This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Agrarian Social Structures, Insurgent Embeddedness, and State Expansion: Evidence from
Colombia
by
Charles Larratt-Smith
A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy
Graduate Department of Political Science University of Toronto
Agrarian Social Structures, Insurgent Embeddedness, and State Expansion: Evidence from Colombia
Doctor of Philosophy 2020
Graduate Department of Political Science University of Toronto
Abstract
In the context of civil war, the efficacy of counterinsurgency strategies varies dramatically
across space and time. While this process of state expansion produces different outcomes between
different national level cases, it also engenders diverging results at the sub-national level.
Counterinsurgent responses can lead to notable reductions in violence, an achievement mirrored by
improved stability and order in contested zones. Quite frequently, however, violence will increase as
stability and order worsen. The fact that state expansion into contested spaces produces such different
results across areas of extremely close proximity begs the following questions: How is the state able
to establish control, and by extension order, in some contested spaces more easily than in others?
Conversely, what enables armed non-state actors to withstand and survive this massive onslaught in
some cases, while failing elsewhere? Since 2002, the Colombian state has embarked upon a massive
state expansion project in many volatile areas of the country that were previously controlled and
governed by armed non-state actors. This projection of military, bureaucratic, and economic power
into these contested spaces has not always brought peace and stability with it, casting into doubt the
efficacy of the central government’s larger attempt at state expansion.
This dissertation explores the above research puzzle through a comparative historical analysis
of two sub-national counterinsurgency laboratories in rural Colombia which demonstrate enormous
variation in counterinsurgent outcomes: Montes de María and Arauca. I provide a longitudinal
iii
qualitative model that highlights the importance of pre-existing agrarian social structures on the
process of insurgent institutionalization in these spaces, or the ability of these armed non-state actors
to embed themselves in rural civilian communities. I find that those actors that are better able to
appropriate local cleavages in favor of specific constituencies will achieve a higher level of
embeddedness in these spaces and thus possess a higher level of populational control over civilians.
These advantages are crucial for insurgents during periods of state expansion, as they are better
equipped to protect civilian populations from counterinsurgent violence and to prevent potential
defection to their rivals.
iv
Acknowledgements
When I started this doctoral program, I never imagined that it would take me eight long years
to finish. My initial plan when I entered the incoming class in 2012 was to complete my degree
requirements in five years, which in hindsight was both profoundly naive and overly ambitious. Little
did I imagine then that the path to this moment would have unfolded as it did, but things happen for
a reason and I am thankful for the fact that I have reached this point. While I have lost some truly
important people along the way, I have also been fortunate enough to meet others who have
unexpectedly changed my life for the better. All in all, there are many people that I need to thank for
making this dissertation and degree possible, yet for the sake of brevity (and given the excessive length
of this dissertation) I will try to keep these acknowledgements as brief as possible.
To all of my friends and family who were genuinely supportive of this journey, and who in
recent years urged me to “stop being a student already”, I salute you. This frank advice provided more
than enough motivation for me to finish this dissertation and to put this chapter of my life behind
me. In academia, it is often easy to lose focus of what really matters in this world and that is those
close relationships which shape one’s life. The time spent with those closest to me, regardless of the
location, has kept me in check (and sane) and given greater clarity to what exactly I have been doing
with my life over the past eight years. While some did not understand and openly admitted as such,
others expressed their admiration for my willingness to chart my own course, regardless of the lack of
stability that this particular career choice presents. For this, I am eternally grateful for your honesty
and for forcing me to think about what it is that I am trying to do in this world. Special thanks to my
mother and father for their perennial moral support, my siblings for their consistent encouragement
and honesty, and my various groups of friends littered throughout the globe for providing me with a
useful distraction when I needed it most.
To Diana “Taba” Leal and Gimli “Pancelón” Leal, you are my heart and soul. From the first
moment we met until the present day, you have both held me down during the tough times and given
me extra motivation to live up to my potential for the greater good of the tabafamilia. Your patience
and willingness to accompany me to Toronto while I drafted and finished this dissertation has been
crucial and I hope that I can return the favor once we enter the next chapter of our life together,
wherever that may be. Words cannot describe how much you both mean to me. It would be remiss
to thank you both without including the extended family in Bogotá, Cúcuta, and Chinácota, who have
always supported me during this endeavor by giving me endless coffee and goodwill when I was
v
working on this project during family visits. I am also indebted to you all for sharing your own unique
stories with me which have helped shape my work in numerous ways.
To all those in Colombia who helped me carry out my fieldwork and develop my dissertation.
First and foremost, I have to thank the incredible people at the Centro de Investigación y Educación
Popular (CINEP) for hosting me during the majority of my time there and for assisting in a variety of
ways with my project. Special thanks go out to the center’s director, Luis Guillermo Guerrero, and
particularly to Padre Fernán González González, the head of the Conflicto y Estado team. It was an
honour and a privilege to develop my dissertation under the guidance of such an eminent scholar of
your standing. Similarly, I have to thank the remaining members of the team, such as Víctor Barrera,
Andrés Aponte, Javier Benavides, Camila Carvajal, and fellow visiting researcher Javier Revelo, for all
of your assistance and late-night revelry over the past couple of years. My debt to the entire team and
CINEP is enormous and one which I will never forget.
There are numerous people outside of CINEP who also made my fieldwork in Colombia
possible. First, I would like to thank the three research assistants who helped me in Montes de María
and Arauca, Camila Carvajal, Diana Rodríguez, and Estefania Forero. Additionally, I would like to
give special thanks to Omar Gutiérrez Lemus, Lucho Celis, and the staff at Federación Luterana
Mundial Programa Colombia for sharing your insight and for putting me into contact with local
contacts on the ground in Arauca and vouching for me in the process. Similarly, I would like to offer
my deepest thanks to Dr. Pablo Abitbol and Eduardo Porras for providing me with the same
assistance in Montes de María. Also, I can’t forget Enrique Peña for his help providing contacts with
ex-military officials in Bogotá. Unfortunately, I cannot directly name those countless individuals and
research participants on the ground in Montes de María and Arauca who helped me out due to the
potential security risks, but the bravery you all possess in trying to confront the past is beyond
inspirational. If there is one thing I have taken away from this entire experience, it is that the human
capacity to overcome the worst forms of adversity knows no bounds, a lesson I have learned from
sharing with you all.
To my doctoral committee, I am enormously thankful for your crucial feedback and support
throughout this extended process. To Shivaji Mukherjee, you joined this project when it was just
beginning and sadly you will be the only original member who is there when it finishes. My deepest
gratitude for your help seeing this over the finishing line. To Catherine LeGrand, even though we met
by way of this dissertation I feel like we have known each other much longer. I will be eternally
appreciative of your willingness to join this committee on the eve of your retirement and your keen
vi
interest in my research progress. To Antoinette Handley, I owe you an immense debt for picking up
the pieces of my dissertation at what was a very bad moment for us both. Even though this dissertation
does not correspond to your regional or topical area of interest, your input and guidance has proven
instrumental to this dissertation and I have learned a great deal working with you over the past two
years. Additionally, I would like to thank both the internal and external readers, Jon Lindsay and Ana
Arjona, for their punctual revisions and vital feedback that has vastly improved my research.
To all those academics located at the University of Toronto and beyond who have assisted in
some shape or form with this dissertation, I owe you all a similar debt of gratitude. Similarly, I would
like to thank the administrative staff at the University of Toronto, in particular at UTM, for their
ceaseless support and for making my working experience at the university that much more memorable.
Special thanks to Norma Dotto and Terri Winchester at UTM, and to Carolynn Branton, Louis
Tentsos, and Sari Sherman at STG. Without you all, the department would cease to function. Of equal
importance, I would like to thank the various donors and grants who provided me with the necessary
financial assistance to carry out my fieldwork and draft my dissertation, including the Department of
Political Science, the School of Graduate Studies, the International Peace Research Grant Association,
the International Development Research Centre of Canada, and the Harry Frank Guggenheim
Foundation.
Finally, I need to express my gratitude to three people who unfortunately passed away over
the course of my time at the University of Toronto. To Ana María Bejarano, if it weren’t for you then
my life would have arguably taken a completely different and unknown course. When I applied for
this program and my application came before your desk in early 2012, you stated that you would be
more than willing to serve as my supervisor and as such the selection committee approved my spot in
the program. I can only speculate where I would be right now if you hadn’t supported my application.
Similarly, your guidance and mentorship early on helped me develop this dissertation and although in
the end you never got a chance to read my research, I like to think that it is influenced a great deal by
your own work. I will be eternally indebted to you for not only this experience, but also for the
opportunity of being your student, a distinction which opened numerous doors in Colombia that
otherwise might have remained shut to an unestablished researcher there. For the duration of the time
that I knew you, your courage battling cancer, while being a single mother and a highly regarded
Colombian academic at an elite North American university, inspired myself and all of those who knew
you.
vii
To Lee Ann Fujii, after Ana María you were the second faculty member in our department
who took me under your wing. Even though you didn’t have to and had no reason to look out for me
other than the fact that I worked for you that initial year, we built a rapport based on our common
research interests and our shared sense of humour. Similarly, I appreciate your honesty and candor
when it came to navigating the often difficult politics of the academic world. When Ana Maria became
sick again, you stepped up at a difficult time and served as my acting supervisor, eventually becoming
my permanent supervisor after her unfortunate passing in early 2017. Even though this was short
lived, your influence on my research has been profound and I like to think that you would be proud
of this dissertation. I feel privileged to say that I was both a student of Ana María and yourself and
will forever be thankful for your assistance during that rough patch of years.
To Andrew “Gillie” Gilchrist, I am writing this on the anniversary of your passing and it seems
like an eternity since we last saw one another given the number of events that have transpired during
that time. Fate has a strange way of working as I never intended to come back to Toronto, yet this
program changed all of that and in hindsight I am thankful for this opportunity which allowed you
and I to live together for your final year before you left us. As crazy as it was, you always served as a
useful and much needed counterbalance to the absurdities of the academic world and I will always be
thankful for our friendship. Not a day goes by that I don’t think about you and all of our ludicrous
interactions, as your absence will always hang over all of those who were fortunate enough to know
you. You were and will always remain my best friend and it pains me that I am finally arriving at this
moment and you aren’t here to see this happen.
