level, carry enormous spill over effect in explaining the
aetiology and dynamics of ethno-political violences at the
national level. Whether at the meso or the micro level, it
is really the same phenomenon being observed.
30
From regional studies, it emerged that the major
direction research should take is in developing and
testing measures on “multi-dimensional polarization”
and test their effects on conflict. In this connection, this
study developed and tested three such measures-ethnic
grievance, quality of life differences, and political
ethnicity. Research done at the sub-national level has
implications for research done at national level, hence
satisfying Fjedle and Østby’s (2012) call for extra
research. More precisely, there is need for research in
sub-national settings of conflict, such as the study area of
Mathare and Kibra, which would shed light to the
situational dynamics of conflict-the factors that define
and animate conflict behaviour. The section reviews
literature on the link between horizontal inequalities and
conflict with the case of Kenya in mind.
2.1.3 National Level Perspectives
At the national level, the general literature has sought to
explain how national political dynamics conspired to
engender political violence. Stewart (2008) has done an
insightful commentary on the social and economic
inequalities. Since its independence, she argues, Kenya
has witnessed conflict-inducing inequalities, both social
and economic as well as political. Observed is the
unbalanced distribution of ethnic groups in senior
executive positions, in the cabinet and civil service
especially. She notes that the representation of ethnic
groups in these positions tends to be associated with
whether or not their kinsman is the president (or for that
31
matter, whether the elite from these groups are deemed
to be supportive of the president). The anomaly in
executive representation, as it were, is in whether the
proportion of cabinet officers from one ethnic group
exceeds the proportion of the ethnic group to which these
political elites belong. Ethnic groups that have had their
political elites underrepresented in the cabinet, at a level
lower than their population proportion, seem to liable to
ethnic grievances. Importantly, Stewart (2008) notes that
the 2007 presidential election was all about addressing
the underrepresentation of political elite among the Luo
and Kalenjin elite especially. These political imbalances
are the basis upon which elite instigate their ethnic
supporters to political actions that include violence.
But Stewart’s (2008) interpretation of some facts in the
narrative above can be questioned. She argues that the
reason the imbalances in representation have not led to
violence, consistent with theoretical models, is that
Kenya’s political system has tended to be inclusive, its
skewed representation notwithstanding. But an equally
appealing argument would be that Kenya, for most of its
history, has had an autocratic polity, making it hard for
ethnic mobilization to occur in respect to inequality-
related grievances. A key trait of the inequality-conflict
model holds that it is semi-democracies –such as the one
Kenya has had since the return of competitive elections –
that grievances are likely to be generated.
Stewart (2008) makes a case for social and economic
inequalities in Kenya between 1993-2000. She uses three
indices - availability of potable water, scope of rural
32
electrification, and level of secondary enrolment - to
make her comparative analysis. These indices are helpful
measures of quality of life, implicating as they do issues
of health and social mobility, the things that matter most
to the quality of life for most ordinary folk. But the
political battle in Kenya has always been about economic
resources, with some communities lamenting that they
have been impoverished by state policies. It might have
been helpful if Stewart’s indices of quality of life included
economic investments, the existence of extractive or
other agricultural industries, and the like. Her data
suggest the reason violence broke out in Nyanza,
Western, and Coast provinces are that these regions have
tended to lag in development terms. These regions fare
rather badly, in comparative terms, to the Central
province about infant mortality, health access, secondary
school enrolment, and per capita outlays on
infrastructure (roads), and asset ownership. The data
reveal that Central and Nairobi provinces, where most
Kikuyu people live, are the most advantaged regions of
the republic in socioeconomic terms. Implicitly, Stewart
makes a case for the existence of in-groups and out-
groups in Kenya, at group and regional level, during the
2007 presidential elections. The Kikuyu mostly voted for
the incumbent and the Luo, Luhya, and Kalenjin voted for
the opposition leader. Importantly, Stewart makes a
loaded statement about the reason the disputed election
led to violence: the opposition’s failure to unseat the
incumbent meant the perpetuation of inequality at the
political and socioeconomic level. This study forwarded
Stewart’s work by exploring in considerable detail why
33
people living in the study area believed the outcome of
the election was important to their welfare and how this
galvanized and energized their conflict behaviour.
