University of Southern California Dissertations and Theses

A theory of status-quo terrorism: democracies in conflict and their proclivity to outsource repression

(USC Thesis Other)

A theory of status-quo terrorism: democracies in conflict and their proclivity to outsource repression

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content

A Theory of Status-Quo Terrorism:
Democracies in Conflict and Their Proclivity to Outsource Repression
By
Cody Brown

A Dissertation Presented to the
FACULTY OF THE USC GRADUATE SCHOOL
UNIVERSITY OF SOUTHERN CALIFORNIA
In Partial Fulfillment of the
Requirements for the Degree
DOCTOR OF PHILOSOPHY
(POLITICAL SCIENCE AND INTERNATIONAL RELATIONS)

December 2021

Copyright 2021 Cody Brown

ii

TABLE OF CONTENTS
ACKNOWLEDGEMENTS ............................................................................................................ iv
LIST OF TABLES ......................................................................................................................... vi
LIST OF FIGURES ...................................................................................................................... vii
ABSTRACT .................................................................................................................................. viii
CHAPTER 1 – STAYING POWER ................................................................................................ 1
SECTION 1.1 – INTRODUCTION ..................................................................................................... 1
SECTION 1.2 – THE IMPORTANCE OF THEORY BUILDING .............................................................. 4
SECTION 1.3 – DISSERTATION OVERVIEW .................................................................................... 5
CHAPTER 2 – A REVIEW OF TERRORISM STUDIES ............................................................ 12
SECTION 2.1 – CHAPTER OVERVIEW .......................................................................................... 12
SECTION 2.2 – DEFINITIONS ....................................................................................................... 12
SECTION 2.3 – TERROR IN THE NAME OF THE STATE .................................................................. 20
SECTION 2.4 – TERROR AND REGIME TYPE ................................................................................ 28
SECTION 2.5 – SUMMARIZING THE LITERATURE ........................................................................ 35
CHAPTER 3 – A COMPREHENSIVE THEORY OF STATUS-QUO TERRORISM ................ 38
SECTION 3.1 - INTRODUCTION .................................................................................................... 38
SECTION 3.2 – SCOPE ................................................................................................................. 38
SECTION 3.3 – STATUS-QUO THREATS AND CONSTRAINTS TO COUNTERING AN INSURGENCY .. 40
SECTION 3.4 – LIBERAL CONSTRAINTS ON COUNTERINSURGENCY ............................................ 45
SECTION 3.5 – STATE DECISION-MAKING AND HARD-LINER REACTIONS ................................... 47
SECTION 3.6 – SUMMARY ........................................................................................................... 55
CHAPTER 4 – STATUS-QUO TERRORISM: A CROSS-NATIONAL ANALYSIS ................ 57
SECTION 4.1 – INTRODUCTION ................................................................................................... 57
SECTION 4.2 – DATA AND METHODOLOGY ................................................................................. 57
SECTION 4.3 – STATUS-QUO REACTIONS TO LIBERAL DEMOCRACY ........................................... 66
SECTION 4.4 – SUMMARY ........................................................................................................... 72
APPENDIX A – CODING PROCEDURE ......................................................................................... 82
APPENDIX B – HISTOGRAM ......................................................................................................... 97
CHAPTER 5 – DEFENDING THE STATUS-QUO IN URABÁ: REACTIONS TO THE
DEMOCRATIC GAINS OF THE LEFT ....................................................................................... 98

iii

SECTION 5.1 – INTRODUCTION ................................................................................................... 98
SECTION 5.2 – CASE STUDY DESIGN .......................................................................................... 99
SECTION 5.3 – COMPARING THE CASES ................................................................................... 103
SECTION 5.4 – A RISING AGAINST THE STATE (1960-1981) ..................................................... 104
SECTION 5.5 – A MIXED APPROACH TO COUNTERINSURGENCY AND STATUS-QUO
TERRORISM IN URABÁ (1982-1993) ......................................................................................... 105
SECTION 5.6 – THE AFTERMATH OF A NEW CONSTITUTION (1994-2003) ................................ 127
SECTION 5.7 – ANALYSIS ......................................................................................................... 145
SECTION 5.8 – SUMMING UP ..................................................................................................... 148
CHAPTER 6 – A SUMMARY ON STATUS-QUO TERRORISM ........................................... 157
SECTION 6.1 – TYING IT ALL TOGETHER .................................................................................. 157
SECTION 6.2 – SUMMARIES ...................................................................................................... 158
SECTION 6.3 – FUTURE RESEARCH ........................................................................................... 163
SECTION 6.4 – FINAL REFLECTIONS ......................................................................................... 167
BIBLIOGRAPHY ........................................................................................................................ 169
NOTE: ALL TABLES AND FIGURES ARE AT THE END OF EACH CHAPTER

iv

ACKNOWLEDGMENTS

I would like to take a moment to acknowledge the great deal of support and assistance that I have
received throughout my time at the University of Southern California, especially in the writing of this
dissertation.
First, I would like to thank my supervisor, Patrick James, whose knowledge and expertise have
been invaluable throughout my time as a graduate student. From your help with setting up field research
in Northern Ireland and Colombia, to the sound direction you have led me in, and the constant emails
about academia (and sports!), your guidance and friendship are truly appreciated. I can’t put into words
how much I have depended on your help throughout this process.
I would also like to thank the rest of my committee for their advice and counsel. Thank you to
Gerardo Munck for sparking my interest in the finer details of democracy and its institutions – your
course on democratization, and your guidance afterwards were crucial to the completion of this
dissertation. Thank you to James Lo for your instruction in all things quantitative, for which I would be
lost without, and for being a great role model for a young scholar. Thank you to Erroll Southers for
forcing me to think critically about what terrorism is, and for providing insight into the real-world policy
issues centered around countering terrorism and violent extremism.
My appreciation and love goes out to my friends and family as well. To my close group of friends
to which I have an incredible bond with, your friendship and love have been the best outlet I could ask
for. To Cindy, thank you for taking us in, and for treating me as if I were your own son. Your love and
kindness have been a gift from God. To Steve, thank you for your love and guidance; you’ve been a great
role model and source of support throughout this whole process. Thank you, Teresa, for not only taking
care of me, but for taking care of Dad as well. I love you and the girls. Thank you, Curtis, for supporting
me in all the ways that you do. The example you set for me has always pushed me to be better. My rock,
my fiancé, and soon to be wife – Shelby Romero – has been here for me every step of the way. It has not
always been easy to be the partner of a PhD student, but your love and support have made me all the more
v

capable of completing this journey I set out to accomplish so long ago. I am truly grateful for your
unwavering support, I love you, and I cannot wait to spend the rest of our lives together and Dr. and Mrs.
(Romero) Brown. And of course, thank you Mom for always pushing me, for always telling me how
proud you are of me, and for raising a son that was always curious and willing to seek new knowledge – I
love you and I could not have done this without you.
Lastly, I would like to thank my late father, who unfortunately passed during the writing of this
dissertation. Dad, you were always my biggest supporter – whether that was in basketball, in school, or in
life in general. I was so privileged to be raised by a father that provided so much stability and love. I have
only ever wanted to make you proud – I hope that this would do that. It breaks my heart that you can’t be
here to witness this accomplishment in person, but I am the man I am today because of your influence as
a father. For this, I will be forever grateful.

vi

LIST OF TABLES
TABLE PAGE
2.1 – DEFINITIONS OF TERRORISM ....................................................................................... 37
4.1 – DESCRIPTIVE STATISTICS ............................................................................................. 74
4.2 – NEGATIVE-BINOMIAL COUNT MODEL ....................................................................... 76
4.3 – ZERO INFLATED MODEL ................................................................................................ 80
5.1 – CAUSAL MECHANISMS ................................................................................................. 150

vii

LIST OF FIGURES
FIGURE PAGE
4.1 – STATUS-QUO TERRORISM AND INSURGENCY ........................................................ 75
4.2 – PREDICTED STATUS-QUO COUNT BY POLITY SCORE ........................................... 78
4.3 – STATUS-QUO ATTACKS BY POLITY REGIME TYPE ................................................. 79
5.1 – MAP OF URABÁ ............................................................................................................... 151
5.2 – STATUS-QUO ATTACKS IN URABÁ BY YEAR .......................................................... 152
5.3 – STATUS-QUO ATTACKS IN URABÁ BY YEAR .......................................................... 153
5.4 – STATUS-QUO ATTACKS IN URABÁ BY YEAR .......................................................... 154
5.5 – STATUS-QUO ATTACKS IN URABÁ BY YEAR .......................................................... 155
5.6 – STATUS-QUO ATTACKS IN URABÁ BY YEAR .......................................................... 156

viii

ABSTRACT
This dissertation seeks to further explain the association between democratic regimes and terrorist
violence. More specifically, it asks the question of how we can better disaggregate the
phenomenon of terrorist violence, and how different types of terrorism may be associated with
democratic institutions. To accomplish this, the typology of terrorism is differentiated between
those attacks perpetrated by a broad range of violent anti-state groups, and those attacks
perpetrated by violent groups committed to maintaining the status quo, focusing on explaining the
latter. Previous literature has established that regime type, and certain subcomponents thereof,
have mixed results with terrorism as a whole. To clarify this relationship between regime type and
terrorist violence, a theory on the democratic tendency to experience status-quo terrorism is
developed and tested using a mixed methodological approach. Utilizing a zero-inflated negative
binomial model, I find that the relationship between democracy - whether measured minimally or
with more context - and status quo terrorism is positive and highly significant. Analyzing the case
of status-quo terrorism within Colombia, I take a closer look at the specific causal mechanisms at
play linking democratic norms to status-quo violence. Both the quantitative and qualitative testing
provide findings which suggest that democracies are more prone to experience status-quo terrorism
specifically.

1

Chapter 1
Staying Power
1.1 Introduction
This research project kicked off with a young graduate student intrigued in the
phenomenon of terrorist violence and democratic politics. I could think of no better case to
immerse myself in than the three-decades long conflict in Northern Ireland known as the
Troubles. This led to field research in Belfast, the epicenter of violence during the conflict, in
order to conduct interviews regarding the determinants of terrorism before the peace deal. The
result was a much more thorough understanding of the conflict as a whole, but also with a new
research question and burgeoning theory to go with it.
The Troubles began in the late 1960s as the minority Catholic population of Northern
Ireland unleashed a massive civil rights campaign. Up until then, the region’s political
institutions had ensured that the majority Protestant population would remain dominant in
Northern Irish politics (Kennedy-Pipe, 1997). Under increasing pressure, the Prime Minister of
Northern Ireland at the time made some concessions to the civil rights groups. However, this
unleashed a violent backlash from the more hardline “loyalists” (extremists on the Protestant side
loyal to protecting the union between Great Britain and Northern Ireland), who felt betrayed by
such concessions. Soon after, violence erupted between the loyalists and the “republicans”
(extremists on the Catholic side wanting to reunite Northern Ireland with the Republic of
Ireland). Eventually, the conflict would transform into one in which both sides would utilize
terrorist violence as their main strategy; the Irish Republican Army (IRA) and the Irish National
Liberation Army (INLA) on the republican side, and the Ulster Defense Association (UDA) and
the Ulster Volunteer Force (UVF) on the loyalist side.
2

Throughout the Troubles, both sides would inflict devastating attacks with terrible
consequences for the civilian population of Northern Ireland. Throughout the 30 years that the
conflict raged, over 3,500 people died as a result of the violence, with the majority of the victims
being civilians (Sutton, 1994). That violence was split among three main actors – the British
security forces, the Loyalist paramilitary groups, and the Republican paramilitary groups. While
the groups fighting to change the status-quo (the Republican paramilitary groups) were
responsible for roughly 2,000 of those deaths, those fighting to maintain the status-quo – the
security forces and Loyalist paramilitary groups – were responsible for 365 and 1,027 deaths,
respectively (Sutton, 1994). All actors in the conflict were responsible for numerous threats, and
attacks against property as well.
To better illustrate the violence each side utilized during the Troubles, a couple of
examples are provided. On November Seventh, 1987 a team of IRA volunteers successfully
transported a 40-pound bomb into the reading rooms of the town of Enniskillen, County
Fermanagh, Northern Ireland. The bomb was hidden in a sports bag, and was prepared to
detonate the next morning, which coincidentally was when a ceremony to commemorate the
fallen British soldiers of the two World Wars was to be held. The IRA bomb exploded the next
day as intended, blowing out the wall of the reading rooms and trapping those who were standing
nearby for the ceremony underneath the rubble (Cullen & Fitzpatrick, 2017). Twelve people
were killed, and 63 were injured due to the blast. Roughly six years later and 63 miles to the
north, The Ulster Defence Association (UDA) was planning an attack on a pub in Greysteel,
County Derry-Londonderry, Northern Ireland. The Rising Sun Bar was holding a big Halloween
party on the night of October 30
th
, 1993 when three men of the Ulster Freedom Fighters (UFF) –
3

the violent cover groups of the UDA – would arrive and shoot up the party (Tourish, 2013).
Eight people were killed and 19 were injured because of the assault on the pub.
The purpose of this dissertation is to explain how, and why, these two events must be
distinguished from one another. These attacks are similar in many respects, yet different in a few
crucial features that are instrumental to making this distinction. First and foremost, both events
are on their face, appallingly deadly incidents which deprived innocent civilians of their good
health or livelihood; they symbolize the devastation which terrorist violence instills upon
undeserving populations across the globe. These instances also share the fact that they are both
coded as domestic terrorist events in the Global Terrorism Database (GTD); that is, both events
are characterized by the targeting of noncombatants, for political goals, and were perpetrated by
non-state actors. These events also took place within a democratic country, as is more likely to
happen according to wide array of literature on the terrorism-democracy nexus (Eubank &
Weinberg 1998; Eyerman 1998; Li 2005; Enders & Sandler 2006; Piazza 2008a; Chenoweth
2010a; Findley & Young 2011; Young & Dugan 2011).
However, the vital distinction between the two events – thus far understudied in the
literature on terrorist violence – is the issue of what type of group perpetrated each attack, and
what sort of broad goals did each pursue? The IRA was an insurgent group which committed
terrorist attacks against Northern Ireland’s non-combatant population. The UFF often did the
same. Both groups were disillusioned with the legal avenues for pursuing their goals, and so
chose violence as the best tool at their disposal. The difference is, the IRA’s attacks were
perpetrated to change the political and socio-economic structures of the state, whereas the UFF
(and other loyalist paramilitary groups in Northern Ireland) chose violence as a means of
4

deterring gains by the IRA and their sympathizers, and to maintain the political and socio-
economic status quo.
These anecdotal cases of terrorist violence serve to highlight the idea that not all
terrorism is the same. Though the attack by the UFF and IRA are coded very similarly by most
terrorism databases, such as the GTD, their motivations for such violence are clearly distinct
from one another. What became apparent as a result of my field research throughout the region
was that the field of terrorism studies does not offer an appropriate analysis of the difference
between these different types of terrorist violence. Indeed, most studies tend to treat these classes
of events as the same phenomenon, and build theoretical models meant to explain both.
Instead, this dissertation will seek to introduce an important dichotomy between the more
general “anti-state” terrorism versus status-quo terrorism. The argument will be made that a
concise distinction between the two is important for future developments in the field.
Specifically, this endeavor looks to establish those factors which are responsible for the latter
type of violence – status-quo terrorism. It finds that democratic regimes are associated with an
increased likelihood of terrorism by status-quo forces which is a relevant and enlightening
finding in the field given the emphasis on democracy and terrorism per se.
The rest of this chapter will continue as follows. The next section will highlight the
importance that qualitative research had on this dissertation for the purpose of theory building.
Field research in Northern Ireland essentially led me to the research question herein and gave me
the building blocks for the theoretical framework to be developed in Chapter 3. The following
sections then provide a brief overview of the remaining contents of the dissertation, chapter by
chapter.
1.2 The Importance of Theory Building
5

Before ending the chapter with a brief overview of the way this dissertation will
unfold, chapter by chapter, it is important to discuss how crucial the theory-building case
of Northern Ireland was to the overall research project herein. As mentioned at the outset,
this project began with field research conducted in Northern Ireland. That research was
originally intended to uncover the causal mechanisms through which democracy and its
institutions led to greater levels of domestic terrorism. However, what came from the
semi-structured interviews with scholars, journalists, politicians, and former terrorists
was a new research question, and a new direction for the dissertation at large.
Designing the interview process in a way in which the interviewees could direct
the conversation as their expertise saw fit led to a new focus on the difference between
anti-state and status-quo terrorism. Furthermore, the process pointed out the important
relationship between the military and elites that became such a crucial part of the
theoretical framework developed later in Chapter 3. Along with the quantitative and
qualitative testing conducted later, the theory-development case facilitates the goal of
learning about status-quo terrorism in a more systematic and mixed-methodological
manner.
1.3 Dissertation Overview
The general research question and theme of the dissertation have now been introduced.
This last section is now meant to provide the reader with a brief outline to the following
chapters that remain.
1.3.1 Chapter 2
Chapter 2, the literature review, inaugurates the reader to a host of studies related
to the research agenda herein. It begins with an honest and thorough review of the
6

dilemma wrapped up in attempting to define exactly what terrorism is. However difficult
that endeavor is, it is crucial to agree upon the core elements of the concept so as to have
the capacity to build a theoretical framework to model its variation. Key among these
core elements is the general agreement that terrorist violence is inherently a non-state
behavior - state terrorism being regarded as a separate class of violence with its own
determinants. Making this distinction clear is critical towards establishing a definition of
status-quo terrorism.
Following this discussion of the definition of terrorist violence, the review shifts
to studies dedicated to explaining state terrorism. Generally, the literature on state terror
has found that such violence is negatively associated with democratic institutions.
However, one major oversight of the state terrorism literature is that scholars of the field
tend to treat violence perpetrated by the formal state apparatus the same as violence
perpetrated by non-state death squads or paramilitary groups. The issue with this is the
potential for the violence of one set of actors to be caused by different factors than the
violence of the other.
The chapter concludes with a review of terrorist violence (in a traditional, non-
state sense) and its association with regime type. This connection between regime type
and terrorism has been ambiguous at times. Many of the pioneering studies in this field
found a positive and strong association between democracy and terrorism. Scholars
theorized that this connection existed for a variety of reasons, whether that was because a
free press gave greater publicity to terrorists, or because the respect for civil liberties
made democracies “soft” targets. Later studies found a somewhat altered relationship,
instead finding an inverted U relationship between the two; weak democracies were most
7

prone to terrorist violence and more consolidated democracies less so. However, the
literature of terrorism and democracy suffers from the same problem as the literature on
state terror - the conceptual stretching of the dependent variable. Some of these studies
lump transnational and domestic terrorism together, and all treat anti-state and status-quo
terrorism as the same. This leaves a need in the field for a thorough analysis of the
relationship between democracy and this new subtype of terrorist violence - status-quo
terrorism.
1.3.2 Chapter 3
Chapter 3 sets out to establish a strong theoretical framework that explains the
manner in which democracy and its institutions may cause status-quo terrorist violence.
The framework begins with a thorough discussion of the necessary condition of some
type of threat to the status-quo. For non-state groups to mobilize and be willing to utilize
terrorist tactics, there must be some perceived threat to their self-interests, and an active
insurgency against the state and its political and socioeconomic status-quo should be high
on any continuum measuring threats. While the theory predicts that most status-quo
terrorism therefore should take place during insurgencies, the logic developed within the
theory does not preclude that such violence may take place due to some lesser threat to
the status-quo - short of an active insurgency.
That logic deals with the serious dilemmas that the state and its security forces
face when confronting a challenge, especially in the case of asymmetric or irregular
warfare. The security forces are pressured to fight an enemy that wears no uniform and
blends in with the civilian population - leading to problems of incomplete information.
Solutions range from indiscriminate violence to relying on local intelligence, oftentimes a
8

combination of both. This is where the effect of regime type comes into the formula. The
theory models in the constraints that democracies tend to place on the ability of militaries
to fight an overly repressive counterinsurgency. Furthermore, when democracies utilize
their unique ability to persuade insurgents to the negotiating table through promises of
political parties and electoral reform, elites within those countries may react harshly to
the potential institutionalization of insurgent preferences.
Lastly, the chapter concludes with a thorough consideration of the precise causal
mechanisms through which democracy causes status-quo terrorist violence. These
mechanisms specifically focus on the ways in which hard-liners within the military, and
hard-liners within a country’s elite class (whether these are businessmen, the landed
class, religious leaders, etc.) react to the democratic tendencies discussed above. When
both actors are thoroughly agitated by such democratic tendencies to a challenge to the
state, the perfect storm is created in which the military relies on their non-state allies to
uphold the status-quo.
1.3.3 Chapter 4
Chapter 4 is meant to provide a cross-national quantitative analysis of the
theoretical propositions developed in Chapter 3. The chapter begins with a thorough
discussion of the data collection process, the operationalization of the variables used in
the model, and the reason why such independent and control variables are utilized in the
testing. Most importantly, there is an in-depth overview which clearly states the way in
which the dependent variable - status-quo terrorism - is coded. Care is also taken to
explain the specific methodology and modeling that is utilized to test for the association
9

between democracy and status-quo terrorism; a zero-inflated negative binomial (ZINB)
model.
The next portion of the chapter provides an analysis of the six different ZINB
models that are run. Much emphasis is put on the robustness checks that are included in
order to alleviate concerns that any association between democracy and status-quo
terrorism is driven by the specific definition or coding of the democracy variable. Instead,
models are run with a dichotomous democracy variable, and two separate interval-level
democracy variables. Through this testing, I find that there is overwhelming support for
the propositions developed in the theoretical framework. More specifically, looking at the
predicted probabilities of the model utilizing the Polity democracy variable, I find that
each level increase in democracy is associated with higher levels of status-quo terrorism;
a clear break from the finding of an inverted U relationship in the traditional terrorism
literature.
1.3.4 Chapter 5
The following chapter is dedicated towards an in-depth examination of a case
study regarding Colombia and its history of status-quo terrorism in the region of Urabá.
Chapter 5 begins with a thorough consideration of the case selection process, the
comparative advantages of qualitative research, and the overall design of the case study.
The goal of the chapter is to provide a mixed-methodological approach for the testing of
the theoretical propositions of Chapter 3. It does this by utilizing process tracing
throughout a comparative case study - one in which the same region is analyzed across
two separate time periods. In essence, by studying the same region over time, the design
hopes to control for variables that would otherwise vary region by region. The big change
10

that occurs between the two periods of time under consideration is the establishment of a
new Colombian constitution - and several institutions soon after, which corresponds with
large spikes in status-quo terrorism. This is further evidence that lends support to the
propositions put forward in the theory.
Furthermore, the chapter carefully examines the potential causal mechanisms
through which this causal process takes place. It specifically finds that certain democratic
institutions and tendencies within the counterinsurgency created overwhelming
frustration for the key actors involved in status-quo terrorism: the security forces and
regional elites (banana plantation owners, cattle ranchers, and the local political elite).
These frustrations were born from the government’s tendency to negotiate with the
insurgents - dangling the promise of political parties and elections - as well as the
constant battle over civilian oversight of the military and its record on human rights
abuses. Disillusioned with such democratic tendencies, the military secretly delegated the
counterinsurgency to those non-state paramilitary groups willing to protect the status-
quo.
1.3.5 Chapter 6
Finally, Chapter 6 provides the reader with a conclusion and summary of the
dissertation. It begins with an overview of the research project, discussing the
accomplishments achieved chapter by chapter, as well as the accomplishments of the
dissertation in an overall synthetic sense. Next, ideas for future research are presented. A
research project is important not only for its findings, but also for the new ideas and
questions it may spark. Therefore, it is important to suggest new avenues in which these
11

findings may be applied. Lastly, the chapter concludes with some final reflections on the
dissertation as a whole.

12

Chapter 2
A Review of Terrorism Studies
2.1 Chapter Overview
The purpose of this chapter is to review both the range of definitions for terrorism and
terrorist violence, and the literature surrounding terror by the state and other status-quo forces.
This study’s main goal is to find those social and political variables that exhibit the strongest
causal relationships with what I term status-quo terrorism. While the definitional issues
surrounding terrorism are nuanced, they are vitally important for the comprehension of those
violent forces set out to protect the existing conditions for society. Just as these definitional
issues play a key role in disaggregating terrorist violence, it is equally pivotal to learn from the
previous research on state terror, and the relationship between regime type and terror -
identifying the gaps in the literature this dissertation seeks to fill along the way.
With that said, this chapter will proceed in the following fashion. Section two will cover
the definitional issues surrounding different types of terrorist violence and how they pertain to
the status-quo terrorism subject of this study. Section three analyzes the literature on state-led
terrorism, searching for explanations that will similarly help to understand terrorist violence by
non-state status-quo forces. In section four, the way in which regime type is generally thought to
interact with terrorism is reviewed - pointing out its potential heterogeneous effects based on the
type of terrorist violence under analysis. Lastly, in section five, the chapter concludes with some
general verdicts regarding the literature and how this dissertation may enhance it.
2.2 Definitions
As the key focus of this dissertation is to explain the occurrence of a particular type of
terrorism, it is important to discuss what that phenomenon is in a more general sense, and how it
13

has already been broken down into subcategories in the literature. In this section, I will first
review the wide variety of definitions of terrorism that both governments and scholars have
asserted, and illustrate the main similarities between those definitions - a minimal classification
that most of those concerned could agree upon. Second, attention will be paid to an essential
distinction that has been made between what is called transnational terrorism and domestic
terrorism. Third, to look further at the disaggregation of terrorism, I analyze the differences
between terrorism performed by the state and terrorism performed by non-state actors. While the
two seem similar, maintaining the distinction does exhibit merit for scholars of political violence.
Lastly, I review the split between the more acknowledged anti-state terrorism and the much less
studied status-quo terrorism. It is with this discussion of the different subtypes of terrorism that it
becomes clear as to why this study focuses on non-state status-quo terrorism.
2.2.1 What is Terrorism?
It is important to begin this chapter with a thorough discussion and analysis surrounding
the definitional problems of terrorism. Without a clear operationalization of the phenomenon, it
may just be violence that ‘we’ don’t like. And, as the ‘we’ in this formula changes, so too does
the classification of whether a violent event is terroristic in nature or not. This of course refers to
the popular adage that one man’s terrorist is another man’s freedom fighter. However, it is
crucial that one clearly specifies what constitutes a terrorist act. Without a clear definition, it
becomes nearly impossible to develop a coherent theory to explain it, or methodology to test for
its relationship with other variables. As these are the goals of this dissertation, I ask: what is
terrorism?
Fortunately, there is plenty of literature on terrorist violence, many that include a
definition of what terrorism is. Unfortunately, many of those scholars (and governments) define
14

terrorism in different ways than each other. In fact, Alex Schmid and Albert J. Jongman found a
total of 109 different definitions of terrorism in their book in 1988 (Schmid, Jongman, et. al,
1988); the number is sure to have grown since that publication. Much of the disagreement is over
how broad or narrow the definition should be: should a terrorist attack have socio-economic or
political goals? Must a terrorist attack be against innocent civilians, or can it target police? Is an
attack against property considered a terrorist attack? Does it include the threat of attack, or only
the attack itself?
Table 2.1 below provides a sample of definitions from heavily referenced scholars in the
field of terrorism studies. The definitions range in date of origin from 1972 to 2006, and while
they vary to a certain extent, they all share certain key elements that are important to expand
upon. The first similarity among definitions of terrorism is they all tend to coalesce around the
idea that terrorist violence is intended to produce a negative psychological effect for an audience
greater than the victims alone. In other words, a terrorist attack serves as a message to others.
Secondly, although it is implicit in the later definitions, another similarity is the idea that
terrorism at its root involves some type of “unacceptable” violence. This is where the question of
targeting becomes important in the classification of terrorism. For example, all experts on
terrorism agree that there is a line that exists between combatants and non-combatants, where
terrorism involves violence solely against the latter. Third, most definitions agree that terrorist
violence must be in pursuit of some political goal. This is what separates terrorism from purely
criminal behavior. Lastly, terrorism is defined as a strategy or a method of violence, and it is of
course possible for groups to have or utilize multiple strategies or methods in pursuit of their
political goals. Therefore, it is important to consider that many different types of groups can
employ terrorism, including insurgents and guerrilla fighters.
15