This dissertation is dedicated to the memory of Ana Maria, Lee Ann, and Andrew.
viii
Table of Contents
List of Figures x List of Tables x-xi List of Maps xi Acronyms xii Introduction 1
A. The Puzzle 3 B. Why This Matters 6
C. Defining Terms 8 D. Literature Review, Theoretical Gaps, and Alternative Explanations 14 E. My Theoretical Model 23 F. The Structure of this Dissertation 38
Chapter Two. Research Design 40
A. Methodology 40 B. Research Design 44
C. Fieldwork 47 D. Rationale for National and Sub-National Case Selection 49 E. A Brief History of Colombia 54 Chapter Three. Pre-Existing Configuration 71 A. Montes de María 74 i. The Territorial Sphere 75 ii. The Economic Sphere 83 iii. The Political Sphere 86 iv. The Civic Sphere 90 B. Arauca 99 i. The Territorial Sphere 100 ii. The Economic Sphere 110 iii. The Political Sphere 117 iv. The Civic Sphere 121
C. Primary Cleavages in Montes de María and Arauca 131 Chapter Four. Insurgent Institutionalization 134
ix
A. Montes de María 137 i. The Territorial Sphere 138 ii. The Economic Sphere 147 iii. The Political Sphere 150 iv. The Civic Sphere 153 B. Arauca 156 i. The Territorial Sphere 157 ii. The Economic Sphere 174 iii. The Political Sphere 182 iv. The Civic Sphere 185
C. Provisional Predation in Montes de María and the Araucan Plains 192 D. Nested Governance in the Araucan Piedmont 199
Chapter Five. Period of Contestation in Montes de María 209
A. Evaluating Counterinsurgency Outcomes in Montes de María 214 B. Counterinsurgent Victory in Montes de María 217
i. The Paramilitary Incursion 219 ii. State Expansion 231 iii. The Insurgent Response 246 C. The Outcome 252 Chapter Six. Period of Contestation in Arauca 263
A. Evaluating Counterinsurgency Outcomes in Arauca 263 B. Short-term COIN Success in the Plains, Insurgent Resilience in the Piedmont 267
i. The Paramilitary Incursion 269 ii. State Expansion 279 iii. The Insurgent Response 296 C. The Outcome 308
Conclusion 318
A. External Validity 313 B. Montes de María 314
C. Arauca 319 D. En fin: Potential Theoretical Contributions and Policy Implications 325
Works Cited 328 Field Interview Guide 348
x
List of Tables
Table 1. Insurgent Annual Finances in 1991 (USD) 61 Table 2. Territorial, Political, Economic, and Civic Spheres
in Montes de María (Pre-1980) 74 Table 3. Territorial, Political, Economic, and Civic Spheres in Arauca (pre-1980) 99 Table 4. Distribution of Public Land by INCORA in Arauca (1961-1989) 107 Table 5. Armed Actor Control in Montes de María (1980-1996) 194 Table 6. Armed Actor Control in the Araucan Plains (1980-2001) 195 Table 7. Armed Actor Control in the Araucan Piedmont (1980-2001) 201 Table 8. State of Conflict and Outcome per Municipality
in Montes de María (2000-2012) 216 Table 9. Armed Actor Control in Montes de María (1997-2007) 253 Table 10. Urban and Rural Population of El Carmen de Bolívar (1993-2005) 254 Table 11. State of Conflict and Outcome per Municipality in Arauca (2000-2012) 266 Table 12. Armed Actor Control in the Araucan Plains (2001-2010) 309 Table 13. Urban and Rural Population of Tame (1993-2005) 313 Table 14. Armed Actor Control in the Araucan Piedmont (2001-2010) 314
List of Figures
Figure 1. Analytical Sequence of Insurgency & Counterinsurgency 24 Figure 2. Key Variables, Critical Junctures, and Outcomes 26 Figure 3. Four Spheres of Contestation 35 Figure 4. Growth in Colombian Armed Forces - Number of Troops (1989-2015) 64 Figure 5. Colombian Homicide Rates per 100 000 Persons (1990-2015) 65 Figure 6. Armed Actions by Armed Actors in Colombia (1990-2011) 68 Figure 7. Character of Social Mobilizations in Montes de María 155 Figure 8. Character of Social Mobilizations in the Araucan Piedmont 190 Figure 9. Character of Social Mobilizations in the Araucan Plains 191 Figure 10. Forced Displacement in Montes de María and Arauca 212
Figure 11. Homicide Rate per 100 000 Residents in Montes de María (1990-2013) 215 Figure 12. Armed Actions and Confrontations in Montes de María (1998-2013) 216 Figure 13. Insurgent Trajectory and Counterinsurgency Outcome in Montes de María 218 Figure 14. Conflict Events Committed by Paramilitary Groups
in Montes de María (1988-2007) 222 Figure 15. Conflict Events per Insurgent Group in Montes de María (1988-2007) 247 Figure 16. Homicide Rate per 100 000 Persons in Arauca (1990-2013) 264 Figure 17. Armed Actions and Confrontations in Arauca (1998-2013) 265 Figure 18. Insurgent Trajectory and Counterinsurgency Outcome
in the Araucan Plains 268 Figure 19. Insurgent Trajectory and Counterinsurgency Outcome
in the Araucan Piedmont 269
xi
Figure 20. Forced Displacement in Arauca (1990-2013) 278 Figure 21. Caño Limón-Coveñas Oil Pipeline Attacks in Arauca (1986 – 2013) 281 Figure 22. Coca Crops Cultivated (Hectares) in Arauca by Municipality (2000-2012) 289 Figure 23. Conflict Actions Committed by Insurgent Groups in Arauca (1996-2013) 297 Figure 24. Mine Casualties in Arauca per Municipality (2000-2013) 298
List of Maps Map 1. Montes de María 76 Map 2. Arauca 101 Map 3. Caño Limon-Coveñas Oil Pipeline 177
xii
Acronyms
ABC - Arauca, Boyacá, and Casanare ACCU - Autodefensas Campesinas de Córdoba y Urabá ACMM - Autodefensas Campesinas de Magdalena Medio ANUC - Asociación Nacional de Usuarios Campesinos ASEDAR - Asociación De Educadores Del Arauca ASOJER - Asociación Juvenil y Estudiantil Regional AUC - Autodefensas Unidas de Colombia BHMM - Bloque Héroes de los Montes de María BVA - Bloque Vencedores de Arauca CCAI - Center for Coordination of Integrated Action CERAC - Centro de Recursos para el Analisis de Conflictos CINEP - Centro de Investigación y Educación Popular COAGROSARARE - Cooperativa Agropecuaria del Sarare COIN - Counterinsurgency CONVIVIR - Cooperativas de Vigilancia y Seguridad Privada para la Defensa Agraria CRS - Corriente de Renovación Socialista CRUCIAGAR - Grupo Cívico Armado de Arauca DAS - Departamento Administrativo de Seguridad DEA – Drug Enforcement Administration DIJIN - Dirección Central de Policía Judicial e Inteligencia DSP - Política de Seguridad Democrática EDA - Estructura de Apoyo ELN - Ejército de Liberación Nacional EPL - Ejército Popular de Liberación ERP - Ejército Revolucionario del Pueblo FANAL - Federación Agraria Nacional FARC - Fuerzas Armadas Revolucionarias de Colombia FENSUAGRO - Federación Nacional Sindical Unitaria Agropecuaria FGO - Frente de Guerra Oriental FGN - Fiscalía General de la Nación HVT - High-Value Targets IDEMA - Instituto de Mercadeo Agropecuario IDP - Internally Displaced Person IHL - International Humanitarian Law INCORA - Instituto Colombiano de la Reforma Agraria JAC - Juntas de Acción Comunal JUCO - Juventud Comunista MIR-PL - Movimiento Izquierdista Revolucionario-Patria Libre PCC - Partido Comunista Colombiano PRT - Partido Revolucionario de los Trabajadores RINCA - Red de Inteligencia Naval del Caribe UAF - Unidad Agrícola Familiar UNDP - United Nations Development Program UP - Unión Patriótica USO - Unión Sindical Obrera de la Industria del Petróleo
1
Introduction
“As usual the problem involves: first, the adversary, and second, the means to destroy him. There are 400 000 Arabs in Algiers. Are they all our enemies? We know they’re not. But a small minority holds sway by means of terror and violence. We must deal with this minority in order to isolate and destroy it. It’s a dangerous
enemy that works in the open and underground, using tried-and-true revolutionary methods as well as original tactics. It’s a faceless enemy, unrecognizable, blending in with hundreds of others. It is everywhere. In
cafés, in the alleys of the Casbah, or in the very streets of the European quarter.”
- Colonel Mathieu, The Battle of Algiers (1966)
“I remember when I was with Special Forces...seems a thousand centuries ago. We went into a camp to
inoculate some children. We left the camp after we had inoculated the children for polio, and this old man came running after us and he was crying. He couldn't see. We went back there, and they had come and
hacked off every inoculated arm. There they were in a pile. A pile of little arms.”
- Colonel Kurtz, Apocalypse Now (1979)
Exhausted from a near sleepless overnight trip, I found myself struggling to stay awake in the
intense heat and discomfort of an old broken-down bus heading west along the main highway
straddling Colombia’s Caribbean coast. Suddenly, the driver shouted the name of my intended
destination, a rural village named Guachaca. Upon disembarking I quickly felt the piercing stares from
different people who were taking in my unexpected arrival. A man on a motorcycle shouted out to
me and asked me where I was going and when in reply I uttered the name of the rural homestead that
my contacts had given me, the man lurched forward and told me to hop on the back of his motorcycle.
As soon as my feet left the ground the driver accelerated dangerously down one of the unpaved access
roads into the jungle. Once we arrived our destination, I paid the moto-taxista and was greeted by Pacho,
the director of the foundation which I had come to the coast to work for. After introducing me to his
wife and children, we sat down and talked for a bit about my journey before we were suddenly
interrupted by the sound of a motorcycle horn from just outside the property. Pacho quickly rose and
2
went outside to converse with the unknown motorcyclist and after a brief chat, he returned and
assured us that everything was in order.
I didn’t think about the incident again for the following two months. It was only on the eve
of my departure that Pacho confided to me what had really transpired that first day. He explained that
word of my sudden and unannounced arrival to Guachaca was quickly relayed back to the local
commander of the neo-paramilitary group that controlled the village and the surrounding zone, a
region coveted by armed non-state actors due to its strategic location as a drug trafficking corridor.
The commander dispatched one of his emissaries to the farm where I was staying to inquire about my
presence in the zone. After satisfying his questions, the emissary reminded Pacho that he was
responsible for me, and anything I did in the community would ultimately have to be answered for by
my host. Prior to this, I had noticed that the locals always seemed to be observing me and that
everything I did in Guachaca was common knowledge in the community. Pacho later explained that
these individuals were in fact key components of an extensive local intelligence network which
supplied the armed group with vital information about any and everything that transpired in the zone.
The local commander was the de facto ruler of Guachaca, more so than any elected politician or police
chief in the area. Even though the National Police and the Colombian military maintained roadblocks
on the coastal highway and even occasionally patrolled the unpaved access roads, Pacho pointed out
that they only controlled the road when they were on it. The neo-paramilitary group in contrast
controlled the entirety of the local territory and the people who resided on it, whether they liked it or
not.
Although this was not my first time in Colombia, this experience served as a starting point in
understanding how land and people are governed in this South American country and how alternative
sources of authority are able to impose their will on entire populations. This realization forced me to
ask how such groups are able to continue to control these communities even when the state expands
3
its presence disproportionately in an attempt to pacify them, whether through the police, the military,
or other agents. Many years and countless hours of research later I have finally come to understand
that in countries like Colombia, Guachaca may not be so much the exception as the rule, as countless
other communities operate according to such situational ethics and the logic of populational control.
A. The Puzzle
In recent years, international security experts have warned against the threat posed by
‘ungoverned spaces’, insofar as these swaths of territory existing beyond the control of any central
government provide natural sanctuary to insurgents, criminals, and international terrorists alike
(Piombo 2007; Raleigh & Dowd 2013). The mere existence of these countries pockmarked with
‘lawless zones’ has been declared a grave security threat and one that: “[…]not only threatens the lives
and livelihoods of their own people but endangers world peace” (Rotberg 2002, 128). While there is
some truth to the claim that armed non-state actors thrive in spaces where the state is conspicuously
absent, a wealth of research has proven that these places are rarely “ungoverned” or “lawless”. Rather,
they are meticulously controlled and administered by alternative sources of authority (Mampilly 2011;
2019). In some cases, states are willing to permit these bifurcated sub-national distributions of power
and authority, as these may prove more effective and less costly forms of governance to the center
(Herbst 2000; Boone 2002; Staniland 2012; Duncan 2014). In others, the central state may feel
compelled by internal or external pressures to expand into these particular spaces in order to
recuperate control and integrate them into the rest of the country (Galula 1964; Leites and Wolf 1970;
Kilcullen 2010). The latter process of state expansion - or counterinsurgency (COIN) as most military
analysts and academics refer to it - produces results that are rarely the same across space and time.