Stewart’s thesis seems to view conflict through the prism
of the out-group, their grievances. This study examined
the contra-thesis, as it were, and explored how the
prospect of electoral defeat for the in-group affected their
attitudes and behaviour.
On this matter of inequality, Muhula (2009) explored
how imbalanced development policies have shaped
political violence behaviour of Kenyans. This study is like
Stewart’s 2008 about tone and writing strategy, including
the narrative of political inequalities among elites. This
issue of patronage has been well canvassed by Stewart
(2008) above, obviating the need for further
deliberations. But Muhula’s thesis is valuable on two
scores. In developing his thesis, the writer amplifies the
role of structural and historical factors in the setting the
infrastructure for asymmetrical developmental in Kenya.
Muhula (2009) views the centralization of the state, with
a powerful executive, has created a scenario in which the
executive often determined the thrust of development
decisions. The decentralization of services through
several agencies and arrangements, including the
constituency development fund, has tended to dilute
these powers. Differently put, Muhula points to unequal
power relations between the state, the regions, and the
people as a cause of political violence. This study build on
Muhula’s work by analysing whether popular fixations
with the executive as the sole agent of socio-economic
34
improvement has diminished, with the emergence of the
new constitution. The section here examines literature
connected to political violences in the study area:
Mathare and Kibra.
Based on work done on political violence at the national
level, some knowledge gaps emerged that research done
at sub-national level can help fill. This study build on
tewart’s work by exploring in considerable detail why
people living in the study area believed the outcome of
the election was critical to their welfare and how this
galvanized and energized their conflict behaviour. The
study tested the in-group and out-group phenomenon
based on quality of life differences between people who
voted for the incumbent and those who voted for the
opposition. Stewart’s thesis seems to view conflict
through the prism of the out-group, their grievances. The
study examined the contra-thesis, as it were, by exploring
how the prospect of electoral defeat for the in-group
affected their attitudes and behaviour and assess how
these contributed to the political violence.
2.1.4 Local Level Perspectives
At the local level, there are a few studies that have
focussed on the causal mechanism of the political
violence in the study area-Mathare and Kibra. Shilloh
(2008:1) has completed a study of what he calls “the
perennial conflicts in Kibra”. Specifically, the study aimed
to tease apart the factors of an escalation of conflict in the
slum area in 2001. The study makes the point that people
35
living in the slum area tend to coalesce around their
ethnic groups; moreover, disputes tend to take on an
ethnic character. For instance, rent disputes have
typically involved a fight between people who not only
belong to different social classes, but also to people who
come from different nationalities, as it were. The Luo and
Luhya have tended to be tenants and the Kikuyu and
Nubians have tended to be the property or landowners.
The other social and economic issue that has taken an
ethnic character is the one of land, with the Nubian
community aggrieved that the State has disregarded its
rights to the land in Kibra. The landlord-tenant
relationship has been a major source of friction in Kibra.
This problem is central to Shilloh’s work. Using
qualitative interviews, Shilloh has done well to tease
apart aspects of this relationship, including showing how
the Nubian community sees itself as a marginalised
group, despite its members being the majority property
owners in some parts of Kibra at least. It is also
surprising that this community should see the Luo, for
example, as belonging to an in-group, given their
numbers and its influence on political power and
representation.
This study pushed forward the analysis done by Shilloh in
Kibra by showing how and why landlords and tenants
interpreted rent disputes in ethnic terms. It also delved
into why economic and social differentiation leads to
ethnic identity formation? Done in 2001, Shilloh’s work is
insightful; but, much of the issues discussed in the work
are dated or old and need fresh light.
36
Kinyanjui and Mutsotso (2002) unpack issues that define
slum life, focussing on Kibra and Mathare slums. One of
the core objectives of this exercise was to examine the
nature and operations of the landlord-tenant
relationship, which is a dummy for social and economic
inequality among ethnic groups in Mathare and Kibra.