As the field of political science has moved more and more towards the quantitative
methodologies, scholars of terrorism and political violence have trended towards utilizing the
large cross-national databases with statistics on terrorist attacks across the globe. As this trend
has continued, much of the field has coalesced around a smaller set of definitions concerning
what terrorism is. Looking at what the majority of definitions share in common, the three main
characteristics of a terrorist attack are as follows: (1) the targeting of noncombatants, (2) for
political goals, (3) perpetrated by non-state actors - much in line with the discussion of terrorism
definitions above, but with the very explicit qualification that terrorism is a non-state
phenomenon.
As this study relies on data from the Global Terrorism Database for its quantitative
analysis, it is simplest to utilize the operationalization of terrorism put forward by that group:
“the threatened or actual use of illegal force and violence by a non-state actor to attain a
political, economic, religious, or social goal through fear, coercion, or intimidation” (LaFree &
Dugan, 2007). The definition provided here has many advantages to it, including that it covers all
three of the main definitional characteristics found in the literature. From a theoretical
standpoint, it is the inclusivity of the dataset that is particularly attractive. The tracking of both
threats and actual uses of force allow for opportunities to test for the full spectrum of terrorism -
for every terrorist attack that uses force must also carry some further threat with it as well. This
is one of the reasons terrorism is such an interesting field of study. For these reasons, the GTD
makes a suitable database to utilize for the analyses carried out in this dissertation.
2.2.2 Domestic versus Transnational Terrorism
Another reason why this dissertation relies on the data from the GTD is its inclusion of
both transnational terrorist incidents and domestic terrorist incidents from a very early time
16

period, something which none of the other major terrorism databases exhibit. The other two
major terrorism databases - International Terrorism: Attributes of Terrorist Events (ITERATE)
and RAND Database of Worldwide Terrorism Incidents (RDWTI) - are either private, cover
transnational terrorism only, or cover domestic terrorist incidents only since 1998 (Enders et. al,
2011). Therefore, to develop and test a theory that seeks to explain patterns of terrorist violence
on a domestic level over a period of time stretching past 1998, the only viable option for scholars
is the GTD.
To further clarify the distinction between these types of terrorism, some examples may
help. The first and most obvious transnational terrorist attack that may come to mind is the flying
of an American Airlines flight into the World Trade Center building on September 11
th
, 2001.
This was quite clearly a transnational terrorist attack for a few reasons: (1) The group that
perpetrated the attack was made up of citizens from several countries, (2) the attack took place in
a country different from that where its perpetrators were from, and (3) the attack killed and/or
injured citizens of many different nations. An example of a domestic terrorist incident is when
Paul Jennings Hill murdered Dr. Britton (an abortion provider) and his bodyguard in Florida in
1994, in protest to abortion. This was a political/religious assassination by an American of an
American. While both incidents are terrorist in nature, they are distinguished by the make-up of
their respective perpetrators and targets.
This is important because there are increasingly major differences between domestic
terrorism and transnational terrorism found in the literature on political violence. Yes, most
domestic groups will have a transnational nature in terms of funding and/or weapons, but the
grievances and goals of domestic terrorism originate in the homeland. Domestic terrorist attacks
are also planned and perpetrated by citizens or residents from within a state, whereas
17

transnational terrorist attacks in one state may originate from motivations, planning, and
individuals outside of that state. The key point here is that the two types of terrorism are unit-
heterogeneous. In other words, theories that explain transnational terrorism may be unable to
explain domestic terrorism; the dependent variables are sufficiently dissimilar so that no
theoretical frameworks are accurate enough to explain or predict both.
The differences between the two can more clearly be seen through the work done by
Walter Enders, Todd Sandler, and Khusrav Gaibulloev to separate the domestic terrorist events
from transnational terrorist events within the GTD (2011). Out of the 66,383 certain terrorist
incidents coded by the GTD between 1970 and 2007, 12,862 of them are coded as transnational,
46,413 as domestic, and the remainder as uncertain. That is to say, over the time period in
question, there were roughly 360% more domestic terrorist incidents than transnational ones.
Interestingly, many of the most popular studies on terrorism have been done with transnational
terrorism data, even though this type of terrorist violence is in the minority. Even as many
scholars have begun to develop and test their theories with data on domestic terrorism, it is often
times done with data that combines both domestic and transnational terrorism data.
Only relatively recently have scholars begun to look specifically at the empirics of
domestic terrorism (Savun & Phillips, 2009; Krieger & Meierrieks, 2010; Aksoy & Carter, 2014;
Aksoy, 2018). Following the direction of these more recent scholars, this study seeks to develop
a theoretical framework to explain a subtype of terrorism that is domestic - where the actors’
primary grievances, motivations, and target audiences are all within their home countries.
2.2.3 Status-Quo Terror versus Anti-state Terror
Within government, the media, and the field of political science, there is an
overwhelming tendency to think of terrorism as a violence that is revolutionary, in opposition to
18

the state, or against the status-quo. Whether the groups are committed to transnational terror (e.g.
al-Qaeda) or domestic terror (e.g. the Irish Republican Army), the goals of revolutionary groups’
attacks have been to overturn pre-existing policies and positions such as the removal of U.S.
troops from Saudi Arabia and the liberation of Northern Ireland from the United Kingdom. It is
safe to think about this type of terrorism as anti-state terror.
However, less talked about, and much less studied is the phenomenon of status-quo
terrorism; terrorist violence perpetrated in order to protect the socioeconomic and political
status-quo of a society (although this type of terrorism is mentioned in some studies, even if not
the sole subject of the study
1
). Oftentimes, this occurs when a group of elites (or other
individuals benefiting from the status-quo) feel threatened by the reformist/revolutionary goals
of another group within society. These individuals then band into violent status-quo groups -
called death squads or paramilitaries for simplicity - and utilize a campaign of terror against
those pushing for reform - called rebels or insurgents for simplicity - in order to dissuade them
from continuing to challenge the status-quo. These death squads often target suspected members
of the population who they deem sympathetic toward the insurgents to instill fear in that
population and discourage their further support of the insurgents. There are numerous examples
to be found of groups that have utilized such a strategy of violence (such as the Ulster Volunteer
Force, Autodefensas Unidas de Colombia, Ku Klux Klan, Ranvir Sena, etc.), and yet the
phenomenon continues to be under-theorized.
2.2.4 State Terror versus Non-State Terror
Part of the issue that confronts a more in-depth analysis of status-quo terror is its
considerable conceptual overlap with state terror. As mentioned previously, the definition of

1
(Jones & Libicki, 2008)
19

terrorism has coalesced around a few primary characteristics of violence - one of them being that
terrorist violence must be perpetrated by non-state actors. This defining characteristic was not
without some debate however, especially as members of the United Nations attempted to
precisely define terrorism in an attempt to increase their cooperation on stemming terrorist
violence. While some states wanted terrorism to be defined in terms of “‘criminal acts’ taken
against governments by individuals or groups” (Stohl & Lopez, 1984), other states looked at
terrorism more broadly “so as to include acts of governmental groups that violate human rights
and reinforce policies such as apartheid” (Stohl & Lopez, 1984). However, even given the strong
debate over precise definitions of terrorism, there was always a relative consensus that acts of
terror by the state are distinct from acts of terror by non-state actors.
Beginning in the 1980s, scholars began to theorize and test propositions relating to terror
by the state (Stohl & Lopez, 1984; Stohl & Lopez, 1986; Mason & Krane, 1989). In these early
works, and in later works as well, state terror includes terrorist violence perpetrated by state
officials and violence perpetrated by surrogates of the state in defense of that state. These
“surrogates” refer to members of society that are not formally part of the state that utilize
violence to protect the status-quo. Put differently, the literature on state terrorism has treated
status-quo terrorism by the state and by non-state actors as the same phenomenon. While this
dissertation agrees that there are similarities between status-quo terror by the state, and by non-
state actors, one of its key purposes is to make the distinction between them clear.
The distinction between the two phenomena of violence may be illustrated by the
following three examples: (1) status-quo groups arise independent of the state out of
dissatisfaction with the current policies directed at the opposition, (2) status-quo groups develop
with some degree of informal collaboration from the state - the perfect “enemy of my enemy”
20

situation, or (3) status-quo groups develop with formal collaboration from the state, which in
essence makes them part of the state.
From this point onward, the terms status-quo terrorism or status-quo terrorist violence
refer specifically to violence perpetrated by groups (1) and (2). While all three scenarios depict
groups that utilize terrorist violence to defend the status-quo, due to the operationalization of a
terrorist attack in terrorism datasets, the third type of group is generally not included in the data.
While these definitions make the phenomenon of terrorism seem quite variegated, it is
this disaggregation of the concept that makes the important study of status-quo terrorism
possible. Again, in agreement with the definition of terrorism as a phenomenon of non-state
actors, this dissertation sees a clear divergence between terror by the state and non-state actors;
marked by separate social and political forces that beg for analysis. Indeed, a question at the
heart of this work is centered around a state’s decision to loosen its monopoly on violence; why
do members of the state tolerate and/or facilitate terrorist violence from autonomous groups
defending the status-quo?
2.3 Terror in the Name of the State
To better understand why a state would tolerate and/or facilitate terrorist violence from
non-state actors, it would be helpful to analyze why a state might utilize terrorist violence at all,
whether in the form of official state repression or through surrogate groups. The literature on
state terror has produced a great number of insights, but three directions merit extra attention for
the purposes of this study. These are approaches to state terror that some believe encompass both
state and non-state status-quo terrorism. The lines of thinking to be discussed include two
internal characteristics of the state - regime type and state weakness - and one characteristic of
the international system - the degree of isolation from the world system.
21

2.3.1 State Terror as an Authoritarian Tool
There is a tendency in the literature to associate state violence with authoritarian or non-
democratic states. While there are many types of state sanctioned violence, this link is thought to
exist with state terrorism as well. This link exists due to a variety of reasons; the military
dominates decision making in authoritarian regimes, leaders in authoritarian states are less
constrained by institutions of accountability and thus more able to implement repressive policies,
and their decision-making process is less conditioned by concerns for the respect of human
rights.
Much of the literature struggles from thinking of the state as a unitary actor (Nicholson,
1986). This oversimplification of the state can clearly become an issue in the theoretical makeup
of state terror when one considers the different motivations and interests of actors across state
agencies. Likewise, when theorizing about state terror encompasses violence by both the state
and non-state actors, it is even more crucial that we do not look at the state as a unitary actor. For
example, in an attempt to formulate a model for explaining political repression, McCamant urges
an emphasis on the “central position of government in society…and the essential part that
coercion plays in ruling” (McCamant, 1984). This focus on the position of government should be
at the expense of too much emphasis on the role of social forces; it must be on "what is truly
political, not social, on the struggle for power and on the processes of ruling” (McCamant,
1984). When the phenomenon under analysis is repression solely by officials of the state and its
security forces, this dissertation sides with McCamant. However, when the phenomenon of state
terror includes violence from non-state actors like the autonomous death squads that are the
focus of this dissertation, understanding the social struggles of those civilians must also be taken
into account.
22

Mason and Krane also look at the ways in which death squads utilize violence as a way to
defend the status-quo of the ruling elites by suppressing those opposition movements that
threaten the system. However, their theory relies on the assumption that death squad violence is
most prevalent in authoritarian regimes where the military dominates. This type of violence
proliferates because “repression is the response for which the requisite institutional machinery
(i.e., the military) is most highly developed” (Mason & Krane, 1989). So where authoritarian
systems are met with resistance and/or calls for reform, the capacity for repression, and
incapacity for accommodation strategies, make it more cost-effective to approach the situation
with the former. Death squads then are a manifestation of the power of the military in the
authoritarian state.
There is a potential problem with this line of thinking. First, there is the common failure
to distinguish between those death squads that are de facto part of the state - consisting of “secret
police, special counterinsurgency units, or regular units of the armed forces engaged in various
forms of institutionalized violence” (Mason & Krane, 1989) - and those death squads that are not
formally or legally part of the state. While the power of the military in an authoritarian system
may help explain the level of state sanctioned death squad activity, it is important to look
elsewhere in order to explain terrorism by non-state death squads.
Not only are authoritarian systems expected to experience more death squad activity due
to the cost-effectiveness and strength of the military, but because those systems of government
lack the democratic values and institutions that serve as constraints on such behaviors of
violence. In democratic systems of government it is expected that:
Democratic political norms emphasize compromise in conflict and participation
and responsiveness in relations between rulers and ruled, traits that are inconsistent
with reliance on violence as an instrument of rule or opposition. Officials in
23

Western democracies who use or condone violence against domestic opponents risk
loss of legitimacy and office as a consequence. (Gurr, 1986)

Therefore, the protection of civil liberties, the liberal norms of compromise, and the
accountability that elections place on elected officials all theoretically serve to reduce the
propensity for the state to utilize terrorist violence against its citizens. In his study on state terror
in Latin America, Sloan concludes that it is the more authoritarian nature of that region’s regimes
that leads to more illegal violence from the state; “greater levels of repression are possible
because of the institutionalized styles of governing; the state of siege provisions in the
constitution; and the usual lack of an independent press, judiciary, and legislature” (Sloan, 1984).
In other words, these authoritarian tendencies reduce the costliness of state terror for government
officials by eliminating democratic constraints.
Likewise, Henderson finds a strong negative relationship between the degree of
democracy a state has and the amount of political repression that it uses against its civilians.
According to Henderson, “the accountability of a democratic government can mean that people
have a chance to acquire their goals and will not become so restive as to attract the policy of
repression.” (Henderson, 1991). Rather, state terror is more likely in authoritarian regimes
because those regimes lead to violent relations between citizens and the state, and therefore
require repressive policies from the latter.
However, it is again essential to consider the distinction between state sanctioned
terrorism and terrorism by semi-autonomous status-quo groups. According to Gurr (1986), Sloan
(1984), and Henderson (1991), democratic institutions make overall state repression less likely.
While it may be true that democracy discourages terrorism through the formal channels of the
state (i.e. the security forces) for the reasons stated above, it may have heterogeneous effects on
different types of status-quo terrorism - between state terror and non-state status-quo terror.
24

Democracy may be ineffective or even counterproductive at constraining terrorist violence from
non-state status-quo groups. In fact, as will be discussed in the theory section of this chapter,
facets of democracy are ineffective at restraining non-state status-quo groups, and in some cases
even encourage the surrogate campaigns of status-quo terror waged by non-state groups.
Fortunately, new data on pro-government militias (PGMs)
2
has allowed researchers to
better understand those violent groups which are less formally connected to the state. In an
analysis of the effect that pro-government militias (PGMs) have on human rights, Mitchell et al.
find that the presence of an informal PGM is associated with an increase in agent-centered
violations of human rights such as torture, disappearances, and extrajudicial killings (2014).
Viewing militia violence through the lens of a principal-agent problem, they theorize that as
militias become less officially linked to the government, they become less constrained by
reputational concerns, and therefore more likely to abuse human rights. Furthermore, countries
that are classified as weak democracies have been found to have the highest probability of
informal PGM activity (Carey et al., 2015). Weak democracies are able to be held accountable
for official state violence, yet are able to exploit a lack of freedoms, and thus informal ties with
PGMs becomes an attractive way to condone repression while maintaining plausible deniability.
These recent studies on PGMs have been transformative in the way scholars understand
these violent groups, and the reasons underlying state interaction with such agents, however, this
paper seeks to pursue this line of research further. While these studies have shown that informal
PGMs are associated with higher human rights abuses, this paper is particularly interested in the
amount of terrorist attacks that are specifically perpetrated by status-quo groups. Furthermore,
while the PGM database is quite extensive, and the groups found in it have some overlap with

2
(Carey et al., 2013)
25

the status-quo groups in this study, their coding misses a segment of the status-quo groups that
perpetrate status-quo terrorism.
2.3.2 State Terror as State Weakness
Another portion of the literature on state terror is devoted to the link between it and the
weakness of the state. The first question that must be addressed in this line of thinking is what
does one mean when they refer to the strength or weakness of the state. Second, how does that
power of the state cause greater or lesser terror by the state?
Mason and Krane hold that in response to social opposition or rebellion, the state may
broadly choose from two strategies; either from repression or from accommodation. According
to their rational choice model, those regimes that commit to more indiscriminate levels of state-
sanctioned terror do so due to state weakness - the inability to accommodate the preferred
reforms of the marginalized communities (Mason & Krane, 1989). Here the understanding of
state weakness is less concerned with the ability of the security forces to protect its citizens and
more focused on the state’s ability to provide satisfactory governance for poor and minority
populations.
Gurr’s emphasis on the link between state weakness and state terror is also less concerned
with the regime’s ability to provide security, and instead focuses on the ability for a state to
respond to challenges in a non-violent manner:
Weak regimes have limited material resources and low levels of political
institutionalization by contrast with strong ones. Both conditions constrain sharply
the policy option of challenged elites. Coercion is relatively cheap and within the
capacity of any rulers who have a standing army. Reform, cooptation, and other
non-violent strategies for dealing with challenges require re-allocable resources and
political institutions capable of inducing or absorbing change (Gurr, 1986).

26

However, yet again this dissertation stresses the importance of disaggregating non-state status-
quo terror from state terror in general. While it is plausible that regimes incapable of reform and
accommodation are more likely to resort to terror and repression, it is less clear why this type of
state weakness would lead to greater violence on the part of non-state status-quo groups. Those
groups interested in defending the status-quo do so because they are happy with the policies of
the state, and are typically actively opposed to the reform and accommodation sought by the
challengers of the state.
Instead, the type of state weakness that may matter for the type of violence under
investigation here would be the capacity of the security forces to maintain law and order. The
definition of state strength or weakness is then taken from Max Weber who claimed that a
defining characteristic of the state is the monopoly of the legitimate use of force (Weber, 2009).
States that are unable to maintain their monopoly on legitimate violence - i.e. cannot stop rebel
or criminal groups from using violence against its civilians - in this definition would be
considered weak. This type of state weakness might be what scholars should look at when
looking at the link to non-state status-quo terrorist violence. A state that provides support for a
particular class of society, and yet is unable to protect that group of people from violence, is a
weak state which may observe the formation and expansion of private groups of individuals
willing to use violence to defend the status-quo. In other words, it is the lack of efficient state
security forces - in combination with a threat to the status-quo - which causes people to take
matters into their own hands by utilizing violence against the challengers to the state.
2.3.3 State Terror and International Norms
Another section of the literature on state terror focuses on the role that the international
community plays on the decision making calculus of the state in its response to threats to the
27

status-quo. Here, the idea is very much in line with the second image reversed (Gourevitch,
1978) where the focus is on how the international system influences interactions at the domestic
level. Much of the international order since the end of World War II has been built by liberal and
humanitarian principles. As these international institutions have grown and become entrenched,
the liberal and humanitarian principles have diffused into those democratic elites most in contact
with such institutions. These norms have done much to promote the respect of human rights, the
peaceful resolution of internal and external conflict, and therefore to discourage certain types of
force - especially terrorist violence - from the state.
These norms, regardless of how they are formed, serve to make it more difficult for states
actively involved in the liberal international order to formally sanction state terror. In his
discussion on why states “are unwilling to accept terrorism as a legitimate tool officially,”
Nicholson claims that it is due to the fact that
they themselves might become the object of it…[and] therefore, they’re likely to
raise a moral standard against it, which, of course, makes it embarrassing if it is
found that the government is aiding and abetting similar activities elsewhere.
(Nicholson, 1986)

While Nicholson’s argument is directed towards state support of foreign terrorists, the reasoning
stands for state sanctioned terror against internal challengers - it makes the issue of denouncing
terrorist violence by rebels or insurgents more difficult when the state itself is utilizing the same
type of violence. In either regard, the point here is that the norms or moral standards against state
sanctioned terror should discourage state officials from terrorizing their publics through formal
state channels (the security forces). Instead, for those civilians and members of the state that
wish to respond to threats to the status-quo with draconian violence, they must do so through
private and informal channels - non-state status-quo terrorism.
28

Not only does the internalization of liberal and humanitarian principles alter the decision-
making process of individuals responsible for countering an insurgency, but it has serious
implications on foreign policy as well. As states democratize, and their officials internalize
democratic norms, more and more national elites become “one way or another accountable to
publics that are opposed to gross violations of human rights…[and thus] elites who practice state
terrorism can usually expect external pressures and informal sanctions to modify their policies”
(Gurr, 1986). Therefore, we should expect that when countering an insurgency, states that are
less involved in the international system are more likely to utilize repressive measures through
the formal channels of the state. Conversely, the external constraints placed on democracies
should make the more hawkish individuals in society more willing to utilize violence through
private and illegal channels.
Thus, the review of the determinants of state terror have focused on three major factors.
The first being regime type, with the assumption that state terror is a tool of authoritarian
regimes. The second factor examining state capacity, points to the idea that state terror is the tool
of weak states. Finally, those states that are most internationally isolated and independent of
other nations are the most likely to utilize state terror. While these studies are useful in
understanding the decision-making behind state terrorism, it is important to note that there hardly
any distinction in these studies between terror by the formal state apparatus and their non-state
surrogates. The next section will focus more in depth on the determinants of terrorism
exclusively by non-state actors.
2.4 Terror and Regime Type
This section analyzes two separate schools of thought on democracy and terrorism: A
review of the literature on (1) the institutional components of regimes and how these produce
29

structures of opportunity for violence, and (2) how regime type affects the degree of grievance
that opposition groups accumulate; distinguishing the two sets of literature into the strategic
school, and the political access school, respecitvely (Wade & Reiter, 2007). Borrowing from the
discipline of economics, both the supply and demand of a product are needed in any
comprehensive understanding of its pricing; likewise, analyzing the grievances and opportunity
structures of any conflict are necessary to understanding the costs and benefits of terrorism.
2.4.1 The Strategic School
According to the strategic school’s way of thinking, democratic institutions make
terrorism more likely. The liberties that democracies grant to their citizens and non-citizens alike
make their territories into soft-targets for those groups and individuals seeking to do harm. A
free press concedes greater publicity to attacks (Drakos & Gofas, 2006), freedom of expression
and organization allows for easier mobilization (Engene, 2004), and a constrained executive
makes counterterrorism more difficult to conduct (Wilkinson, 2001).
Regimes, regardless of their classification, have two main strategies for combating
potential terrorist groups: coercion and cooptation. For example, coercive strategies apply force
to degrade the capabilities of terrorist organizations either through raids, detainment, or active
combat. Cooptation strategies can take the form of anything from payoffs and rent-seeking, to
appointments to the government and inclusion in the electoral process.
Democracies are more likely to experience terrorism than single party authoritarian
regimes because they rely almost exclusively on cooptation with a low capacity for coercion
(Wilson & Piazza, 2013). The limitations placed on the coercive apparatus decreases the ability
of democracies to quell dissent, and therefore creates an environment in which terrorists are
more capable of utilizing violence. Furthermore, due to the constitutional protections typically
30

enshrined in democracies, terrorist groups expect limited consequences from their activity in
such regimes.
The media is an important factor in the argument put forth by the strategic school. The
freedom and independence of a country’s media can affect both the likelihood of terrorist attack
coverage, and the repression-protest nexus between a government and its opposition groups. As
Drakos and Gofas point out, the freedom of press within democracies may play dual roles in
accounting for increased terrorism. Terrorist attacks are more likely to be reported on in
democracies, as a controlled press may be forced to ignore events which make an authoritarian
regime look weak. Likewise, terrorist actors may choose to attack democracies for this very same
reason; they choose their targets based partially on the amount of publicity their attacks will
create (Drakos & Gofas, 2006). The media may also play a key role in facilitating protest against
a government by watchdogging the actions of political and economic elites. Yet, the media is
less likely to play this role in democracies, and when the probability of successful state
repression is certain (Kim, Whitten-Woodring, & James, 2014). Taken together, these findings
illustrate the importance in understanding how violence and disruption may be aided, albeit
unintentionally for the most part, by a free media, and therefore indirectly with regime type as
well.
Also relevant to this dissertation is the idea that democracies should be more probable
targets of terrorist attacks due to their respect for civil liberties (Wade & Reiter, 2007;
Chenoweth, 2010; Eyerman, 1998; Eubank & Weinberg, 2001). This variable, while intuitive at
first, is strikingly under-tested or incorrectly specified in the statistical literature to date.
Authors either lump the idea of respect for civil liberties with minimalist and procedural
definitions of democracy (Eyerman, 1998), or measure this construct through Freedom House
31

and Polity scores looking at levels of freedom and institutional constraints, or composite scores
of executive recruitment, constraints on the executive, and the competitiveness of participation
(Wade & Reiter, 2007; Chenoweth, 2010).
I argue that these data do not accurately measure the degree to which subpopulations
have their liberties disregarded or infringed upon. In other words, inaccuracies and errors in the
data used in these statistical analyses may bias the findings by failing to capture the motivating
grievances sparked by maltreatment from the state. For example, Freedom House creates a
measure for a country’s respect for civil liberties ranking states on an interval from 1 being the
best, to 7 being the worst. Questions included in the checklist referenced in the score include
ideas of an independent judiciary, protection from unjustified imprisonment, equal treatment of
laws and policies for various segments of the population, among many others. However, despite
the United Kingdom’s massive experiment with internment without trial between 1972 and 1975
in Northern Ireland, the country continually received the highest possible score on civil rights.
Instead, it is more important to look at a country’s record of respect for physical integrity rights,
in which the abuse of that respect has been shown to be correlated with increased terrorism
(Walsh & Piazza, 2010). Torture, extrajudicial killings, disappearances, and political
imprisonment disenfranchise segments of the population who then may provide much needed
support to terrorist groups.
2.4.2 The Political Access School
According to the political access school, democratic institutions allow subgroups within a
country a larger variety of channels through which they may channel their dissent non-violently,
making terrorism less likely. Institutions that promote inclusion and deliberation, such as free and
fair elections, or legislative bodies, should prevent the formation of grievances, or provide
32

aggrieved groups the opportunity to ameliorate their grievances already formed. If government
violent overreactions can alienate moderate civilians, then democratic protections of civil
liberties may restrict the available pool of individuals that terrorist groups are able to recruit
from. In other words, the rules that require democratic governments to be accountable to their
publics may also render them more capable of proficiently deterring, or countering, terrorist
violence.
Testing the conventional wisdom of the strategic school, Abrahms finds that democracies
– and specifically liberal democracies
3
– are the most effective regimes for counterterrorism. The
results suggest that the liberal aversion to civilian losses not only make these regimes more
motivated to fight terrorists, but it forces them to avoid the overreaction that further radicalizes a
public (Abrahms, 2007). The more internalized the norms of liberal ideology are in a country, the
more likely it is that leaders develop strategies of counterterrorism that are adept at maintaining
the support of its moderate and at-risk populations.
Draconian or repressive counter terror policies may disenfranchise civilians with current
political institutions, and violence may result when these civilians, “inadequately embedded in
institutionalized political life[,] seek political ends directly” (Rule, 109, 1989). In fact,
indiscriminate violence by the government is noted to be one of the leading factors in insurgent
recruitment, highlighted by the “tendency of insurgents to actually welcome incumbent
reprisals… because such reprisals bring in recruits.” (Kalyvas, 151, 2006). There are some
indications that there may even be a unique cost that democracies pay for enacting repressive
policies, centered around discord over the constitutionality of said measures (Foster, 2017;

3
These liberal democracies, according to Abrahms, are primarily defined by their commitment to the protection of
civil liberties.
33

Schmid, 1992; Wilkinson 2001). When democracies enact the policies that terrorism scholars of
the strategic school claim are most effective, the effects may be the opposite of what is expected.
Likewise, Collier finds that those democracies who are incapable of providing legitimacy
or accountability to their citizens are those that are most susceptible to political violence; as the
government becomes more legitimate and accountable, extremists are less able to attain
community support (Collier, 2011). In other words, democracy and political inclusion can act as
a conflict resolution mechanism (Mazzuca & Munck, 2014). This is supported by case studies
and quantitative studies which provide evidence that inclusion in political processes allowed for
successful negotiations between terrorist groups and governments (Dalacoura, 2006; Jones &
Libicki, 2008). So, not only do increased political freedoms and free elections make terrorism
less likely by providing alternative paths to political change, but they are useful tools for
managing or resolving terrorist conflict. In other words, in democracies experiencing insurgency
and terrorism, utilizing methods “acceptable to democratic societies” may prove more effective
in combatting such violence than “undemocratic” methods.
2.4.3 Heterogeneous Effects of Democracy and Different Types of Terrorists
Thus far, the literature purports that democracy effects levels of terrorism in opposite
directions. In some cases, this is due to the fact that distinct aspects of democratic regimes may
increase the likelihood of terrorism – such as a free press – or when certain institutions decrease
the likelihood of terrorism – such as greater inclusion in the legislative process (Li, 2005). These
heterogeneous effects of democracy are not irreconcilable, but others are a bit more problematic,
particularly when the same institutions are theorized to have contradictory effects. For example,
while the strategic school claims that executive constraints increase terrorism by providing
34

greater opportunities to terrorists, the political access school argues that these constraints
decrease terrorism by increasing the proficiency of counterterrorism strategies.
Therefore, it is important to take both schools of thought into account. For example, only
looking at the constraints and the opportunities that institutions provide fails to specify where the
motivations for terrorism originate. Furthermore, “Institutional constraints may provide a
permissive environment for terrorism to thrive in democracies, but they do not necessarily
prohibit democracies from pursuing repressive counterterrorist tactics in practice. Indeed, some
democratic electorates quickly and easily grant their executives extra powers in responding to
terrorism” (Chenoweth, 2010). Put differently, although democracies tend to respect civil
liberties more than autocracies, this does not mean that they cannot engage in harsh and targeted
violence against terrorist groups in times of emergencies.
In fact, an analysis of the theory on civil liberties and their relationship with terrorism
presents a logic which is unnecessarily zero-sum. Generally, the question is posed as to whether
we, “want to sacrifice some democratic substance in order to be effective against terrorism or do
we have to tolerate a certain level of terrorism for the sake of maintaining the civil liberties and
political rights we cherish?” (Schmid, 15, 1993). Yet, surely the two issues are not mutually
exclusive. It seems that in some cases, when democracies do not respect civil liberties, the results
may be detrimental to their counter-terrorism results.
These issues arise due to the relative lack of recognition of the heterogeneous nature of
terrorist violence, and the counter-terrorism campaigns that are formulated to ameliorate said
violence. Given this, the purpose of this dissertation is to further our understanding of a more
conservative type of terrorism, that violence which seeks to further the political and
socioeconomic status quo. It is important to note the scarcity of literature regarding this variety
35

of violence and its significance, if any, in the regime type-terrorism relationship. What are the
shared characteristics of these status quo groups, and what drives them to utilize their violent
strategies? Looking at the relationship of pro-government militias during times of civil war, the
state may exert control over them as a way of outsourcing the more draconian methods of
counterinsurgency (Campbell and Brenner 2000; Carey, Colaresi and Mitchell 2015). Likewise,
Stanton points out that these militias are often composed of individuals from outside the
insurgent communities, and for that reason, their violence may be likely to target the insurgent
communities’ civilian base – not being sure who is a rebel and who is not (Stanton 2015).
Questions remain as to how different regimes interact with status quo groups, and the degree to
which they outsource or turn a blind eye to violence committed by such organizations.
So, there seems to be a gap in the literature that needs to be addressed when it comes to
the thinking behind trends in democratic terrorism. The type of regime, and the institutions that a
state possesses, quite surely influences terrorist tendencies. However, questions still remain as to
the consistency of these relationships, and the direction in which they hold. Furthermore, do the
same general trends found in the literature thus far hold when looking specifically at a subset of
terrorist violence: status quo terrorism? Components of democratic rules and norms still shape
this variety of terror, but as this dissertation will show, the theoretical mechanisms that drive
these relationships may differ.
2.5 Summarizing the Literature
This section summarizes the general lessons and findings from the literature review. First
and foremost, this chapter delivers a broad overview of the developments made toward defining
what terrorism is. While there is still disagreement over the exact wording, a general definition
of terrorist violence can be adduced through its three core elements; it is violence defined by (1)
the targeting of noncombatants, (2) for political goals, (3) and perpetrated by non-state actors.
36

Furthermore, the literature review seeks to highlight the major trends in the field in regards to
both state terror and terrorism by non-state actors. More specifically, much care is taken to focus
on how regime type has been regarded as a key determinant in the decision to utilize such
violence.
However, in both fields a glaring gap has been identified which is necessary to address
going forward - the lack of a consideration that terrorist violence by non-state groups in defense
of the status-quo is separate from broader conceptions of terrorist violence. We must
disaggregate the concept of terrorism in order to truly understand its association with important
stimuli such as democracy and its institutions. The following chapter does just this – creating a
theoretical framework which models the ways in which regime type ultimately influences status-
quo terrorist violence.
37

Tables and Figures

Table 2.1
Definitions of Terrorism
Author Definition
Martha Crenshaw (1972) Terrorism is part of a revolutionary strategy--a method used by insurgents to
seize political power from an existing government...manifested in acts of socially
and politically unacceptable violence. There is a consistent pattern of symbolic or
representative selection of the victims or objects of acts of terrorism. The
revolutionary movement deliberately intends these actions to create a
psychological effect on specific groups and thereby to change their political
behavior and attitudes.
Alex Schmid and Albert Jongman (1988) Terrorism is an anxiety-inspiring method of repeated violent action, employed by
(semi-)clandestine individual, group, or state actors, for idiosyncratic, criminal, or
political reasons, whereby—in contrast to assassination—the direct targets of
violence are not the main targets. The immediate human victims of violence are
generally chosen randomly (targets of opportunity) or selectively (representative
or symbolic targets) from a target population, and serve as message generators.
Bruce Hoffman (2006) Terrorism is ineluctably political in aims and motives, violent—or, equally
important, threatens violence, designed to have far-reaching psychological
repercussions beyond the immediate victim or target, conducted by an
organization with an identifiable chain of command or conspiratorial cell
structure (whose members wear no uniform or identifying insignia), and
perpetrated by a subnational group or non-state entity.
38