State expansion sometimes produces notable reductions in violence, an achievement mirrored by
improved stability and order in contested zones. Quite frequently however, violence will increase as
4
stability and order worsen. The fact that state expansion into contested spaces produces such different
results across areas of extremely close proximity begs the following questions: How is the state able
to establish control, and by extension order, in some contested spaces more easily than in others?
Conversely, what enables armed non-state actors to withstand and survive this massive onslaught in
some cases, while failing elsewhere?
Since 2002, the Colombian state has embarked upon a massive state expansion project in many
volatile areas of the country that were previously controlled and governed by armed non-state actors.
This projection of military, bureaucratic, and economic power into these contested spaces has not
always brought peace and stability with it, casting into doubt the efficacy of the central government’s
larger attempt at state expansion. Perhaps nowhere else is this range of outcomes more evident than
in a comparison of Montes de María and Arauca. Having served as counterinsurgent laboratories for
the Colombian government from 2002 onwards, these two particular sites demonstrate the mixed
results that accompany state expansion projects on a micro-level across disparate municipalities and
regions of the country. The incursion by the Colombian military into Montes de María in northern
Colombia saw previously astronomical levels of violence reduced almost completely within five years,
while armed insurgents were also expelled from this once heavily contested zone. Simultaneously, the
same military strategy was employed in Arauca and this state expansion produced a dramatic variation
of results between the department’s two sub-regions, the Araucan plains and the Araucan
piedmont. In the case of the former, there was a permanent decline in violent deaths, forced
displacement, and armed actions from 2002 onwards, an accomplishment mirrored by a dramatic
reduction in the armed insurgent presence in the sub-region. Meanwhile, the Araucan piedmont
experienced initial increases in the same indicators of violence, while insurgent groups fiercely resisted
the state’s onslaught. They have maintained a consolidated presence there until the present.
5
I hypothesize that the Colombian state defeated the insurgents in Montes de María and the
Araucan plains by allying with paramilitary groups and landed elites to undermine weak insurgent
embeddedness in local civilian communities to physically dislocate the peasantry from the rebels, a
strategy that allowed it to bind civilians to the nascent counterinsurgent order. This proved successful
insofar as it effectively reclaimed control of these disparate territories (permanently in the case of
Montes de María and temporarily in the case of the Araucan plains). In contrast, the lack of success
achieved by the Colombian military in the Araucan piedmont was the result of its inability to weaken
insurgent embeddedness in civilian communities in the sub-region and to establish a viable alternative
to rebel rule. Quite simply, the sub-region lacked both landed elites and any substantial paramilitary
presence for the Colombian state to find any local buy-in to its counterinsurgent project. Armed
insurgent groups were heavily embedded in peasant communities, which in turn engendered a high
degree of populational control, binding peasants to the insurgent social order and preventing potential
civilian collaboration with counterinsurgent forces that were widely perceived as more predatory than
the guerrilla groups.
My theoretical model examines holistically the strategies of both competing sides amidst the
particular structural conditions in each regional theater of war, and rests on four core assumptions
about the efficacy of state expansion into spaces controlled by armed insurgent groups:
i. The success of counterinsurgency efforts is determined by their ability to weaken insurgent
embeddedness in civilian populations.
ii. Insurgent embeddedness is determined by the ability of these groups to appropriate local cleavages
shaped by pre-existing agrarian social structures.
iii. The greater the level of insurgent embeddedness, the greater level of populational control.
iv. High levels of insurgent population control provide these groups with greater monitoring and
enforcement mechanisms to prevent civilian collaboration with counterinsurgent forces.
6
B. Why This Matters
According to the Correlates of War database on interstate and intrastate armed conflicts,
between 1816 and 2000, there were 464 such conflicts around the world, 79 of which (or 17% of the
grand total) were interstate while the remaining 385 are classified as intrastate (Kilcullen 2010). Since
the end of the Second World War in 1945, international (e.g. interstate) conflicts have subsided
dramatically (Lacina, Gleditsch, and Russett 2006). International geopolitical phenomena such as the
Cold War and the process of decolonization brought about a marked increase in internal conflicts and
civil wars around the globe (Brubaker and Laitin 1998). Whereas many of these armed conflicts were
in fact struggles for independence, others were driven by economic motivations, particular group
grievances, or a combination of both (Collier and Hoeffler 2004; Cramer 2003; Sambanis 2001; Stewart
2008). The costs of these civil wars have been extreme, as Fearon and Laitin’s (2003) tally between
1945 and 1999 counts: “[…]127 civil wars that killed at least 1,000, 25 of which were ongoing in 1999”,
with a “[…]conservative estimate of the total dead as a direct result of these conflicts is 16.2 million,
five times the interstate toll” (75). Beyond battlefield deaths, the collateral damage wrought by these
intrastate conflicts in incalculable, with the legacy costs including widespread poverty, high mortality
rates, and a greater likelihood of future conflict (Bayer and Rupert 2004; Black, Morris, and Bryce
2003; Fortna 2004).
The sudden increase in civil wars during the latter half of the 20 th century prompted various
colonial and emerging powers to study the causes of, effects of, and solutions to these constant and
seemingly endless civil wars throughout the globe. Driven by competition and imperial overreach,
these militaries amassed a wealth of counterinsurgent experience, whether the French in Algeria and
Indochina, the British in Malaya and Kenya, the Americans in Vietnam, or the Russians in Afghanistan
(Trinquier 1964; Nagl 2002; Fitzgerald 2013). The accumulation of COIN knowhow has generated an
expansive literature which focuses on past successes and failures, all of which supposedly provides
7
important lessons for present and future state expansion campaigns. Unfortunately, many
‘COINdanistas’ or ex-field commanders, military instructors, and academics who study
counterinsurgency operate under the misguided assumption that all of these cases are essentially
similar and can therefore be addressed through a common solution. This analytical fallacy has caused
many to overlook the inconvenient fact that genuine COIN successes have been rare, and those that
do occur are often the result of insurgent weakness, external pressures, or the extremity and
ruthlessness of the tactics employed. There is a greater need for COIN experts to pay heed to Leites
and Wolf’s (1970) simple yet critical observation about modern insurgency:
Each major insurgency is, in some sense, unique, as suggested by the diversity of areas and circumstances in the list. But most of them have shared many features – organization, tactics, violence, coercion, persuasion, ideology, internal grievance, external influence. The common features make insurgency a proper subject for more general analysis. The diversity warrants caution to avoid pushing generalizations too far. (2)
Perhaps most worrying of all is that those individuals who wield enormous influence over
contemporary military policy continue to advocate a vigorous state intervention and expansion into
contested spaces in order to securitize and, by extension, pacify them of perceived threats, without
first taking other non-military options into consideration. Supporters of this strategy invoke its
successes in places as disparate as Kenya, Malaysia, Iraq, Sri Lanka, and even Colombia, all the while
ignoring its failures elsewhere. This is not to say that all state interventions into such spaces are either
complete successes or abject failures, rather that the results often vary dramatically even when within
areas of extremely close proximity and warrant greater evaluation. This dissertation examines the role
of the state when it expands into territory controlled by other armed non-state actors in order to
develop a better understanding of how military interventions into contested spaces can produce such
radically different outcomes during the course of civil wars. Although my research is rooted in an
inter-regional comparison of two different cases in northern Colombia, it may also provide important
insights into other cases of violent internal conflicts (i.e. the Democratic Republic of Congo, the
8
Philippines, Somalia, Syria, Mali, Myanmar, Sudan, Iraq, and Afghanistan) or states hosting a myriad
of armed groups with high levels of violence (i.e. México, Nigeria, El Salvador, Libya, Jamaica,
Honduras, South Africa, Venezuela, and Brazil).
C. Defining Terms
My use of the term state expansion is to address the shortcomings of the conventionally used
term counterinsurgency. While seemingly interchangeable, state expansion provides a more holistic
balance in terms of its meaning and definition when describing what the state is effectively attempting
to achieve when it inserts itself into spaces controlled by armed non-state actors. To paraphrase the
U.S. Army Field Manual 3-24, counterinsurgency “[…]is military, paramilitary, political, economic,
psychological, and civic actions taken by a government to defeat insurgency” (Petraeus et al. 2006, 2).
As expansive and clear as this definition may appear, it fails to capture the logic of insurgency, an
omission which demonstrates the need for a more holistic interpretation. According to Kalyvas (2006):
“[…]civil war is, at its core, a process of integration and statebuilding” (14). This is reflected better in
some cases than others. For example, civil wars in developed countries such as Greece, Bosnia, or
Northern Ireland occurred in contexts where state formation had largely been achieved under previous
regimes, political orders which seemingly exercised a monopoly of violence throughout the entirety
of the national territory. However, most insurgencies and civil wars have emerged in states
characterized by fragmented sovereignty and these are still far more likely to experience internal
conflict than other states that have effectively achieved the monopoly of the means of coercion within
their territory (Fearon and Laitin 2003). The meaning of counterinsurgency in historically fragmented
countries such as Colombia is better captured by state expansion, meaning the belated attempt by the
central government to seize control and integrate regions of the country previously developed and
administered by non-state sources of authority back into the national fold. Although “state expansion”
is a more accurate term than “counterinsurgency” in the case of Colombia, it is complementary and
9
interchangeable with that term, rather than conflictive, and as such both terms are used throughout
this dissertation.
The concept of nested governance follows in the steps of previous works that highlight the
complex layering of different yet interconnected modes of governance at the national and sub-national
levels, all of which constrain political actors and their policies (Ostrom 1990; Tsebelis 1990; Sinha
2005). Historically, Colombia has been characterized by different levels of political competition in
which national political elites “[…]effectively delegated the running of the countryside and other
peripheral areas to local elites”, powerful individuals who governed these regional fiefdoms as they
saw fit, all “[…]in exchange for political support and not challenging the center” (Robinson 2013, 44).
Therefore, nested governance refers to the manner in which armed non-state actors insert themselves
between communities of people and the state in order to control and regulate formal and informal
processes between the two parties according to their own interest and objectives. Nested governance
effectively enables armed groups to carry out both informal processes typically ascribed to criminal
organizations such as the indirect rule of local populations, territory, and illegal rackets, and formal
processes centered on legal economic activities, electoral contests, and bureaucratic-administrative
institutions, all of which are nominally administered by the state (Duncan 2014).
Over the course of the current civil war, armed units from leftist insurgent groups and right-
wing paramilitary organizations have inserted themselves into these sub-national dynamics in an
attempt to control people and territory, sometimes forging alliances with malleable regional elites, in
other cases bypassing them entirely. Garay et al. (2008) label this phenomena “the co-opted
reconfiguration of the state”, a term which refers to legal and illegal organizations that employ
illegitimate practices to change the regional political order in order: “[…]to obtain sustainable benefits
for their own advantage and to ensure that their interests are validated politically and legally” (96).