Employing a mixed methods study, the scholars used a
sample of 372 respondents. These scholars point out that
the largest proportion of people living in the slum area at
the time were from the Luo community, roughly one in
three residents. Also emphasised is the fact that most
people living in the slum area (85%) were tenants, and
that most property owners (73%) lived outside the slum
area, unexposed then to the dynamics of life in the slum
area. Most respondents (57%) had a favourable
relationship with their property owners and only (36%)
of them had an unfriendly relationship. Even so,
unresolved rent disputes is one of the leading reasons
conflicts break out in the study area; what is more, “on
the whole, the landlord-tenant relationship is one
characterised by suspicion, hostility, and hatred” (p.7). In
this relationship, tenants perceive property owners as
drawing financial benefits from them in an exploitative
relationship. Substantially, then, this relationship
assumes a class fight, in which landowners and tenants
are draped in ethnic mantles, as it were.
The general literature on the causes of political violence
addresses several key issues. Highlighted in most of the
commentary on the subject, including the best known,
37
the most influential, and the most widely cited analyses
on political violence or violence from an international
perspective, is the causal mechanism. Involving as they
do the examination of these variables in different
contexts and over considerable time, such studies
identify shared factors of political violence and even
explain differences in patterns or trends in specific cases
of conflict. These studies afford analysts to make
meaningful cross-national comparisons of data, among
other things. These studies supply a strong ground for
justifying further research in national and even sub-
national levels.
Existing studies that examine horizontal inequalities and
political violence have only just begun sprouting
worldwide, no less in Kenya. Research has, however,
affirmed the claim that such inequalities trigger conflict.
This discourse needs further clarification. First, existing
studies dwell on explaining political violence involving an
armed insurgency, with the state serving as a combatant.
With a few exceptions, these studies do not address inter-
ethnic violence, such as the post-election violence in
Kenya in which the violence did not involve an armed
insurgency against the state, conflicts in which the setting
was a disputed election.
Secondly, very few studies examine the outbreak of
violence, specifically in relation to an election. The third
one, which is related to the aforementioned one, is the
paucity of knowledge on political violence occurring in
local theatres of conflict. The most authoritative works
38
on inequality and conflict use the nation as a unit of
analysis. But there is merit too in a bottom-up approach,
with analytical attention being aimed at regions or
locales, within countries, where political violence breaks
out. There was need to ascertain the meanings that out-
groups have about their social and cultural inequalities
and gauge how these forms of social discrimination
impact on their conflict behaviour.
2.2 Role of Ethnic Grievances in FuellingEthnic
Mobilization
2.2.1 Global Level Perspectives
At the global or international level, work has been done
to test whether the grievance or greed model is adequate
in explaining why horizontal inequalities contribute to
the build of political violence. Besides the grievance
model, there is a contra or competing theory, the
feasibility or greed model, in which civil war occurs
simply because conditions exist in which it can—the
availability of financial and military resources. In such a
scenario, elites would merely use grievance as rallying
points (Collier, Hoeffler, and Rohner, 2009). The greed
model has it that rebel groups would sponsor and
insurgency to score economic (control areas of a country
rich in resources) or political goals, improve their access
to political offices. But this model is rather narrowly
defined. There was need to expand the concept of greed
to include the actions and postures taken by elite in
power to keep their privileged stations. To be sure, the
“justice-seeking” model, laden in the inequality-conflict
model, predicated on such elite postures and adventures.
39
But even if the greed model, involving “rebel” elite
seeking economic profit were true, then one needs to
explain why these leaders chose costly and risky
mechanisms to realize their objectives. But probably, the
decision to engage in conflict is not a rational one,
devolving on economic explanations or rationale alone.
Instead, such a decision, or reaction, could be the
overflow of emotions associated with ethnic grievances,
hence the need to explore how elites make use of ethnic
grievances. All the same, the grievance model has
relevance and usefulness because ethnic grievances
entail conflict behaviour, some of which operate
insidiously, by forging identities that makes groups view
each other with animosity. The level of animosity makes
it easier for political operatives to excite ethnic passions
to conflict.
The greed theory and mechanism is useful because it
supposes that elite from the in-group and out-group
merely use ordinary ethnic supporters to achieve their
commercial and other benefits. If so, the motivations and
reasoning associated with this elite contest was
important to unravel, including from the point of view of
ethnic supporters in whose name these conflicts tend to
be justified.