Chapter 3
A Comprehensive Theory of Status-Quo Terrorism
3.1 Introduction
In this chapter, I lay out the theoretical model for status quo terrorism. As was
discussed in Chapter 2, while important findings have been made in the field of terrorism
studies – leading scholars to have a much higher degree of understanding regarding
terrorist violence – there are still crucial dilemmas that are ready to be solved. Key
among these dilemmas is the importance of further disaggregation of the phenomenon of
terrorist violence. Many of the theoretical models that have been developed to explain
such violence are too broad and treat certain political and socioeconomic variables as
determinants for what are meaningfully different types of violence. Thus, it is necessary
to formulate a new theory to explain a key disaggregated form of terrorism – status-quo
terrorist violence – including how it is different, and what its main determinants are.
I address the scope of the research project, again specifying the type of terrorism
under study, and the specific conditions under which the theory applies. Next, issues of
incomplete information and community support during internal conflict are discussed -
laying out the blueprint for why insurgency makes status-quo terrorism more probable. I
then develop the theoretical framework surrounding non-state status-quo terrorism and its
relationship with democracy. To do so, it is essential to look at both liberal constraints to
counterinsurgency, and democratic opportunities provided towards status-quo groups.
3.2 Scope
As noted in the preceding chapter, the purpose of this dissertation is to study a
very specific and under-theorized subtype of political violence - status-quo terrorism. The
39

purpose is to clearly distinguish terrorism perpetrated by groups seeking to maintain the
status-quo from the more thoroughly researched anti-state terrorism. These status-quo
groups may have varying degrees of ties with their states, from almost entirely
independent of the state to close cooperation with it. The unifying factor between these
groups is that they are characterized by both their violent opposition to anti-state groups,
as well as their intention to protect certain institutions or norms they fear to be threatened.
While such groups may at times bump heads with elected officials or parts of the formal
security apparatus, their primary goal is to not allow the political or socioeconomic
foundations of the state, which benefit them, to be altered. It is crucial to study these
groups because much of the quantitative work done on terrorism has treated status-quo
terrorism as the same phenomenon as anti-state terrorism. Likewise very few of the
qualitative studies in the field have focused on terrorism perpetrated in favor of the
status-quo. This leaves an obvious oversight in the field that is ripe for rectifying.
The goal is to better understand the terrorist violence of these status-quo groups.
Thus the theory laid out below takes as one of its major assumptions that these groups are
most likely to be violent in response to an insurgency or civil war that challenges the
state. It is in these situations of domestic conflict where one would expect the pressures
for status-quo violence to be highest - as the threat to the status-quo has culminated in an
attempt to violently overturn the government. This condition of insurgency or civil war is
an essential point of logic in the theory, and the relationship between insurgency and
status-quo terrorism is tested for in the following chapter. To briefly foreshadow the
results, I find evidence that status-quo terrorism is indeed more likely in countries
experiencing a prolonged campaign of violence from challengers to the state. However,
40

the theory and its propositions apply to any challenge to the state and its status-quo, and
its findings are not dependent upon analyzing cases only in which the state is
experiencing an insurgency or civil war. Instead, much of the logic expressed in the
theoretical model may be present in non-violent threats to the status-quo. Thus, while the
majority of status-quo terrorist violence takes place in country-years experiencing an
insurgency or civil war, the testing is not limited to only such cases.
3.3 Status-Quo Threats and Constraints to Countering an Insurgency
So why does insurgency increase the prevalence of status-quo terrorist violence
by groups autonomous from the state? There may be many reasons why internal warfare
makes status-quo terrorism more likely, yet it is important to look both at the rationale of
non-state status-quo groups themselves, as well as the decision-making process of the
state. In regards to the former, a threat to the status-quo is a necessary condition for
private groups to rise in defense of the existing state of affairs - it is the motivation for
their violence. Concerning the state, much of the explanation lies with problems of
incomplete information and the constraints this places on the state in its efforts at
counterinsurgency.
First and foremost, a perceived threat to the socioeconomic and political status-
quo is a necessary condition for non-state status-quo terrorist violence. Moreover, an
increased severity of that perceived threat makes the likelihood of status-quo terrorism
more probable. One way in which those groups concerned with protecting the status-quo
gage the severity of this threat is based on the capability of the challengers to sustain a
campaign of insurgent and terrorist violence against the state. For these intuitive reasons,
41

an insurgency should increase the motivations of private individuals interested in the
status-quo to utilize more extreme means, and even violence, against anti-state groups.
On the state side of this issue (whether they should tolerate and/or cooperate with
these status-quo groups) one must look at the constraints it faces when conducting
counterinsurgency. For simplicity, it helps to think of a government’s strategy for counter
terrorism as a continuum, ranging from repressive on one end to inclusive on the other.
On the inclusive end of the spectrum, incumbents can attempt to resolve conflict
primarily through engaging with aggrieved or marginalized communities. This type of
strategy may take the form of a variety of policies. For example, governments may
commit to infrastructure projects in areas that need them the most, they can emphasize
community policing policies which engage with at-risk populations, and they may even
take part in negotiations for peace.
On the repressive end of the spectrum, incumbents can commit to tactics which
can inflict heavy costs on insurgents and provide for short term security, but which may
come at the detriment of community support and long-term security. Such tactics may
include internment of suspected terrorists, surveillance, illegal searches, and
indiscriminate violence against communities suspected of harboring terrorist insurgents
or terrorist rebel sympathizers. These policies are able to degrade the immediate
capabilities of insurgent groups by eliminating leaders, reducing the numbers of these
groups, destroying their bases of operations, and cutting off their resources.
When choosing its particular amalgamation of repressive and inclusive policies to
counter an insurgency, building and maintaining trust among affected communities is
ever-present in the mind of the state. One reason why community support is so crucial in
42

counterinsurgency is due to the specific role that incomplete information plays in
irregular or asymmetric warfare that is characteristic of intrastate conflict. Both the
insurgent groups who utilize terrorism, and the government forces who try to prevent that
violence, rely heavily on local communities for intelligence (or lack thereof) in order to
effectively fight against the other.
Terrorism is primarily a strategy of the weak, utilized when one side is incapable
of confronting the other head on in conventional warfare. A major part of its utility
derives from the commitment to secrecy and underground tactics (especially when used
by challengers to the state), making it extremely difficult for governments to distinguish
between combatants and non-combatants. The state faces an adversary which is difficult
to identify; insurgents and groups that commit terrorist attacks do not typically wear a
uniform or wave a flag as they plant bombs, and their active supporters do not make it
public that they are providing safe haven or meals and resources to the insurgents. In
other words, the state typically has incomplete information on who exactly they are
fighting.
Those who utilize terrorist violence take advantage of these issues of incomplete
information through strategies of provocation (Kydd & Walter, 2006). In essence,
challengers of the state use relatively low levels of violence to provoke disproportionate
and/or indiscriminate attacks by the state in retaliation. Without the ability to distinguish
between the challengers and non-combatants, such retaliation risks the chance of
aggrieving previously neutral individuals into supporting the challengers. Such support,
whether active or passive, is often the lifeblood of insurgent groups, providing them with
manpower, material resources, places to train and plan operations, and maybe most
43

importantly the gift of silence. The key supporters are the ones living in those
communities where the insurgents are most active, giving them access to crucial
intelligence for the state who wishes to disrupt the insurgents and prevent their attacks.
Their silence to the security forces is a key asset for the insurgents' chances against the
state.
Part of any successful counterinsurgency by the state then must rely on the ability
to collect intelligence from community members with knowledge of insurgent operations.
Unless the security forces are able to maintain a permanent presence where the insurgents
are present, in order to coerce the intelligence from community members, they need to
collect intelligence some other way. To do this, the security forces must be careful not to
alienate these potential sources of information with indiscriminate attacks. Therefore, the
state is faced with a dilemma constraining it from utilizing the brute force it may
otherwise wish to employ.
Given these obstacles in counterinsurgency, states become more likely to partially
relinquish their monopoly on violence - so long as those they are relinquishing it to hold
similar interests in upholding the socioeconomic and political status-quo. There are two
benefits to the counterinsurgency aims of the state in tolerating and/or facilitating the
violence of non-state status-quo groups.
Firstly, these groups consist of local community members with access to the type
of information the state needs for successful counterinsurgency. These groups are
typically made up of the local elites most rooted in the status-quo, the victims of
insurgent/terrorist violence, and poorer members of society who can be enticed by the
pay that such organizations can offer. One of the comparative advantages of status-quo
44

groups in the fight against insurgents is the direct access they have to local intelligence.
Therefore, states faced with the obstacles inherent in irregular warfare should be willing
to deal with status-quo groups as an important source of information.
Secondly, the toleration of and/or cooperation with status-quo groups gives the
state a degree of plausible deniability while still advancing the defense of the status-quo.
Part of the assumption in this theory is that part of the state (and mostly in the security
forces) will have a general proclivity towards a more repressive and violent response to
insurgent and terrorist violence. As discussed just above, the state must always be
concerned with the way such draconian responses can affect community support, and
therefore their ability to collect important intelligence. One way around this obstacle is to
allow violent status-quo groups to carry out these violent campaigns against the anti-state
groups while remaining officially separate from the former. In other words, the status-quo
terrorist violence that comes from these groups is akin to the brute force that many in the
state wish they could implement as part of their formal counterinsurgency strategy - with
the ability to deny any involvement in such violence and avoid alienating important
community support.
Therefore, the pressures that come as a result of countering an insurgency make
status-quo terrorism more probable. The threat to the status-quo is very high, as
insurgencies are typically fought by rebels committed to overturning the current system
of government - making status-quo forces more likely to fight. Furthermore, issues of
community support and incomplete information incentivize the state to either allow or
facilitate counterinsurgency efforts from their non-state allies. Thus, the first proposition
is as such:
45

Insurgency Proposition: States experiencing an insurgency are more likely to
experience non-state status-quo terrorism.
To summarize, violent uprisings against the state – such as an insurgency –
present a clear and dangerous threat to the self-interest of those benefitting from a
country’s status-quo. This serves to radicalize certain segments of the country’s elite,
creating a force separate from the state ready and willing to utilize violence to protect
their interests. Furthermore, internal conflicts are often asymmetric and irregular, plagued
by problems of incomplete information. More specifically, state security forces are often
fighting individuals who don’t wear uniforms and tend to blend in with the civilian
population. In turn, the military should be more willing to cooperate with those non-state
groups who share similar goals in defeating the insurgents, and that are more capable of
gathering local knowledge through any means necessary. Therefore, it is expected that
insurgencies increase the likelihood of status-quo terrorist violence.
3.4 Liberal Constraints on Counterinsurgency
Given that a state is experiencing an insurgency, the next question this chapter
addresses is why democracies are more prone to experience status-quo terrorism as a
response to such a challenge. To solve such a puzzle, one must first look at the
constraints a democracy places on a state’s counterinsurgency options. How do the
formal and informal institutions of democracy alter the solutions for challenges to the
status-quo, and how do hard-liners (both within and outside of the state apparatus)
respond to such limitations?
The characteristics normally attributed toward democracy, and especially toward
liberal democracy, constrain the state’s ability to counter an insurgency in an overly
46

repressive manner - in a way that the hard-liners or “hawks'' in a country may prefer.
What are these characteristics - the formal and informal institutions - that limit the ability
of democratic states to enact repressive policies? First and foremost, one must look at the
democratic norms that promote peaceful resolution to conflict and the respect for human
rights. Furthermore, given that a sizable portion of the public shares these values, it is
then essential to examine the formal institutions that increase the accountability of
democratic leaders.
Generally speaking, democracies promote and instill norms within their societies
that disavow violence. This is one of the tenets of the democratic peace theory, which
holds that democracies do not fight against each other due to similar values that favor
negotiation over violence. The same logic applies to situations of internal violence - what
some have called the domestic democratic peace
4
. In other words, the same values that
constrain democracies from using violence against each other act to limit the state from
using excessive violence within democracies. Specifically, democracies instill in their
citizens respect for human rights and civil liberties - key among them being the freedoms
from extrajudicial killings, torture, and arbitrary detention, the right to free and fair
elections, among many others. Taken together, these democratic values promote the idea
that conflicts should be solved in a rules-based manner, if not through the ballot box.
Moreover, democratic states are set up so as to disincentivize the infringement of
these rights. Democracies have a variety of formal institutions which are meant to hold

4
There is a growing amount of literature concerning the idea of the domestic democratic peace (Conrad & DeMeritt,
2012; Davenport, 2007; Davenport & Armstrong, 2004; Poe & Tate, 1994). The general ideas is that key democratic
institutions decrease the probability that states will utilize repression against their citizens to violate their civil and
personal integrity rights. However, not all democratic institutions reduce state repression the same way and to the
same degree.
47

public officials accountable, punishing those leaders who breach the system of norms
which encourage peaceful resolution to conflict. The most obvious of these institutions is
the election of public officials - making political leaders directly accountable to the
public who vote them in, and out of office. When a sizable portion of the population
shares the democratic values centered around human rights, it is reasonable to assume
that those politicians who support a campaign of terror and indiscriminate violence
against fellow citizens may face an increased likelihood of losing their next electoral
race. Assuming that a politician’s primary goal is reelection - to maintain their position of
power - this electoral accountability should motivate elected officials to mitigate the
extent of repression within a counterinsurgency campaign.
On the opposite end of the spectrum of regime types, it is useful to think about
how fully authoritarian regimes may or may not be disincentivized to utilize violence
against their citizens. For example, most authoritarian regimes do not bother with
creating any institutionalized human rights standards. Furthermore, even for the publics
of such regimes which are instilled with a belief in human and personal integrity rights,
they typically lack the ability to hold their leaders accountable for such transgressions.
For example, there are usually no meaningfully competitive elections through which
citizens of authoritarian regimes can vote their leaders out of office. Without such
mechanisms for accountability, hard-liners within the government and security forces are
much more likely to utilize repressive tactics against challengers to the state as they see
fit.
3.5 State Decision-Making and Hard-Liner Reactions
48

It is not enough to understand how democracies may be constrained in their
counterinsurgency efforts alone. It is essential to also theorize the way in which different
actors within these democracies will respond to these constraints. This section of theory
focuses in depth on state decision-making in response to such structural forces, as well as
the reactions of different types of hard-liners within a democracy - both within and
outside of the state apparatus. After describing the ways in which democratic limitations
on counterinsurgency clash with the interests of the hard-liners, I hypothesize the ways in
which these hard-liners respond to such limitations - ultimately leading to status-quo
terrorist violence.
So how do the democratic institutions discussed in the previous section impact
upon state decision-making in regards to counterinsurgency? In essence, the institutions
under consideration either increase the portion of the population’s (and state’s) belief in
non-violent resolution of conflict, increase the costs of implementing overly violent or
repressive policies, or a mixture of both. As either condition increases, the likelihood that
the state will commit to draconian policies will decrease. Whether these are policies of
internment, systematic torture, or extrajudicial killings by the security forces, democratic
institutions make the repercussions from such actions especially costly. Compared to
their non-democratic counterparts, leaders must continuously fear the electoral
repercussions of their actions. Therefore, when leaders work against public opinion and
violate the human rights of their own citizens, they are much more likely to be held
accountable than they would be otherwise as part of a different regime type.
Likewise, these democratic institutions should increase the likelihood that the
state will try to find a peaceful resolution to the insurgency, as democratic regimes have
49

the unique ability to use the promise of electoral institutions as tools for conflict
resolution. In other words, as opposed to their non-democratic counterparts, democratic
regimes can offer carrots to insurgents in the form of political parties, guaranteed seats in
the legislature, amendments to electoral rules to favor minorities, among other forms.
Combine this with the proclivity of democracies to have a greater degree of respect for
human rights, and you have a state that takes an overall more inclusive approach to
counterinsurgency than hard-liners would pursue.
The first hardline actors that require analysis are the individuals and groups, not
formally part of the state, which have a vested interest in maintaining and protecting the
socioeconomic and political status-quo. That is to say, these are the private citizens
within a democracy which benefit from the current system in place. They benefit so much
from the status-quo that they are willing to utilize violence in order to protect it. It is fair
to think of this set of actors as extremists.
While the particular status-quo may vary from case to case, what matters is that
the existing state of affairs is challenged. When that status-quo is challenged, it is
essential to look at the way in which hard-line citizens respond to the actions of the state
and its security forces. Regardless of the type of democracy, there will always be some
portion of the population that perceives the liberal rules of engagement and criminal
justice to be too soft on domestic rebels. When these extremists perceive the response to
the status-quo threat as being limited by democratic institutions, or perceive new
democratic institutions as potentially detracting from their power, the hard-liners will be
more likely to take matters into their own hands. Ultimately this may end up with the
50

hard-liners ready and willing to fight an irregular war against those they see as the chief
danger to the system in place.
Furthermore, just as one must examine the reactions of hard-liner citizens, one
must also analyze the attitudes and responses of hard-liners within the state. More
specifically, how do the hawkish politicians, bureaucrats, and members of the security
forces behave when they find their options to counterinsurgency limited by democratic
norms and institutions? Similar to hard-liner citizens, it is safe to assume that there will
be a non-zero portion of the security forces who are disillusioned with these liberal
constraints, and thus will also be willing to consider extra-legal means to defeat the
insurgency in the way they see fit. When the liberal constraints are strong enough to
dissuade these hard-liners within the state from taking such action in their formal
capacities, then the phenomenon under consideration leaves the realm of state terror.
Furthermore, when a sizable population of hard-liner citizens present themselves as an
alternative force for counterinsurgency, hard-liners within the state should seek to
provide assistance to such autonomous forces, or at the very least push the state to
tolerate that violence. An enemy of my enemy is my friend!
It is the interaction between these two groups of actors which plays the most
significant explanatory role in understanding status-quo terrorist violence. Analyzing
only one or the other misses half of the equation. Variation in status-quo terrorism, which
is explicitly a behavior of non-state actors, is dependent upon the military’s willingness to
allow a breakdown in their monopoly on violence. This willingness depends upon the
position of, and constraints put on the security forces - a strong and unrestrained military
51

has less need to create shadowy alliances with paramilitary groups or to be permissive of
the violence such groups perpetrate.
Furthermore, the relative autonomy that the paramilitary groups have is crucial to
understanding such violence. It is this autonomy of the non-state actors that give the
hard-liners within the state an opportunity to channel their draconian preferences for
counterinsurgency. The perceived independence of the paramilitary groups allows the
security forces plausible deniability in any violence that originates from these groups.
Such plausible deniability is necessary precisely because of the costs that democracy
places on state terror. So, whether these paramilitary groups are truly autonomous from
the state or not, the state has a vested interest in maintaining the perception that they are.
Such a distinction is what ensures that such violence is not categorized as state terror, but
rather what this study defines as status-quo terrorism. This interplay between democracy
and hawkish members of society leads to this dissertation’s main hypothesis - to be tested
in the following chapter:
Democracy Proposition: Democracies are expected to encounter more status-quo
terrorism than non-democracies.
Again, it is highly useful to apply the logic of the theory to authoritarian regimes.
Elites concerned with the status-quo may be less likely to mobilize and utilize violence
themselves if they perceive that the military is not constrained by concerns over human
and personal integrity rights. Furthermore, even in situations in which non-state actors do
decide to mobilize in favor of the status-quo, the state and its security forces are much
more likely to establish a formal relationship with those allies. Without such a need to
maintain plausible deniability over the transgressions these groups may cause, an
52

authoritarian state will seek to maintain formal control over such paramilitary groups.
However, as terrorist violence (status-quo terrorist violence included) is inherently a non-
state phenomenon, the more formal control that authoritarian governments tend to keep
over these groups transform such violence into state terror or state repression.
3.5.1 Causal Mechanisms
This theory also develops potential causal mechanisms through which the process
of causation takes place between a very broad and structural variable - democracy - and
status-quo terrorist violence. Again, the theory developed here emphasizes the
importance of understanding how democracy affects two sets of actors - those within the
state and those outside of the state.
The first proposed causal mechanism to explain status-quo terrorism relates to
how the military reacts to democratic tendencies when countering challenges to the state
or status-quo. Assuming the interests of the military as an actor - due to its training and
general mission as an organization – it is likely that the security forces prefer a violent or
draconian reaction to such challenges, relative to other sectors of the government.
Therefore, more inclusive approaches that democracies are anticipated to adopt for such
challenges should alienate at least part of the security forces. Policies such as cease-fires
and peace negotiations with insurgents and an emphasis on human rights standards within
an ongoing insurgency frustrate the security forces by “tying their hands.” Agitated by
such democratic constraints, hardline members of the military will search for alternative,
and potentially illegal routes to accomplish their counterinsurgency goals. Thus, under
such circumstances they are more likely to pursue their preferred counterinsurgency
methods through shadowy alliances with non-state groups. Without such alliances, it is
53

difficult for paramilitary organizations to operate freely throughout and enact high levels
of violence. Therefore, democracy leads to higher levels of status-quo terrorism by
pushing the military to facilitate and/or allow the violence of non-state paramilitary
organizations.
The second and third proposed causal mechanisms regard the ways in which
democracy affects a country’s elites and other individuals outside of the state apparatus
that are interested in maintaining the status-quo. Again, democratic approaches to
challengers of the state will frustrate and alienate these elites, and push the most extreme
of them to violence. The second causal mechanism addresses the anger and fear that
elites feel when challengers to the status-quo attempt to institutionalize their reformist
ideas. For example, when pushes for greater political representation displace political
elites, challengers to the status-quo may gain the ability to change the system from
within. The elites that feel most threatened (depending on the makeup of the country in
question), will then look for alternative, more violent paths to maintain power. The third
causal mechanism regards these actors’ frustration with limitations placed on the security
forces in their handling of insurgents (or protestors for example). When these individuals
perceive the military as being too constrained to fight the insurgents properly, tempers
boil over even further, making it more likely that these actors will mobilize and fight their
own war to protect their way of life.
Taking these last two mechanisms together, democratic approaches to solving an
insurgency can lead to the institutionalization of the preferences of the challengers, with
seemingly too soft of a reaction. These issues culminate in significant frustrations within
the hardline non-state actors, motivating them to mobilize and utilize violence to protect
54

their own interests. Thus, democracy leads to higher levels of status-quo terrorism by
allowing the institutionalization of insurgent preferences and constraining the military’s
response to the insurgents. This in turn convinces individuals, who benefit the most from
the status-quo, that they must use violence themselves in order to stop such advances.
One caveat to emphasize is that this logic does not solely rely on the condition of
there being an insurgency present within a state. The main prerequisite for status-quo
terrorist violence is that there is some sort of perceived threat to the political and
socioeconomic status-quo on the part of those who benefit most from it. Likewise, the
threat needs to be perceived of as being dangerous enough that some subset of elites are
willing to mobilize and use violence to protect themselves from it. While some sort of
violent uprising against the state – such as an insurgency – is on the high end of any
continuum measuring threats to the status-quo, other less violent threats would be
sufficient as well. For example, some nascent political movement ideologically opposed
to the current status-quo, which is gaining strength through elections and protest
movements may be sufficiently threatening to some individuals and groups. Coupled with
restrictions on the police and military to respond to members of such a movement with
violence, the same formula for status-quo terrorist violence is very much applicable in
such cases.
To briefly summarize the theory above, status-quo terrorist violence is more likely
to occur in democracies for several reasons. Recall that terrorist violence, by its very
nature, is defined as a non-state phenomenon. Due to the formal and informal institutions
of democracy, those governments are disincentivized to utilize state repression to counter
challenges to the state. Democracies are also more likely to attempt to solve such
55

challenges in a peaceful manner utilizing their unique institutions as carrots. Therefore,
hard-liners outside of the state apparatus become more likely to utilize violence against
such challengers themselves, and hard-liners within the state apparatus become more
likely to either help or allow said violence.
3.6 Summary
This section is meant to integrate the discourse above into a single unified theory.
To do so, it is essential to first clarify the limited scope of this study; to account
specifically for the phenomenon of status-quo terrorism. This type of violence refers to
terrorist violence perpetrated only by those groups perceived to be independent enough
from the state so that their actions are not coded as state-sponsored terrorism. What this
dissertation is modeling then, is how and why the state’s monopoly on violence breaks
down and non-state status-quo forces fill part of the counterinsurgency vacuum.
To accomplish such a task, this dissertation identifies the primary influence that is
expected for regime type on state decision-making in counter-insurgency. Specifically,
due to the liberal constraints typically found in democracies, states under such a regime
type are less able to counter an insurgency in a draconian or repressive manner.
Assuming a portion of the security forces which prefers such aggressive policies, and a
portion of the population which is willing to fight the insurgents on their own, liberal
constraints promote a subtle relationship between the two - one where violence is
perpetuated to protect the status-quo, and yet plausible deniability is maintained.
Furthermore, compared to democracies, non-democracies are much less
constrained by domestic institutions in establishing formal connections with such status-
quo groups. Without the need for plausible deniability, the state is able to maintain
56

control over violence by such groups, and thus that violence is normally considered part
of state-led violence or terror. Therefore, the expectation is that democracies experience
more status-quo terrorism than non-democracies. This proposition is tested in Chapter 4
through a cross-national quantitative analysis, analyzing whether there is support for the
theorized relationship between regime type and status-quo terrorist violence. Moreover,
to understand the true causal processes through which democracy affects status-quo
terrorism, Chapter Five explores the causal mechanisms leading to the varying levels of
status-quo terrorist violence in the region of Urabá, Colombia from 1982-2003.

57

Chapter 4
Status-Quo Terrorism:
A Cross-National Analysis
4.1 Introduction
This chapter provides quantitative analyses to discover whether there is
supporting empirical evidence for the theoretical model developed in the preceding
chapter. Different methodologies offer their own comparative advantages, and utilizing as
many of these advantages as possible is important for gaining sufficient causal leverage
in a research program. The cross-national analysis employed in this chapter gives the
dissertation the chance to test for statistical associations between different measures of
democracy and status-quo terrorist violence, over time, and across many states - all while
controlling for several important variables found to affect political violence throughout
the literature.
The chapter will unfold in the following fashion. First, the data used in the
analysis will be discussed. This includes a definition of, and coding process employed
for, the dependent variable - status-quo terrorist attacks. This section also covers
descriptions of the different measures of democracy used throughout the statistical
models, as well as the control variables and the reasons for their inclusion. Second, I
describe the specific tests that are run, and why each is chosen for this specific research
question. Third, the results of the quantitative analysis are discussed. Lastly, I summarize
the general findings of the chapter and how they relate to the overall research project.
4.2 Data and Methodology
58

I construct an empirical analysis to test the propositions put forward in Chapter 3.
The first of those propositions deals with the relationship between insurgency and status-
quo terrorism. More specifically, the insurgency proposition expects that states
experiencing an insurgency are more likely to experience non-state status-quo terrorism.
The second proposition to be tested for in this chapter deals with democracy’s association
with status-quo terrorism. The democracy proposition predicts that democracies are
expected to encounter more status-quo terrorism than non-democracies.
The unit of analysis for the forthcoming tests is the country-year. The dependent
variable is taken from the Global Terrorism Database (Lafree & Dugan, 2007), and
reshaped to exclude transnational attacks according to the methods developed by Enders,
Sandler, and Gaibulloev (Enders et al., 2011). The dependent variable, SQ attacks, is a
count variable measuring the number of status-quo terrorist attacks in a given country-
year.
In order to determine a status-quo attack, I coded terrorist groups from the GTD
as status-quo or not. To do so, the dataset was filtered to identify all of the groups who
committed ten or more attacks during the sample years. Next, each group was hand coded
as status-quo or not based on the following conditions
5
. First, each group was coded as
status-quo if it was included in the Pro Government Militia Database (PGMD) (Carey et
al., 2013), and showed up as a violent non-state group in the GTD. Second, for those
groups that were not found in the PGMD, secondary search was conducted to determine
whether the groups had either (1) explicit relationships with the government, military, or
police forces, (2) an implicit relationship with the government, military, or police forces,

5
To alleviate fears of coder-bias in this process, one of my priorities in the future – before publication of my results
– is to run a set of inter-coder reliability tests.
59

or (3) a violent strategy focused on anti-state groups, or their communities, as a means to
support the status quo. The point of emphasis here is that groups are coded as status-quo
based on their explicit or implicit goals. A relationship with the government is a
sufficient condition, but not a necessary one. The necessary precondition is the group’s
motivation to protect the status quo violently from those seeking change.
6

To better illustrate this coding process, a few concrete examples are included.
First, as an example of a group that is included in this study’s data from the PGMD
database is the United Self-Defense Forces of Colombia (AUC). The group was a right-
wing paramilitary organization formed to defend themselves against the threat of the
guerrillas. They maintained explicit ties to segments of the Colombian state (military,
police, and legislators), yet independent enough from it to be considered a non-state
group within the GTD. In other words, those involved in the coding process for the GTD
did not categorize the AUC as a formal part of the state, and so their violence was coded
as terrorism.
Second, as an example of a group that is included in this study’s data that is
excluded from the PGMD database is the Ulster Volunteer Force (UVF) in Northern
Ireland. Also a right wing-paramilitary organization, this group was always suspected of
having implicit ties with the military, police, and intelligence services of the United
Kingdom. Most important is their clear motivation for mobilization and sustained
violence: an attempt to maintain the protestant domination of the Northern Ireland state

6
An example to illustrate the thinking behind this coding process can be found by looking at a statement made by
Ian Paisley, the leader of a protestant paramilitary organization, and later First Minister of Northern Ireland. At a
public rally in Newtownards, Paisley stated that “My men are ready to be recruited under the crown to destroy the
vermin of the IRA. But if they refuse to recruit them, then we will have no other decision to make but to destroy the
IRA ourselves!” Here we can see that the Third Force’s main motivation was to defend the state against agitation
from the IRA, whether or not they developed a relationship with the government.
60

and halt the encroaching liberal democratic norms therein (Edwards, 2017). Likewise, the
violence perpetrated by the UVF is included in the GTD, as they were non-state actors.
Lastly, I provide an example of a group that is included in the PGMD but
excluded from this study’s database; the Basij Militia of Iran. Originally formed as a
volunteer force in the Iran-Iraq War, the Basij Militia has now become “an entrenched,
and feared, part of the state” (Berger, 2020). It is a paramilitary force headed by the
Iranian Revolutionary Guard Corps (IRGC) responsible for enforcing order, especially on
display during their violent repression of protests following the 2009 Iranian presidential
elections. The point here is that while the group utilizes terrorist violence to maintain the
status-quo, it is too much a part of the state
7
, and so its violence is not realized in the
GTD.
In fact, of the 30 groups that are coded as status-quo terrorist groups, only 11 of
those groups are also found in the PGMD. Therefore, while there is some overlap
between the pro-government militias in the PGMD and the status-quo terrorist groups in
this chapter, they are significantly separate from each other that the latter groups warrant
their own analysis.
Data for my main independent variables are taken from the Dichotomous Coding
of Democracy project (Boix, 2018), the Polity IV Project (Marshall et al., 2017), and the
Varieties of Democracy (VDEM) (Coppedge et al., 2018). Three separate independent
variables are used to test the democracy proposition: Democracy, Polity, and Polyarchy.
The Democracy variable is a dichotomous variable where 1 means that a country is