Nested governance adheres to a similar logic. However rather than focusing solely on how armed
10
non-state actors place themselves as interlocutors between macro and meso-level political institutions
and the communities they purportedly govern, this concept also takes into account how these groups
navigate formal and informal processes of a social and economic character (Arjona, Kasfir and
Mampilly 2015). In regards to informal processes, armed actors can control local territory and
populations through the effective deployment of violence against both civilians and the state, while
they can also explicitly regulate illegal economies such as extortion, kidnapping, contraband, and drug
trafficking (Beckert and Dewey 2017). However, as Gustavo Duncan (2014) highlights, such groups
are typically unable to invest the proceeds from such economic activities into the legal economy due
to the likelihood of state forfeiture. It would be similarly problematic for such groups to participate
directly in formal processes related to legal economic activities, electoral contests, and bureaucratic-
administrative institutions. Ostensibly, these are all administered by the state and thus armed non-state
actors need another means to capture power in these dynamics.
Provisional predation refers to a model of governance adopted by armed non-state actors that
prioritizes short-term group goals and objectives over the long-term consolidation of a comprehensive
social order in a given territory. Typically, a group that adopts provisional predation does so because
the objective conditions are not conducive to establishing nested governance – either due to the
collective rejection of such actors by local communities, or the lack of local networks to latch onto –
yet the result is that these groups end up controlling territory and economic resources, but lack any
meaningful embeddedness in the human terrain. Jeremy Weinstein (2007) describes such forms of
armed non-state rule: “The short-term orientation of opportunistic insurgencies, on the other hand,
tends to be detrimental to civilian populations…A constant demand for short-term rewards also drives
combatants to loot, destroy property, and attack indiscriminately” (10-11). While similar to Mancur
Olson’s (1993) work which distinguishes between the incentives guiding “stationary” and “roving”
bandits, the relationship between nested governance and provisional predation does not necessarily
11
constitute a binary. Rather, these are two points on a broader spectrum of possible insurgent-civilian
relations which are liable to evolve over time given the changing structural conditions in which both
combatants and non-combatants exist.
According to my explanatory model, armed actors that successfully appropriate the pre-
existing cleavage (e.g. synchronize insurgent goals and interests with those of the local peasantry) are
far more likely to establish nested governance which enables such actors to better embed themselves
into local communities. Stronger insurgent embeddedness generates greater populational control, a
dynamic which binds civilians to the existing insurgent social order in these spaces. Conversely,
insurgent social orders characterized by provisional predation are generally the result of an inability by
armed actors to appropriate the primary cleavage, a shortcoming which limits the extent to which such
groups are embedded in such spaces, and weakens the degree to which civilians are bound to their
order. This reduced level of populational control makes it difficult for insurgent groups to repel and
defeat counterinsurgent challenges. The strength of inter-civilian linkages, or embeddedness as this
concept has been refashioned by Granovetter (1985), are integral to outcomes ranging from the
economic productivity of immigrant communities in the Global North, to the success of post-war
recovery efforts (Portes 1998; Colletta and Cullen 2000). However, while there is a plethora of
literature discussing the role of these linkages in enabling disparate outcomes in a variety of contexts,
there are few convincing explanations about how embeddedness is produced at the individual or group
level. The importance of this concept in the study of social networks is best articulated by Moody and
White’s (2003) observation that: “Embeddedness indicates that actors who are integrated in dense
clusters or multiplex relations of social networks face different sets of resources and constraints than
those who are not embedded in such networks” (105). The pre-existing conditions found in a specific
territory can serve as “anchors” or “barriers” to insurgent institutionalization which in turn great
affects meso and micro-level outcomes in civil wars. Anoop Sarbahi (2014) operationalizes
12
embeddedness in the context of violent competition between armed non-state actors and highlights
its importance:
In view of gross power asymmetry between the warring parties, social embeddedness is crucial to the rebel’s ability to resist the powerful state and impose a costly stalemate - whether a rebel group is favorably inclined to negotiate, is susceptible to co-optation, presents a serious military challenge or marginally survives, and is highly influenced by this characteristic. (1474)
All communities possess varying degrees of inter-civilian embeddedness and the manner in which
these relations are conformed is unique to each one of them. Similarly, the manner in which armed
groups embed themselves in such communities depends on the pre-existing agrarian social structure,
the compatibility of their interests with those of local habitants, and the manner in which they attempt
govern and control both populations and territory.
My conceptualization of populational control bears much in common with the established
research from Criminology on social control. Although Political Science and Criminology have
remained fairly detached, many prominent scholars maintain that modern armed non-state actors have
more in common with criminal groups than with traditional insurgent forces (Kaldor 1999; Ross 2004;
Weinstein 2007). While this discussion lacks consensus and has already been debated exhaustively
within the social sciences, Criminology-based research that examines the societal constraints placed
on social deviancy is quite useful when analyzing how some armed groups are able to control local
populations even when they do not maintain complete territorial control in a particular space. Travis
Hirschi’s (1969) social bond theory argues that an individual’s attachment, commitment, involvement,
and belief in the established social order ultimately shape their conformity to its rules, values, and its
legitimacy. Individuals who live under a social order that permeates every facet of their daily lives and
places a great number of constraints on their agency: “[…]are unlikely to place their good standing in
society at risk through acts of crime and deviance” (Thyne and Schroeder 2012, 1069).
In sub-national theatres of conflict, deviation from the established rules means defying the
social code established by the dominant armed actor, and by extension, breaking from the normative
13
constraints adhered to by the larger community. In spaces where insurgents possess a wide range of
enforcement and monitoring mechanisms, the probability of an insurgent sanction for violating the
established social code is all but guaranteed. Whereas Kalyvas (2006) argues that territorial control
“[…]allows the effective use of violence, thus deterring defection; opponents are identified and flee,
are neutralized, or switch sides” (124), this fails to capture the extensive socialization of local
communities which have developed for extended periods of time under traditional forms of authority
or armed non-state actors (Sluka 1989; Souleimanov and Aliyev 2017). De jure territorial control by the
state does not automatically destroy pre-existing constraints and the commitments that local habitants
may possess towards the long-standing social order, nor does it necessarily weaken or neutralize the
enforcement and monitoring mechanisms utilized by insurgent groups in these contexts.
Finally, the term agrarian social structure represents the convergence of various structural
conditions (land distribution, mode of production, state presence/absence) which shape regional
social orders. There is a substantial body of work on the importance of agrarian social structures on
rural conflict dynamics, and it is clear that these hierarchies do not materialize from thin air. Rather,
they are the product of the historical settlement and usage of the land, the social hierarchies which
develop over time, and finally the relations between peasants and landed elites (Tilly 1992; Wolf 1969;
Skocpol 1979). Agrarian social structures are important to processes such as insurgent formation and
state expansion insofar as they represent the human terrain in which armed actors attempt to embed
themselves by forging strategic alliances with specific groups, whether smallholding peasants,
merchants, or large landholders (Scott 1976; Paige 1978; Gutiérrez-Sanín and Vargas 2017; Brewer
2010). The pre-existing cleavages which emerge in distinct agrarian social structures affect the ability
of armed insurgent groups, paramilitary structures, and the formal military to penetrate and control
these rural spaces. These hierarchies can be vertical, characterized by highly stratified pyramid structures
with landed elites at the top and landless peasants at the bottom, or these can be horizontal, representing
14
egalitarian structures with no discernible ruling class or elite constituency. In some cases, agrarian
social structures can trace their configuration back to colonial institutions designed to administer
agrarian communities (Boone 2002; Banerjee and Iyer 2005; Mukherjee 2018a, 2018b). In others, these
structures are shaped by top-down state policy directives or spontaneous migrations to inhabit
unsettled frontier zones (LeGrand 1986; Gootenberg 2003; Ballvé 2020).
D. Literature Review, Theoretical Gaps, and Alternative Explanations
Previous scholarship on historic contentious events ranging from the French Revolution to
the Arab Spring has examined the relation between state capacity and counterinsurgent efforts to
reclaim and control contested spaces (Tilly 1964; Greer 1935; Josua and Edel 2015; Alley 2013). My
research challenges many of the established arguments found in the dual literatures on
counterinsurgency and civil wars, suggesting that many of the potential alternative explanations used
to explain this particular research puzzle, whether counterinsurgency strategy, pre-existing institutions
and networks, territorial control and civilian collaboration, economic conditions, geography, have
analytical and theoretical limitations in regards to my cases. Additionally, my theory offers a way to
fuse the existing counterinsurgency literature with the emergent research agenda on civil wars and
armed non-state actors. In order to assess and evaluate the strength and resilience of insurgent social
orders, it is difficult to assess the strength and durability of these governance models until they are
challenged by counterinsurgent forces. It is during these critical junctures, I argue, that insurgent
embeddedness is tested and ultimately revealed. Given that embeddedness is what ultimately
determines counterinsurgency outcomes, there is a pressing need for the COIN literature to better
examine how such embeddedness is generated and weakened in peripheral spaces.
Despite more than sixty years of counterinsurgency knowledge derived from past conflicts,
military commanders and counterinsurgent forces continue to repeat past errors by assuming that a
fixed tactical approach can succeed anywhere, regardless of the context. Counterinsurgent military
15
historian Douglas Porch (2014) describes this hubris: “For counterinsurgents, tactics became the end-
all, a formula that if properly applied would win such contests, irrespective of the strategic
environment, just so long as governments have the stamina to see the enterprise through to a
successful conclusion” (174). Numerous military experts and scholars alike have arrived at this
conclusion based on large-N cross-national comparative works, which fail to appreciate the meso and
micro-level dynamics on the ground (Staniland 2014b; Berman and Matanock 2015). More recently,
proponents of the pre-eminent population-centric theory believe that “hearts and minds” can be won
over solely with select discriminate violence and material inducements, a belief that overlooks strong
pre-existing ties that may exist between insurgents and civilians within a specific space (Petraeus 2006;
Kilcullen 2010; Dixon 2009; Lyall 2009; Fjelde and DeSoysa 2009; Taydas and Peksen 2012).
While my research suggests that it is extremely difficult for counterinsurgents to militarily
defeat insurgent groups while they remain embedded in local populations, I diverge from the core
assumptions of population-centric theories that civilian support or obedience is a by-product of
material inducements and/or good governance. I posit that robust insurgent embeddedness engenders
a strong level of populational control, a dynamic which impedes civilian collaboration with the state
during periods of contestation, regardless of potential rewards and punishments being offered by
counterinsurgent forces. In this sense, my theory gravitates more towards Hazleton’s (2017) emergent
“coercion theory”, which holds that counterinsurgent success or failure is largely shaped by “[…]the
application of brute force” against both civilian populations and insurgents themselves (81). I find that
in Montes de María and Arauca, there were two interconnected factors which proved instrumental for
COIN success or failure, paramilitary collusion and indiscriminate violence against civilians
(particularly in the form of forced displacement). My findings diverge from Hazleton’s model insofar
as they seek to explain sub-national variation in counterinsurgency and my theory operationalizes
agrarian social structures rather than elite linkages as the key variable in shaping these outcomes.