To help answer the searching question of what makes a
country vulnerable to political violence, Collier, Hoeffler,
and Rohner (2006) examined 208 countries and 84 cases
of civil war. Using regression analysis, this analysis
examined outbreak of conflict from 1969 to 2004, in five-
40
year phases. Collier, Hoeffler, and Rohner (2006) found
that low-income countries are more at risk of civil war
than better of countries. Low economic growth and
unequal resource allocations are primary determinants
of civil conflict. Nevertheless, these writers observed that
resulting grievances only matter when acted upon by the
political elite to advance their political agenda. Collier,
Hoeffler, and Rohner (2006) explain why the grievance
model is adequate in explaining civil war. Whatever the
case, whether conflict is spontaneous or premeditated, it
is noteworthy that elites scarcely engage or support civil
wars on a social justice agenda; but rather, these people
pursue elite interests. There was need to understand how
elite grievances intersect with group grievances and
assess what this meant for the onset and escalation of
conflict.
Accordingly, conflict studies at the national or sub-
national level need to train on assessing elite grievances.
Analysts need to pay attention on elite, rather than group
grievances. In the case of Mathare and Kibera, the elite
who contributed to conflict escalation were national
figures, who called for mass action. The PEV was not just
a slum conflict; instead, it was a national conflict enacted
in a slum area. To this end, there was need to probe the
case of Mathare and Kibra to find out who the actors of
conflict were, what broader elite interests they pursued,
and what roles they played in escalating conflict and with
what goals in mind. In the section a critical review of the
literature is presented regardingthe relationship between
41
ethnic grievances and ethnic mobilization using regional
perspectives.
2.2.2 Regional Level Perspectives
At the regional level, Murshed and Tadjoeddin (2007) did
a survey of developing countries worldwide, with special
attention given to 17 countries with the highest incidence
of political violence. With half the countries surveyed in
Africa, this study was thus more immediately useful in
giving perspectives and insights on the anatomy and
mechanism of political violence than perhaps Collier,
Hoeffler, and Rohner’s work (2009). Murshed and
Tadjoeddin affirmed that the greed and grievance model
of political violence are not competing but
complementary. Murshed and Tadjoeddin (2007)
developed a mechanism with which to measure, not just
social and economic inequalities, but also ethnic
grievances. Among the areas of research proposed by
Murshed and Tadjoeddin (2007) is the “need for data on
regionally based indices for horizontal inequalities
requiring information not only on income, assets, health,
and education, but also on subjective measures of well-
being”. It is important then, Murshed and Tadjoeddin
(2007) note, to develop a mechanism with which to
measure, not just social and economic inequalities, but
also ethnic grievances, at national and sub-national
levels. To this end, scales were developed to assess the
subjective views of ethnic groups in the study area on
their living standards, when viewed through the prism of
political actions of elite.
42
Ethnic grievances often give rise to high levels of ethnic
consciousness and identification. Understanding what
impedes the development of a national identification is a
way of understanding the impact of ethnic grievances. In
this connection, Robinson (2009) examined issues of
ethnic identity are connected to (or actually are a dummy
for) inter-group relationships, structure of national
economies, state of development of democracy, and
political violence. The in-group and out-group mentality
should be seen as a stark challenge to the building of a
national identity. Robinson (2009) suggests that such a
bifurcation of identities is injurious to political co-
operation needed to develop an economy and distribute
the ensuing resources. She asserts rightly, that the
“correlates of ethnic salience should be negative
predictors of national attachment” (p.6). Robinson
(2009) tested nine hypotheses of which four are directly
related to the rhetorical objectives of this research. These
hypotheses are enumerated as follows: (1) formal sector
employment and education should be positively related
to an individual’s propensity to identify with the state;
(2) members of poorer ethnic groups would be less likely
to identify with the state than relatively richer ethnic
groups; (3) the provision of basic public services/goods
by the state will be positively correlated with
nationalism. The study under review was done in 16
African countries, including Kenya. Nationally
representative samples were applied, with 1200 to 2400
people interviewed in each of these countries. The
themes discussed included “democracy, governance,
43
social capital and political identity” (p.12). Among the
variables tested was one of national identity, which
required respondents to report on which identity they
preferred ethnic or national. The findings of this study
were that people with higher income tend to identify
more with the state than their ethnic group, and the
converse holds. People with a higher standard of living,
with greater access to public goods, are more likely to
identify with the state than people who have a lower
standard of living. In the main, Robinson’s work (2009)
reflects existing studies on horizontal inequalities and
the prospect of political violence. The in-group and out-
group phenomenon seems to apply in the tension
between ethnic and national identities. But an important
finding in Robinson’s work is that in-groups, which have
one of their own as head of the executive, would have
high levels of ethnic identification. If the correlates of
ethnic identification are factors of political violence, then
deconstructing ethnic identity is a wholesome project.