7
This study does not determine which groups are “too much a part of the state” to no longer be considered non-state
actors. The purpose is to rely on the coding process of the GTD and how they categorize terrorist groups.
61

considered a democratic regime, and 0 means that a country is non-democratic. This
variable is considered to utilize a somewhat minimalist definition of democracy, focusing
on two dimensions - free and fair elections, and suffrage. More specifically, democracies
must meet the following conditions: “(1) The executive is directly or indirectly elected in
popular elections and is responsible either directly to voters or to a legislature. (2) The
legislature (or the executive if elected directly) is chosen in free and fair elections. (3) A
majority of adult men has the right to vote” (Boix, 2018). Using this minimalist definition
affords the ability to test for empirical relationships between democracy and other
variables that are typically lumped into broader characterizations of democracy.
However, to provide robustness checks for the findings, the Polity IV democracy
variable is used in a separate set of empirical tests. The Polity IV democracy variable is
calculated by subtracting the country’s institutionalized autocracy score from the
institutionalized democracy score which originally ranges from -10 to +10. Put together,
the variable used in this analysis measures ideas regarding competitiveness of political
participation, the regulation of participation, the openness and competitiveness of
executive recruitment, and constraints on the chief executive (Marshall et al., 2016). For
simplicity, 10 is added to every score resulting in a range of 0 to 20, with 0 being strongly
autocratic and 20 being strongly democratic. Importantly, this offers the opportunity to
categorize regimes into more than two types to analyze whether the connection between
hybrid regimes and terrorism carries over to status-quo terrorism in particular. In other
words, with this measure of democracy it becomes possible to explore whether status-quo
terrorist violence is associated mostly with “weak” democracies rather than consolidated
democracies.
62

To offer one further robustness check regarding democracy’s relationship with
status-quo terrorism I look to the electoral democracy index created by VDEM. The
Polyarchy variable is also utilized in the ZINB model separate from the other two
measures of democracy. It is an interval level variable ranging from 0-1, with lower
levels signaling lower levels of democracy. The operationalization of the variable is
meant to cover the “core value of making rulers responsive to citizens, achieved through
electoral competition for the electorate’s approval under circumstances when suffrage is
extensive” (Coppedge et al., 2018). The score is an aggregation of ideas relating to
freedom of association, clean elections, freedom of expression, elected officials, and
suffrage. Again, this gives the overall quantitative analysis another measure of
democracy with a slightly different operationalization than the previous two - making any
finding that associates democracy with status-quo terrorism more plausible.
My control variables come from a variety of sources. The control variables
include measures of human rights standards of international partners, state control of
territory, logged GDP per capita, internal conflict, and a moving average of past terrorist
incidents for each state. These are all included as they generally have a strong track
record of helping to explain a broad spectrum of political violence (Fearon & Laitin,
2003; Li, 2005; Kalyvas, 2006; Carey et al., 2015).
The first two control variables come from Greenhill’s replication data on IGO
context and references to human rights protections within free trade agreements
(Greenhill, 2010). The IGO Context variable is an interval level variable which measures
the average Personal Integrity Rights (PIR) index score (Cingranelli & Richards, 2010) of
IGOs that a state is a part of each year. The PIR index considers the degree to which a
63

state violates human rights related to torture, political imprisonment, extrajudicial
killings, and disappearances. The IGO Context variable is constructed in two stages. First,
an average PIR score for each IGO that a country belongs to is calculated, and secondly a
mean score across all IGOs that a country belongs to becomes the final calculation for a
given country-year (Greenhill, 2010). The idea behind this variable is to capture the
degree to which a state is socializing with “good” or “bad” human rights actors on the
international level. The scale of the variable is from 0, signaling socialization with bad
human rights actors, to 8, signaling socialization with good human rights actors.
The other control variable that comes from Greenhill’s data is Soft PTA. To
construct this variable, a content analysis of preferential trade agreement (PTA) treaties
found from the World Trade Organization was conducted to determine whether a PTA
contained references to human rights standards. Soft PTA is a dummy variable where 1
signals that a state belongs to one or more PTAs with human rights standards in a given
country-year, and 0 signals that a state belongs to no PTAs with human rights standards.
There are two reasons as to why these international level variables are included.
First, the socialization effect from participation in the international order should increase
a state’s interest in promoting liberal democratic norms and protecting human rights. As
more elites become familiar with these norms abroad, we should expect those same
norms to diffuse to elites at home. In essence, this should lower the likelihood and/or
amount of members of the state which are willing to endorse a draconian approach to
counterinsurgency, frustrating any remaining hawkish members of the state and
increasing their likelihood to tolerate and/or collaborate with violent status-quo groups.
Second, commitments to the protection of human rights at the international level can
64

reduce the likelihood of official state-sanctioned terror, and increase the levels of status-
quo terror in another way. These commitments increase the costs of human rights abuses
by tying their behavior to explicit punishments written into different international
institutions and/or implicit reputation costs. It follows that to avoid such costs, these
states may want to maintain plausible deniability between the state and any groups
utilizing terror in the name of the status-quo.
The Territorial Control variable comes from the Varieties of Democracy database
(Coppedge et al., 2018), which measures the amount of territory that the state has control
over within its own borders. The control that this variable is measuring is hegemonic, in
that states must be recognized as preeminent and capable of asserting its control over
anti-state groups in these zones. This is an interval variable, with a scale of 1-100, with
higher levels denoting greater levels of state control of territory. The more territory a
state has effective control of, the less anti-state terrorism we should observe, and
therefore there should be less of a threat to the status-quo. Hence, I expect to find a
negative relationship between Territorial Control and status quo terrorism.
GDP Per Capita is another control variable, this time taken from the World Bank and
logged for each state. The idea is that GDP per capita is an indicator for both the economic
development and capabilities of the state. The higher this value is, the less economic grievances
may be present, and the more capable that state may be at addressing those that do come about. I
therefore expect a negative relationship between GDP Per Capita and the number of terrorist
events.
Internal Conflict is a variable taken from the UCDP/PRIO Armed Conflict Dataset
(Gleditsch et al., 2002). It is a dummy variable that is coded one if the state experiences 25 or
65

more battle related deaths in a given year, and those deaths are a result of either (1) internal
conflicts, or (2) internationalized internal conflicts. If any of these conditions are not met, the
variable is coded as zero. The inclusion of this control variable mirrors some of Gurr’s thinking
on the conditions for state terrorism - that the greater the threat posed by a group of challengers
to the state, the more likely we are to observe violence from the state (Gurr, 1986). Given the
threat that civil conflict can pose to the status-quo, and the obstacles that a state faces when it
comes to counterinsurgency decision-making, it is reasonable to expect status-quo terrorist
violence to be most prevalent during periods of civil conflict. Lastly, following the example of Li
(2005), I add a variable called Past Incidents. This variable has a moving average, and is
calculated as the average number of attacks a state has experienced since 1970. The idea here is
to control for the influence of the temporal dependence that future attacks may have on past
attacks.
The data for the dependent variable – SQ attacks – is based on count data. Therefore, the
dependent variable cannot take on a negative value. Such data therefore tend to have a strong
right skew and are characterized by either “a discrete probability distribution such as the Poisson
or negative binomial distribution” (Monogan III, 2015). To distinguish between the two
distributions, it is key to look at the mean and variance of the dependent variable. With SQ
attacks, the variance is greater than the mean, which signals that there is a case of
overdispersion
8
. As many other studies have exhibited, the intuition in such circumstances is to
apply a negative binomial model to this data to control for the inefficiencies, inconsistencies, or
biases that may occur as a result of applying OLS to such count data.

8
A histogram illustrating this variance is provided in Appendix B (Figure B)
66

However, it is also important to note that the data has an excess of zeroes, and that it is
likely that there are different types of zeros within the data. There are certainly zeros that
represent a country-year in which there are violent non-state status-quo groups and they fail to
commit a terrorist attack. There are also quite certainly zeros that represent a country-year in
which there are simply no violent non-state status-quo groups mobilized to commit a terrorist
attack. In other words, this second set of zeros in the data represent country-years in which the
state has a monopoly on the use of status-quo violence, and so any such violence fails to be
coded as a terrorist attack. Therefore, the decision was taken to run a zero-inflated negative
binomial model to be able to account for the overdispersion in the count data as well as two sets
of zeros. Essentially the zero-inflated negative binomial regression is meant to model the excess
zeroes independently with a logit model, and then account for the count data with a negative
binomial model (Bruin, 2006).
Table 4.1 provided below lays out the descriptive statistics for the data. The sample
includes 3,162 total observations. The descriptive statistics illustrate that only about 15% of the
country-years in the sample experience an insurgency. Furthermore, 44.5% of the country-years
involve states that would be considered democratic under Boix’s dichotomous coding of
democracy project.
4.3 Status-Quo Reactions to Liberal Democracy
In this section, results will be presented regarding the different tests done to
ascertain the relationship between democracy and status-quo terrorism. Efforts are made
to make sure the findings are as robust as possible. This includes examining the
relationship between status-quo terrorism and insurgency, as well as using different
67

measures of democracy in order to support the idea that the findings are not dependent
upon the specific coding of democracy used.
Thus, the first part of this section is dedicated towards analyzing status-quo
terrorism’s relationship with insurgency. Recall that the theory is developed around the
idea that there is some assumed threat to the status-quo. Insurgencies and internal
conflicts are generally on the high end of any spectrum of threats to a country’s status-
quo. Typically, the violence is aimed at overturning a particular government or a set of
policies it has enacted. This threat should increase the stimuli required to make the
theorized causal processes behind status-quo terrorism more likely.
To provide support for whether this holds true, I first examine the occurrence of
status-quo attacks within my sample. Figure 4.1 below illustrates the relationship
between insurgency and status-quo terrorist attacks. In my sample of 179 countries which
covers 50 years - from 1970 until 2019 - status-quo attacks were much more likely to
occur in country-years experiencing an insurgency. While there are 1,178 country-years
experiencing an insurgency, and 6,655 which experience no insurgency in the sample,
1,315 status-quo attacks occur during an insurgency compared to only 433 attacks during
periods of no insurgency. So there are roughly 1.12 status-quo attacks per country-year
during insurgency, and only 0.07 per non-insurgency country-year. In other words,
status-quo attacks occur disproportionately during country-years experiencing internal
conflict. This is somewhat intuitive and unsurprising, as many subtypes of violence occur
at higher rates during an internal conflict. However it does provides reassurance that the
logic involved in the theory surrounding the pressures of insurgency and status-quo
terrorism hold true.
68

Table 4.2 and 4.3, which are presented below, illustrate the findings of the statistical
analysis. First, I will discuss the findings of the negative binomial portion of the model (the
count model) in regards to the hypothesized independent variables, including the robustness
check done with both the Polity and the VDEM polyarchy variable. Lastly, I discuss what the
results have to say about the control variables’ relationship with status-quo terrorism.
To start with, the findings from the negative binomial portion (the count portion - Table
4.2) of the ZINB will be discussed. I find support for the democracy proposition in five of the six
models (models 1-4, and 6). Those five models result in the expected positive and highly
significant relationship between democracy and status-quo terrorist attacks. In model 5, the result
is still a positive relationship between democracy and status-quo terrorist attacks; however, the
relationship avoids being a significant one by .006. Therefore, in all models but one, and holding
all other variables constant, I find support for the proposition that democratic states experience
greater levels of status-quo terrorism. On average, between the first two models, democratic
states are expected to experience anywhere between 0.85 and 1.12 more status-quo terrorist
attacks per year than their non-democratic counterparts. Between models three and four, as a
country increases its Polity score by one, it can expect to experience between 0.113 and 0.156
more status-quo attacks per year. According to model six, a state with a one on the polyarchy
score is expected to experience 0.616 more status-quo attacks per year than a state with a zero on
the polyarchy score.
To check the robustness of the findings for the democracy proposition, models three and
four substitute the minimalist definition of democracy for the broader and widely used Polity
score. In this model too, the findings provide support for the Democracy Proposition, with a
positive and significant relationship between a state’s polity score and the count of status-quo
69

attacks in a given year. Since Democracy is a dichotomous variable, and Polity is an interval
level variable, the latter also provides a meaningful way in which to illustrate the predicted
amount of status-quo terrorist attacks within a state at different levels of democracy - holding all
other variables at their average (Figure 4.2 below). As the figure demonstrates, there is no
predicted curvilinear – or inverted U – relationship between regime type and status-quo terrorist
violence - a relationship that has been found in other studies of terrorist violence more
generally. Instead, the model predicts with every move higher on the Polity scale, a state can
expect to experience more status-quo terrorist violence.
Furthermore, Figure 4.3 (below) gives a visual illustration to the trends in status-quo
terrorism over time by three separate regime types: autocracy, anocracy, and democracy.
Autocracies are those states that scored less than -5 on the Polity score, anocracies in between -5
and 5, and democracies are anything higher than 5 on the Polity score. This provides further
evidence that for most of the sample years, democracies have experienced the most status-quo
terrorism; more so than anocracies or “weak democracies”. Interestingly however, beginning in
2013 this trend switches, and anocracies begin to see a significant uptick in such violence. There
could be a variety of reasons for such a shift, and further research on this issue could provide a
fruitful research agenda. For example, it might be that weak democracies have increasingly
adopted those democratic institutions which are specifically linked to status-quo terrorism, while
failing to adopt other institutions which would qualify them as fully democratic.
As for the control variables in the count models, the findings for three of them are as
expected throughout all six models. Soft PTA is statistically significant and positively correlated
with status-quo terrorist attacks in each model. Recall that this is a dummy variable where a
value of 1 means that a state is part of at least one preferential trade agreement with human rights
70

standards, and a score of 0 means that state is not part of any such PTA. According to the
models, states which are part of such PTAs average anywhere from 0.292 to 1.577 more status-
quo terrorist attacks than their counterparts, depending on the democracy variable utilized and
holding all else constant. Territorial Control is also statistically significant throughout all six
models, and in the expected negative relationship. As the state controls less of its territory, it may
be that status-quo groups feel more of a need to take matters into their own hands, increasing
their amount of attacks. Last of these three control variables is Internal Conflict, which is also
found to have its expected positive relationship with status-quo terrorist violence. States
experiencing some type of domestic conflict can expect to see on average from 0.447 to 1.1 more
status-quo terrorist attacks than their counterparts, again depending on the democracy variable
utilized and holding all else constant.
Two of the control variables have mixed results. IGO Context is only significantly and
positively related to status-quo terrorist attacks in one of the three models it is utilized in. This
may mean that as far as international level explanations go for status-quo terrorist violence,
whether a country is involved in a PTA with human rights standards is more important than the
overall level of human rights protections of a state’s IGO partners. GDP Per Capita is negatively
correlated with status-quo attacks in four of the six models, but only statistically significant in
models 1, 3 and 4. This lends some support to an idea like that associated with territorial control
- that as a state’s economic capabilities decrease, certain actors interested in the preservation of
the status-quo feel more compelled to protect it themselves.
The last control variable to mention is Past Incidents. This control variable is positively
associated with status-quo attacks in all six models, but only achieves statistical significance in
models five and six. However, as for the zero-inflated logit models of the ZINB (table 4.3
71

below), which are important for predicting the excess zeros in the data, Past Incidents becomes a
highly significant and negative predictor of excess zeros in all six models. This is a relatively
intuitive finding, where greater levels of past status-quo attacks are associated with a lower
likelihood of zeroes associated with the lack of a status-quo group.
Overall, the findings presented above reinforce the theory developed throughout this
dissertation. The quantitative analysis provides support for the democracy proposition; that
democracies encounter more status-quo terrorism than non-democracies. This finding is robust
and for the most part, does not depend on the type of definition for democracy that is utilized.
There is also significant support for the inclusion of most of the control variables in the analysis
of this subtype of terrorist violence.
A few lessons can be applied from these findings to the theory-building case of status-
quo terrorism in Northern Ireland, introduced at the beginning of this dissertation. To understand
the violence perpetrated by the UFF in Northern Ireland, it should be analyzed as its own sub-
category of terrorism to avoid conceptual stretching. The incentives and motivations for the
UFF’s status-quo terrorism are divergent from other terrorist groups such as the IRA.
Furthermore, the United Kingdom’s democratic tendencies increased the likelihood that the UFF
and other groups like it would utilize such violent strategies. The formal and informal institutions
that restrained the state’s security forces in its fight against the IRA and other republican groups
frustrated hard-liners both within and outside of the state apparatus. Those frustrations
culminated in a push by hardline citizens to take violent matters into their own hands, and a
decision by some hard-liners within the security forces either to look the other way or facilitate
such violence. Thinking in terms of a counterfactual, the findings of this analysis suggest that
72

had the UK been a non-democracy, there would have been less status-quo terrorism; perhaps
because the state would have been more prone to utilize repressive acts itself.
Previous literature on terrorism has left the distinction between different types of terrorist
groups understudied. Just as we have come to expect that transnational and domestic terrorism
are different phenomena, and therefore require different variables to predict them, status-quo and
anti-state terrorism are also distinct, and therefore need the same academic scrutiny. As can be
seen from the analysis above, there are indeed unique characteristics of status-quo terrorism. For
example, instead of the inverted U relationship found between democracy and terrorism more
generally, I find that democracy has a consistently positive association with status-quo terrorism.
In other words, status-quo terrorism becomes more probable even as a country moves from a
weak democracy to a strong democracy.
4.4 Summary
This chapter has sought to take a first step into the examination of status-quo terrorism
and its determinants. If democracy has had mixed results regarding its effects on terrorism in
general, how do the findings hold when the phenomenon of terrorism is further broken down into
its subtypes? This study demonstrates that democracy does indeed increase the likelihood of
status-quo terrorist attacks. The chapter has also provided evidence that the association between
democracy and status-quo terrorist violence is not a curvilinear relationship as has been recently
found in the literature on regime type and terrorism more generally. These are important findings
to build upon, as states from around the world continue to experience internal conflict and seek
to find the best resolutions to these issues. While there is certainly no suggestion for abandoning
democracy or its institutions, it is nonetheless helpful to know the potential negative
consequences in order to resolve such problems. Lastly, this analysis is part of an advancement
73

that this dissertation seeks to address in the field of terrorism studies for its serious consideration
of the problem of conceptual stretching (Sartori, 1970) prevalent in how many of us think about
terrorist violence.
Yet, further work must be done to distinguish between different types of terrorist groups.
Unfortunately, the different datasets on terrorist violence do not seem to be adequate to
completely capture all status-quo terrorist incidents; of the over 100,000 domestic terrorist events
that have taken place over the past 47 years, over 50,000 of those events have been perpetrated
by “unknown” groups in the GTD. As the coding procedure for determining status-quo terrorism
in this paper relies on an actor-centric model, the data are obviously very troubling. To help
correct for some of these issues, a suggestion for future research on this topic is for greater
qualitative research on specific cases of status-quo terrorism.
Chapter 5 of this dissertation seeks to do just that, by conducting an in-depth case study
of status-quo terrorist violence overtime in the region of Urabá, in Colombia. To complement
key findings of this chapter, distinguishing status-quo terrorist violence from the broader
phenomenon, Chapter 5 focuses on testing for and discovering the unique causal mechanisms
which lead to such violence. To do so, the comparative advantages of process tracing and
comparative case studies are utilized. More specifically, two time periods with varying levels of
status-quo terrorism in Urabá are compared. Due to the specific actors under analysis (the
military and regional elites), both the national and regional circumstances are traced to better
understand the key stimuli which account for the region’s violent experience with paramilitary
groups.

74

Tables and Figures
TABLE 4.1
Descriptive Statistics
Statistic N Mean
St.
Dev.
Min Pctl(25) Pctl(75) Max
Democracy 7,360 0.445 0.497 0.00 0.000 1.000 1.000
Polity 7,281 11.34 7.450 0.00 3.000 19.000 20.000
IGO_Context 3,162 4.903 0.470 3.67 4.570 5.187 6.267
softpta 4,076 0.200 0.400 0.00 0.000 0.000 1.000
territorial_control 7,648 91.40 10.70 33.7 87.333 99.000
100.00
0
loggdp 3,647 7.422 1.596 4.03 6.039 8.600 10.830
past_incidents 7,906 0.148 0.881 0.00 0.000 0.000 11.174
Insurgency 7,906 0.150 0.357 0.00 0.000 0.000 1.000

75

76

Table 4.2
Negative-Binomial Count Model

Dependent variable: Count Model
SQ_attacks
(1) (2) (3) (4) (5) (6)
Democracy 1.120
***
0.851
**

(0.406) (0.355)
Polity 0.156
***
0.113
***

(0.040) (0.033)
Polyarchy 0.407 0.616
***

(0.252) (0.236)
IGO Context 1.371
**
0.716 -0.226
(0.601) (0.626) (0.150)
Soft PTA 1.085
**
1.135
***
1.554
***
1.577
***
0.327
***
0.292
***

(0.485) (0.396) (0.496) (0.406) (0.123) (0.097)
Territorial Control -0.044
**
-0.025
*
-0.039
**
-0.039
**
-0.034
***
-0.034
***

(0.018) (0.015) (0.017) (0.015) (0.004) (0.003)
GDP Per Capita -0.512
***
-0.097 -0.507
***
-0.235
**
0.029 -0.010
(0.186) (0.096) (0.184) (0.103) (0.051) (0.036)
Past Incidents 0.060 0.100 0.039 0.067 0.038
***
0.032
***

(0.069) (0.062) (0.066) (0.060) (0.013) (0.012)
Internal Conflict 0.809
**
1.100
***
1.089
***
1.088
***
0.447
***
0.655
***

(0.325) (0.293) (0.328) (0.297) (0.102) (0.089)
Constant 0.422 2.079 1.351 3.296
**
5.031
***
3.971
***

(2.060) (1.427) (1.942) (1.414) (0.486) (0.303)
Observations 2,814 3,565 2,641 3,337 2,812 3,561
Log Likelihood -511.937 -648.991 -473.514 -575.335 -1,050.255 -1,264.376
77

Note:
*
p<0.1;
**
p<0.05;
***
p<0.01

78

79

80

Table 4.3
Zero Inflated Model
Dependent variable: Zero Inflated Model
SQ_attacks
(1) (2) (3) (4) (5) (6)
Democracy -0.819 -0.320
(0.521) (0.472)
Polity -0.082 -0.056
(0.052) (0.044)
Polyarchy -1.559
***
-0.482
(0.569) (0.528)
IGO Context 1.863
***
1.588
**
1.257
***

(0.700) (0.710) (0.331)
Soft PTA -0.214 0.083 0.312 0.724 -0.351 -0.354
(0.559) (0.462) (0.649) (0.589) (0.287) (0.234)
Territorial Control -0.023 0.009 -0.032 -0.018 -0.049
***
-0.033
***

(0.021) (0.016) (0.025) (0.023) (0.014) (0.012)
GDP Per Capita -0.202 0.189 -0.115 0.120 -0.005 0.114
(0.224) (0.165) (0.237) (0.180) (0.111) (0.091)
Past Incidents -24.600
***
-27.663
***
-23.301
***
-22.492
***
-0.616
***
-0.615
***

(4.873) (5.330) (4.572) (4.318) (0.064) (0.056)
Internal Conflict -2.576
***
-1.711
***
-2.383
***
-2.171
***
-2.398
***
-2.116
***

(0.517) (0.445) (0.531) (0.487) (0.260) (0.229)
Constant -0.037 2.426 1.885 6.095
***
3.727
**
6.787
***

(2.982) (1.638) (3.164) (2.128) (1.613) (1.096)
Observations 2,814 3,565 2,641 3,337 2,812 3,561
Log Likelihood -511.937 -648.991 -473.514 -575.335 -1,050.25 -1,264.37
81

Note:
*
p<0.1;
**
p<0.05;
***
p<0.01

82

Appendix A
Coding Procedure
Each group that was listed by the Global Terrorism Database (GTD) with 10 terrorist
attacks or more was hand coded as status-quo or not based on the following conditions. First,
each group was coded as status-quo if it was included in the Pro Government Militia Database
(Carey, Mitchell & Lowe 2013), and showed up as a violent non-state group in the GTD.
Second, for those groups that were not found in the PGMD, secondary search was conducted to
determine whether the groups had either (1) explicit relationships with the government, military,
or police forces, (2) an implicit relationship with the government, military, or police forces, or
(3) a violent strategy focused on anti-state groups, or their communities, as a means to support
the status quo. The point of emphasis here is that groups are coded as status-quo based on their
explicit or implicit goals. A relationship with the government is a sufficient condition, but not a
necessary one. The necessary precondition is the group’s motivation to maintain the status quo
violently from those seeking change.

83

A List of Groups Coded as Pro-State

Awami League, Bangladesh Nationalist Party (BNP), Death Squad, Harkatul Jihad-e-Islami,
Inkatha Freedom Party (IFP), Institutional Revolution Party (PRI), Israeli extremists, Israeli
Settlers, Jamaat-E-Islami (Bangladesh), Janjaweed, Jewish Extremists, Ku Klux Klan, Lord's
Resistance Army (LRA), Loyalist Volunteer Forces (LVF), Loyalists, Muttahida Qami
Movement (MQM), Paramilitaries, Peasant Self-Defense Group (ACCU), Protestant Extremists,
Ranbir Sena, Red Hand Defenders (RHD), Right-Wing Paramilitaries, Secret Anti-Communist
Army (ESA), Seleka, Ulster Freedom Fighters (UFF), Ulster Volunteer Force (UFF), Union
Guerrera Blanca (UGB), United Self Defense Units of Colombia (AUC), Vigilante Groups

84

1. Awami League (After 2013, yes)

Ruling party in Bangladesh after the election in 2013.

https://www.nytimes.com/2016/05/20/opinion/the-real-source-of-terror-in-bangladesh.html

2. Bangladesh Nationalist Party (BNP) (Prior to 2013, yes)

After 2013, they do not seem to be acting with the support of the government (Awami League).
They actually protested the elections, and are targets of the government afterwards.

https://www.nytimes.com/2016/05/20/opinion/the-real-source-of-terror-in-bangladesh.html

3. Death Squad

Colombia:

Right-wing paramilitary groups that conspired with military, police, and government officials to
fight against the left-wing rebel groups.

https://www.nytimes.com/2016/09/11/world/americas/colombia-cocaine-human-rights.html

El Salvador:
85

Right-wing paramilitary groups that conspired with military, police, and government officials to
fight against the left-wing rebel groups. Some squads still colluding with police today.

https://www.businessinsider.com/el-salvador-death-squad-international-2017-9?r=UK&IR=T

Guatemala:

Right-wing paramilitary groups that conspired with military, police, and government officials to
fight against the left-wing rebel groups.

https://nsarchive2.gwu.edu/NSAEBB/NSAEBB15/

4. Harkatul Jihad-e-Islami

Ties to Pakistani Army, helped with plans to overthrow Benazir government, also helps with
attacks on Kashmir. Discreetly released from prison by authorities.

https://www.sipri.org/sites/default/files/files/PP/SIPRIPP11.pdf

5. Inkatha Freedom Party (IFP)

86

Violence perpetrated by the IPF intensified as apartheid was coming to an end, and multiracial
elections were becoming more and more a reality. As the elections became closer, the
government and the IFP attempted to create chaos and tarnish the ANC.

https://www.telegraph.co.uk/news/worldnews/africaandindianocean/southafrica/1368739/ANC-
was-target-of-terror-campaign.html

6. Institutional Revolutionary Party (PRI)

They were the ruling party of Mexico from the time it was considered a democracy all the way
until 2000. Therefore, any terrorist incidents taking place during this time, under their group
name, must be pro-state.

7. Israeli extremists

These are individuals within Israel who attack Palestinians in Israel, and the West Bank/Gaza
which is considered part of Israel in this study. Israeli extremists are considered pro-state as they
direct their attacks towards what they see as the enemy of the Jewish state, and their actions are
for the maintenance, or expansion of the status quo.

8. Israeli Settlers

87

Much the same as the individuals above. The difference is that these individuals are specifically
using violence as a means of acquiring or defending land in traditionally Palestinian land.

9. Jamaat-E-Islami (Bangladesh) (Yes, sometimes)

From 1991-1996 and from 2001-2006, JeI was part of the ruling coalition in Bangladesh. There
are some attacks by them during this time, which are considered status-quo. However, the
majority of their attacks come in 2013 or later, at which time they are no longer part of the ruling
government, and fighting for the status quo.

10. Janjaweed

A government supported militia in Sudan. Plenty of evidence of collaboration during the Darfur
genocide.

https://www.hrw.org/legacy/features/darfur/fiveyearson/report4.html

11. Jewish Extremists (Yes, in Israel)

These are individuals within Israel who attack Palestinians in Israel, and the West Bank/Gaza
which is considered part of Israel in this study. Israeli extremists are considered pro-state as they
direct their attacks towards what they see as the enemy of the Jewish state, and their actions are
88

for the maintenance, or expansion of the status quo. However, there are 4 attacks that happen
within the United States, and these are are unable to be coded as pro-state

12. Ku Klux Klan

“Within a year of initiating the program, the FBI estimate ‘about 15% of entire Klan was
comprised of informants’ and that ‘about half were elected to leadership positions.’ At one point,
the Klan was so heavily infiltrated that ‘Hoover briefly considered installing an informant at the
top of the Klan, and thus making Klan policy.’ Pg 140-1

“In contrast to earlier periods, during the 1950s and 1960s, both national political parties openly
shunned the Klan and there is no evidence to suggest that the federal government provided the
Klan with either information or material. An entirely different story emerges when we consider
the support of local law enforcement organizations. Wherever police either turned a blind eye or
offered a willing hand, coordinated Klan terrorism surged. Wherever it was actively opposed, the
Klan was either destroyed or forced to decentralize in order to survive.” Pg 154

Support in the 70’s and 80’s for the Klan was at the local government / police level, not at the
national level as it was in the 20’s.

http://www.dtic.mil/dtic/tr/fulltext/u2/a551880.pdf

89

13. Lord’s Resistance Army (LRA) (in Sudan Yes, in other countries no)

The LRA started as an insurgent group in Uganda.

“Uganda and Sudan were at a state of low-level, undeclared war: Uganda was supporting rebels
groups in southern Sudan who fought that country’s Arab, north-based government. Sudan
responded by adopting the LRA as its own, offering them safe havens in Sudan and truckloads of
Kalashnikov rifles.”

“The Sudanese government asked Kony to spread chaos in Uganda as well as in southern
Sudan.”

“A second Iron Fist campaign in 2004 pushed much of the LRA into Sudan and the neighboring
Democratic Republic of Congo but did little else. Kony was later pushed even further west in the
Central African Republic.”

https://www.theatlantic.com/international/archive/2011/10/the-bizarre-and-horrifying-story-of-
the-lords-resistance-army/246836/

The other countries that the LRA uses violence in are anti-LRA

https://www.hrw.org/news/2012/03/21/qa-joseph-kony-and-lords-resistance-army#9

90

14. Loyalist Volunteer Forces (LVF)

“Loyalists want Northern Ireland to remain part of the United Kingdom, and many are willing to
support the use of violence to keep the Protestant-majority province, also known as Ulster, under
British rule. Young Protestant men from Ulster’s most downtrodden neighborhoods make up the
core membership of loyalist paramilitary groups, which are effectively pro-state terrorist
organizations.”

https://www.cfr.org/backgrounder/northern-ireland-loyalist-paramilitaries-uk-extremists

15. Loyalists

“Loyalists want Northern Ireland to remain part of the United Kingdom, and many are willing to
support the use of violence to keep the Protestant-majority province, also known as Ulster, under
British rule. Young Protestant men from Ulster’s most downtrodden neighborhoods make up the
core membership of loyalist paramilitary groups, which are effectively pro-state terrorist
organizations.”

https://www.cfr.org/backgrounder/northern-ireland-loyalist-paramilitaries-uk-extremists

16. Paramilitaries

91

All of the attacks perpetrated by “paramilitaries” in the GTD happened in Colombia, from 1997-
2001. At this time, paramilitaries were primarily those groups that were formed to defend
against, or attack guerilla groups throughout Colombia, later merging into AUC.

https://www.jstor.org/stable/24590184?seq=1#page_scan_tab_contents

17. Peasant Self-Defense Group (ACCU)

The ACCU is the most famous/public paramilitary in the bigger grouping of Colombian
paramilitaries known as the United Self Defense Units of Colombia (AUC). These groups act
frequently in coordination with the Colombian security forces.

https://www.hrw.org/legacy/reports98/colombia/Colom989-04.htm

18. Protestant Extremists

“Loyalists want Northern Ireland to remain part of the United Kingdom, and many are willing to
support the use of violence to keep the Protestant-majority province, also known as Ulster, under
British rule. Young Protestant men from Ulster’s most downtrodden neighborhoods make up the
core membership of loyalist paramilitary groups, which are effectively pro-state terrorist
organizations.”