16
The recent growth in literature on civil wars and armed non-state governance has seen the
emergence of a modest research agenda on the role played by pro-state militias and paramilitary
groups, a canon of work that has challenged the role of the state as a unitary actor in armed conflicts
(Romero 2003; Duncan 2006; Jentzsch et al. 2015; Daly 2016; Gutiérrez-Sanín 2019). This is important
insofar as more than two-thirds of civil wars over the past three decades have seen the deployment of
paramilitary groups, reflective of the changing nature of armed conflict and irregular warfare, and of
the utility of these groups in performing key counterinsurgency functions (Stanton 2015; Cigar 2014;
Staniland 2015). According to Aliyev (2016), however, there has been a theoretical bias in this
emergent literature on militias and paramilitary groups: “[…]although state-manipulated militias have
been featured prominently in the existing theory of paramilitary violence, it is the state-parallel groups
that have thus far remained unnoticed and undertheorized” (500). Addressing this theoretical deficit,
my work contributes to the growing literature the role played by independently created militia groups
in counterinsurgency operations. Specifically, I highlight how state-paramilitary collusion can create
the conditions conducive to successful state COIN outcomes through the delegation of indiscriminate
violence and human rights violations to these parastatal groups (Ron 2002; Alvarez 2006; Mitchell,
Carey, and Butler 2014).
Any serious undertaking which seeks to analyze counterinsurgency in Colombia, past or
present, needs to assess the role played by paramilitary groups in the development of anti-subversive
security initiatives, a phenomena which owes much to the prevailing U.S. counterinsurgency doctrine
during the Cold War (Thomson 2018). Even though Colombia has historically been characterized as
a democracy, albeit for long periods an exclusionary one, there has been a long history of Faustian
bargains between the official armed forces and these paramilitary structures, counterinsurgent
marriages of convenience which are normally attributed to authoritarian regimes during civil wars
(HRW 1996; Lyall 2010, Byman 2016). Whereas these groups composed in the 1960s could be
17
classified as “state-manipulated militias”, from the 1980s onwards paramilitary groups in Colombia
have been “state-parallel groups”, entities composed of different constellations of landed elites and
drug traffickers which had: “[…]slightly different components…tied to disparate grievances and
opportunities” (Gutiérrez-Sanín and Vargas 2017, 746). This variation aside, regional paramilitary
blocs were largely counterinsurgent expressions formed in response to insurgent predation against
landed elites, which almost always counted on some form of direct collusion with the Colombian
military (Gutiérrez-Sanín 2014). At the nadir of the armed conflict in the 1990s and 2000s, “[…]the
vast majority of non-combat politically-motivated killings, disappearances, and cases of torture {were}
carried out by army-backed paramilitaries” (Amnesty International 2005, 3-4). Other statistical
grounded research has found that while increased U.S. military aid to the Colombian armed forces
failed to reduce insurgent violence, it did lead to “[…]the diversion of foreign military aid from the
Colombian military to illicit paramilitary groups”, thereby strengthening their military capacity (Dube
and Naidu 2015, 266). Therefore, my theory posits that the repertoires of violence deployed by
paramilitary groups in Montes de María and Arauca was fundamental in explaining divergent sub-
national counterinsurgency outcomes in these two regions, particularly their usage of indiscriminate
violence against civilians.
State collusion with paramilitary structures and massive forced displacement clearly do not fall
in the best practices of the population-centric theory of counterinsurgency (Watts et al. 2014). Most
COIN successes, historic and modern, owe much to the deployment of unethical methods, often
violating International Humanitarian Law (IHL), rather than good governance or any prioritization of
the wellbeing of civilian populations (Hazleton 2017; Valentino, Huth, and Balch-Lindsay 2004;
Böhmelt et al. 2019). My work suggests that most counterinsurgent victories can be attributed partly
or fully to such tactics, an argument that is supported by many important cases from the distant and
recent past. The forced relocation of civilian populations has occurred in “[…]almost a third of all
18
counterinsurgency campaigns since 1816”, and has proven particularly useful in a variety of conflict
settings, whether deployed against Boer settlers in the Transvaal, ethnic Chinese migrants in British
Malaya, or anti-colonial nationalists in Algeria (Zhukov 2015, 1155; Lackman 1985; Hack 1999). This
tactic constitutes one of many violations of IHL seen in numerous “successful” scorched earth
counterinsurgency campaigns, indiscriminate and disproportionate state violence conducted against
groups as diverse as Mayan peasants in Guatemala, ethnic Chechens in the North Caucasus, and Tamil
civilians in Sri Lanka (Flynn 1984; Kramer 2005; Staniland 2014b).
For reasons of state weakness or plausible deniability, professional militaries often outsource
these questionable functions to paramilitary groups, a delegation of dirty war tactics that the
Colombian paramilitary boss, Fidel Castaño, once described in an interview:
The military can’t eradicate the guerrillas. I tell you that myself as I have waged war with them. The only way to defeat them is by finishing off the guerrilla’s social bases in every region and to create paramilitaries with what remains. The difference with the guerrillas is that they enter zones for the first time, where there is no violence, and they can fraternize with people and win them over without having to sacrifice anybody…Paramilitaries enter zones that are ravaged by violence and then they aren’t able to distinguish between guerrillas and peasants. They carry out a general cleansing and only after can they begin to talk with people. (Reyes Posada 2009, 131; emphasis added)
The cleansing operations referred to above involved the violent elimination or displacement of
any civilian suspected of being an insurgent sympathizer, as these counterinsurgent campaigns of
“draining the sea” through massive forced displacement was largely delegated to paramilitary groups
(Romero 2003; Tate 2003; Gutiérrez-Sanín 2019). Most, if not all, successful counterinsurgent
outcomes at the sub-national level owed much to the brutal efficiency of the paramilitary dirty war
conducted against civilian populations in collusion with the armed forces, a symbiotic campaign which
saw Colombia become one of the largest producers of internally displaced persons (IDPs) in the world
(CODHES 2010; Ibañez 2008; Daly 2016). Similar to other phenomena seen in civil wars such as
recruitment and collaboration, it is hard to ascribe any one underlying motive for why forced
displacement occurs. The expansive literature conducted on the topic lists socio-political loyalties
(perceived and real), land redistribution, economic modernization, private accumulation, institutional
19
incentives, and armed violence as the main drivers behind forced displacement in Colombia (Vargas
and Uribe 2017; Gutiérrez-Sanín 2019; Steele 2017; Ballvé 2020). My work builds on this canon by
emphasizing the importance of direct indiscriminate displacement, or the intentional yet indiscriminate
forced displacement of entire communities of people, for successful counterinsurgency outcomes.
The study of how pre-existing organizational, institutional, and social endowments in
communities governed by armed non-state actors can influence their relations with civilians possesses
clear analytical limitations. While the networks that insurgent groups employ and depend upon to
build national-level organizations are of critical importance to their ability to challenge the state and
withstand the expected violent backlash from the ruling classes, there is also dramatic variation at the
sub-national level between disparate units of the same insurgent organizations (Staniland 2014a;
Sarbahi 2014; Arjona 2016a; Daly 2016; Gutiérrez Sanín 2019). It has been clearly established that pre-
existing endowments shape and determine the degree to which armed non-state actors are able to
embed themselves into particular spaces, although there is conflicting evidence as to the nature of this
relationship. Some scholars maintain that stronger social cohesion and organizational networks
engenders greater collective resistance to non-state rule and thus limits the extent of armed non-state
actors’ ability to govern local communities (Rappaport 2007; Brewer 2010; Arjona 2016a; Kaplan 2013;
2017). Others have demonstrated that armed actors can co-opt, infiltrate, and develop organizations,
formal and informal institutions, and networks to further consolidate control over populations and
2020). My research speaks to both literatures, as I find that the ability of armed groups to co-opt these
organizational endowments depends largely on the structure and density of these networks, as well as
the ability of armed groups to appropriate the primary cleavage as their own and successfully resolve
the dominant peasant grievance in a particular space.
20
Stathis Kalyvas’s (2006) control-collaboration model greatly advanced our understanding of
how and why violence is generated in local contexts during civil war. It similarly offers valuable insight
into the conditions under which counterinsurgents are able to gain the crucial civilian support needed
to defeat insurgent groups. However, this model overlooks the role of socio-historic conditions which
may facilitate or impede cooperation with armed actors in contested spaces. The assumption that
violence is contingent on control of territory and the level of collaboration that armed groups obtain
from local populations is largely correct, but distant and recent history is replete with examples where
armed actors in contested spaces found local cooperation (and reliable intelligence) extremely hard to
come by despite their overwhelming territorial dominance, whether the Union forces in the
Shenandoah Valley of Confederate Virginia, the British Army in Western Transvaal during the Boer
War, or the 32 Battalion of the South African Defense Forces in Owamboland in South West Africa
(Eland 2013; Pakenham 1979; Steenkamp 1989). Similarly, armed groups with weaker levels of
territorial control have been able to prevent such civilian collaboration quite effectively due to other
local endogenous mechanisms (Vargas 2009; Bhavnani et al. 2011). As demonstrated by the cases of
Montes de María and Arauca, the degree of territorial control does not always produce a
commensurate level of collaboration and information which is required to defeat opposing actors.
High levels of insurgent populational control can make the cost of collaboration or defection to the
state substantially higher and hence unfeasible or unappealing, regardless of the potentially personal
risks they may assume for non-compliance with whatever actor has the greatest degree of control
within said space.
Many scholars highlight the prevalence of economic factors such as poverty and local resource
endowments, or “[…]whether insurgent groups have access to material resources”, as the primary
factors shaping insurgent-civilian relations (Weinstein 2007, 327; Ross 2004; Humphreys and
Weinstein 2006). The existing literature on the greed and grievance drivers of armed conflict has
21
examined the relationship between resource endowments, civil war onset, and insurgent mobilization
in great detail (Collier and Hoeffler 1998; Keen 1998; Hegre 2002). Most quantitative and case study
based research suggests that countries with resources such as oil and coca invariably are more likely
to experience the onset of civil war onset than those that do not (Collier and Hoeffler 2004; de Soysa
2002). In the case of Arauca, insurgent groups arrived and had already consolidated their presence
before the discovery of oil or the start of coca boom in the 1990s and 2000s. More importantly, if
economic endowments were in fact the sole source of insurgent strength and embeddeness in civilian
communities then it would have been reasonable to expect the insurgent groups to have been strongest
in the Araucan plains, which clearly benefitted from oil royalties more than all of the other
municipalities combined. While insurgents in Montes de María did not possess the same resources as
their colleagues elsewhere in Colombia, they nevertheless managed to control numerous lucrative
economic activities in the region until 2005, yet this did not enable the group to resist counterinsurgent
efforts to physically separate them from their civilian bases in the years prior.
Of equal importance, these cases suggest that the finances available to insurgent groups can
actually strengthen relations with local civilian bases, if they channel these resources back into the
communities in the form of social services and public works. Furthermore, when insurgent groups do
not depend on local individual contributions to finance their operations, peasants are more likely to
support them or tolerate their presence than in those regions where rebels resort to predation and
banditry to sustain themselves (Gutiérrez-Sanín 2008; Evans 2014; Olson 1993). My research finds
that insurgent groups that are able to capture and control the “economic sphere” are able to further
bind civilian communities into their established order due to dependence mechanisms, an additional
means of generating insurgent-civilian embeddedness in such contexts. Socio-economic factors such
as poverty and lack of economic opportunity almost certainly helped insurgent recruitment in both
Montes de María and Arauca, yet the fact that there was no substantial difference between these socio-
22
economic indicators in the two regions suggests that this was not a key variable in producing different
counterinsurgency outcomes. This is especially notable in the extremely high rates of rural poverty
found across both regions, corresponding with Leites and Wolf’s (1970) assertion that: “Historically
the success or failure of insurgency has not borne a simple relationship to the degree of poverty” (17).