Furthermore, Robinson brings out some issues that are
important in understanding ethnic grievances too. Ethnic
voting means that electoral processes can become
mechanisms of perpetuating injustice and inequality. If
this applies, some ethnic groups can despair of this
centrepiece of democratic governance and consider it an
inbuilt or malleable mechanism with which some ethnic
groups perpetuate their control of public resources
unfairly or unilaterally. In frustration, such groups could
resort to other means of political action, ethnic
mobilization that can result in violence. It would be
44
counterintuitive to expect an out-group to have more
nationalist beliefs or commitments, relative to the in-
group. For this group, the state to which they should owe
supreme loyalty vis a vis their ethnicity, is hostile rather
than a friendly actor.The researcher, at the same time,
analyse the perspectives of the out-group on the kind of
institutional arrangements that they feel would produce
for them lifestyles that will reduce the sense of grievance.
This study expanded Robinson’s work by exploring the
subjective reflections of people, who live in a theatre of
political violence, on what informs and sustains their
primary identities. Correspondingly, there was need to
analyse the vision of national identity that respondents
had, and what they consider its components to be. This
line of inquiry can contribute to a framework that can be
used to measure and interpret political identity in Kenya.
Ethnic grievances and ethnic identification are correlates
and need to be treated as such in any analysis on political
violence - that is, they are the basis from which conflict
sprout. Reviewed below is the literature on ethnic
grievances and ethnic mobilization from the national
viewpoint.
2.2.3 National Level Perspectives
The Media Focus on Africa Foundation MFAF have made a
useful inquiry into the factors, actor, and remedies for
conflict in Kenya. This study grew out of the 2007 post-
election violence and its contents should then be treated
in the context of political violence. Importantly, the
45
findings of this report are disaggregated by ethnic group
and by region. This study was a national survey of 1600
respondents and was done in all the country’s eight
provinces. Some of the key issues addressed were as
follows: in regions of the country that have experienced
political violence, during the 2007 post-election violence,
ethnicity was found to be the leading cause of conflict. By
far, the link between tribalism, politics, and poverty is
most to blame for political violence, especially among the
ethnic groups, such as the Luo, Luhya, and Kalenjin,
groups that one would consider to have been an out-
group during the 2007 election. In other regions, poverty
was deemed to be basic to conflicts. The respondents felt
that historical injustices and partisan governance
contributed to negative ethnicity, a key driver of conflict.
Indeed, even in regions that cited poverty as the leading
cause of conflict, such as Nyanza, poverty was seen
through an ethnic prism. Other notable findings of this
study that are consistent with the instrumentalist theory
are as follows. Politicians and their agents are the main
drivers of conflict. The other major actors in conflicts are
idle youth. For the researcher, one area that analysts
needed to pay attention was the role of youth
unemployment in facilitating ethnic mobilization to
violence (Muhula 2009). This dimension of conflictwould
help form intelligent opinions on the actors of conflict
and their motives, including activities of ethnic militia,
which have tended to be instrumental to political
violence in Kenya.