92

All of the protestant extremist attacks take place in the United Kingdom and from the data seem
to be related to the Northern Ireland conflict. Therefore, it is more than likely that these
individuals are Loyalists, acting for the protection of protestant superiority in Northern Ireland.

https://www.cfr.org/backgrounder/northern-ireland-loyalist-paramilitaries-uk-extremists

19. Ranbir Sena

Ranbir Sena is a “state-backed private army” operating in the Bihar state of India, with the
support of the political and economic elites of the region.

https://www.jstor.org/stable/pdf/23251760.pdf?seq=1#page_scan_tab_contents

20. Red Hand Defenders (RHD)

“Loyalists want Northern Ireland to remain part of the United Kingdom, and many are willing to
support the use of violence to keep the Protestant-majority province, also known as Ulster, under
British rule. Young Protestant men from Ulster’s most downtrodden neighborhoods make up the
core membership of loyalist paramilitary groups, which are effectively pro-state terrorist
organizations.”

https://www.cfr.org/backgrounder/northern-ireland-loyalist-paramilitaries-uk-extremists

93

21. Right-Wing Paramilitaries

All of the attacks perpetrated by “paramilitaries” in the GTD happened in Colombia, from 1997-
2001. At this time, paramilitaries were primarily those groups that were formed to defend
against, or attack guerilla groups throughout Colombia, later merging into AUC.

https://www.jstor.org/stable/24590184?seq=1#page_scan_tab_contents

22. Secret Anti-Communist Army (ESA)

The ESA was a right wing terrorist group in Guatemala during the early 1980s. They targeted
opposition leaders, union leaders, academics, and guerilla groups throughout Guatemala. Their
assassinations served as an extra counter terrorism element for the government…”the
Guatemalan military has developed a corporate tradition of independent conservatism, supported
by an equally conservative private sector that is disposed to assist in the repression of popular
movements in order to retain its monopoly over the benefits of economic growth.” (Premo 1981).

https://www-jstor-org.libproxy1.usc.edu/stable/165453?seq=1#metadata_info_tab_contents

23. Anti-Balaka Militia

94

The Anti-Balaka are village defense groups that formed after the coup in the Central African
Republic, set up by President Bozize, and backed by the National Army. They were used at first
to push back the Seleka forces, and after retaking the capital, they continued their violent
campaigns in the North and East of the country.

https://institute.global/insight/co-existence/what-seleka

24. Ulster Freedom Fighters (UFF)

“Loyalists want Northern Ireland to remain part of the United Kingdom, and many are willing to
support the use of violence to keep the Protestant-majority province, also known as Ulster, under
British rule. Young Protestant men from Ulster’s most downtrodden neighborhoods make up the
core membership of loyalist paramilitary groups, which are effectively pro-state terrorist
organizations.”

https://www.cfr.org/backgrounder/northern-ireland-loyalist-paramilitaries-uk-extremists

25. Ulster Volunteer Force (UFF)

“Loyalists want Northern Ireland to remain part of the United Kingdom, and many are willing to
support the use of violence to keep the Protestant-majority province, also known as Ulster, under
British rule. Young Protestant men from Ulster’s most downtrodden neighborhoods make up the
95

core membership of loyalist paramilitary groups, which are effectively pro-state terrorist
organizations.”

https://www.cfr.org/backgrounder/northern-ireland-loyalist-paramilitaries-uk-extremists

26. Union Guerrera Blanca (UGB)

“Taking their cue from the highly successful activities of the extreme right in Guatemala, a group
of conservative Salvadorans had formed a right-wing terrorist organization known as the White
Warrior Union (Union Guerrera Blanca, UGB), to assassinate those believed associated with the
left-wing guerrilla groups or the popular movements. Like its Guatemalan counterparts, the UGB
was rumored to work closely with the security forces, especially with the Policia Nacional, and
with Col. Rene Chacon, the chief of army intelligence under Molina.” (Anderson 1988)

https://books.google.com/books?id=uar_iWVhSaIC&pg=PA81&lpg=PA81&dq=Union+Guerrer
a+Blanca+(UGB)&source=bl&ots=FHtYNOXsKX&sig=bqdEXs5CVjtqGS19uJVcQRGR3zw&
hl=en&sa=X&ved=2ahUKEwj-
kJuUr8PdAhVR6Z8KHTAyA1cQ6AEwB3oECAcQAQ#v=onepage&q=Union%20Guerrera%2
0Blanca%20(UGB)&f=false

27. United Self Defense Units of Colombia (AUC)

96

The AUC was a grouping of several right-wing paramilitary groups throughout Colombia. These
groups acted frequently in coordination with the Colombian security forces.

https://www.hrw.org/legacy/reports98/colombia/Colom989-04.htm

28. Vigilante Groups

All attacks attributed to “vigilante groups” are for incidents in Sri Lanka in 1989. During this
time, reporting of government linked human rights abuses were attributed to vigilante groups, or
death squads.

https://www.hrw.org/reports/1990/WR90/ASIA.BOU-11.htm

97

Appendix B
Histogram
Table B

98

Chapter 5
Defending the Status-Quo in Urabá:
Reactions to the Democratic Gains of the Left
5.1 Introduction
The purpose of this chapter is to explore the causal mechanisms at play within a specific
case study of status-quo terrorist violence. To do so, it is important to review the hypothesized
mechanisms through which democracy leads to higher levels of such violence. Recall from the
theory chapter that in order to understand status-quo terrorism, one must analyze hard-liners both
within and separate from the state apparatus. Both sets of actors are motivated by national and
regional circumstances, and therefore it is crucial to trace the causal processes between such
circumstances to the decisions to commit status-quo terrorism.
National policies are those that will be most telling in whether and how the military is
restrained in their counterinsurgency; government initiated cease-fires and peace negotiations,
statutes of emergency, the emphasis on human rights, and the juridical powers of the military all
affect the ability of the armed forces to fight the guerrillas as they see fit. All other things being
equal, the military will be more likely to prefer a more draconian approach to counterinsurgency,
relative to other sectors of the government. On the national spectrum, it is also important to look
at the way in which regional elites view the consequences of peace negotiations with the
guerrillas - specifically the introduction of new leftist political parties that are synonymous with
active rebels or insurgents, and the institutionalization of their priorities.
On the regional spectrum, one must take into account the interests of the local elites, the
demands/tactics of the guerrillas there, and the relative strength of a community sympathetic to
99

them. These regional variables will determine the threat that local elites perceive from
insurgents, and consequently the methods they are willing to use to counter said threat.
The remainder of this chapter will unfold in the following fashion. First, there will be a
brief review of the case selection process. This includes a discussion of the specific type of case
study that is conducted, explaining the use of temporal variation and the merits of process
tracing, and lastly how this qualitative chapter fills in gaps of a purely quantitative project to
make for a more complete study of status-quo terrorism. Second, the first time period under
consideration is analyzed, tracing the theorized independent variables in the Urabá region of
Colombia during an era where status-quo terrorist violence is relatively low. Next, the second
period of time in Urabá is examined, where status-quo terrorism is high, including a thorough
tracing of the same variables from the first time period and a consideration of other potentially
confounding variables which may account for the variation in the dependent variable. Lastly, the
chapter’s conclusion discusses the process through which status-quo terrorist violence is
affected, summarizing the findings of the comparison between the two cases.
5.2 Case Study Design
At the heart of every qualitative study lies the important set of questions surrounding
research design, most notably, case selection. This chapter utilizes two separate research designs:
conducting both a comparative case study and process tracing. The primary purpose of the case
study is to trace the process by which democracy influences the severity of status-quo terrorist
violence, specifically by locating the precise causal mechanisms which go from the structural
level down to the group or individual level. Secondarily, this case study seeks to provide a
further source of hypothesis testing for the theoretical framework built in chapter 3. To do so, I
leverage the causal inference from a longitudinal comparative case study. To accomplish these
100

research objectives, I have chosen to focus on the case of Colombia, and more specifically on
two time periods of status-quo terrorism within a northern region of Colombia called Urabá.
Colombia is a country located in the north of South America, bordering Panama,
Venezuela, Ecuador, Peru, and Brazil. It is a democratic country with a presidential system of
government somewhat similar to that of the United States. It is a highly urbanized society with
over 81% of its citizens living in urban areas. It is also plagued by high economic inequality,
especially pronounced when comparing the urban areas with the rural areas, registering a Gini
index coefficient of 53.5 in 2014 (CIA factbook). Its economy is highly dependent on primary
commodities, such as oil, coal, coffee, cut flowers, and bananas (particularly important in the
region of Urabá). The country has also experienced what some consider the longest ongoing
insurgency, which can be traced back to the end of Colombia’s civil war known as La Violencia.
Urabá is a region in the northwest of the country, which borders Panama and gives Colombia
access to both the Pacific and Atlantic oceans. Importantly, Urabá was one of the main regions in
Colombia where the phenomenon of status-quo terrorism really took hold. Colombia, and the
region of Urabá within it, was picked as the focal point of this chapter for several reasons. As
mentioned, the primary purpose of this case study is to specify the causal mechanisms through
which democracy influences the levels of status-quo terrorism. An in-depth analysis of this case
will offer the opportunity to “peer into the box of causality to locate the intermediate factors
lying between some structural cause and its purported effect” (Gerring, 2007). In order to trace
the process through which this occurs it is necessary for the case under analysis to shed light on
the hypothesized variables from the theoretical framework under consideration. In this sense,
Colombia was selected as a case where the variables of interest - specifically levels of
democracy (IV) and amount of status-quo terrorist attacks (DV) - vary through time and reach
101

high/extreme values. It is in these cases where the “causal mechanisms are starkly evident,” and
the method of process tracing may be especially fruitful (George and Bennett, 2005).
For example, in the middle of the Colombian insurgency a new constitution was put in
place that increased that state’s level of democracy on a range of measures, effectively
liberalizing the country in several aspects of concern for this study. Furthermore, Colombia as a
whole, and Urabá as a region experience relatively high levels of status-quo terrorism, although
the rate varies through time. It is precisely in a case such as this that any causal mechanisms
between democracy and status-quo terrorist violence would be starkly evident as per the wording
above.
Moreover, this chapter’s process tracing takes place within a comparative case study
utilizing temporal variation in order to provide further testing for the hypothesis developed in the
theory section. Rather than selecting two cases with spatial variation, I decided comparing the
same region over time would be the most appropriate method for controlling for as many
confounding variables as possible. In essence the design of the study follows the logic behind
John Stuart Mill’s method of difference (1843) and Przeworski and Teune’s most similar
system’s design (1970), where the comparison cases vary on their values of the dependent
variable, and are similar in regards to as many explanatory variables as possible except for the
main independent variable of concern. The basic logic behind the design is that by identifying
two cases that vary on the dependent variable, the researcher is able to eliminate from
consideration those alternative explanatory variables that do not covary with it. Much of the
value that comes from the “before-after'' comparative case study design is that when studying
one region over time, “political culture, political structure, history, rivalries, historical lessons,
etc. change very slowly if at all,” and therefore the goal of matching cases on confounding
102

variables is much easier to approximate than other cross-case designs (Levy, 2008). Thus,
selecting the region of Urabá, Colombia during two time periods (1982-1993 & 1994-2003)
structures the comparison in a way that strengthens the ability to make a causal inference.
Regardless of the design, within the social sciences it is hard to imagine two cases that
perfectly match each other on all potential confounding variables but one. Fortunately, process
tracing has a comparative advantage when it comes to keeping track of the potential confounding
variables that are not perfectly matched between cases. Combining process tracing with the
comparative case study “can help establish whether the variables of interest were causal and
whether the other variables that changed in the same period were not, or at least that they do not
account for all of the change in the outcome” (George & Bennett, 2004). To utilize such
advantages from the combination of the two methods, it is beneficial to provide a clear and
concise index of the proposed causal mechanisms between democracy and status-quo terrorism,
as well as alternative factors that may affect the dependent variable.
The dependent variable – status-quo terrorist attacks – is put together from a group of
datasets created by the Grupo de Memoria Histórica (GMH)
9
. The idea behind utilizing this data
is that it has been collected by sources within the country, and is much more fine-grained than
the data in the GTD. One key asset is that the attacks are coded by the municipality in which they
occurred and so I am able to identify all the attacks that took place during my case study within
Urabá. The data that comes from the GMH comes from a mix of human rights reports and from
news reports. Fortunately, one of the datasets from the GMH includes terrorist attacks – however
the coding utilized for terrorist attacks is quite exclusive. Therefore, in forming the dependent
variable for this chapter, I also include data collected on massacres, kidnappings, selective

9
This group was created by the Colombian government to help contribute to the remembering the events and
atrocities of the Colombian armed conflict.
103

assassinations, and attacks on property so as to make the dependent variable as identical as
possible to the one in Chapter 4.
Table 5.1 below provides a brief guide to the variables that are traced throughout the case
study. As can be seen from the table, this case study will be examining three potential causal
processes through which lead to status-quo terrorism. Two of these are from the theory chapter,
which focus on democracy’s effect on status-quo terrorism, the other is an alternative factor
which also may explain such violence. The focus here is on how the main explanatory variables
(democracy for example) affect the dependent variable through some intermediate factor. In
essence, the purpose of these case studies is to “peer into the box of causality…between some
structural cause and its purported effect” (Gerring, 2007). Thus, each case will analyze how such
important broad level variables, such as democracy or insurgent strength, actually affect the
decision-making processes of the important actors regarding status-quo terrorist violence.
5.3 Comparing the Cases
In this section, the national and regional circumstances surrounding Colombia’s status-
quo terrorist violence are considered in depth. First, a very brief overview of the insurgency that
the Colombian state was facing is presented – specifically, who the main actors were that rose up
against the government, and what they were fighting for. The next section traces the first case for
comparison, focusing specifically on the time period between 1982-1993. This case is an
instance in which levels of democracy are relatively low and status-quo terrorist violence is too.
The third section provides a process tracing for the second comparison case: the region of Urabá
during the years of 1994-2003. These analyses include context from both the national and the
regional level, and include data from semi-structured interviews of individuals from Colombia.
On the national side of the equation, special attention is placed on policies regarding peace
104

negotiations, the constitutional assembly, and the institutional transformation of civil-military
relations. In regards to the regional circumstances, the focus of the comparative case studies will
be on the regional elites in Urabá, the strength and tactics of the insurgents, and developments at
the regional political level.
5.4 A Rising Against the State (1960-1981)
The world’s longest ongoing insurgency started in the early 1960s in Colombia not long
after that country had finished its civil war called “La Violencia.” What became the most
powerful guerrilla group, known as the Fuerzas Armadas Revolucionarios de Colombia (FARC),
formed out of the communist peasant-farmer self-defense groups that were left out of the bi-party
reconciliation plan (the National Front) that ended the civil war. On the political margins of
Colombia’s political spectrum, the peasant-farmer self-defense groups felt discriminated against
by the National Front’s two-party system, and saw this as a justification to opt for an armed
struggle (GMH, 2016). In response to the perceived violent threat from the “independent
republics” of peasant self-defense groups, Colombia implemented Plan Lazo as a way of giving
much greater autonomy to the military in eliminating the subversive threat and ensuring public
order. The guerrillas reacted to this more aggressive approach on their part by expanding their
national outreach and officially becoming the FARC in 1964. The smaller guerrilla groups who
fought an insurgency against the Colombian state also formed around this time, including the
National Liberation Army (Ejército de Liberación Nacional – ELN) and the Popular Liberation
Army (Ejército Popular de Liberación – EPL). The EPL and FARC were the insurgent groups
strongest in the region of Urabá, the specific area under consideration in this chapter. While there
were differences between the groups, their goals were to gradually take over territorial control of
Colombia, overcome the central government, and implement their leftist policies (CISAC, 2019).
105

The FARC remained relatively small, and by one account only reached 1,000 combatants
around 1978 (GMH, 2016). The EPL and the ELN followed behind the FARC in numbers at this
time as well. However, Colombia is sometimes referred to as “two countries” with a highly
urbanized and relatively developed center surrounded by an underdeveloped and unequal rural
periphery where the guerrillas are easily able to find safe-haven and support for their ideology
(Davis, 2015). This combined with the country’s topography and rugged terrain has led to an
uneasy territorial stalemate between insurgents and the state, despite the relatively low numbers
of the former. Most importantly, the country was faced with an insurgency capable of inflicting
damage and interested in overturning the status-quo on both a regional and national scale.
Leading up to the election of President Belisario Betancur in 1982 the armed forces,
faced with ensuring public safety, had steadily increased their power and autonomy to fight
against the insurgency in the ways they saw fit. With the election of President Turban Ayala
(1978-82) the military found a way to further extend their authority. The greatest example of this
is the National Security Statute (Estatuto de Seguridad Nacional), which was passed as a way to
give the army new judicial powers and further restrict citizen rights (Carroll, 2011). More
specifically, this new statute “raised the sentences for the crimes of kidnapping, extortion and
armed attacks; it allowed military courts to judge civilians and extended the category of
‘subversion’ to extremist propaganda inciting riots and civil disobedience” (GMH, 2016). This
autonomy from the civilian government in how to fight the insurgency led to higher levels of
human rights abuses from an already draconian approach to the counterinsurgency.
5.5 A Mixed Approach to Counterinsurgency and Status-Quo Terrorism in Urabá
(1982-1993)
106

However, in 1982 the Betancur administration came into power with a more inclusive
and democratic approach to the counterinsurgency in mind. His administration immediately
created a Peace Commission which included members from a wide range of society - including
those representing the insurgency. Betancur took the approach of negotiating with different
insurgent groups on their own terms, rather than insisting that they negotiate together as a whole
under the same terms. The most prominent of the negotiations took place with the largest of the
insurgent groups at the time - the FARC. The talks between the government and the FARC
initiated with promises from the latter to cease their use of kidnapping and extortion. In return,
the FARC insisted that the government commit to “modernise the political institutions, enable
agrarian reform, facilitate the mobilisation of campesinos and indigenous groups, [and]
strengthen education, health, housing and labour policy” (Durán, 2004). Especially in regard to
the rural areas throughout the country, many of these commitments by the government were
promises to alter the socio-economic status-quo dominated by wealthy landowners.
These talks eventually lead to the Uribe Accord in April of 1984. The agreement
established a cease-fire between the two sides and the establishment of a new political party
associated with the FARC called the Patriotic Union (UP). Importantly, the UP would be capable
of representing the far left of the political spectrum that had previously been absent in the two-
party system of the National Front put in place to end La Violencia. Taking advantage of this
push for democratization, the FARC then dedicated itself to boosting “its legislative influence
and potential alliances with urban workers, middle class and ‘nationalist business’ sectors”
(Durán, 2004). As another part of the agreement, the government promised to protect the
fledgling political party.
107

However, even with the formation of this new political party, the FARC would not
commit to disarmament until further progress was made in the peace negotiations. In actuality,
Jacobo Arenas - the political commander of the FARC - called for an offensive plan to build an
army of 28,000 troops capable of surrounding the country’s major cities by 1990. This
concurrent strategy of building its military capabilities while also utilizing the new political
avenues opened up through the accord. was known as la combinacion de todos las formas de
lucha - the combination of all the forms of struggle (Dudley, 2004). While the government hoped
that the formation of the UP would bring the FARC into the mainstream and lead to its
abandonment of violence, FARC’s leadership found the new political party to be a
complimentary tool in its violent and revolutionary goals.
Besides its dealings specifically with the FARC, the Betancur administration reached a
separate ceasefire agreement with the EPL in August 1984, but the ceasefire between the two
sides lasted only a little over a year. The decision to negotiate with the different insurgent groups
mentioned above coincided with Betancur’s repealing of the National Security Statute - an effort
to limit the independence of the military and give the civilian government a larger role in
handling public security (GMH, 2016). Lastly, in 1985. the government “amended the
constitution to allow for the direct election of municipal mayors (alcaldes),” a political position
previously appointed by the governor of each department (Steele and Schubiger, 2018). These
new policies demonstrated a significant turn in counterinsurgency strategy towards an emphasis
on peaceful resolution, utilizing democratic institutions - such as political parties and electoral
reforms - as a unique tool to address the grievances of the insurgents.
Yet, in its attempt to solve the problems of one sector of society, the state irritated its
more traditional allies inside the military and those regional elites that were most threatened by
108

the violence and aims of the guerrillas. Frustrated with the initial shift towards a more
democratic approach with the guerrillas signaled by President Betancur’s promise to negotiate
with the rebels, then-Defense Minister Fernando Landazabal set in motion the beginning of
Colombia’s modern paramilitary and status-quo terrorism problem. Betancur’s policies were
seen to create a “struggle over autonomy-subordination between the presidency and armed
forces” (Romero, 2003). His willingness to talk with the FARC with very few preconditions, and
especially his granting of amnesty to their fighters, led Landazabal and others in the military to
formulate a counterinsurgency plan separate from the administration’s official policy.
Meanwhile, many of the regional elite around the country - especially those most affected
by the extortion, kidnappings, and violence from the guerrillas - were similarly frustrated by
Betancur’s peace policies. For example, the FARC’s “combination of all forms of struggle” and
unwillingness to disarm created fear and frustration among those benefiting from the status-quo.
The government had pledged to issues of land reform and improve the economic situation of the
campesinos - policies perceived to disadvantage the regional elite in the rural areas of the
country. Furthermore, these members of the regional elite felt as if the negotiations were a signal
that the central state was abandoning them to the very real threat of violence from the insurgents
(GMH, 2016). So not only did many within the military feel betrayed by the more liberal
approach of Betancur, but so too did many regional elites, and this had the unintended
consequence of pushing the two actors to meet with each other (Romero, 2003).
Thus, in his attempts to circumvent the new restrictions placed on the military,
Landazabal recognized an ally to which the army could turn. He began to meet with similarly
frustrated prominent cattle ranchers, large businessmen, and powerful politicians in the middle
Magdalena valley (Dudley, 2004). These men were known to be harshly anti-communist and
109

particularly upset with the peace policies towards the insurgents. In essence, Landazabal and
other hardliners within the armed forces parlayed with individuals who shared their interest in
maintaining a draconian and non-conciliatory approach to countering the rebels.
This concurrence of interests was based on the traditional relationship between the
armed forces and regional elites, based on the latter’s logistical, economic and
political support for counter-insurgency offensives and the State’s encouragement
of their territorial control in outlying areas where the State’s meager financial
resources limited its military action. Added to that was the fact that the armed forces
counter-insurgency mission had shaped their relationship with the civilian
population, poor and marginal sectors whom they stigmatized as the enemy, in
contrast with the elites, who were regarded as allies in their defense of the State
(GMH, 2016).

In other words, this burgeoning counterinsurgency partnership was based on a relationship
conditioned by a historical dependency between the security forces and the country’s regional
elites.
What came from these meetings was a pact between the military and the regional elites to
create the self-defense groups that were the precursor to the deadly perpetrators of terrorist
violence throughout Colombia. The large businessmen, local politicians, and cattle ranchers
decided to ally themselves with not only the military, but with the drug traffickers as well - a
necessary but uneasy combination of the latter two. However, the drug traffickers were an
indispensable part of the group due to their immense resources, as well as their ruthlessness and
willingness to engage in violent tactics the military had been prevented from engaging in
(Romero, 2000). This was a specific concern of the military, as not only had the Betancur
administration begun peace negotiations with the insurgents, but it had also restricted the
autonomy of the military - for example, through the repealing of the National Security Statute.
Military leaders now feared the legal consequences of their fight against the insurgents -
in their duty to secure public safety. According to Romero, “the absence of a legal status such as
110

a state of exception to protect military practices that ignore individual and collective rights has
been the reason adduced by the Ministry of Defense to account for military inefficiency in
combating the guerrillas” (2003). Doble Cero, an ally of the Castaño brothers and the powerful
paramilitary groups they formed, stressed the dilemma that soldiers faced; members of the army
had to decide between acting based on the rules of engagement and becoming just another
statistic, or putting aside the rules in order to survive in the dangerous situation that is Colombia
(Civico, 2016). According to him, in the midst of Betancur’s negotiations with the FARC, the
armed forces were perpetuating the idea that in order to defend their democracy, they “had to
temporarily transgress the rules imposed by the same system” (Civico, 2016). For these reasons,
hardliners within the military decided it was in their best interest to act clandestinely on their
own, or in collaboration with regional non-state actors in order to preserve the status-quo.
Thus, during the initial years of Betancur’s tenure as President of Colombia, we see an
intense struggle between the executive and the military surrounding the question of public
security. Eventually, as the first general election with the FARC’s political party (UP)
approached, violence against UP members and supporters increased. This increased violence
against the UP was despite the cease-fire in place, as the hardliners within the country saw the
UP “as nothing more than rebels in civilian clothes” (Dudley, 2004). The violence, perpetrated
by both state and non-state actors culminated into a full on politicide against the UP, many of
which had nothing to do with the FARC’s insurgency, but rather found solace in the political
ideology the party professed.
However, beginning in 1985, events would have it that the military would gain back
some of its autonomy in the fight against the insurgents. Peace negotiations between the
government and the other insurgent groups, such as the EPL and M-19, had failed to maintain a
111

ceasefire. Then on November 6, 1985 rebels from M-19 staged a siege on the Palace of Justice -
the equivalent of the Supreme Court in the United States - in which over 100 people lost their
lives, including eleven Supreme Court justices (Davis et al., 2016). Directly following this tragic
incident, we can witness an implicit resignation by Betancur to the leaders of the military in
matters of public safety yet again. Especially after the events at the Palace of Justice it became
increasingly clear that Betancur’s peace efforts were not indicative of the national consensus, and
as a consequence “the Peace and Verification Commission gradually became a body without real
power and at best an organization for ‘good offices’” (Dudley, 2004). With the election of
President Virgilio Barco Vargas in 1986, the approach that the government would take towards
the counterinsurgency would shift, in general, to a more hawkish method of dealing with the
guerrillas.
5.5.1 Regional Reactions to Betancur in Urabá
The region of Urabá is located within the Northwestern corner of Colombia, bordering
both the Atlantic and Pacific oceans, and spanning three different departments - Choco,
Antioquia, and Cordoba. A map of the region is provided below in Figure 5.1. Economically
speaking, the region is well known for its specialization in large banana plantations, as well as
the prevalence of large land-owning cattle ranchers. The banana industry was one of Colombia’s
leading exports, and so those involved in the business had a lot of political leverage on the
national stage. The banana plantation owners were part of a trade association named the
Association of Banana Plantation Owners and Farmers of Urabá (AUGURA), allowing them a
high degree of organization which they utilized to push for anti-labor policies (Carroll, 2011).
Due to their converging interests in anti-labor and anti-communist measures, as well as their
growing economic clout, AUGURA and the plantation owners became the central constituency
112

of Antioquia’s most prominent political party - the Guerrista faction of the Liberal Party in
Antioquia. (Carroll, 2011). Willing to employ their economic and political power at the national
level, these regional elites successfully lobbied for the military to intervene in their favor against
any labor mobilization in Urabá. Especially under Ayala’s Statute of Security, the alliance
between Urabá’s land-owning class and the army were able to keep labor and peasant
movements in the region relatively weak.
Due in part to the prevailing socioeconomic conditions in the region, guerrillas dedicated
to different strands of communist ideology found safe-haven in Urabá. More specifically, both
the EPL and the FARC established strong holds throughout the region, creating and maintaining
relationships with the peasant population holding strong grievances against the status-quo. Many
within the region were upset not only with the repressive backlash toward labor movements, but
at the increasing concentration of land bought and held by the elites. So, when the guerrillas took
hold in the region and were able to provide both a sense of security and help with the political
goals of the peasant workers of Urabá, the two became natural allies. For example, both the
FARC and the EPL helped to lead land squatter invasions of large landholdings throughout the
region, much to the dismay of the surrounding landowners worried about the same thing (Carroll,
2011). The forging of this new alliance between the guerrillas and the peasant population
signaled a new era of struggle against the interests of the large landowning class.
Eventually, the guerrillas upped the ante against the ruling class of the region as they
increased the war ‘taxes’ that they collected from these individuals. In an attempt to expand the
taxable population for example, the EPL in the early 1980s began to include small and middle-
size landowners in their required taxes, which unintentionally gave common cause for a wider
range of landowners to “share goals and promote a common action, paving the way for the
113

emergence of the self-defense and paramilitary groups” (Romero, 2000). Furthermore, at the
same time that the guerrillas had started to increase the taxes they were forcibly collecting from
the landowners, Betancur was elected as President campaigning on a promise to negotiate for
peace with the guerrillas. Along with the army generals, the cattle ranchers, banana exporters,
and other elites were dead set against such negotiations, as they believed the talks “implied
giving credence to the notion that armed insurrection reflected the need for social and political
reform” (Romero, 2000). In other words, any negotiations with the guerrillas constituted an
implicit recognition that the status-quo needed to be changed, a status-quo already noted to
distinctly benefit these very elites. Even worse, the reform that the guerrillas advocated most for
- agrarian and land reform - were especially harmful to the interests of the cattle ranchers and
banana plantation owners.
It is also important to note that Urabá saw the introduction and expansion of the drug
trade around the same time that the threat of the guerrillas was becoming overwhelming for the
regional elites. While the narco-traffickers did not immediately become the driving force in the
region, they did provide the army and regional elites with a new ally with crucial resources and a
willingness to fight a dirty war against the guerrillas. Furthermore, as a way to both keep their
money laundered and increase their legitimacy/reputation, the narco-traffickers took to buying
massive amounts of land in order to cattle ranch. Therefore, the efforts by the guerrillas to
violently change the status-quo surrounding land ownership further solidified this triangular and
deadly relationship.
Interestingly enough, this perception of an increasing and unbearable threat from the
regional elites of Urabá came about as the guerrillas agreed to ceasefires with the national
government. As the ceasefires came into place in 1982, violence from both the EPL and the
114

FARC had subsided substantially, and thus in a traditional sense, the threat of the guerrillas had
too. However, key to understanding the sense of urgency that the regional elites had in
confronting the guerrillas after 1982 was the newfound fear of the guerrillas’ political power.
Even before the negotiations resulted in the creation of new political parties for the different
insurgent groups, the banana plantation owners witnessed the guerrillas’ abilities to organize and
mobilize the peasant worker community of Urabá. Although the two main labor unions for the
banana industry - Sintrabanano and Sintagro - had been around since 1964 and 1972
respectively, neither union had been able to very successfully mobilize their workers or persuade
the plantation owners to secede to their demands. As a result of the weakness of the labor unions,
working conditions were abysmal,
characterized by piecework and day labor, the total absence of benefits, the
employers’ use of subcontractors to evade their legal obligations to their workers,
failure to pay overtime for workdays over eight hours, and, in general, the near-
total disregard of labor regulations by employers, often with the complicity of the
Ministry of Labor (Carroll, 2011).