Geographic and topographical considerations are frequently operationalized as an explanatory
variable in the study of insurgent mobilization and civil war. According to Fearon and Laitin (2003):
“Mountainous terrain is significantly related to higher rates of civil war” (85). The most famous
practitioners of 20th century guerrilla warfare plied their trade in such terrain, from Mao in the Jinggang
mountains, to Vo Nguyen Giap in Dien Bien Phu, to Fidel Castro in the Sierra Maestra, and it has
become conventional wisdom in the study of armed insurgencies that “[…]fighting on favorable
ground and particularly in the mountains presents many advantages” (Guevara 2002, 63). The history
of the Colombian civil war bears this observation out, but upon closer inspection it may not reveal
much about the role of topography on counterinsurgency outcomes. As Tollefsen and Buhaug (2014)
point out, mountainous terrain covered in forest or jungle is even more advantageous, as they
“[…]present major obstacles to armored vehicles and other heavy equipment as well as putting a strain
on supply lines, and dense forest canopies hinder aerial detection”(6). Thus, it is tempting to assume
that the labyrinthine hills covered in tropical jungle that Montes de María’s topography provides would
be ideal terrain for armed insurgents, whereas the virtually flat lowlands and open plains found
throughout Arauca would be considerably less suitable for the formation of a rebel movement.
However, as demonstrated by the results of Democratic Security Policy in both of these regions,
topography neither explains the variation in outcomes between two regions with dramatically different
physical terrain, nor between the sub-regions themselves found in Arauca.
Of similar importance, the availability and proximity of border refuges is of enormous value
to insurgent groups, as “[b]y moving from one side of the border to the other, the insurgent is often
23
able to escape pressure or, at least, to complicate operations for his opponent” (Galula 1964, 35;
Trinquier 1964; Scott 2000; Goodhand 2005; Salehyan 2009). As my cases demonstrate, insurgents in
Arauca were able to take advantage of the sanctuary made available to them in neighboring Venezuela,
a geographic endowment that was not available to insurgents from the very same groups in Montes
de María given its geographic location. Yet, it is difficult to ascribe any causal determinism to this
explanation due to the simple fact that state expansion was successful in expelling insurgent groups
from plains municipalities sharing extensive borders with Venezuela (Arauca municipality, Cravo
Norte). However, the very same strategy failed to dislodge and expel the same insurgent fronts based
in those piedmont municipalities located on the border (Arauquita, Saravena), and in the interior
(Fortul, Tame).
E. My Theoretical Model
My theory is based on explaining two causal processes key to understanding counterinsurgency
outcomes: (i) insurgent embeddedness in the human terrain, and (ii) counterinsurgent efforts to
disembed insurgent groups from civilian populations. Following state-centric frameworks established
by important works on insurgent formation and revolution in the Global South, I propose a similarly
holistic analysis of state expansion, albeit one that is inverted. As Jeff Goodwin (2001) notes:
“[…]successful revolutions necessarily involve the breakdown or incapacitation of states” (24).
Timothy Wickham-Crowley (1992) maintains a similar position by arguing that specific regime
characteristics are key to understanding the strength and efficacy of revolutionary movements that are
pitted against them: “Indeed, it was the nature of the regimes themselves that increased the likelihood
that the opposition would unite across classes and despite ideological differences; hence the regime
itself served to strengthen the opposition” (7-8). The study of counterinsurgency, however, focuses
on the inverse puzzle, or how the state confronts insurgencies militarily at the national and sub-
national levels. Therefore, I examine both the counterinsurgency strategy to expand its presence into
24
spaces controlled by armed non-state actors, and more importantly, the specific characteristics of these
insurgent factions and their capacity to resist these counterinsurgent efforts. My model adheres to
historical and contemporary works on insurgency and counterinsurgency by arguing that the strength
and cohesion of a particular insurgent structure is contingent on its level of embeddedness in local
civilian communities (Zedong 2005; Guevara 1961; Trinquier 1964). The strength or weakness of such
insurgent-civilian linkages plays a deterministic role in whether an insurgent group can survive a
substantial counterinsurgent challenge or not. As Mampilly (2011) notes, most modern state expansion
campaigns are guided by a fundamental principle which focuses on “[…]severing the ties between an
insurgent organization and its civilian support base” (55).
Figure 1. Analytical Sequence of Insurgency & Counterinsurgency
Source: Author’s Elaboration
The structure of this historical-institutional model is quite simple. There are three phases and
two critical junctures which shape the overall sequence of each case examined. The pre-existing
configuration of a territory (Phase 1) is key to understanding the resources and constraints which
insurgent groups encountered upon their insertion into these spaces. The pre-existing configuration
phase primarily focuses on the formation and evolution of the agrarian social structure, or the
predominant rural class hierarchy which consolidated in these peripheral spaces prior to the arrival of
25
armed non-state actors. Leon Zamosc (1986) highlights how rural social structures that emerge during
this period ultimately shape subsequent interactions between key actors and outcomes:
The socioeconomic structure of the countryside shapes the conditions for economic performance, defines the relative strength of the rural social classes, and sets the stage for their participation in the political processes linked to the development of capitalism. The way in which the agrarian question is resolved depends, in turn, upon a complex mix of historical, socio-economic, and political factors unique to each country. (7)
Agrarian social structures are characterized by different logics that are shaped by competing
demands from different social groups and their representative organizations which each seek to
structure and order local societies according to their interests. On the one side are landed elites, or the
dominant landholding class that promotes latifundia by seeking to accumulate and maximize their
control of land through an assortment of legal, political, and coercive mechanisms, generally at the
expense of the local peasantry who are forced to work for them under unfavorable terms (Boone
2014; Cramer and Wood 2017; Woods 2020). On the other side are peasants, or those who work in
agriculture as: “[…]a landless day laborer, a permanent wage employee, or a farmer working a small
holding” (Wood 2003, 5). In contrast to landed elites, peasants promote either a collective or
smallholding system of agriculture, seek access and title to land, a source of credit, and basic
infrastructure to develop external markets for their products (Giddens and Held 1982; Horowitz
2002). Although these are other important rural constituencies (merchants, middle-peasants, etc.) in
agrarian social structures which play a role in shaping these hierarchies, my model prioritizes
parsimony in light of the fact that rural land tenure patterns in the Global South tend to be highly
stratified between landed elites and peasants, demonstrating a need for analytical simplicity (Griffin et
al. 2002; Margulis et al. 2013).
The expansive study of civil wars has demonstrated that a deficiency of state presence, or the
relationship of groups to structures of power, in a specific territory makes it easier for other actors to
take root and develop territorial and populational control in the absence of the central government’s
26
presence (Tilly 1969; Humphreys 2005; Reno 2003). In many cases, the uneven distribution of state
power throughout a national territory owes much to the colonial legacy of direct and indirect rule
(Mamdani 1996; Naseemullah and Staniland 2016; Besley and Reynal-Querol 2014; Wucherpfennig et
al. 2015). In the virtual absence of the state, different rural constituencies compete to shape the
character of local society, deploying distinct regulatory mechanisms used to resolve collective action
problems (Brewer 2010; Balcells 2017; Staniland 2014a; Arjona 2016a). These mechanisms are often
organizational endowments, both formal and informal, which reflect the collective action capacity of
a particular community (Oberschall 1973). However, these organizational endowments also tell us a
lot about the composition of agrarian social structures, and more importantly, the primary cleavages
circumscribing them (Gould 1995). Landed elites form their own clubs and associations to both
regulate their agrarian social structures, and to protect their interests from potential state reforms
and/or peasant unrest. Peasants similarly mobilize behind organizations created to administer and
protect their communities from external predation, and to articulate their demands vis-à-vis landed
elites and the state (Zamosc 1986).
Figure 2. Key Variables, Critical Junctures, and Outcomes
Source: Author’s Elaboration
Agrarian social structures strongly influence how insurgent groups decide to insert themselves
into a particular space. Such groups are attracted by conditions which appear conducive to
constructing an active and prosperous rebellion, a phenomenon Michael Rubin (2018) observes: “Civil
27
war is fought on human terrain, in which the local population influences the belligerents’ interests in
seizing territory, the costs of doing so, and the capability to win and retain control” (4). David Kilcullen
(2010) places special emphasis on these primary cleavages in processes related to insurgent formation
and consolidation:
The center of gravity of an insurgent movement – the source of power from which it derives its morale, its physical strength, its freedom of action, and its will to act – is its connectivity with the local population in a given area. Insurgents tend to ride and manipulate a social wave of grievances, often legitimate ones, and they draw their fighting power from their connection to a mass base. (8)
Agrarian social structures ultimately shape and determine the primary cleavage that
circumscribes local spaces, social ruptures that represent local grievances and disputes rather than the
master cleavage of a civil war (Kalyvas 2003; Roldán 2002). According to Lubkemann (2005): “In
fragmented war contexts…the violence of military actors is appropriated and deployed in the service
of local-level social conflicts” (501). My model is based on the Maoist assumption that insurgencies
prioritize peasant support over that of landed elites, although it allows for the possibility that civilian
support does not necessarily function as a binary and in many cases armed actors will simultaneously
attempt to solicit support from multiple rural constituencies. However, this can prove difficult in
agrarian social structures where the primary cleavage exists between landed elites and the peasantry,
particularly if it is tied to the volatile “land question” (Mason 1998; Wood 2003; Albertus and Kaplan
2013).
Although some insurgent groups arrive to a zone abruptly and in large numbers, others slowly
infiltrate and build up their presence progressively. Regardless of the pace, this specific moment
(Critical Juncture A) begins the period of insurgent institutionalization (Phase 2). This phase represents a
causal process, or the “[…]particular type of sequence in which the temporally ordered events belong
to a single coherent mode of activity” where an insurgent group attempts to assume control of various
functions related to governance, whether it be the regulation of rural land tenure patterns, local
political contests, agrarian economies, or civic participation (Falleti and Mahoney 2015, 214). Some
28
groups limit their responsibilities to providing security and order and little else, while others intervene
in the most intimate aspects of civilians lives in areas under their control (Mampilly 2007; Wiegand
2010; Arjona 2016b). The extent to which armed groups immerse themselves in local societies depends
on their immediate and long-term objectives, group ideology, and the extent and the degree to which
civilian communities collectively accept or resist their rule (Weinstein 2007; Gutiérrez Sanín and Wood
2017; Kaplan 2013).
The insurgent institutionalization phase indicates the arrival and consolidation of an insurgent
group into a particular space, and ultimately represents the period where insurgent embeddedness is
formed between these armed groups and the civilian communities in which they are operating. Given
how crucial insurgent embeddedness is in determining state expansion outcomes, it is necessary to
examine how this factor is generated in such contexts. David Galula (1964) claims that embeddedness
itself is determined by “[…]the tacit or explicit agreement of the population or, at worst, on its
submissiveness” (8), while his contemporary Roger Trinquier notes that “[…]the sine qua non of victory
in modern warfare is the unconditional support of a population” (1964, 8). Robert Taber (1965)
explains why civilian buy-in is so deterministic in forming insurgent embeddedness:
{The}population is the key to the entire struggle. Indeed, although western analysts seem to dislike entertaining this idea, it is the population which is doing the struggling. The guerrilla, who is of the people in a way which the government soldier cannot be, fights with the support of the non-combatant civilian population; it is his camouflage, his quartermaster, his recruiting office, his communications network, his efficient, all-seeing intelligence service. Without the consent and aid of the people, the guerrilla would merely be a bandit, and could not long survive. If, on the other hand, the counter-insurgent could claim this support, the guerrilla would not exist. (23)
Numerous works have demonstrated how civilian support in civil wars actually functions
(Petersen 2001; Wood 2003; Kalyvas 2006). My model builds on these various established
conceptualizations of civilian support that armed actors depend on. Jeffrey Sluka (1989) offers a binary
of “hard support’ and “soft support” in which the former represents those civilians that “[...]are
prepared to act legally or illegally in support of {insurgents}”, while the latter indicates those who
“[…]may do nothing at all other than tolerate the presence of the guerrillas” (137, 143). Ana Arjona’s
29
(2017) more recent typology of civilian support and non-support for armed actors breaks down hard
support by distinguishing between “enlistment” and “spontaneous support”, while soft support is
categorized as “obedience”, or behavior that “[…]entails any action by a civilian after an armed group
has ordered her to do so, either directly or by establishing a general rule” (762). My model similarly
holds that while most armed groups attempt to maximize hard support for their rule, they expect
obedience as a minimum requirement of civilian populations living under their control.