46
The causal link between horizontal inequalities and
conflicts is only implied in the MFAF report. However,
probably this link need not be investigated. Regarding the
perception of fairness in the provision of public goods,
only 3 percent of Kenyan felt that the government
supplied public goods fairly. In other words, horizontal
inequalities and ethnic grievances appear to characterise
the public psyche. It is interesting that the Meru felt most
strongly about this asymmetrical distribution of
resources, yet these groups are deemed to have belonged
to the in-group at least indirectly. This should be
unsurprising though. The inequalities and grievances
that matter are those that involve the elite. The MAF
report also makes an important contribution to
knowledge by shedding light on questions of self-identity
and inter-group relationships, which are an illuminating
index of ethnic grievances. Regarding identity, 25 per
cent of the respondents identified themselves primary
based on their ethnic group, followed by (20%) based on
their occupation, and (12%) by religion. About (40%) of
the respondents identified themselves based on these
and other criteria, implying rather than demonstrating
that being Kenyan was their primary identity. The
findings seem to undermine, if not belie, theoretical
models on the following score. Whereas one would
expect Luos, Luhya, and the Kalenjin to feel that their
communities were treated worse than others were - after
all, they can be deemed as an out-group, given their
support for the opposition party - one would not expect
this from the Kikuyu, the presumed in-group. Yet the
observations suggest that the Kikuyu felt most that they
47
were treated worse than other communities, with an
index at (54%) compared to the Luo at (48.3%). But the
writers of the report under review reconcile this seeming
anomaly. The Kikuyu likely saw themselves as subjects of
antipathy by other communities during the 2007
election. In other words, they considered themselves
victims of being an in-group. On the other hand, the Luo
felt victimised, not by other communities, but by the
state. Scales for ethnic grievances should estimate the
levels of satisfaction or dissatisfaction of ethnic groups
based on access or otherwise to public goods. Even so,
the characterization of the in-groups and out-groups is
not a linear phenomenon, but a complicated one, and
should be explored as such. Hence, there was need to find
out what made in-group and out-groups identify
themselves as they did.
On the question of identity or political consciousness,
Kimenyi and Romero (2008) assessed voting behaviour
and ascertained the factors that contribute most to the
decision of voters. Done just ahead of the 2007 election,
the study included a sample of 1207 Kenyan, drawn from
all the country’s provinces. The writers note that voting
in Kenya is identity-based, with ethnically based political
parties in force. This situation implicates ethnic
identification and grievances as key animators in
elections, hence the ethnic caricature of elections. Even
so, only (1%) of the respondents reported that ethnicity
was their sole determinant in making the voting decision.
At the same time, (40%) of the respondents would vote
for the leaders of the leading political parties, meaning
48
that they would vote, as they did, along ethnic lines. Yet,
only (20%) of respondents said that they identified
themselves primarily according to their ethnic groups.
The writers give several reasons why these gaps exist
include the belief that many people believed life had
worsened (71%) under the government of President
Kibaki. Inequalities and grievances then also played a
role in the clannish voting patterns observed, with (55%)
of Kenyans feeling that the government had been unfair
in its development policies. The overwhelming rejection
of the government-endorsed constitution, during the
referendum, has been associated with the fear of out-
groups that skewed distribution of resources and
opportunities would persist under the new constitution.
Some of the key observations of this report were as
follows. Ethnic identification and polarization were
significant factors in the clannish voting patterns seen.
The reason, in part, why such polarization existed was
due to horizontal inequalities and the resultant
grievances. This report has ably identified several
determinants of electoral choices during the 2007
national election. The report sheds light into the factors
that inform or condition voting behaviour, including
ethnicity. What this report has left out is making a
direction connection between polarization and the post-
election violence (Kimenyi and Romeno, 2008). This
study examined how ethnic polarization influenced the
ethnic groups in Mathare and Kibra towards violence.
Furthermore, the bulk of inequality-conflict studies do
not adequately explain a trait of the violence to which the
49
2007 post-election violence bore: its genocidal
properties. Studies that view the 2007 post-election
violence as meaning that Kenya is fitting into the script of
probable outbreak of genocide are emerging yet. There is
evidence that ethnic polarization and grievances are
driving Kenya to genocide. The authors of this report
assert that “perhaps the most important contributing
factor to Kenya’s elevated riskfor genocide are the
strained and rivalry-prone social and cultural
relationships between tribes and the recent violent
aftermath of the December 2007 national elections
(Sentinel Project, 2012:1). In this report, several risk
factors identified are of relevance to the grievance-
conflict discussion. Since the return of competitive
politics in 1992, Kenya has been a semi-democracy. The
country’s fledgling democracy has meant that ethnic-
ethnic mobilization and polarization have been prevalent.