Furthermore, many attempts by the unions to strike or bargain with the banana plantations were
met by the militarization of those plantations. Workers within the region were in need of change,
and unfortunately the normal avenues of reform seemed all but blocked.
It was among these conditions that the guerrillas started to solidify their relationship with
the banana unions. In other words, the blockage that these workers sensed in the traditional
means of advocating for their rights created a space in which the armed insurgents became
attractive allies. In the early 1980s, before the formal peace negotiations took place, the
guerrillas played the essential role in organizing the unions, protecting them against repression,
and pressuring the plantation owners to negotiate (Carroll, 2011). However, once the peace
negotiations between the guerrillas and the government began, and the Betancur administration
115

relaxed its restrictions on labor unions, then the insurgents could declandestinize and labor
activists could openly organize (Carroll, 2011). Due to Betancur’s new approach to relations
with labor and the armed insurgents, regional elites could no longer fall back to their traditional
methods of dealing with threats to their status-quo - for example, relying on the army to use brute
force in retaliation to labor demands.
With the introduction of the new leftist political parties associated with the FARC (the
Patriotic Union (UP)) and the EPL (Frente Popular), the left now had reasonable means with
which to organize around and push for a new system of economic relations within Urabá. Most
importantly, with the passing of the new law in 1985 which allowed for the direct election of
county executives, the regional elites could no longer rely on the patronage of party elites at the
departmental level to appoint them to key positions at the county level. Although the first direct
elections did not take place until 1988, “in counties where the Patriotic Union won more county
council votes than any other party the governor would appoint a county executive from the UP in
1986” (Carroll, 2011). Due to this provision in the law, the UP was able to win several county
executive positions throughout the country. Notably, the the county of Apartado - a major banana
producing area of Urabá with a relatively strong economy - had a leftist county executive
appointed at this time.
The worst fears of banana plantation owners and other regional elites had now been
realized. Not only had the guerrillas been able to remain armed throughout the peace process,
and therefore retained the capability to return to violence, but now their popularity with the left
of the country had been formally institutionalized - especially in counties where they had been
appointed to the executive based on their electoral results. With this newfound power, the county
administration - led by the UP - was now able to utilize legal and financial resources to the
116

advantage of the worker and peasant population in Apartado. For example, they legalized the
squatter invasions of large properties by giving squatters access to property titles (Carroll, 2011).
It is no wonder that the regional elites at this time felt that they needed a way to intervene in
favor of the status-quo.
With the military’s hands officially tied due to the cease-fire, any violence perpetrated
against the guerrillas and their supporters needed to be conducted informally and with plausible
deniability. Therefore, it is no surprise that it was in the years of formal negotiations, cease-fires,
and institutionalized political gains for the guerrillas that the region saw its first recorded uptick
in status-quo terrorist violence. According to Figure 5.2 presented below, in 1983-86, the region
averaged 10.5 status-quo terrorist attacks per year, up from an average of 2.5 between 1981-82
(the first years where the data is available).
Henriquez Gallo, a banana plantation owner, and member of the House of
Representatives, offered a telling statement representing the thinking behind many of the
region’s elites regarding the guerrillas and the peace process. In a presentation at the AUGURA
congress in 1986:
Security, which we request daily from the government, which we demand from the
Armed Forces of the Republic, which we cry for from every trade organization, is
something...that we must make the government aware of at every opportunity...As
long as [the president, the ministers, the governor, and the armed forces] give us
slow, late, solutions that make us despair, we must form our own strategies.
(Carroll, 2011).

In other words, the typical militarization of the region that Gallo and his companions had relied
on to solve their issues with the labor force was no longer a solution they could count on.
Instead, the elites of Urabá would need to take matters into their own hands if there was
to be any hope of preserving the status quo of the region. What Gallo’s comments represent is a
common line of thinking at the time which justified the increasingly dirty war being fought
117

against the left. The hardliners within the military, similarly frustrated by Betancur’s new
restrictions on the fight against the guerrillas, were more than happy with this development. They
now had the perfect outlet which they could use to funnel their repressive tactics through. This
came in the form of both informal collaboration with the newly formed self-defense groups and
paramilitaries, as well as a happy ignorance of any independent violence perpetrated by such
groups. In essence, as attempts at democratization from the central government increasingly
became seen as an impediment and threat to the status quo, hardliners both within and outside
the state became increasingly willing to utilize terrorist violence.
However, in the context of Colombia’s recent history, the levels of status-quo terrorist
violence seen in this period were at relatively moderate levels. What can be observed is a definite
willingness on the part of the regional elites and other non-state actors to utilize violence against
the left. Yet, much of the violence could still be attributed to the “clandestine acts by radical
sectors of the armed forces, or simply the work of hired killers employed by temporary alliances,
with short-term aims, of different economic, political, and military agents who did not intend to
form permanent groups or fronts” (GMH, 2016). In other words, while the early-to-mid 1980s
saw a definite uptick in status-quo terrorism - attributable to certain democratic pressures - parts
of the armed forces were still willing and able to conduct some of their repressive tactics against
the left in Urabá at this time (albeit more discreetly than normal). That need for discreteness
would almost disappear as a new administration came to power.
5.5.2 A New Strategy for Peace under Presidents Barco and Gaviria (1986-1993)
Colombia saw the election of a new president in 1986 with a different view on
negotiations with the guerrillas. President Virgilio Barco and his new administration continued to
search for peace with the guerrillas but with different stipulations to any deal that may have been
118

reached between the two sides. Barco’s approach to peace was outlined in what he called the
National Rehabilitation Plan (PNR). Much of the PNR was dedicated towards expanding the
state’s civic presence throughout the country, most specifically the rural areas of Colombia that
were in need of investments in infrastructure. The extension of the state’s presence to these
regions was to be achieved partially through attempts at decentralization.
Decentralization was to continue both politically and administratively, as it had begun
with Betancur’s decision to allow members of the UP and other leftist groups to be appointed as
county executives. In 1988 the country would see its first popular election of mayors, attempting
to strengthen the population’s connection to civil society - especially in Colombia’s outlying
regions (GMH, 2016). This decentralization and increased competition in the local political
arenas was a sincere worry of the traditional political elites throughout the country. For example,
the UP was able to present a challenge for power in 280 of the country’s 1,098 municipalities;
representing only a small proportion of the country’s population but more than half the country’s
surface area (Dudley, 2004). Therefore, while the far left was not a viable threat to win many
seats in the national government, the new focus on decentralized politics offered them a path
towards institutionalized control over many of the country’s natural economic resources.
A major difference between Barco and his predecessor Betancur when it came to
negotiations with the guerrillas was the belief that military pressure could be put on the guerrillas
while continuing to search for peace. Within a year of taking power, the final remaining cease-
fire from Betancur’s presidency came to an end. As the dirty war against the UP continued
across the country, and the military was seen to be partially responsible for the attacks, and/or
unwilling to protect members of the political party, the FARC began to ramp up its attacks once
again. While there was no formal end to the ceasefire, an ambush by the FARC which killed 26
119

military personnel and one civilian marked the full opening of hostilities between the two sides
after years of negotiations (Durán, 2004). Now with all the cease-fires broken, the state security
forces had a freer hand to pursue the guerrillas in a violent manner.
Eventually, President Barco would seek out his own peace initiative with the guerrillas,
but in a few key different ways. The most important of these differences in strategy was the idea
that to allow the guerrillas a cease-fire from the government’s security forces while not requiring
them to demobilize and disarm was an unwise strategy. This is well illustrated by Barco’s speech
regarding his own peace initiative with several guerrilla groups (but not including the FARC):
Our reconciliation policy is inspired by the principle of ‘an outstretched hand and
a firm, steady pulse’...In the first place, the government, with firmness, generously
reaches out to the armed insurrection groups that convincingly demonstrate that
they have a sincere desire to reintegrate themselves into civil life. Furthermore, if
Congress approves the corresponding law, those members of the armed rebel
groups that accept the terms of the Peace Initiative will be pardoned. But the pardon
will not be conceded at the beginning of the conversations, only later, after they
have given up their arms. The development of this Peace Initiative is not, nor will
it ever be, an obstacle when it comes to having public forces and the Justice
Administration fulfill their constitutional functions fully in confronting and
punishing those who carry out acts of violence, terrorism or subversion. Much to
the contrary, the serene presence of governmental authority throughout all of the
national territory is an indispensable complementary to this initiative. (Barco, 1988)

Barco therefore sought to emphasize to the nation that while he continued to search for peace,
that would by no means be accomplished by tying the military’s hands. Instead, the military’s
ability to confront those actors who carried out violence would be used as leverage to persuade
rebel groups to disarm and demobilize.
It is important to think about the ways in which the security forces reacted to such a shift
in negotiation strategy with the guerrillas. While certain hardliners within the military would
have rather not negotiated with the insurgents at all, the requirement that any parties to the peace
initiative would need to disarm and demobilize was seen as the correct approach as compared to
120

Betancur’s allowance of the guerrillas to negotiate while building their military capabilities.
Furthermore, the country’s largest insurgent group - the FARC - was not a part of the peace
initiative, and so the security forces were still very much engaged in counterinsurgency tactics
throughout the process.
Concurrent with the ongoing peace talks with the M-19, much of the evidence and
statistics coming out about violence in Colombia was showing that the paramilitaries were
killing more civilians than the guerrillas were. In response to these findings, Barco announced
several new presidential decrees that were anti-paramilitary in nature. For example, according to
Decree 813, an independent commission was to study the paramilitary issue, and according to
Decree 814, a special police unit would be created to specifically police paramilitary groups.
Most importantly, overturning the former legal status of armed self-defense groups, Decrees 815
and 1194 established “criminal penalties for the formation or operation of such groups, and
require[d] the approval of the prescient before any type of self-defence group was established”
(Avilés, 2006). These criminal penalties were established to apply to anyone operating or helping
such groups, including members of the armed forces.
Meanwhile, the M-19 Movement agreed to lay down their arms and become reintegrated
into civilian life as part of a peace deal reached with the Barco government. After reintegration,
members of the M-19 were granted amnesty and allowed to create a new political party named
the Democratic Alliance of M-19. Around this time, pressure from many of the country’s citizens
for a new Colombian Constitution intensified - especially after the assassination of Liberal
presidential candidate Luis Carlos Galan in 1989. Building off of the increased public support for
the crafting of a new Constitution, President Barco hatched a plan. As his term was ending,
Barco decreed a state of emergency in order to formalize “the proposal of a vote to summon a
121

Constituent Assembly to be included in the presidential elections in May 1990” (GMH, 2016).
With the election of President Cesar Gaviria as the successor to Barco, and an overwhelming
majority of voters who voted in favor of the initiative, Gaviria announced a presidential decree
for an election to appoint members to the Constituent Assembly - within which the Democratic
Alliance of M-19 won the second most seats.
The delegations of the Constituent Assembly over the formation of a new Colombian
constitution would take place from January to July of 1991. In the shadows of these talks over a
new constitution, many of the guerrilla groups would come to believe in the institutional changes
being discussed. Recognizing the significant potential to reshape the system through the formal
political process, the EPL, and the Movement Quintin Lame both decided to lay down their arms
and become legal political parties in 1991. This shift allowed these groups to participate in the
upcoming elections that would take place after the creation of the new constitution. Now with the
demobilization of M-19 and the EPL, and the creation of the Democratic Alliance of M-19 and
Esperanza, Paz y Libertad (Hope, Peace and Liberty), two of the three largest insurgent groups
had committed to Colombia’s new democratic institutions. While this may have lessened the
violent threat of the left in the short term, the new institutions were seen to have put unnecessary
constraints on government in the fight against the remaining insurgents.
The constitution of 1991 won much of its support from the far left because of its promises
of decentralization, a revamp of the electoral system, and a mandate to encourage much broader
access to the ownership of land. The new constitution also laid out several new ways in which
the legislature would check the executive branch and reduce the president’s authority, including
veto power over cabinet nominees, obligations to report to Congress, and a limitation to state-of-
emergency powers - obvious attempts at ensuring there were more constraints on power. There
122

were also very explicit references to human rights in the new document; that all Colombian
citizens are born with such rights, and that it is the responsibility of the government to protect
said rights.
However, when analyzing the effects that the constitution had on hardline sectors of the
armed forces, one must look at how the new institutions sought to constrain their autonomy. It is
important to remember that there had been an ongoing struggle between the civilian government
and the armed forces over who was ultimately in charge of public order. The founders of the new
constitution, and those that took power under the new system of government made clear attempts
to establish civilian control over such issues. The most obvious and symbolic of this push by the
new government was the appointment of a civilian as the Minister of Defense for the first time in
over half a century (GMH, 2016). Along with the appointment of a civilian defense minister,
there was “greater interference from the executive branch in matters of security, financing, and
organization of the armed forces” (Romero, 2003). Ultimately, the new system put in place was
meant to subordinate the armed forces to the civilian government:
Even the military yielded to the possibility that the ordinary judicial system might
investigate and judge members of the Armed Forces involved in violations of
human rights and drug-trafficking. To that was added the creation of the National
Security Office, to strengthen the influence of the civilian power in the planning of
strategies to secure public order. (GMH, 2016)

Hence, the new constitution and the institutions established immediately afterwards signaled a
real shift towards the military’s accountability to the civilian-led government – one that had
prioritized strengthening the country’s human rights record.
There was a relative tranquility for the time between President Barco’s peace talks and
the formation of a new constitution and government under President Gaviria. However, as it will
become clear in the mid-1990s and onward, during this time period the foundations were laid for
123

a severe uptick in status-quo terrorism throughout the country (and especially the region of
Urabá). Several attempts at the liberalization of the Colombian democracy, particularly in 1989
and onward, proved to influence members of the armed forces, regional elites, and drug
traffickers toward a disenfranchisement with the direction of the country which would eventually
lead to the consolidation of paramilitary power throughout much of Colombia. Key among these
developments were the criminalization of self-defense groups, institutionalization of the
military’s subordination to the civilian government, and an emphasis on the protection of human
rights in an ongoing counterinsurgency.
5.5.3 A Militarized Urabá Experiences Low Levels of Status-Quo Terrorism
With an understanding of the national circumstances regarding the Barco and Gaviria
presidential administrations, I now seek to explain the important regional developments
throughout Urabá and how the combination of national and regional circumstances led to the
relatively low levels of status-quo terrorism during this time period (1986-1993).
First and foremost, the key to understanding why the region of Urabá experienced a
relative lull in status-quo terrorist violence during this period is due to the extreme military
buildup in the region after the breakdown in talks between the government and the EPL in 1984.
Barco’s decision to markedly increase the militarization of Urabá reflects both the reopening of
hostilities in the region, as well as the importance of the region politically and economically:
Whereas in 1984 there was only the Voltigeros Battalion in Carepa, in 1987 the XI
Brigade was created in Monteria (the capital of the contiguous department,
Cordoba); by decision of the national Ministry of Defense, its jurisdiction included
Urabá. At the insistence of AUGURA, in April 1988 the Military Headquarters of
Urabá was created and was maintained through July 1990…With the creation of
the Jefatura Military, Urabá became the region with the greatest number of troops
per square kilometer. As part of the changes in 1988, the X Brigade was also
created in Carepa, and in 1989 yet another battalion was added: the Batallón
Francisco Paula de Vélez in Chigorodó. This brought the total number of soldiers
in the banana axis alone to about 1,600 by 1989. (Carroll, 2011)
124

Now that the cease-fire had ended with the region’s biggest insurgent group (and soon thereafter
with the second biggest - the FARC), and then President Barco authorized a massive military
buildup, there was much less of a demand from the armed forces for status-quo terrorist violence
from those actors outside the state.
However, it is important to note that there was still plenty of violence against the left in
Urabá during this period - just not officially perpetrated by non-state actors. Instead, the violence
that characterized this period was a mix of legitimate military actions, human rights abuses by
the armed forces, and a campaign of violence against different leftist political parties and the
sympathizers perpetrated by a shadowy collaboration between security forces and (up until 1989)
legal self-defense groups.
Allowing the military to have such a large presence and important mission, with very
little oversight over their transgressions, detracted from their need to rely on the paramilitaries
for plausible deniability. Analyzing the military’s relationship with the paramilitaries in the sense
of a principal-agent problem, the military is more inclined to delegate tasks and autonomy to
their agents when they feel that the agents may be able to accomplish those tasks in a way that is
less costly, or in a way that the military is unable to do. To give their agents autonomy over a
mission or task, gives the military real plausible deniability as to the methods that are executed in
pursuit of those goals. However, there are tangible risks to delegating such autonomy to their
agents; for example, the military loses full operational control of the mission at hand. They may
not agree with all of the strategies their agents utilize, and once the military’s monopoly on force
is willingly given away it may be hard to rein that authority back in. Therefore, in Urabá, when
the military was given more of a free hand to pursue the guerrillas and their supporters in the
methods they preferred, members of the armed forces were less willing to gift such autonomy to
125

the paramilitaries. This was especially the case in the mid to late 1980s as the drug cartels, who
very often were an essential part of the country’s different paramilitary groups, began to cause
havoc. In essence, even hardline members of the state began to see the risks involved with
allowing too much autonomy to certain paramilitary groups.
However, what we see in Urabá at that particular moment in Colombian history is the rise
of the Castaño brothers and their paramilitary groups throughout the region. Just as the
government’s attention was pointed towards the Cali and Medellin drug cartels, the Castaño
brothers formed and led a paramilitary group directly allied with the government in order to
capture Pablo Escobar called (Persuigidos por Pablo Escobar (Persecuted by Pablos Escobar -
Los Pepes)). This ultimately led to a reordering of the paramilitary pecking order throughout the
country, with the Castaño brothers sitting toward the top.
The Castaño brothers got into the paramilitary business after their father had been killed
by the guerrillas, making them staunchly anticommunist. The older brother, Fidel, had taken over
the family business and eventually became one of the region’s most wealthy cattle ranchers.
Furthermore, he had brought it upon himself to form the biggest paramilitary group in his region:
Through connections to a vast network of local rancher-sponsored self-defense
groups, Castaño’s paramilitary infrastructure extended its radius of action to the
south of Cordoba and the north of Urabá. By 1990 the police listed paramilitary
bases in all of the banana axis counties of Urabá, as well as Arboletes and Necocli
in the north of Urabá. (Carroll, 2011)

The group would eventually be named the Peasant Self-Defenders of Córdoba and Urabá
(ACCU), the precursor to the even larger paramilitary umbrella group that his brother Carlos
would create, the United Self-Defense Forces of Colombia (AUC). There is plenty of evidence
that the ACCU at this time was “tolerated and even aided by the armed forces” (Carroll, 2011).
In essence, we can observe Carlos Castaño reorganizing the different, formally legal self-defense
126

groups, into a new more coherent paramilitary organization intent on establishing close ties with
certain sectors of the military.
While the Castaños and the paramilitaries they led would cause even more havoc in the
upcoming years, the early 1990s saw a peace between such forces and many of the guerrilla
fighters throughout Urabá. It is important to recall that around this time Colombia is
experiencing its new peace initiative with a ceasefire and demobilization of the country’s biggest
urban guerrilla group - the M-19. As the progress with the Constituent Assembly over the writing
of a new Colombian constitution is witnessed by the other insurgent groups, more come to the
negotiating table with the Barco and Gaviria administrations. Eventually the EPL, Urabá’s
strongest guerrilla group, decides to disarm, demobilize, and create a new political party - Hope,
Peace and Liberty. This creates a regional peace between the paramilitaries and the left from
about mid-1990 to 1992:
In turn, in the regions influenced by the demobilized guerrilla groups, the
paramilitaries also began to disarm, at least partially. A significant case with effects
at the national level was seen in the department of Cordoba and the neighboring
banana growing zone of Urabá . Here the EPL had its main area of influence, as did
the Castaño family and its private army, with its older brother Fidel at the head. The
demobilization of the EPL and its conversion to a legal movement - Hope, Peace
and Freedom - was followed by the announcement of Fidel Castaño of the
distribution of near 16,000 hectares of its property to poor peasants and victims of
the armed confrontation. (Romero, 2003).

However, even though the EPL had demobilized, the FARC still maintained a threatening
presence in the region, and there would soon be a dissident group of former EPL members that
the government - and the paramilitaries of the region – would turn their attention to.
After the relatively moderate levels of status-quo terrorist violence in Urabá during the
Betancur years and unconditional negotiations with the guerrillas, the region experienced a
transformation in the type of violence that plagued its citizens. Under president Barco, the
military severely increased its presence in Urabá, and was given a much freer hand to pursue the
127

guerrillas in the way they saw fit. As a result, the armed forces were less dependent on utilizing
paramilitary groups to achieve those same goals while maintaining plausible deniability for any
human rights abuses that occurred. Therefore, as can be seen in Figure 5.3, the region observed a
major reduction in status-quo terrorism during the years from 1986-1993, averaging only 3.37
status-quo terrorist incidents annually (only 2.28 attacks annually if we look at 1987-1993).
Yet as will be discussed in the following section, many of the foundations had been laid
in the final years of this particular period for a substantial spike in such violence. Specifically,
the creation of a new and more democratic constitution which institutionalized means of land
distribution and constrained the military’s autonomy in its role of ensuring public order.
Furthermore, certain moves made by the Barco administration criminalized the formerly legal
self-defense groups, pushing many of the individuals involved to join paramilitary groups that
were even more vicious.
5.6 The Aftermath of a New Constitution (1994-2003)
President Gaviria, by his second year in office, had now signed into law a new
constitution in order to further democratize the nation. Within the context of one of the world’s
longest ongoing insurgencies, there were many changes to the power and autonomy of the armed
forces. The reason that such reforms were necessary was that there was overwhelming evidence
of human rights abuses within the military. The struggle over improvements in human rights
practices would come to define much of military’s disenfranchisement with the institutions of the
new constitution.
The new institutions affecting the armed forces did not stop with the signing of the
constitution. Instead, the civilian government would continue to put into place new laws and
128

standard operating procedures that the military themselves found quite constraining. According
to a declassified CIA report on the Colombian Counterinsurgency during this time:
Bogota is striving to improve its human rights record. The Gaviria administration
is beginning to institutionalize human rights training in the security forces.
[Redacted] government human rights ombudsmen are instructing in various
military and police training courses. In addition, [redacted] Gaviria endorsed the
appointment of a UN High Commissioner to examine human rights in Colombia
and announced disciplinary actions against several military personnel charged with
violations” (CIA, 1994).

While this emphasis on reigning in the human rights abuses of the armed forces did eventually
lead to an improvement in the official human rights standards of the military, it also led to severe
frustration among many within the security forces that felt they were being unfairly constrained
against a continuously deadly threat.
A declassified U.S. Department of State intelligence report illustrates this mounting
frustration and the effects it had:
Although human rights violations attributed to the military have declined, the
military’s frustration with its ability to defeat the guerrillas has contributed to a
jump in paramilitary violations. Many guerrilla suspects detained by the military
are quickly released by judges for a lack of evidence, which leads to military
complaints about having to fight ‘with both hands tied behind its back.’ The result
has been decreased military aggressiveness in the field, and at least tacit support for
paramilitaries, which liquidate suspected guerrillas and sympathizers without legal
concerns (U.S. Department of State, 1998).

In other words, as these new institutions increasingly limited the military’s ability to wage war as
they pleased, they turned their support even more towards the paramilitaries which were not tied
down in the process.
This line of thinking was echoed among individuals I interviewed during my time in
Colombia. In an interview with a military intelligence commander of the Colombian Army, we
talked about the many difficulties that the security forces faced in the counterinsurgency against
the guerrillas over the years. According to him, the number one issue that the security forces
129

faced was the omnipresent problem of their enemies blending in with civilians. The second
biggest issue was the “restrictive rules of engagement.”
When I heard this, I decided to push this topic a bit further and asked him to elaborate.
This is when he became most animated throughout our interview - signaling the sincere
frustration that the phenomenon creates within the armed forces. He dove into the ideas behind
the international human rights regime, and how when confronting the insurgents it means that the
military must say, “stop, we are the military, put your hands up!” After this, he showed me a
video of members of the military that are confronted with knives, punched, thrown rocks at by
civilians, and how they can do nothing in return. “Can you imagine the outrage if we shot at
these people?” I then asked why they could not just shoot to injure in defense, and he answered,
“that is not how we are trained, our shooting is lethal.” Thus, the interview illustrates how the
first and second issues are overlapping. As a force that is trained to fight and engage in lethal and
regular warfare, the army was oftentimes ill-equipped and unprepared to fight against a guerrilla
force that blended in with unarmed civilians.
With the perception that the military was becoming more and more constrained by the
nation’s new institutions, those hardline commanders most invested in a repressive
counterinsurgency became more likely to delegate the fight to their paramilitary allies. Due to
the illegal nature of these paramilitary groups, members of the military had to make sure to
maintain plausible deniability. A declassified Defense Intelligence Agency report on the subject
demonstrates this logic:
There is no known official connection between the Colombian (COLAR) and the
various paramilitary organizations which operate throughout parts of Colombia.
Nevertheless, at the local level, COLAR commanders often enter into discrete
marriages of convenience with these groups in order to obtain important
information regarding ongoing guerrilla activity occurring in their area of
130

responsibility. Local commanders generally find it foolhardy not to maintain a
dialogue with these groups (D.I.A. 1996).

Not only did commanders maintain discrete relationships with paramilitary groups for pure
information or intelligence advantages, but they did so for the methods these groups could utilize
as well.
Asking the military intelligence commander of the Colombian Army about the
collaboration that parts of the Army offered the paramilitaries, he responded that “it was not a
plan, but an opportunity...a result of a tactical situation….an enemy of my enemy is a friend.” In
order to further justify and explain the logic behind the collaboration he asked, “what would you
do if you’re in a zone, get hit by a mine and don’t know who to kill, or can’t due to the rules of
engagement, and someone comes and says they will kill them for you, or does it and doesn’t
ask?” Thus, the ways in which democracy constrains the security forces in a counterinsurgency
contributed to the willingness of members of the military and their commanders to enter these
discrete marriages with the paramilitaries.
Yet, the democratization of the Colombian government also conditioned the ways in
which the military structured its relationships with the paramilitaries. Recall that in 1989, in an
effort to reign in the severe human rights violations being perpetrated against the political left of
the country, criminalized the formation and facilitation of self-defense groups. Before this
presidential decree, commanders of the military could legally collaborate with such groups in
their efforts against the insurgents. However, in the 1990s the military now had to keep such
relationships secret:
Local COLAR (Colombian Army) commanders usually meet discreetly with
members of these groups. Because of these discreet, local meetings, there is little
fear by either side of the association becoming known publicly (‘making the
newspaper’). Both parties can claim ignorance of any official association. All
131

parties involved are aware that due to the illegal nature of these groups the military
cannot openly associate with them (D.I.A., 1996).

In other words, the new democratic institutions put in place to limit state violence and human rights
violations by the military essentially shifted the responsibility of such violence to different actors.
Now, instead of the violence being a primarily state-led phenomenon, non-state actors were
increasingly being given the autonomy to fight their own counterinsurgency using the methods
they saw fit.
This all took place during the rise of the strengthening of the FARC in its violent
capabilities across the country. While the peace initiative and Constituent Assembly took place,
and the M-19 and EPL disarmed and demobilized, the FARC continued to gain strength.
Following a full-scale offensive against Casa Verde where the FARC had first begun their peace
talks with the government, the FARC greatly expanded its recruiting efforts. Their goal was to
execute a plan initiated by Jacobo Arenas in the FARC’s VII Conference (1982) to build a large
enough army to surround Colombia’s major cities, splitting their forces into military blocs; as the
1990s progressed, the FARC found that they were forming an army capable of fighting and
holding territory (Dudley, 2004).
The advances by the FARC rightly unnerved the Gaviria administration - and the rest of
the country. At the end of his term, Gaviria announced a presidential decree to form the Special
Vigilance and Private Security Services (Servicios Especiales de Vigilancia y Seguridad Privada,
or CONVIVIR), which were very much akin to another form of legalized self-defense groups. As
President Ernesto Samper came into power in 1994, he developed the CONVIVIR groups even
further. These CONVIVIR groups were meant to operate similarly to something like a
neighborhood watch group, protecting communities from the increasing threat of the FARC, and
providing local intelligence to the government. However, according to a declassified Defense
132

Intelligence Agency assessment, many of the paramilitaries saw the establishment of
CONVIVIR as a greenlight and used the institution to their advantage:
Many illegal armed groups were granted convivir licenses even as they continued
operating outside the scope of convivir responsibilities. Although Colombian Army
unit commanders could legally provide handguns to convivir members for personal
defense, officers were accused of arming some with automatic rifles and including
them in military operations (D.I.A., 2000).