Similar to the existing research on insurgent recruitment, it is enormously difficult to broadly
ascribe civilian support, both hard and soft, to any particular motive whether sympathy, revenge,
personal association, ideology, financial gain, or the prevailing set of socio-cultural norms (Wood 2003;
Lyall 2009; Reno and Matisek 2018). However, what ultimately matters for insurgent groups at the
community level “[…]is their unwillingness to act in any way that would interfere with the guerrillas”,
or for civilians living under their rule to obey their rules and not deviate from the established social
code (Sluka 1989, 137). My model maintains that agrarian social structures circumscribed by state
absence and/or local class conflicts provide armed insurgent groups with an opportunity to “[…]fill
a vacuum and cultivate all forms of cooperation” by appropriating these local cleavages, whereby
“[…]the armed group gains the opportunity to shape beliefs in ways that render both obedience and
spontaneous support” (Arjona 2017, 767). The most effective way of achieving local buy-in is through
the co-optation of local organizational endowments which provide potential anchors for armed
insurgents to foster widespread civilian adhesion to their nascent order. As Gould (1995) notes:
“[…]formal organizations – including clubs, correspondence committees, and militias – can exert an
enormous influence on the scale at which group identities are convincing to potential participants in
collective action” (21). Apart from deeper insurgent embeddedness, these mechanisms provide a
greater capacity for populational control, as insurgents can better monitor and enforce civilian
obedience in spaces under their control.
30
The true test of insurgent embeddedness arises when these groups are challenged by
counterinsurgent rivals who seek to disembed them from the civilian communities in which they
operate, permanently displacing them these spaces. The moment (Critical Juncture B) demarcating the
end of the insurgent institutionalization phase and the beginning of the period of contestation (Phase 3)
occurs when a counterinsurgent rival appears in sufficient numbers to violently contest the insurgent
social order by attempting to take control of territory, resources, and local populations. There is
substantial variation in the manner in which counterinsurgent groups attempt to do this and the period
of contestation almost always escalates and intensifies the local armed conflict, thereby leading to
higher levels of violence in the process (Kalyvas 2006; Metelits 2011; Wood 2010). My model argues
that while insurgent groups require some hard support in relation to intelligence and manpower to
resist counterinsurgent attempts to expel them, what really matters to them during these conflicts is
widespread soft support in the form of obedience so as to prevent civilians from collaborating with
their opponents. Conversely, counterinsurgents attempt to weaken insurgent embeddedness in order
to prevent civilians from cooperating with these groups in an attempt to isolate and destroy them.
Therefore, the conflict between insurgents and counterinsurgents becomes a struggle for control of
the civilian population. The insurgents’ capacity to resist and withstand the counterinsurgent onslaught
depends in large part on their ability to remain embedded in local civilian populations, whereas the
counterinsurgents seek to weaken this embeddedness either through coercion and physical separation,
or good governance and material incentives.
Given that a central goal of counterinsurgency strategy is to disembed insurgent groups from
the civilian populations in which they operate, it is only logical that these armed non-state actors will
both attempt to protect civilians from counterinsurgent violence deployed to weaken this
embeddedness, and to deter civilians from collaborating with their opponents. The ability of insurgent
groups to protect rural populations from counterinsurgent attempts to forcibly separate the two is of
31
crucial importance to the eventual outcome. While some groups prioritize protecting territory and
resources over people, my theory adheres to the maxim that: “The richest source of power to wage
war lies in the masses of the people” (Zedong 2005, 186). As David Kilcullen (2010) points out:
“[…]insurgents cannot operate without the support – active, passive, or enforced – of the local
population” (4). Echoing this, Suykens (2014) argues that: “Any rebel group interested in maintaining
legitimate relations with the population must offer protection. Without protection, at least against
counterinsurgency forces, civilians have little incentive to cooperate with rebels” (145). More than any
other resource endowment available to insurgents, civilian support will determine their ability to resist
and challenge the state in contested spaces, a strategic military benefit highlighted by Galula (1964):
“As long as the population remains under his control, the insurgent retains his liberty to refuse battle
except on his own terms” (15).
Continued civilian sympathy for insurgent groups during periods of contestation can manifest
itself in important ways that differ from the usual contribution of manpower, supplies, or information.
Timothy Wickham-Crowley (1992) emphasizes this fact: “Peasants, like others, have practical
resources to offer to the guerrillas. If peasant sympathy means that a fleeing guerrilla will be offered
shelter, or that, under even torture, peasants will not reveal guerrilla locations, then we may consider
those actions as indicators of “support”, rather than the feelings themselves” (53). Under the logic of
asymmetrical civil war, counterinsurgents possess a military and coercive advantage over insurgent
groups, which may or may not be sufficient to overcome the superior informational capacity
maintained by insurgent groups (Kalyvas 2006; Zhukov 2015). In the face of overwhelming
counterinsurgent pressures, insurgent groups seek “[…]not to win battles, but to avoid defeat, not to
end the war, but to prolong it” (Taber 1965, 147). However, for insurgent groups to avoid defeat and
prolong an armed conflict is extremely difficult, if not impossible, without peasant support. As T.E.
Lawrence (2005) highlights, insurgencies “[…]must have a friendly population, not actively friendly,
32
but sympathetic to the point of not betraying rebel movements to the enemy” (284). In order to
achieve these tactical goals, insurgent groups need to prevent counterinsurgent attempts designed to
physically detach them from their social bases of support, while also providing a credible deterrent
against civilian collaboration with their rivals. My model suggests that insurgent groups that are more
heavily embedded in civilian populations have greater populational control due to the number of
enforcement and monitoring mechanisms available to them to better protect populations under their
control and to impede civilian defection to the state.
On the other side of the coin, the efficacy of counterinsurgent forces to disembed their
insurgent rivals depends on both the embeddedness of their insurgent rivals in civilian populations
and counterinsurgent linkages to local constituencies disaffected with rebel rule. As Dasgupta (2009)
highlights: “Local allies are central to counterinsurgency campaigns, but their role is not well
understood beyond the provision of general political support and access to local knowledge” (1). Apart
from these two obvious benefits, local allies who are willing to collaborate and even join
counterinsurgency operations are often those on the losing end of primary cleavage, whether large
landholders or ethno-religious minority groups (Gutiérrez-Sanín 2019; Abbs et al. 2020). Whereas they
represent potential local buy-in for counterinsurgents to appropriate meso and micro-level conflicts,
counterinsurgent forces appear as a way for these local constituencies to address the primary cleavage
circumscribing their communities. Local allies provide more than just material resources, potential
recruits, and local intelligence; they offer an expansive pre-existing social network that can be
mobilized in any number of ways against insurgent groups and their civilian bases of support (Brewer
2010; Sarbahi 2014; Daly 2016).
Although it may seem intuitive, the ability of counterinsurgent forces to separate their
insurgent rivals from civilian populations is shaped by the level of insurgent embeddedness in these
very same communities. Quite simply, the more embedded insurgents are in a specific space, the
33
harder it will be to disembed them from it. In contrast to the good governance and material
inducements prescribed by proponents of population-centric COIN theory, my model argues that
coercion and forced relocation has proven to be the most effective counterinsurgency method,
historically and presently (Plakoudas 2016; Hazleton 2017). Whether this tactic is planned and directed
by military commanders who forcibly relocate civilian populations to state-controlled spaces, or third-
party militias or paramilitary groups who forcibly displace entire “enemy communities” with selective
and indiscriminate violence, the result deprives insurgent groups of familiar human terrain, while
allowing counterinsurgent forces to develop more aggressive military operations against their
opponents (Zhukov 2015; Valentino, Huth, and Balch-Lindsay 2004).
However, separating civilian populations from insurgent groups through massive forced
relocation may succeed in “draining the sea” for counterinsurgent forces to better locate and target
their opponents, but it is not necessarily a guarantee of military success in itself. While Kilcullen (2010)
is correct when he argues that “[…]cutting the insurgent off from the population is a critical task in
counterinsurgency” (8), John Nagl (2002) posits that anti-guerrilla forces still need to persuade civilians
“[…]not to fight on behalf of, nor even support, the insurgents” (25). Similar to insurgent groups,
counterinsurgents can bind civilians to their own nascent order to establish populational control,
providing basic protection and “[…]human security to the population, where they live, 24 hours a
day” (Kilcullen 2009, 486). The sudden appearance of an armed competitor into contested spaces can
alter the pre-existing structural and social landscape, a sea-change that provides counterinsurgents with
an opportunity to appropriate shifting cleavages and to bind civilians to its own order by employing
“[…]a variety of means that have the effect of convincing the populace that their interests are better
achieved through siding with or acquiescing to the counter-insurgents” (Evans 2014, 259). Over the
course of violent local conflicts in which civilians are caught between armed actors trying to control
them, the primary cleavage can quickly change from pre-conflict grievances to more pressing concerns
34
such as safety and survival (Galula 1964). In such contexts, civilians are often willing to deny or shift
their support away from insurgent groups if another actor can better guarantee their lives, yet this
largely depends if they are physically separated from their former protectors or not.
The focus of the existing literature on counterinsurgency and civil war has centered on a
limited array of potential outcomes. These tend to gravitate towards military victory (incumbent or
insurgent), stalemate, and negotiated settlement (DeRouen and Sobek 2004; Licklider 1993; Mason
and Fett 1996; Walter 2002). As the cases of Arauca and Montes de María demonstrate, there is a
wider range of potential outcomes in such instances of state expansion. In regards to my own range
of outcomes these can either represent a definitive and conclusive result as a product of this state
expansion (e.g. victory/defeat), or are classified in more ambiguous terms
(advantage/stalemate/disadvantage) after a sufficient period of time has transpired without a clear
victor emerging. These more indefinite and fluid outcomes bear much in common with other
conceptualizations of fragmented power structures such as Vladimir Lenin’s (1975) “dual power”,
Charles Tilly’s (1978) “multiple sovereignty”, Paul Staniland’s (2012) typology on wartime political
orders, or more recently Gustavo Duncan’s (2014) “oligopolies of coercion”. Stalemate clearly
indicates a parity of control and power between competing actors. Advantage refers to when an actor
enjoys superior power capabilities over a rival who may be at a disadvantage yet is still able to compete.