Ethnic intolerance and grievance are a mark of political
culture in Kenya, the Sentinel Project report adds.
Political elite and their ethnic groups were made to
understand that their social, economic, and political
privileges depended on their keeping of cordial
relationship with the executive. Importantly, the Kenyatta
government practiced ethnic intolerance against the Luo
community, and the Moi Government applied this form of
marginalization against the Kikuyu community. Ethnic
grievances and resentments have a long history. The
ethnic hatred displayed in inter-ethnic rivalry between
the Luo and Kikuyu in Mathare and Kibra is of long-
standing and needs to be treated, in analytical terms, as a
50
sort of generational feud. The conflict analysis in this
research aims to prepare the way for dialogue between
these ethnic groups by seeking to find out what is it that
groups believe about each other that is helping to cause
conflict. Kimenyi and Romero (2008) give no indication
of the social and cultural beliefs and attitudes that are
attached to the political violence in the slum area.
Kimenyi and Romero’s report addresses, however, an
important theme of conflict - that is, its socio-cultural
dimension. Inter-group hatred has a long history in
Kenya. Even so, this report argues rightly that the in-
group and out-groups tend to change over time; as such;
the more helpful analytical posture to take is to address
the structural factors of this form of political violence
segmentation. There is need to look at political violence
hatred in general, rather than narrow in one or two
groups that are presumed to be either the in-group or
out-group. This approach is sensible. The structures and
dynamics of conflict then would be the social divisions
among ethnic groups, which are at the heart of the
polarizing effect of political life in Kenya. The report
makes a good case of the role of ethnic nationalism in
exciting conflict at the social and cultural levels. The
devaluation of out-groups as unfit to govern—as
backward, primitive or warlike, as lazy and anti-
development, as outsiders and as people who do not
belong-is an important face of political violence. Political
leaders can persuade their ethnic supporters that the
members of other ethnic groups are inferior, threatening,
and even dangerous, the report rightly observes. Why the
51
message of such elites would receive a receptive audience
was intriguing. What is that in-groups and out-groups
already believe about each other or want to believe about
each other, attitudes that make them accept implicitly, it
seems messages of hate from political elites. In this
regard, the study focussed on an area the Sentinel report
has only alluded too (rather than delved into) that is,
what is it that in-groups and out-groups, in the study
area, believe about themselves and other groups,
attitudes and beliefs that are basic to their conflict
behaviour, even when the conflict situation appears to
have de-escalated. Little is known about how these
beliefs and attitudes are formed and perpetuated. In what
follows, literature on ethnic grievance and ethnic
mobilization with a focus on the study area, Mathare and
Kibra.
2.2.4 Local Level Perspectives
The (CIPEV, 2008) Report attributes the violence that
occurred in Mathare and Kibra to ethnic grievances. The
slums areas of the city, where most people live, are a
testimony of social and economic inequality. The slum
areas are deprived of public goods and services, the
report highlights. Nevertheless, this begs the question of
how one is to understand the grievances that have been
observed in the slum areas, grievances that are presumed
to devolve based on horizontal inequalities. The report
isolates the landlord-tenant relationship as a major cause
of the poor state of ethnic relations in the study area. Just
before the national election of 2007, Kikuyu property
52
owners pushed out Luo tenants, under the pretext of
doing repairs, because the property owners feared that if
the opposition leader came to power, the government
would regulate rents. The suspicions and acrimony that
ensued was a significant factor in the 2007 post-election
violence. The land-lord and tenant relationship is
associated with ethnic grievance because it is a class
contest involving ethnic groups. It is noteworthy that the
violence in Kibra and Mathare targeted not just members
of the in-group, but the property and means of livelihood
of the perceived in-group. Arson attacks in Kibra and
Mathare and the destruction of the Toi Market, from
where 3,000 people (mostly Kikuyu) made a living are
illustrative. The destruction of livelihoods, by the
presumed out-group, was a means of correcting the
social and economic differences by “other
means”.Qualitative in texture, the CIPEV report has
assessed the quantum of ethnic grievance, measured as a
variable, among perceived in-group and out-groups. The
other important domain of knowledge that needs to be
understood is differentiating the observed levels of
ethnic grievance between historical and situation factors,
or at least explain the role of historical memories and
grievances in observed levels of ethnic grievance
Okombo and Sana (2010) justify their inquiry into
political violence in Nairobi slums by asserting that the
deprived areas, including Mathare and Kibra, were at the
epicentre of the post-election violence in the city.