In fact, what would become apparent in Urabá - and in other parts of the country as well - was
that the paramilitaries were using CONVIVIR groups as fronts for their illegal activity.
Speaking with a former member of the military and then member of the paramilitary
group led by the Castaño brothers named Veloza, Civico gets confirmation that this is how the
paramilitaries viewed the new institution: “The Convivir was a legal organization basically
established by the state. We used it for illegal activities” (Civico, 2016). However, as opposed to
the previous period when self-defense groups were legal, the paramilitaries had consolidated
much of their power by now. Whereas much of the violence that occurred through the self-
defense groups in the 1980s was state-led, now most of that violence was planned and executed
primarily by the paramilitaries. As this facade became more and more apparent, and the
Constitutional Court began to take actions to rule the decree unconstitutional, the Samper
administration was forced to dismantle the Convivir groups.
Finally, as the FARC ratcheted up its attacks against the state, and increasingly captured
and held territory across Colombia, the Samper administration was forced to negotiate with the
group. By the end of August 1996, the FARC initiated a brutal offensive with all 7 of its military
blocs launching twenty-six simultaneous attacks across the country (A Great Perhaps). After
major losses on the part of the government, Samper was forced to create the Demilitarized Zone
133

(DMZ) in Cartagena del Chaira where the government and the FARC could negotiate for a peace
deal (Dudley, 2004).
5.6.1 Consolidation of Power by the ACCU in Urabá
The beginning of the 1990s was characterized by a relatively short peace between the
guerrillas and status-quo groups in the region. Much of this was due to the disarmament and
demobilization of the EPL during the writing of the new constitution. However, as a new
breakoff dissident group emerged to take its place, and the FARC expanded its power and
capabilities in the region, the Castaños executed a brutal plan to consolidate their power
throughout Urabá.
Following the success of the private justice group (Los Pepes) that Fidel had organized to
hunt down Pablo Escobar, he and his brother Carlos became guides for the Colombian Army:
But when they located suspected guerrillas or collaborators, the army seemed to
have little power to arrest or punish these people. So, Carlos explained to me,’ We
would get on our civilian clothes, grab our rifles or whatever, and —tan! tan! tan!—
we would kill them.’ The Castaño cadre was relentless. The first to fall were Don
Jesus’s kidnappers, but the House of Castaño didn’t stop there. One witness
reported a weeklong killing spree in a neighboring province by men ‘dressed in
ponchos, white hats and carrying new machetes, rifles, knives, pistols, and
grenades.’ Dozens were killed, including women and children.” pg 145 (Dudley
2004)

Soon Fidel and his paramilitary forces became an indispensable weapon in the counterinsurgency
efforts in the region.
However, just as he was planning his all-out war in Urabá, he died in 1994 under
uncertain circumstances. In the aftermath, his brother Carlos took over the lead role in the
upcoming paramilitary-led war against the guerrillas and their sympathizers in Urabá. Under
Carlos Castaño, the ACCU “served as the spearhead of the regional political and economic
opposition to the democratizing reforms of the new Constitution, but also responded to the
134

military realignment of the FARC,” as it attempted to take over the territories previously held by
EPL militants (GMH, 2016). The strategy that Carlos pursued during this time was to not only
militarily defeat the guerrillas, but to “seek the acquiescence of the Armed Forces and other
institutions, establish alliances with local political groups…[and] make some economic
concessions to strengthen social support” (Valencia, 2007). Essentially, these goals were
accomplished primarily through the demonstration of the effectiveness of the paramilitaries’
fighting methods.
By 1997, Carlos Castaño had declared victory over the guerrillas in the region of Urabá.
He would soon after call upon the remaining paramilitary groups throughout the country to unite
into a single paramilitary confederation called the United Self-Defense Forces of Colombia
(Autodefensas Unidas de Colombia, AUC). Castaño and the ACCU maintained the leading role
in the new paramilitary umbrella organization, however, and sought to push their strategy onto
other groups across the country on a national scale. These groups increasingly took over the
more repressive tactics that the military had used in earlier years, using “forced disappearances
to complement other methods of terror and hide the real dimensions of the violence they
inflicted” (GMH, 2016). Thus, the paramilitaries had finally consolidated their power, first at the
regional level and then at the national level. If the military could not pursue the guerrillas in an
“effective” manner, the paramilitaries would.
Due to the manner in which the paramilitaries could fight the guerrillas, in an
unconstrained way, and unafraid of the costs of human rights violations, hardliners within the
armed forces did in fact cooperate or look the other way regarding their status-quo terrorist
violence. This is amply demonstrated by the example of General Rito Alejo del Rio and his
tenure as the Brigadier General of the Urabá -based 17th Brigade. In 1996 Del Rio’s deputy,
135

Colonel Carlos Alfonso Velasquez wrote an internal report to complain about his boss’
behavior,
Calling on the Army to investigate the unit’s paramilitary ties and accusing Del Rio
of turning a blind eye to paramilitary activity. Rather than heed his warning, the
Army fired Velasquez, forcing him into early retirement for insubordination.
(Evans, 2010).

In a subsequent interview with Velasquez at the U.S. embassy, he stressed that under the
command of Del Rio, the brigade had begun to collaborate much more with the paramilitaries of
the region.
The embassy had reportedly begun to notice a correlation between areas under Del Rio’s
command and paramilitary activity. These events led the embassy to conclude that Del Rio had
“contributed to a command climate conducive to turning a blind eye to paramilitaries, or worse”
(Evans, 2010). Importantly, the levels of status-quo terrorism in Urabá reach their highest point
in the entire period under analysis the first year that the commander takes his post in the 17th
brigade, tallying 29 total status-quo terrorist attacks in 1995. Furthermore, when Del Rio leaves
the region in 1998, the total number of attacks that year drops all the way down to 5. In fact, a
declassified memo from the US Embassy to the State Department discusses Del Rio’s time in
Urabá:
Although Brigade Commands are generally rotated every year, general Del Rio was
allowed to remain in command of the 17th Brigade in highly-conflicted Urabá
region for two years, apparently because he had been very successful in bloodying
the FARC’s nose during the period of his command. His systematic arming and
equipping of aggressive regional paramilitaries was pivotal to his military success
at the time (U.S. State Department, 1998)

What becomes clear in observing General Del Rio’s behavior and unsavory relationship with the
paramilitaries in Urabá was that hardline members of the military by this point had come to rely
on the violence of the former.
136

Given Urabá’s importance both politically and economically, it was crucial for the
government to establish control of the region. To do so required confronting an increasingly
strong militant force that could blend in with ideologically sympathetic civilians. Respecting the
human rights of Urabá’s population and successfully pushing the guerrillas out of the region was
deemed to be too difficult to do alone. Therefore, Del Rio and other members of the 17th
Brigade were happy to play a facilitating role in the paramilitaries’ status-quo terrorist violence
in pursuit of such goals. As they shared a common enemy, the success of the ACCU translated to
the success of the careers of many in the military.
The economic elites of the region participated in the schemes of the ACCU and later the
AUC as well. Employees for Banadex, which was a subsidiary company owned by Chiquita
Brands International, have been caught up in a scandal in which they paid the paramilitaries for
protection and a common interest in pushing the FARC out of the banana axis region of Urabá.
There is evidence that before they began payment and collaboration with the paramilitaries, they
were in negotiation with the guerrillas of Urabá. They were particularly tired of the advance of
the banana labor unions in the region which the guerrillas had helped to organize and strengthen.
In 1993 for example, there was a memo from the general manager of Banadex - Charles Kaiser -
requesting funds to pay the FARC saying “we should get from them an understanding not to
block anything we do with the sindicato [labor union]” (Evans, 2011). In other words, the FARC
was pushing an economic ideology in Urabá that the banana companies felt were directly against
their self-interest.
Later memos from the company show that by 1997 the situation in the region had
changed to one in which Banadex looked to the AUC for protection and their ability to drive the
FARC out of the region. For example, there is now evidence that Keiser and a Banadex attorney
137

named Reinaldo Escobar de la Hoz both had a secret meeting with none other than the leader of
AUC Carlos Castaño. According to a company memo, the meeting with Carlos Castaño went as
follows:
[redacted] recounted the well-known killings and property destruction carried out
in Urabá by various guerilla groups, and he told [redacted] and [redacted] that
Banadex shared a common interest with the government, military, and business
community in driving the guerrillas out of Urabá (my emphasis added). It was well-
known at the time that senior officers of the Colombian military and the Governor
of the Department of Antioquia were campaigning for the establishment of a
Convivir organization in Urabá. [redacted] said he was sure [redacted] agreed the
guerrillas needed to be driven out of Urabá and looked forward to their support of
Convivir (Evans, 2011).

Thus, according to the thinking of the main paramilitary leader himself, members of the military,
business community, and government were all united in their violent opposition to the guerrillas.
Looking into the payments that were eventually made by Banadex to the Convivir group
(called La Tagua del Darien), Chiquita senior counsel Robert Thomas when Keiser and Escobar
first attended the meeting with Carlos Castaño, they felt that had no choice but to go, as
“refusing to meet would antagonize the Colombian military, local and state government officials,
and Autodefensas [paramilitaries]” (Evans, 2011). This meeting is quite illustrative of the
ongoing alliance between parts of the government, military, business community, and
paramilitaries. This is a story that demonstrates just how much all of the actors involved shared a
common interest in opposing the violence and goals of the FARC. It also shows that the military
and the Governor of Antioquia (later President of Colombia Uribe) were hoping to establish a
Convivir group in Urabá through which they could provide support for the AUC. In other words,
it is apparent just how dependent parts of the government and military had become on the
capabilities and viciousness of the paramilitaries for their counterinsurgency strategy.
138

For the reasons laid out above, we see the biggest surge in status-quo terrorist violence in
Urabá during this time period, visualized through Figure 5.4 below. Over the 4 years between
1994 and 1997, the annual average of status-quo attacks jumped all the way up to 20.5 per year.
Furthermore, if we look only at the years of 1995-1997, that number skyrockets even further to
25.3. Clearly, the region of Urabá had experienced the culmination of events that made status-
quo terrorism the violent method of choice. Economically, the region stayed very much the same
as in the previous years in terms of the importance of banana exports and cattle ranching, and the
levels of economic inequality.
However, the Colombian military and the regional elites of Urabá reacted intensely to
influential democratic changes made at the end of the previous period of study. What we see
specifically is that the military reacted to the new constraints put in place by the new constitution
and subsequent institutions developed by the Gaviria administration. Whereas the armed forces
were given practically free reign in Urabá in the late 1980s, if they were to continue to fight their
counterinsurgency without regard for human rights they now had to rely heavily on
paramilitaries. As for the regional elites - banana plantation owners, cattle ranchers, local
politicians, and narco-traffickers - the increasing threat posed by the FARC, and the military’s
growing deference to the paramilitaries increased their willingness and ability to mobilize and
perpetrate violence.
5.6.2 President Pastrana’s Peace - The Last Straw (1998-2003)
By the end of President Samper’s tenure in office, the FARC had made sweeping strides
in their military capabilities, and the amount of territory they controlled or threatened to control.
When President Pastrana was elected into office in 1998, he felt forced to negotiate for peace
with the guerrillas as the best way forward for the government. Just as Betancur and his
139

administration had done in the early 1980s in their negotiations with the FARC, Pastrana would
pursue the peace talks without the precondition that the guerrillas would need to disarm and
demobilize. In fact, he continued to allow the FARC their demilitarized (DMZ) zone, which had
been put in place at the end of the Samper administration. It was in this DMZ where the two
sides agreed not to fight each other, and instead would be the seat of peace negotiations starting
in 1999.
President Pastrana felt so compelled to seek for peace and permit the guerrillas their
DMZ because their military offensive had been so overwhelming. Not only had the FARC
increased their attacks on the armed forces with growing precision and deadliness, but they had
stepped up all forms of violence and extortion against a wide spectrum of Colombian citizens.
To those were added acts of sabotage in the form of attacks on the electricity and
roads infrastructure, the blockage and restriction of movement throughout the
country with illegal road blocks and the expulsion of the State on regional and local
levels by means of attacking police stations in towns, pressuring civilian authorities
to resign their posts and obstructing local and regional elites (GMH, 2016).

Thus, after decades of building its military capabilities, the FARC was finally flexing its strength
- oftentimes at the great expense of the country’s businessmen, cattle ranchers, and politicians.
However, the damage inflicted by the FARC was not exclusive to the regional elites but
extended to ordinary citizens as well - something that would shift public opinion sharply against
the guerrillas and the peace talks as they progressed.
As for the peace talks, the FARC had several goals in mind regarding what such
negotiations with the government could achieve. The FARC and their leaders were explicit in
their expectation that at least three things needed to happen in order for there to be progress in
the talks. These included exchanging members of the armed forces that they had kidnapped for
FARC members that had been imprisoned, a promise on the part of the state to fight against the
140

paramilitaries, and the protection of the DMZ without transgression by the military (GMH,
2016). These preconditions for progress in the negotiations were added to the ideological goals
the FARC would push for in the talks seeking systemic reform in the Colombian government.
However, the talks would be plagued by constant violence between the FARC, the
paramilitaries, and to a lesser extent, the armed forces during the talks. On behalf of the latter
two actors, there was a strong belief that the negotiations - and especially the insistence on the
protection of the DMZ - were ways in which the guerrillas were seeking to strengthen their cause
even further. For example, the FARC maintained training camps for the guerrilla fighters in the
DMZ and kept their kidnapped victims there as well. Eventually, as violence between the armed
actors of the conflict continued to occur outside the DMZ, and progress could not be made on the
substantial issues, the peace talks broke down. The kidnapping of Senator Jorge Eduardo
Gechem Turbay – president of the peace commission of the Senate – by the FARC in 2002
turned out to be the final breaking point - after which President Pastrana ended the peace talks
and sent the military into the DMZ (GMH, 2016).
The military towards the end of the Samper administration and the beginning of the
Pastrana administration was reeling from a series of military setbacks. Just as the FARC had
expanded the amount of Colombian territory under its control, the armed forces were forced to
withdraw from these areas due to substantial losses and the inability to conquer one place and
maintain their presence. Not only was the military experiencing military defeats at the hands of
the FARC, but they were still facing significant backlash from the civilian government over the
severe human rights abuses as part of the government’s sincere attempts to curb such violations
with the passing of the new constitution. Moreover, the military was dead set against the idea of
negotiations with the guerrillas, especially the concession of the demilitarized zone which would
141

allow them respite from military pressure. These tensions between the government and the
military over human rights reached a high point in 1999 when the Minister of Defense Rodrigo
Caicedo resigned from his position - with the support of at least 40 high ranking military officers
- over his dissatisfaction with Pastrana’s overtures to the FARC (GMH, 2016). Once again, it is
apparent that constraining the military in its counterinsurgency efforts leads to agitation
throughout the ranks of the armed forces. Unfortunately, that agitation translated to increased
collaboration with, and dependence on the paramilitary groups unconstrained by such efforts.
For their part, the paramilitaries were also in direct opposition to the idea of Pastrana’s
peace talks with the guerrillas. The regional elites that led, formed, and supported the
paramilitaries:
Regarded the peace policy of the government of Andres Pastrana as a political
arrangement which went against the status quo; they felt that that process
accelerated the growth of the FARC and that an eventual institutionalization of the
political and military power of that guerrilla group would dismantle their power in
the regions. In addition, they saw that their protests against being abandoned by the
State were not being heard, but, on the contrary, it not only enabled the FARC to
attack them but the expansion of that guerrilla was also being encouraged by the
central government in Bogotá (GMH, 2016).

In other words, the Pastrana peace process demonstrated a betrayal of the elites by the new
government. The Pastrana administration was not only bargaining for institutions that would
eventually degrade the power of the regional elites, but it was also negotiating with the FARC
even as they continued their violent campaign throughout the country.
After witnessing the appeal that the UP had exhibited after their formation during the
Betancur administration, and their ability to contest for political office in important rural areas
across Colombia - the regional elites were ready to prevent similar advancements of the
guerrillas and the leftist ideology they fought for. With the threat of any institutions developing
142

out of those peace talks, combined with the military’s growing inability to confront the guerrillas
throughout different regions of Colombia, the paramilitaries evolved further.
In response to such pressures, the regional elites behind the paramilitaries adopted a new
military and political strategy. On the military front, with the success of the ACCU’s model of
violence in Urabá, and the formation of a larger national paramilitary organization - the AUC -
the paramilitaries would engage in an offensive against the guerrillas on a national level. While
the AUC initially had established itself in the northern regions of the country, it would now
begin to execute a violent campaign against the guerrillas and those pushing for reform to the
rural status-quo in the south of Colombia (GMH, 2016). Where the military was unable or
unwilling to put pressure on the guerrillas, the paramilitaries would step in to do the dirty work.
Running parallel to their violent campaign in the south, the paramilitaries also began to kidnap
politicians from the central government in favor of Pastrana’s peace efforts. In May of 1999 and
in November of 2000, the paramilitaries captured several congressmen with statements signaling
their direct protests of the negotiations - even asking for a separate and parallel negotiating
process for the AUC (GMH, 2016).
Since the formation of the new constitution, the growth of the paramilitaries had become
blatantly obvious for all to see. With estimates of fewer than a thousand paramilitary troops
beginning in 1992, that number had risen to more than eight thousand by the year 2000, with
very little done by the government to tackle the problem (Carroll, 2011). However, with the
events of September 11th, 2001 in the United States (Colombia’s biggest military partner), the
government was forced to take a more confrontational approach towards the paramilitaries. Thus,
the political strategy of the AUC came into full effect. In an attempt to become a greater part of
the country’s mainstream,
143

The paramilitaries therefore sought to move from local and regional groups of
armed counter-insurgents to a national movement backed by political, social and
economic resources. That objective implied strengthening their influence on
regional and local politics, no longer in the de facto way of the past, but to
participate in a more or less open manner in elections with candidates of their own
and control over aspects of governance which the State delegated to territorial
agencies, like health and education. (GMH, 2016).

Eventually, this political strategy paid off for the paramilitaries. By 2002 and 2003, individuals
affiliated with the AUC won substantial representation in the local and national elections. This
included control of 250 mayoralties, nine governors’ offices, and the election of 26 senators
(Carroll, 2011). This representation of the paramilitaries and their allies in government, plus the
election of longtime ally President Alvaro Uribe in 2002, helped to pave the way for negotiations
in 2003 over their eventual formal demobilization in 2006 as part of the Justice and Peace Law.
While paramilitary violence continued after the demobilization of the AUC, the groups - known
as bandas criminales (BACRIM) - were made up of slightly different actors, with a much bigger
emphasis on narco-trafficking and maintaining less support from the regional elites they had
previously relied on.
5.6.3 Carlos Castaño Looks to the South
Although Carlos Castaño had declared victory in Urabá for his paramilitary group the
ACCU against the guerrillas in 1997, violence in the following period continued. With the
formation of the AUC, the establishment of the DMZ, and the ongoing peace talks led by the
Pastrana administration, the paramilitaries began to fight a war across Colombia rather than just
solely focused on the northern regions as observed previously. While status-quo terrorist
violence in Urabá declined as a result of this new national outlook, consequences still led to a
relatively high level of such violence in the region nonetheless.
144

Castaño and the AUC still relied on status-quo terrorist violence in Urabá during this
period, albeit to a lesser extent, due to the pressure that the FARC and other groups (like the
dissident ELN) were able to exert on him and other regional elites. Even though the
paramilitaries of the region had just about completely decimated the presence of leftist parties
such as the Communist Party and the UP, and the different banana workers’ unions, the FARC
were determined not to give the region up. From 1998 and onward, the FARC carried out a
variety of operations “aimed at recovering zones that they had dominated before, like the
Antioquia and Choco parts of the Urabá region” (GMH, 2016). Therefore, despite the earlier
claims of victory in the region, the AUC felt compelled to continue its campaign of violence.
Especially telling of the FARC’s continued presence in the region was the attack and
violent capture of a central encampment in Urabá. Carlos Castaño personally claimed that this
attack was a blow for the self-defense movement, and afterwards led to a large amount of
revenge massacres, assassinations, and forced displacements in January 1999 throughout
Colombia - including within Urabá itself (GMH, 2016). This violent campaign led to a
withdrawal by the FARC from negotiations with the government just as a deal was about to be
made regarding prisoner exchanges between the two sides. Apparent here is the transformation
of the conflict in which the status-quo violence in Urabá became one part of a much larger
national level conflict. While the military’s influence on status-quo terrorism had always been
driven by a national counterinsurgency logic, much of the early violence had been driven by
local and regional concerns for the perpetrators of said violence. Now, with a growing threat
from the guerrillas, a government seemingly catering to the FARC, and a military constrained by
peace talks, the paramilitaries would take it upon themselves to bear the brunt of the national
counterinsurgency efforts.
145

How did the events of this time period affect the levels of status-quo terrorist violence?
Figure 5.5 below illustrates the variation in status-quo terrorism over this particular time period.
1998 saw a relatively low level of status-quo terrorism (only five attacks), as this is the year after
Carlos Castaño had declared victory in Urabá. However, recall the FARC’s efforts to recapture
territory in the region, and subsequent raids against the AUC themselves in December of that
year. That renewed threat from the left, and the attempts by the AUC to spoil the peace process
between the FARC and the government, would lead to another spike in status-quo violence over
the following three years (1999-2001), with an average of 14.3 status-quo attacks per year.
Finally, with the breakdown in peace talks, and the election of president Uribe - who empowered
the military and negotiated the demobilization of the AUC - status-quo terrorism dropped again
to an average of 5.5 attacks per year between 2002-2003. Overall, compared to the 1980s and
early 1990s, Urabá experienced a high level of status-quo terrorism, albeit a bit less than what
was observed in the mid-1990s.
5.7 Analysis
With the comparisons and process tracing completed for both time periods, the focus now
is to summarize the findings from this chapter; providing an analysis of the qualitative study and
how it fits within the overall research project. The purpose of this chapter was to explore -
through a concrete historical case - the processes through which status-quo terrorist violence
varies over time according to different social and political stimuli. While the study has surely
demonstrated that the strength of the insurgent threat increased status-quo terrorism in Colombia
over time, it has also highlighted the ways in which certain democratic tendencies have had a
similar and compounding effect on such nonstate violence. Figure 5.6 below gives a great visual
146

representation of the activity of status-quo groups over the different cases under analysis in this
chapter.
In the initial instance or case of status-quo terrorism in the region of Urabá, we observe
moderate to low levels of such violence. The early-to-mid-1980s was characterized by a sudden
shift in policy from the national government; one that tried to reign in the autonomy of the
military and sought a peaceful resolution to the insurgency with promises of democratic reforms
to allow the left greater representation - especially at the local level. This significant shift in how
to counter the Colombian insurgency frustrated hardline members of the military and those
regional elites that had the most to lose from the new democratic reforms. Members of the
security forces then looked to ally with self-defense groups as a way to get around the cease-fires
with the guerrillas. While the military still violated the human rights of the guerrillas and their
sympathizers during this time, due to their constraints and the willingness to fight on the part of
key regional actors, the monopoly on state violence began to voluntarily break down. As a result,
we observe moderate levels of status-quo terrorist violence in the region.
After the breakdown of Betancur’s peace talks with the FARC in 1984, the mid-to-late
1980s witnessed a new attempt at negotiations with Colombia’s other guerrilla groups. However
this time there were key preconditions to such talks, and the guerrillas interested in peace talks
were required to disarm and demobilize before they were allowed to form a political party and
join the political process. The M-19 agreed to those conditions, and won the second most seats
on the Constituent Assembly tasked with the formation of a new Colombian Constitution. Soon
after, Urabá’s biggest guerrilla group at the time - the EPL - disarmed and demobilized in order
to join Colombia’s new system of government. Most importantly for Urabá, the late 1980s
experienced a massive military buildup in the region, where the military was once again given
147

free rein to fight the insurgents in a draconian manner. Consequently, the late 1980s and early
1990s saw very low levels of status-quo terrorist violence.
However, in the second instance - or case - of status-quo terrorism violence in Urabá, we
observe relatively high levels of such violence. Whereas the military still retained much of its
autonomy in the first case, the new constitution of 1991 and consequent laws enacted gave the
civilian government the responsibility of ensuring public order. Furthermore, new institutions
were put in place to scale back the human rights violations perpetrated by the military - making
the fight against insurgents who blended in with the population exceptionally difficult for them.
At the same time, the new democratic competition and attempts by the left to push for pro-labor
laws at the local level pushed the regional elites to partner with illegal paramilitary groups such
as the Castaño’s ACCU. On top of all of this, even with the demobilization of the EPL, the threat
of the FARC became more severe during this period. The outcome was a giant spike in status-
quo terrorism, as the military was more than willing to allow the ACCU to take the lead in
fighting the FARC and its political supporters throughout the region.
The late 1990s to early 2000s also experienced a high level of status-quo terrorism,
although events in 2002 and 2003 led to the gradual tapering off of such violence. This period
experienced another attempt at peace negotiations with the FARC without the precondition for
disarmament or demobilization. Even more, the government had agreed to the establishment of a
demilitarized zone as the seat of negotiations, a stronghold of the FARC where the military could
not operate. This concession by the government was due to the ongoing violent campaign of the
FARC and their significant territorial expansion. However, president Pastrana truly believed the
best way to solve the issue of violence was to bring the FARC and its supporters into the formal
148

political system – willing to reform the Colombian system of government further in order to
achieve such buy-in.
Although Carlos Castaño and the ACCU had declared victory in Urabá in 1997, a
subsequent push by the FARC in the region would pressure the AUC to retaliate in kind.
Furthermore, with the military constrained yet again by ongoing peace talks, much of the
national counterinsurgency efforts would fall to the paramilitaries - something they could not
achieve without maintaining the defense of their base of operations in Urabá. Not until the
election of long time paramilitary-ally president Uribe did the military regain its autonomy.
Consequently, many of the paramilitary organizations willing to demobilize, leading to a
tapering off of status-quo terrorist violence at the end of the case.
5.8 Summing Up
Thus, this qualitative study fits into the overall dissertation by allowing the research
project to benefit from the comparative advantages of a mixed-methodological approach. As was
seen in the quantitative chapter, a cross-national analysis found a significant and positive
relationship between democracy and status-quo terrorist violence. This comparative case study
further contributes evidence to the idea that the more democratic a state is, the more status-quo
terrorism it experiences. By studying the same region over multiple time periods, the design
attempted to isolate the effect that the democratic institutions of the new Colombian constitution
had on subsequent status-quo terrorism.
However, potentially even more important, is the evidence surrounding the causal
mechanisms inherent to such a causal relationship. The process tracing in this chapter seeks to
illuminate those mechanisms by examining the logic and reactions of specific actors that are
important for understanding the occurrence of status-quo terrorist violence. The finding that can
149

be taken away from this process tracing is that in the face of some perceived threat to the status-
quo, there are key actors who feel frustrated and disenfranchised by certain democratic
tendencies. More specifically, it is the reactions of key individuals and groups within the military
and club of regional elites to such stimuli that were responsible for the varying rates of terrorist
incidents within Urabá.
With this in mind, the dissertation has progressed through both the quantitative and
qualitative tests. Utilizing a mixed-methods approach, this research project has been able to shed
light on the primary question at hand in a manner that would be otherwise impossible using only
one method or the other. In the following chapter, the work accomplished thus far will be
summarized, offering a final prognosis as to how the theoretical model developed herein
represents the empirical evidence of status-quo terrorist violence.

150

Tables and Figures

Table 5.1
Causal Mechanisms

Explanatory Variable Intermediate Factor Dependent Variable
Democracy Security forces restrained More status-quo terrorism
Democracy Hard-line citizens frustrated More status-quo terrorism
Insurgents strength/tactics Hard-liners motivated to fight More status-quo terrorism

151

Figure 5.1
Map of Urabá

152

153

154

155

156

157

Chapter 6
A Summary on Status-Quo Terrorism
6.1 Tying it All Together
The reader now has a better grasp of what makes the forces of status-quo terrorism tick.
The literature in the field of terrorism studies has been reviewed, emphasizing the serious need
for an in-depth analysis of terrorist violence utilized to protect the system, rather than tear it
down. This dissertation has attempted to do just that, by developing a theoretical framework for
the occurrence of such violence, and then testing for its propositions through a mixed-
methodological approach. Much in the way of new knowledge can be taken from this project.
Yet a study is important not only for the knowledge it produces, but also for the way in which it
points scholars towards new avenues of research. I hope this dissertation does just that by
forming an important building block within the field of terrorism studies, and by contributing to
the field’s more general understanding of human conflict.
This chapter serves to summarize the route this dissertation took to achieve such goals. It
is structured so as to emphasize what was accomplished throughout each chapter, offering brief
conclusions regarding key concepts and findings. Furthermore, a more synthetic synopsis of the
entire research project is offered; seeking to answer the question of what has this dissertation
accomplished in a broader, more overall sense. In the next section of this chapter, I spend time
discussing thoughts about future research - about how this project may point myself and other
scholars towards new and important questions. What is offered here are preliminary ideas, and
other intellectuals will surely take these findings in different directions than I. Lastly, this chapter
concludes with a few final thoughts and reflections on the dissertation as a whole - what the
findings from this research project teach us about political conflict and violence more generally.
158