Historically, these areas have been hotspots for political
rivalries among elite, campaigns that have been couched
as ethnic rivalry. Okombo and Sana argue that ethnic
53
groups in the area are politically conscious and seem to
take strong personal positions about political events,
including the treatment of their ethnic elite in national
politics.
The reason: the slum dwellers believe that these elite
hold the key to their social and economic improvement.
But Okombo and Sana aver that this belief is misplaced.
On this ground, the grievances of ordinary people should
be distinguished from the grievances of the elite before
any attempt is made to assess how the former grows out
of or facilitates the latter. Jacobs (2011) has noted that
the slum areas of Nairobi are theatre where the political
elite conducted political manoeuvres using poor people
as pawns. If what matters are elite grievances, then
analysts ought to focus on this form of grievance and
consider the group grievance as incidental to
understanding the escalation of conflict. Elite
manoeuvring maybe the reason ethnic grievances are
necessary, but it is not a sufficient condition for conflict.
Conflict studies at the national or sub-national level
needs to train on assessing elite grievances and identify
who the actors of conflict were, what broader elite
interests they pursued, and what roles they played in
escalating conflict and with what goals in mind.
As Okombo and Sana (2010) note elsewhere disputes
over rents often assume an ethnic dimension. The long-
standing differences political differences between the
Luo and the Kikuyu, in the slum areas, seem to energise
these confrontations. Of the three theories articulated by
54
Okombo and Sana (2010) to explain political violence,
the social identity theory and the social dominance
theory are most helpful in elucidating the landlord-
tenant relationship, which is often central to political
violence in slum areas. These theories speak to the sense
of discrimination and the “us” versus “them” mentality
that often characterises conflicts in slum areas. But in the
landlord-tenant relationship, the “us” would be ordinary
members of the out-group and the “them”, would be
people who one can characterised as belonging to an
economic elite. Indeed, Oucho 2010 notes that slum areas
tend to serve as ethnic enclaves, in which people from
marginalised areas, such as the former Nyanza and
Western provinces, congregate and are yet exposed to
marginalization and underdevelopment. Suppositions of
this sort lead meant issues of in-group and out-group in
the study area needed to be treated with tentativeness.
The qualitative portion of his work will seek to unpack
the renditions that ethnic groups have about each other
vis a vis the in-group and out-group mentality.
For Okombo and Sana (2010), another key factor in
outbreak of violence in slum areas are vigilante groups,
ethnic militia that provide informal employment to youth
who provide security to ethnic enclaves. The
privatization of security is an index of neglect of slum
areas by the government and or unresolved inter-
political violence. This is importantbecause it connotes a
duality of identity: even the presumed members of the in-
group might perceive themselves as an out-group in a
sense.
55
This study improved Okombo and Sana’s analysis by
compiling a compendium of cultural markers and tags
that the presumed in-group and out-group have about
each other, especially about the tenant and landlord
relationship. These cultural distinctive are often at the
core of conflict behaviour and need to be fleshed out and
interrogated.
There was need to tease apart the socio-political and
socio-cultural dimensions of the conflict as impediments
or obstacles to conflict transformation. The other
important domain of knowledge that needs to be
understood is differentiating the observed levels of
ethnic grievance between historical and situation factors,
or at least explain the role of historical memories and
grievances in observed levels of ethnic grievance. This
research endeavoured to shed light on these questions,
specifically, and ascertained the relationship between
ethnic grievance and quality of life differences.
The major contribution of Oucho (2010) to political
violence is his conceptual framework that allows one to
see the probable causal chain that brought about the
escalation of the conflict. The legacy of the colonial
enterprise falls, as it were, on the fringes of the circle of
Oucho’s theoretical model. The live issues of conflict, in
this regard, are administrative units that have tended to
serve as ethnic enclaves. The arrangement of these has
meant that it has been easy for successive governments
to realize partisan deployment of resources. The
consequences of such state operations become
56
pronounced when people living in a marginalised region
o