6.2 Summaries
This section seeks to briefly recap the content of each chapter. Along with a brief
summary of each chapter, a point will be made to emphasize what was accomplished in a
broader sense as well.
6.2.1 Chapter 1
Anecdotal cases of terrorist violence throughout Northern Ireland during the Troubles are
put forward in Chapter 1. An eye towards theory development played an important role in
highlighting the need to distinguish the violence between the Catholic extremists - generally
found in groups such as the IRA or the INLA - and Protestant extremists - generally found in
groups such as the UVF or the UDA. The idea put forward here was that the field of terrorism
studies has long suffered from the problem of conceptual stretching. I had gone to Northern
Ireland to study the effects of democracy on terrorist violence, but soon realized that democracy
and its institutions surely affected the violence from each side in a different manner. In essence,
Chapter 1 was utilized to highlight a research dilemma and why it is important to learn more
about it in a systematic and scientific way.
6.2.2 Chapter 2
The next chapter in the dissertation reviewed the general literature that has been produced
surrounding this newly introduced research dilemma. The chapter starts by addressing the ever-
important issue of how one defines what terrorism, or a terrorist attack, really is. While this
question has been tackled by hundreds of scholars, understanding the nuances of the definition
are quite vital to nailing down what status-quo terrorism is, and why it is separate and distinct
from state terrorism and anti-state terrorism. Next, through a review of the literature on
democracy’s relationship with state terror and terrorism more broadly, important gaps in those
159

works are highlighted. These gaps are based around the issues that come with conceptual
stretching, or failing to recognize key distinctions within the phenomenon they are trying to
better explain. In the literature on state terror, the issue is that the scholars treat violence by the
formal state apparatus and shadowy death squads not formally part of the state as the same
phenomenon - a key distinction that this dissertation seeks to underscore. Similarly, much of the
literature on non-state terrorism fails to distinguish between terrorist violence against the state
and status-quo, and terrorist violence in defense of it. This leads to the ambiguous relationship
between democracy and terrorist violence in general, as the former’s influence may depend on
the type of terrorism under analysis. All together, this chapter identifies some of the potential
failings in the previous literature and why a research project focused specifically on status-quo
terrorism can fill in for the gaps.
6.2.3 Chapter 3
The third chapter of the dissertation worked towards developing a theoretical framework
to explain status-quo terrorist violence. This framework takes as its jumping off point that, for
such violence to take place, something must be so perceivably threatening to the status-quo that
certain individuals are willing to fight to protect it. As an active insurgency or violent internal
conflict seeking to overtake the government is conceptually high on any continuum measuring
such a threat, the theory is developed with that type of danger in mind. Its logic and consequent
propositions however, do not apply only to cases in which there is some violent internal conflict,
and instead are consistent with any perceived threat to the status quo.
The theory ultimately proposes that democracies are more likely to experience status-quo
terrorist violence than their non-democratic counterparts. This proposition rests upon the idea
that democracies possess institutions and tendencies that constrain important actors in their
160

ability to counter a challenge to the state, especially in a hardline or more draconian manner.
Such constraints push certain members of the security forces to look for alternative methods
through which they can implement the more repressive counterinsurgency strategies that they
prefer. Furthermore, democratic responses to such challenges may rely on utilizing their unique
institutions - such as political parties and elections - to entice the challengers into pursuing their
goals through the legal and political process. These democratic tendencies then create the perfect
storm for a military with their hands tied to cooperate with, and facilitate the violence of,
disillusioned elites throughout a country. Therefore, not only does the theory chapter propose a
specific association between democracy and status-quo terrorism, but it assembles specific
mechanisms through which the causal process takes place.
6.2.4 Chapter 4
With the theory development complete, Chapter 4 set out to create an appropriate cross-
national quantitative analysis of the specified theoretical propositions. It accomplished this by
first setting out to describe the data collection process, the reasoning for the inclusion of an array
of important control variables, and explaining why the different models would take the form of a
zero-inflated negative binomial (ZINB) model. Thereafter, evidence was provided to support the
assumption that country-years experiencing some type of violent internal conflict were most
prone to experience status-quo terrorist violence; while the vast majority of country-years in the
sample had not experienced an insurgency, the vast majority of status-quo terrorist attacks took
place in country-years experiencing an insurgency.
Moreover, the results of the six separate ZINB models were presented. Most noticeably,
in all six of those models democracy was positively correlated with status-quo terrorist attacks,
as was hypothesized; in one of those models, however, the association between the two missed
161

out on statistical significance very narrowly. The results of the tests provide robustness checks to
alleviate any concerns that the findings depended on the specific operationalization of
democracy. To do so, one dichotomous and two interval-level measures of democracy were
utilized. The one finding that really stands out from the analysis is that status-quo terrorist
attacks are predicted to increase at each increasing level of democracy. This is inconsistent with
the findings from other terrorism studies where the relationship often takes the shape of an
inverted U - signaling that weak democracies are most prone to terrorist violence in general. This
provides further evidence of the need to disaggregate the concept of terror.
6.2.5 Chapter 5
The following chapter pivots from the quantitative analysis to conduct an in-depth case
study on varying levels of status-quo terrorism throughout the Urab á region of Colombia. This
transforms the dissertation into a mixed-methodological study, giving the overall project a higher
degree of causal leverage over the question at hand. It accomplishes that goal by process-tracing
for distinct causal mechanisms, and utilizing a temporal comparative analysis.
The comparative case study traces why the levels of status-quo terrorist attacks vary
between two distinct time periods in Colombian history. The first period under study is 1982-
1993, and in that time, Urabá experienced relatively low levels of status-quo terrorism. The
second period is from 1994-2003, where the region experienced a huge spike in the levels of
such violence. Since the two cases measure the same region over time, many potential
confounding variables are controlled for by the design itself. The main change between the two
periods is the signing of a new Colombian constitution, and the several institutions put in place
soon afterward. Through process-tracing, the democratic tendencies to constrain the military and
to offer political opportunities for the insurgents were revealed to play an important causal role
162

in the actions of key violent status-quo actors. Between the hardliners within the military and the
regional elites (businessmen, cattle ranchers, and political class), it was evident that the
frustrations they had with these democratic impulses led to their cooperation and ultimate
campaign of status-quo terrorism.
6.2.6 Synthesis
What are the accomplishments of this dissertation in an overall synthetic sense? I believe
the major takeaways are twofold. First, this study has clearly demonstrated the importance of
disaggregating the concept of terrorist violence - not all terrorism is the same. Second, this
research project also takes an important first step in providing a well-rounded analysis into the
determinants of status-quo terrorism.
As for the first takeaway, the emphasis has been to highlight the danger of conceptual
stretching. As introduced by the anecdotal case of terrorist violence in Northern Ireland, and later
expanded upon by the case study of Colombia, the motivations for violence by anti-state groups
(such as the FARC) are clearly different from those of groups interested in maintaining the
status-quo (such as the AUC). When specifically theorizing and testing for democracy’s effect on
terrorist violence, it is crucial that scholars think more in depth about the type of terrorist
violence they are wanting to measure. For example, recall from Chapter 2 that the strategic
school of terrorism studies claims that the constraints of democracy increase terrorist violence by
making those states into soft or weak targets. However, as is evident from the findings herein,
the mechanism through which democracy is associated with status-quo terrorism is through the
agitation of the military and hardline elites, not because they perceive the government as an easy
target to attack. The same rings true for studies of state terrorism - this dissertation has
demonstrated a need to distinguish grave human rights abuses perpetrated by members of the
163

state’s security forces, and that same violence perpetrated by non-state groups. Democracy has
clearly been shown to alter the military’s ability to commit such abuses, and their willingness to
create shadowy alliances with non-state allies.
Regarding the second takeaway, there is now established a first look into status-quo
terrorism and the key determinants which influence its intensity. Many scholars will now be able
to build off of the knowledge developed throughout this study. Most specifically, the design of
the dissertation has facilitated the search for an association between democracy and status-quo
terrorism, but also the specific causal mechanisms through which this relationship takes place.
While many of the studies that have looked specifically at violence from death squads or
paramilitary groups have analyzed the variables that affect these groups and individuals directly,
this dissertation makes it a point to bring in the interests and decision-making process of the state
itself as a clear contributor to status-quo terrorist violence.
6.3 Future Research
While the accomplishments of the dissertation are exciting, the ways in which the project
can lead to new research are just as important. As many articles and books have sparked
discussion before, questions may arise regarding this work as to how to further refine the results,
how other potentially confounding variables might come into play, or how generalizable the
findings are to a wide array of seemingly dissimilar cases. Provoking these questions, and
finding answers to them, can help to establish a new sub-field of research with critical findings
for a range of issues relating to political violence. This section is dedicated towards suggesting a
few avenues through which I, or other scholars, can refine and build upon this dissertation in
future research.
164

One avenue for future research as it relates to the agenda put forward by this dissertation
is to look more in-depth at types of democracy, more specificically democratic institutions, and
how these affect status-quo terrorist violence. Now armed with the knowledge that democracy in
general is associated with such violence, it is important to further disaggregate the concept of
democracy to look for differential effects. For example, one particularly fruitful research path to
take would be to take a closer look at any potential differences between presidential and
parliamentary democracies and their relationship with status-quo terrorism. Another potential
analysis could involve the electoral institutions of a democracy - for example whether a country
has a proportional representation or first past the post voting system. There is a long list of
variables that may lead to variation in status-quo terrorist attacks within democracies that should
be explored.
Another important issue to be tackled in future research is the issue of the data itself. As
was discussed in the data section of Chapter 4, the data for status-quo terrorism was taken from
the Global Terrorism Database (GTD). This is necessary for any cross-national study of status-
quo terrorism as it is currently the only large scale database on terrorism that includes domestic
terrorist attacks for all years, and allows researchers the ability to separate domestic attacks from
international attacks. However, the GTD does not codify terrorist attacks by their goals, or
whether each attack is directed against the state or in defense of the status-quo. Instead, recall
that this dissertation uses an actor-centric model that relies on the name of the perpetrator,
classifying that group as status-quo or not, and then looking at all of the terrorist attacks by
status-quo groups. Although this is the best way of analyzing status-quo terrorist attacks
currently, two major issues arise from this.
165

The first, which is extremely difficult to solve, is the fact that every single attack by a
“status-quo group” is not necessarily aimed at protecting the status-quo. There are plenty of
instances of status-quo groups feuding with each other for example. However, without
reexamining every single terrorist attack in the database and reclassifying each attack based on
its intent, there is no resolution to this. The second issue relates to the actual coding of the
perpetrators themselves. As mentioned at the end of Chapter 4, over 50,000 of the over 100,000
terrorist attacks are perpetrated by unknown groups, and thus we are potentially missing out on
large amounts of status-quo terrorist attacks. One way in which to solve this issue is to replicate
the type of qualitative case study conducted in Chapter 5, where country level data is oftentimes
much more detailed than the data collected in a large cross-national database like the GTD.
There are even more reasons to conduct further qualitative research on status-quo
terrorism than issues of data. While quantitative analyses such as the ones done in this
dissertation provide great support for the statistical relationships between hypothesized variables
over time and across a large variety of states, they must be reinforced by methodologies capable
of teasing out the causal mechanisms between such variables. The case study conducted on
Colombia in this dissertation is a great first step, but without other studies of other countries, it is
impossible to know whether the mechanisms found in the former will apply to the latter. Instead,
as the field generates more qualitative research regarding status-quo terrorism, a greater degree
of certainty can be gained regarding the true causal processes behind such violence, and how
broadly applicable the theory developed here really is.
Besides the finding that democracy is associated with higher levels of status-quo terrorist
violence, this dissertation serves to generate future lines of research in connection with the
evidence from the process-tracing in the Colombian case study. As a by-product of this study,
166

new propositions are ripe for testing. Two avenues of research come to mind. One would be a
closer examination of the ways in which different regime types decide on when and how to
negotiate with insurgent groups. One of the major assumptions of this dissertation is that it is a
democratic tendency to negotiate with insurgent groups due to their ability to utilize unique
carrots, such as the promise of new political parties or electoral institutions. However, it would
be very fruitful to conduct some sort of cross-national analysis of the frequency of such
negotiations by regime type. Furthermore, in accordance with the frustrations that status-quo
actors had with negotiations without preconditions in particular, it would also be interesting to
see how different regime types tend to negotiate with insurgent groups. This line of research
would provide further knowledge as to how democracies are conditioned to resolve conflicts
peacefully.
The other line of research, already burgeoning, would be the linkage between state
security forces and violent non-state actors. There is plenty of work done in this field, especially
regarding state security forces fostering relationships with violent non-state actors in other
nations - knowledge that is relevant to issues of proxy wars and internationalized civil conflicts.
More specifically, following the findings of this study, there could be a renewed focus on the
linkage between such groups in a more domestic and democratic context. In essence, more
research should be done to explore who the military tends to establish and maintain relationships
with inside of democratic countries. Research questions might include what common interests
tend to drive such relationships, how consistent are these relationships, or what causes these
linkages to fray? This future research agenda would give scholars a more in-depth of the types of
domestic groups that the military turns to in the midst of a crisis, among a range of other
interesting findings.
167

Lastly, while this dissertation has attempted to add in certain international-level control
variables in the modeling of what causes status-quo terrorism, important work remains to be
done in this regard. Certainly, the motivations for the violence under consideration is primarily a
domestic affair, influenced by political and socioeconomic conditions at home. However, as the
world has become an increasingly globalized system of states, domestic affairs are increasingly
subject to international pressures. For example, what are the effects of large international media
companies, non-governmental organizations, and transnational advocacy networks in uncovering
the abuses of state security forces across the world? As their ability to do so increases, it is
possible that militaries are more and more willing to delegate repressive techniques so as to
maintain plausible deniability. Or, is it the opposite, and international attention uncovers such
linkages between the state and paramilitary groups - making it less likely that violence by the
latter is considered to be ‘non-state’ in nature. Even further, it might be particularly illuminating
to look at the effects of economic interdependence interacted with democracy, or even
involvement in key international governmental organizations on status-quo terrorism. Achieving
a better understanding of the key international pressures on such violence is an important next
step in producing well-rounded knowledge regarding status-quo terrorism.
6.4 Final Reflections
This final section is meant to provide a sense of how the dissertation and research within
has contributed to our knowledge of human behavior within political science. More specifically,
this project has facilitated our goal of understanding human conflict in general. One of the keys
to unlock such understanding has always been located in the study of politics, of how groups
compete with one another over the power to govern. As has been seen throughout history, many
168

of those who compete for such power do so with the protection of their own self-interest and
preservation foremost in their minds.
The findings contained in this dissertation help to uncover just how violent some groups
and individuals are willing to become in protection of the status-quo, especially when the
consequences of politics threaten such powerful interests. This project illustrates the breakdown
of the state’s monopoly on legitimate violence. The most surprising part of this breakdown is that
it is a somewhat voluntary transfer of the responsibility for violence from the security forces to
individuals not formally part of the state.
Finally, this dissertation reveals the very troubling effects that democracy may have on
violence. This is troubling because democracy, as a certain set of institutions, is linked with so
many positive outcomes for a wide range of actors that may otherwise be neglected or punished
in non-democracies. By no means are the findings from this dissertation meant to be some type
of repudiation of democracy. Instead, the study is meant to call attention to the potential violent
and negative effects of particular democratic tendencies that come about when there are
pressures to alter the status-quo. Equipped with such knowledge, democratic governments have
the capacity to make changes in pursuance of the peaceful resolution to conflict. This is highly
valuable because the only consistent thing about the status-quo, is that at some point, it is bound
to change.

169

Bibliography
Abrahms, Max. "Why democracies make superior counterterrorists." Security Studies 16.2
(2007): 223-253.
Aksoy, Deniz. "Electoral and Partisan Cycles in Counterterrorism." The Journal of Politics 80.4
(2018): 1239-1253.
Aksoy, Deniz, and David B. Carter. "Electoral institutions and the emergence of terrorist
groups." British Journal of Political Science (2014): 181-204.
Ash, Konstantin. "Representative democracy and fighting domestic terrorism." Terrorism and
Political Violence 28.1 (2016): 114-134.
Avilés, William. "Paramilitarism and Colombia's low-intensity democracy." Journal of Latin
American Studies 38.2 (2006): 379-408.
Barco, Virgilio. “Presenting the Peace Initiative to the Country.” 1 Sep. 1988, Colombia.
Televised Speech.
Berger, Miriam. “What Is Iran’s Revolutionary Guard Corps That Soleimani Helped to Lead?”
The Washington Post, WP Company, 4 Jan. 2020,
www.washingtonpost.com/world/2020/01/04/what-is-irans-revolutionary-guard-corps-
that-soleimani-helped-lead/.
Boix, Carles. "Boix-miller-rosato dichotomous coding of democracy, 1800-2015." Harvard
Dataverse 3 (2018).
Bruin, J. 2006. newtest: command to compute new test. UCLA: Statistical Consulting Group.
https://stats.idre.ucla.edu/stata/ado/analysis/.
Campbell, Bruce B. "Death squads: Definition, problems, and historical context." Death squads
in global perspective. Palgrave Macmillan, New York, 2000. 1-26.
170

Carey, Sabine C., Neil J. Mitchell, and Will Lowe. “States, the security sector, and the monopoly
of violence: A new database on pro-government militias.” Journal of Peace Research
50.2 (2013): 249-258.
Carey, Sabine C., Michael P. Colaresi, and Neil J. Mitchell. "Governments, informal links to
militias, and accountability." Journal of Conflict Resolution 59.5 (2015): 850-876.
Carroll, Leah. "Violent democratization." Social Movements, Elites, and Politics in Colombia’s
Rural War Zones (2011).
Chenoweth, Erica. “Democratic competition and terrorist activity.” The Journal of Politics 72.1
(2010): 16-30.
Cingranelli, David L., and David L. Richards. "The Cingranelli and Richads (CIRI) human rights
data project." Human Rights Qaurterly 32 (2010): 401.
Civico, Aldo. The Para-state: An Ethnography of Colombia’s Death Squads. University of
California Press, 2016.
Collier, Paul. Wars, guns and votes: Democracy in dangerous places. Random House, 2011.
Colombian Counterinsurgency: Steps in the Right Direction, Central Intelligence Agency, 1994.
Colombia: A Violent Backdrop, U.S. Department of State. Bureau of Intelligence and Research,
1998.
Conrad, Courtenay R., and Jacqueline HR DeMeritt. "Constrained by the bank and the ballot:
Unearned revenue, democracy, and state incentives to repress." Journal of Peace
Research 50.1 (2013): 105-119.

171

Coppedge, Michael, John Gerring, Carl Henrik Knutsen, Staffan I. Lindberg, Svend-Erik
Skaaning, Jan Teorell, David Altman, Michael Bernhard, Agnes Cornell, M. Steven Fish,
Haakon Gjerløw, Adam Glynn, Allen Hicken, Joshua Krussel, Anna Lührmann, Kyle L.
Marquardt, Kelly McMann, Valeriya Mechkova, Moa Olin, Pamela Paxton, Daniel
Pemstein, Brigitte Seim, Rachel Sigman, Jeffrey Staton, Aksel Sundstrøm, Eitan Tzelgov,
Luca Uberti, Yi-ting Wang, Tore Wig, and Daniel Ziblatt. 2018. “V-Dem Codebook v8”
Varieties of Democracy (V-Dem) Project. https://doi.org/10.23696/vdemcy18
Crenshaw, Martha. Revolutionary Terrorism: The FLN in Algeria, 1954-1962. Vol. 196. Hoover
Institution Press, Stanford University, 1978.
Cullen, Louise and Michael Fitzpatrick, “Enniskillen Poppy Day bomb memorial unveiled,” 8
November 2017, accessed at https://www.bbc.com/news/uk-northern-ireland-41899718,
28 May 2021.
Dalacoura, Katerina. “Islamist terrorism and the Middle East democratic deficit: Political
exclusion, repression and the causes of extremism.” Democratization 13.3 (2006): 508-
525.
Davenport, Christian. State repression and the domestic democratic peace. Cambridge
University Press, 2007.
Davenport, Christian, and David A. Armstrong. "Democracy and the violation of human rights:
A statistical analysis from 1976 to 1996." American Journal of Political Science 48.3
(2004): 538-554.
Davis, Dickie, David Kilcullen, Greg Mills, and David E. A Great Perhaps?: Colombia: Conflict
and Convergence. Hurst, 2016.
De Mesquita, Bruce Bueno, et al. The logic of political survival. MIT press, 2005
172

Doc. 9. Update on Colombian Army Relations with Paramilitary Organizations. Defense
Intelligence Agency, 1996
Doc. 11. Colombian Autodefensas: History and Organization. Defense Intelligence Agency,
2000.
Drakos, Konstantinos, and Andreas Gofas. “The Devil You Know but Are Afraid to Face
Underreporting Bias and its Distorting Effects on the Study of Terrorism.” Journal of
Conflict Resolution 50.5 (2006): 714-735.
Dudley, Steven. Walking ghosts: murder and guerrilla politics in Colombia. Routledge, 2004.
Durán, Mauricio García, ed. Alternatives to war: Colombia's peace processes. Conciliation
Resources in collaboration with CINEP, 2004.
Edwards, Aaron. UVF: behind the mask. Merrion Press, 2017.
Enders, Walter, Todd Sandler, and Khusrav Gaibulloev. “Domestic versus transnational
terrorism: Data, decomposition, and dynamics.” Journal of Peace Research 48.3 (2011):
319-337.
Enders, Walter, and Todd Sandler. "Distribution of transnational terrorism among countries by
income class and geography after 9/11." International Studies Quarterly 50.2 (2006):
367-393
Enders, Walter, Todd Sandler, and Khusrav Gaibulloev. “Domestic versus transnational
terrorism: Data, decomposition, and dynamics.” Journal of Peace Research 48.3 (2011):
319-337.
Engene, Jan Oskar. Terrorism in Western Europe: Explaining the trends since 1950. Edward
Elgar Publishing, 2004.

173

Eubank, William, and Leonard Weinberg. “Terrorism and democracy: Perpetrators and victims.”
Terrorism and Political Violence 13.1 (2001): 155-164.
Evans, Michael. “The United States vs. Rito Alejo del Río.” The National Security Archive,
Washington, D.C. (2010)
Evans, Michael. “The Chiquita Papers.” The National Security Archive, Washington, D.C.
(2011)
Eyerman, Joe. “Terrorism and democratic states: Soft targets or accessible systems.”
International Interactions 24.2 (1998): 151-170.
Fearon, James D., and David D. Laitin. “Ethnicity, insurgency, and civil war.” American
political science review 97.1 (2003): 75-90.
Findley, Michael G., and Joseph K. Young. "Terrorism, democracy, and credible commitments."
International Studies Quarterly 55.2 (2011): 357-378
Foster, Dennis M. "Inter Arma Silent Leges? Democracy, Domestic Terrorism, and Diversion."
Journal of Conflict Resolution 61.7 (2017): 1371-1400.
George, A. L. and Bennett, A. (2005). Case Studies and Theory Development in the Social
Science. Cambridge: MIT Press.
Gerring, John. Case study research: Principles and practices. Cambridge university press, 2006.
Gerring, John. "The case study: what it is and what it does." The Oxford handbook of
comparative politics. 2007.
Gleditsch, Nils Petter, et al. “Armed conflict 1946-2001: A new dataset.” Journal of peace
research 39.5 (2002): 615-637.
GMH. BASTA YA! Colombia: Memories of War and Dignity, Bogotá. CNMH, 2016.

174

Gourevitch, Peter. "The second image reversed: the international sources of domestic politics."
International organization 32.4 (1978): 881-912.
Greenhill, Brian. "The company you keep: International socialization and the diffusion of human
rights norms." International studies quarterly 54.1 (2010): 127-145.
Greysteel ‘Trick or Treat’ Massacre 20 Years On (2013, October 30), BBC. Retrieved from
https://www.bbc.com/news/uk-northern-ireland-24736060
Gurr, Ted Robert. "The political origins of state violence and terror: A theoretical analysis."
Government violence and repression: An agenda for research 45 (1986).
Henderson, Conway W. "Conditions affecting the use of political repression." Journal of
Conflict Resolution 35.1 (1991): 120-142.
Hoffman, Bruce. "Insurgency and counterinsurgency in Iraq." Studies in Conflict & Terrorism
29.2 (2006): 103-121.
Jones, Adam. Genocide: A comprehensive introduction. Routledge, 2010.
Jones, Seth G., and Martin C. Libicki. How terrorist groups end: Lessons for countering al
Qa’ida. Rand Corporation, 2008.
Kalyvas, Stathis N. The logic of violence in civil war. Cambridge University Press, 2006.
Kim, HeeMin, Jenifer Whitten-Woodring, and Patrick James. "The Role of Media in the
Repression–Protest Nexus: A Game-theoretic Model." Journal of Conflict Resolution
59.6 (2015): 1017-1042.
Krieger, Tim, and Daniel Meierrieks. "Terrorism in the worlds of welfare capitalism." Journal of
Conflict Resolution 54.6 (2010): 902-939.
Kydd, Andrew H., and Barbara F. Walter. "The strategies of terrorism." International security
31.1 (2006): 49-80.
175

LaFree, Gary, and Laura Dugan. “Introducing the global terrorism database.” Terrorism and
Political Violence 19.2 (2007): 181-204.
Li, Quan. “Does democracy promote or reduce transnational terrorist incidents?.” Journal of
Conflict resolution 49.2 (2005): 278-297.
Levy, Jack S. "Case studies: Types, designs, and logics of inference." Conflict management and
peace science 25.1 (2008): 1-18.
Mapping Militant Organizations. “Revolutionary Armed Forces of Colombia.” Stanford
University. Last modified July 2019.
https://cisac.fsi.stanford.edu/mappingmilitants/profiles/revolutionary-armed-forces-
colombia-farc.
Mason, T. David, and Dale A. Krane. "The political economy of death squads: Toward a theory
of the impact of state-sanctioned terror." International Studies Quarterly 33.2 (1989):
175-198.
Marshall, Monty G., Ted Robert Gurr, and Keith Jaggers. "Polity IV project: Political regime
characteristics and transitions, 1800–2016. Dataset Users’ Manual. Center for Systemic
Peace." (2017).
Mazzuca, Sebastián L., and Gerardo L. Munck. "State or democracy first? Alternative
perspectives on the state-democracy nexus." Democratization 21.7 (2014): 1221-1243.
McCamant, John F. "Governance Without Blood: Social Science's Antiseptic View of Rule; Or,
The Neglect Of Political Repression'." The state as terrorist (1984): 11-42.
Midlarsky, Manus I. The killing trap: Genocide in the twentieth century. Cambridge University
Press, 2005.

176

Mitchell, Neil J., Sabine C. Carey, and Christopher K. Butler. "The impact of pro-government
militias on human rights violations." International Interactions 40.5 (2014): 812-836.
Monogan III, James E. Political analysis using R. Springer, 2015.
Nicholson, Michael. "Conceptual problems of studying state terrorism." Government violence
and repression: an agenda for research (1986): 27-44.
Poe, Steven C., and C. Neal Tate. "Repression of human rights to personal integrity in the 1980s:
A global analysis." American political science review (1994): 853-872.
Romero, Mauricio. "Changing identities and contested settings: Regional elites and the
paramilitaries in Colombia." International Journal of Politics, Culture, and Society
(2000): 51-69.
Romero, Mauricio. "Reform and reaction: the paramilitary groups in Colombia." Irregular
Armed Forces and Their Role in Politics and State Formation (2003).
Rule, James B. Theories of civil violence. Univ of California Press, 1989.
Rummel, Rudolph J. "Democracy, power, genocide, and mass murder." Journal of Conflict
Resolution 39.1 (1995): 3-26.
Sartori, Giovanni. "Concept misformation in comparative politics." The American political
science review 64.4 (1970): 1033-1053.
Savun, Burcu, and Brian J. Phillips. "Democracy, foreign policy, and terrorism." Journal of
Conflict Resolution 53.6 (2009): 878-904.
Schmid, Alex P. "Terrorism and democracy." Terrorism and Political Violence 4.4 (1992): 14-
25.
Schmid, Alex P., and Albert J. Jongman. "Political Terrorism: A New Guide to Actors." Authors,
Concepts, Data Bases (1988).
177

Sil, Rudra, and Peter J. Katzenstein. Beyond paradigms: Analytic eclecticism in the study of
world politics. Macmillan International Higher Education, 2010
Sloan, John W. "State repression and enforcement terrorism in Latin America." The State as
Terrorist: The Dynamics of Governmental Violence and Repression (1984): 83-98.
“South America: Colombia.” The World Factbook, Central Intelligence Agency, 20 June 2021,
https://www.cia.gov/the-world-factbook/countries/colombia/.
Stanton, Jessica A. "Regulating militias: Governments, militias, and civilian targeting in civil
war." Journal of Conflict Resolution 59.5 (2015): 899-923.
Steele, Abbey, and Livia I. Schubiger. "Democracy and civil war: The case of Colombia."
Conflict Management and Peace Science 35.6 (2018): 587-600.
Stohl M, Lopez GA. The State as Terrorist : the Dynamics of Governmental Violence and
Repression . Greenwood Press; 1984.
Stohl M, Lopez GA. Government Violence and Repression : an Agenda for Research .
Greenwood Press; 1986.
Stuart Mill, John. "A system of logic." Raciocinative and inductive (1843).
Sutton, Malcolm, ed. Bear in Mind these Dead: an Index of Deaths from the Conflict in Ireland,
1969-1993. Beyond the Pale Publications, 1994.
Teune, Henry, and Adam Przeworski. The logic of comparative social inquiry. New York:
Wiley-Interscience, 1970.
Tourish, Keiron, “Greysteel ‘Trick or Treat’ massacre 20 years on,” 30 October 2013, accessed
at https://www.bbc.com/news/uk-northern-ireland-24736060, 28 May 2021.
Valencia, León. "Los caminos de la alianza entre los paramilitares y los políticos." Parapolítica.
La ruta de la expansión paramilitar y los acuerdos políticos (2007): 13-47.
178

Wade, Sara Jackson, and Dan Reiter. “Does Democracy Matter? Regime Type and Suicide
Terrorism.” The Journal of Conflict Resolution, vol. 51, no. 2, 2007, pp. 329–348.
Walsh, James I., and James A. Piazza. “Why respecting physical integrity rights reduces
terrorism.” Comparative Political Studies 43.5 (2010): 551-577.
Weber, Max. From Max Weber: essays in sociology. Routledge, 2009.
Wilkinson, Paul. "Current and future trends in domestic and international terrorism: Implications
for democratic government and the international community." Strategic Review for
Southern Africa 23.2 (2001): 106.
Wilson, Matthew C., and James A. Piazza. “Autocracies and Terrorism: Conditioning Effects of
Authoritarian Regime Type on Terrorist Attacks.” American Journal of Political Science,
vol. 57, no. 4, 2013, pp. 941–955.
Young, Joseph K., and Laura Dugan. "Veto players and terror." Journal of Peace Research 48.1
(2011): 19-33.

Abstract (if available)

Abstract This dissertation seeks to further explain the association between democratic regimes and terrorist violence. More specifically, it asks the question of how we can better disaggregate the phenomenon of terrorist violence, and how different types of terrorism may be associated with democratic institutions. To accomplish this, the typology of terrorism is differentiated between those attacks perpetrated by a broad range of violent anti-state groups, and those attacks perpetrated by violent groups committed to maintaining the status quo, focusing on explaining the latter. Previous literature has established that regime type, and certain subcomponents thereof, have mixed results with terrorism as a whole. To clarify this relationship between regime type and terrorist violence, a theory on the democratic tendency to experience status-quo terrorism is developed and tested using a mixed methodological approach. Utilizing a zero-inflated negative binomial model, I find that the relationship between democracy?whether measured minimally or with more context?and status quo terrorism is positive and highly significant. Analyzing the case of status-quo terrorism within Colombia, I take a closer look at the specific causal mechanisms at play linking democratic norms to status-quo violence. Both the quantitative and qualitative testing provide findings which suggest that democracies are more prone to experience status-quo terrorism specifically.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The erosion of democracy in the post–Cold War world: understanding the phenomenon and its causes

PDF

Flinging the boomerang: locating instability and the threat potential of identity-bias in US national security policy

PDF

Romancing the bomb: Gothic terror and terrorism in late nineteenth- and early twentieth-century literature

PDF

Identity salience and political violence: centering social identity in the study of intrastate conflict

PDF

Power to the masses? The rise of direct democracy in Latin America

PDF

Politics behind appointments: policy, patronage, and party systems in the Andes

PDF

The partial Good Samaritan states: China and Japan in the international relations of autocracy and democracy

PDF

The strategy of firms towards political elites and consumers in polarized western democracies

PDF

Challenges from below: the origins of status competitions in world politics

PDF

Refugee protection and empowerment joint venture: understanding how and why public and private actors cooperate to tackle the global refugee crisis

PDF

The IO contestation project: a multi-method approach towards understanding the ways states contest international organizations.

PDF

Developing countries in the digital era: state and business interactions for industrial upgrading

PDF

The political psychology of democracy promotion, or how democracies democratically promote democracy promotion

PDF

Code 'war' theorizing: information and communication technology's impact on international relations theorizing, negotiation, and cyber relations

PDF

Authoritarian religion: explaining when and why authoritarian governments instrumentalize religion

PDF

Forced migration as a cause and consequence of conflict

PDF

Habits and policy: the socio-cognitive foundations of foreign policy stability

PDF

The changing state of tolerance and tolerance in changing states

PDF

Pathways to nuclear weapon reversal: exploring mechanisms and understanding non-proliferation policy

PDF

Are you as old as you feel? A quantitative and qualitative analysis of leader age and foreign policy decision-making

Asset Metadata

Creator Brown, Cody Louis (author)

Core Title A theory of status-quo terrorism: democracies in conflict and their proclivity to outsource repression

School College of Letters, Arts and Sciences

Degree Doctor of Philosophy

Degree Program Political Science and International Relations

Degree Conferral Date 2021-12

Publication Date 11/09/2021

Defense Date 08/06/2021

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag Colombia,democracy,insurgency,mixed-methods,OAI-PMH Harvest,paramilitary,status-quo terrorism,terrorism

Format application/pdf (imt)

Language English

Contributor Electronically uploaded by the author (provenance)

Advisor James, Patrick (committee chair), Lo, James (committee member), Munck, Gerardo (committee member), Southers, Erroll (committee member)

Creator Email codybrow@usc.edu,codyl.brown@yahoo.com

Permanent Link (DOI) https://doi.org/10.25549/usctheses-oUC16661209

Unique identifier UC16661209

Legacy Identifier etd-BrownCodyL-10209

Document Type Dissertation

Format application/pdf (imt)

Rights Brown, Cody Louis

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the author, as the original true and official version of the work, but does not grant the reader permission to use the work if the desired use is covered by copyright. It is the author, as rights holder, who must provide use permission if such use is covered by copyright. The original signature page accompanying the original submission of the work to the USC Libraries is retained by the USC Libraries and a copy of it may be obtained by authorized requesters contacting the repository e-mail address given.

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus MC 2810, 3434 South Grand Avenue, 2nd Floor, Los Angeles, California 90089-2810, USA

Repository Email cisadmin@lib.usc.edu