Given the centrality of insurgent embeddedness to this theory, it is imperative to highlight the
different ways in which this variable can manifest itself in conflict settings. As demonstrated in Figure
3, there are four spheres of contestation which I use to assess the strength of competing armed actors
in a particular zone, all of which are intrinsically connected. The territorial sphere represents the
geographical space or landmass of a specific zone which different actors seek to control. The
economic sphere refers to the modes of production, extraction, and exchange which occur within the
zone in question. The political sphere encompasses formal and informal political institutions, electoral
35
contests, and political movements. The civic sphere is perhaps the most ambiguous of the four, yet
for purposes of parsimony and clarity it refers to the public space where civilians interact and organize.
People and the civilian populations they form are present across all four spheres and cannot be
disaggregated nor solely confined to any one of these. Armed groups compete with one another for
control of territory, economic resources, political power, and civil society, often in a bid to maximize
their level of populational control over local civilian communities. Whereas people draw all of these
spheres together, the numerous ways in which these spheres overlap is complex.
Figure 3. Four Spheres of Contestation
Source: Author’s Elaboration
The territorial sphere serves as a fundamental arena for insurgent-civilian interactions during
the period in which armed groups insert themselves into regional dynamics where they attempt to
control “[…]civilians and the territory upon which they reside” in order to construct and protect their
non-state social orders (Arjona 2015, 1). The pre-existing land tenure pattern, the prevailing local
systems of production, and the historic settlement pattern all shape the manner in which armed actors
will attempt to control these spaces, and the extent to which peasants and landed elites will allow them
to (Gutiérrez 2017). In some instances, armed groups will leave the pre-existing agrarian system
relatively untouched, whereas in others these actors will dramatically alter the prior system by violently
displacing targeted groups, whether peasants or landed elites, in an attempt to recalibrate these rural
36
dynamics according to their own ideological preferences (CNMH 2014a; Ballvé 2013; Gutiérrez-Sanín
and Barón 2005; Barbosa 2015). Territorial control is inextricably connected to the protection of those
populations residing within these spaces. As Mampilly (2011) notes, “[…]controlling territory allows
insurgents to offer utilitarian benefits to civilians in ways that groups without territory could never
do” (54). The importance of this is captured by Galula (1964), who observes that: “Once the insurgent
has succeeded in acquiring stable geographic bases…he becomes ipso facto a strong promoter of order
within his area, in order to show the difference between the effectiveness of his rule and the
inadequacy of his opponent’s” (12).
In contrast to Carl Von Clausewitz’s (1918) maxim that war is a continuation of politics by
other means, my research suggests that in many civil wars capturing power in the political sphere is a
means to an end. Operationalizing politics as a sphere of competition itself rather than as an
ideological impetus is thus a fairly recent phenomenon and somewhat unique in the study of
contentious politics. Various armed conflicts have emerged in countries endowed with long-
established democratic institutions and a constitutional rule of law. In such contexts, the political
sphere can also become another battleground between armed non-state actors and their opponents.
Matanock and Staniland (2018) describe this phenomenon: “[…]armed groups pursue a surprisingly
diverse array of electoral strategies. Some groups choose nonparticipation, ignoring electoral politics
or not trying to shift the balance of electoral outcomes. But many—perhaps most—armed groups
seek to use electoral politics” (710). Therefore, political power at the national and sub-national levels
can be operationzalized as a mechanism for competing actors to fortify their ranks and to gain an
advantage over their rivals.
The importance of the economic sphere cannot be overstated in the study of armed
mobilization and counterinsurgency. For any insurgency to consolidate a meaningful presence in a
particular community, it needs material support in order to house, arm, and feed its cadres (Zedong
37
2005; Guevara 1960; Galula 1964). Conversely, for any counterinsurgency campaign to succeed, it
“[…] requires that attention be devoted to counterproduction efforts…rather than counterforce
efforts alone” (Leites and Wolf 1970, 78). Contending actors in civil wars compete for control over
local economies, legal and illegal, by offering security to peasants, merchants, landed elites, and
corporate interests alike in exchange for material support (Richani 2005; Gutiérrez-Sanín and Vargas
2017). In rural based armed conflicts, local agriculture serves as the basis for all economic activity, as
“[…]the real bases of economic and social life were located in the countryside, where haciendas and
peasants produced surpluses for the limited regional markets as part of their traditional pattern of
cattle raising and agricultural production” (Zamosc 1986, 9). Other economic activities may co-exist
alongside the agricultural economy, some legal (oil production, mining, external trade), and others
illegal (drug production/trafficking, illegal mining, contraband), yet all of these potential revenue
streams are fiercely contested over by rival actors in sub-national theatres of war (Duncan 2014).
Those armed actors that are able to control, regulate, and ultimately monopolize the economic sphere
often benefit from ensuing dependence this creates, a mechanism which further binds civilians to their
incipient social order.
The civic sphere represents what we broadly refer to as civil society, or the space outside of
formal political institutions where individuals and groups “[…]undertake collective action for
normative and substantive purposes”, one which is “[…]relatively independent of government and
the market” (Edwards 2011, 5). In the context of civil war, armed actors generally attempt to either
co-opt or negate the civic sphere outright (Ramírez 2011; DeGregori 1990; Romero 2000; Mustafa
and Brown 2010). Failure to control or destroy the civic sphere leaves strong repositories for civilian
collective action intact. These in turn can potentially serve as counterweights to the ability of these
actors to impose their will on civilian communities by mobilizing popular resistance against such
coerced forms of rule (Sauders 2011; Kaplan 2013, 2017; Arjona 2016a). Control of the civic sphere
38
is quite advantageous in terms of establishing populational control as it serves a fundamental role in
providing anchors or barriers to aspiring insurgent groups and counterinsurgent forces alike, while
also offering the means to mobilize populations and achieve immediate and long-term group goals.
The civic sphere provides the space for the creation of organizations, both formal and informal, to
resolve immediate collective action problems at the local level. As Gould (1995) points out, such
organizations foster “[…]the creation of social ties that encourage the recognition of commonalities
on a scale considerably broader than what would be expected on the basis of informal social networks
alone” (22). According to Oberschall (1973), these organizational endowments provide vital resources
for groups seeking to mobilize such as: “[…]a pre-established communications networks, resources
already partially mobilized, the presence of individual and leadership skills, and a tradition of
participation among members of the collectivity” (125).
F. The Structure of this Dissertation
Following the introduction, my dissertation is divided into six subsequent chapters. In Chapter
Two, I detail my research design, methodology, case selection, before providing a brief national level
history of Colombia in order to provide the appropriate context for readers unfamiliar with the
Colombian case. Chapter Three examines the Pre-existing Configuration phase in both Montes de
María and Arauca and highlights the antecedent conditions related to agrarian social structures, all of
which reveal the source and character of the primary cleavage that would eventually emerge and
circumscribe social conflicts in both zones. Chapter Four focuses on the arrival, consolidation, and
‘institutionalization’ of insurgent governance in the two zones, and more specifically how these
differed enormously across time and space. In Chapters Five and Six, I break down the period of
contestation which emerged as a result of the arrival of counterinsurgent forces into both Montes de
María and Arauca, and more specifically how this escalation of conflict altered the distribution of
39
control of the territorial, political, economic, and civic spheres between competing actors. These dual
chapters evaluate the outcome of state expansion in qualitative and quantitative terms and then explore
how these different outcomes were produced in these two zones according to my theoretical model.
Finally, Chapter Seven offers a cursory examination of the historical legacies of the Colombian state’s
intervention into Arauca and Montes de María, the external validity of my theory, and the theoretical
contributions towards other cases and the potential policy implications of my findings.
40
Chapter Two. Research Design
“In Colombia, the question is: Who is going to kill us? The guerrillas, the paramilitaries, the narcos, or the politicians?”
- Jaime Garzón 1
A. Methodology
My research and explanatory model follows the lead of other historical-institutional based
works in the social sciences (Steinmo, Thelen, and Longstreth 1992; Mahoney 2000; Thelen 1999;
Fioretos, Falleti, and Sheingate 2016; Hall and Taylor 1996; Campbell 2012; Pierson 2004; Mahoney
and Thelen 2010; Pierson and Skocpol 2002; Streeck and Thelen 2005). It would be impossible to
answer the principal research questions of this dissertation by merely examining the period in which
the outcome was finally determined. Rather my explanation as to how the Colombian state succeeded
in expelling insurgent groups from some spaces - permanently in the case of Montes de María and
temporarily in that of the Araucan plains - while failing to do so in others such as the Araucan
piedmont owes a great deal to the manner in which these territories were settled, the organizational
endowments which emerged, and the manner in which leftist guerrilla groups arrived and
institutionalized their rule over the course of many years. Much path dependency scholarship is
dedicated to the creation, consolidation, change, and breakdown of formal political institutions over
an extended period of time (Dahl 1971; Collier and Collier 1991; Rueschemeyer, Stephens, and
Stephens 1992). The research of non-state social orders bears much in common with the above-
mentioned line of social inquiry, yet has not been given the same analytical priority in the discipline.
1 Jaime Garzón (1960-1999) was Colombia’s most famous comedian and television personality, also celebrated for his sharp social commentary and political activism. He was murdered on August 13th, 1999, by paramilitary hitmen in Bogotá. In the years following his death, several judicial proceedings determined that Colombia’s top paramilitary boss, Carlos Castaño, had ordered Garzón’s assassination at the behest of several top ranking military and intelligence officials.
41
Tsai (2016) highlights this disparity, arguing that the “[…]direct theorization of informal institutions
represents a fertile frontier for comparative politics, particularly when combined with insights from
historical institutionalism about institutional stability and change” (270). Established research has
shown that armed non-state actors create and administer informal institutions in order to govern
civilian populations, a phenomenon which does not differ much from how formal political institutions
are utilized for the same purposes in non-conflict settings. As Ana Arjona (2016a) notes:
In war zones, such institutions can vary greatly as they prescribe different conducts for civilians, combatants, or both. With a variety of rules comes a variety of expectations. Civilians’ and combatants’ expectations about others’ behavior create specific patterns of social, economic, and political interaction. I define “wartime social order” as the particular set of institutions that underlie order in a war zone, giving place to distinct patterns of being and relating. (22)
My explanatory model accommodates both the role of structure and agency on the outcome
of interest by adopting what historical-institutional scholars refer to as ‘structured contingency’ (Karl
1990, 1997). Ana María Bejarano (2011) summarizes this approach succinctly: “[…]socioeconomic
structural transformations, as channeled by preexisting institutions (i.e., states and parties), add up to
a historical-institutional account of the kinds of struggles, the nature of the setting, and the types of
decisions that become available to political actors in times of change” (15). Structured contingency
provides an adequate comparative historical framework to analyze armed conflicts holistically by
focusing on the courses of action pursued by both insurgents and counterinsurgent forces following
critical junctures. According to Slater and Simmonds (2010), critical junctures are “[…]periods in
history when the presence or absence of a specified causal force pushes multiple cases onto divergent
long-term pathways, or pushes a single case onto a new political trajectory that diverges significantly
from the old” (888). My interpretation of the term adheres to this logic, especially in regards to
Mahoney’s (2000) corresponding assertion that “[…]once a particular option is selected it becomes
progressively more difficult to return to the initial point when multiple alternatives were still available”
(513). The role of agency, whether it be insurgent, counterinsurgent, or even civilian, following these
42
critical junctures is of the utmost importance. As Capoccia and Keleman (2007) point out, “[…]the
range of plausible choices open to powerful political actors expands substantially and the
consequences of their decisions for the outcome of interest are potentially much more momentous”
(343). The strategies selected by actors to govern local populations, to combat their rivals, and to
maintain internal cohesion during critical junctures can ultimately shape the outcome of interest.
Jeffrey Haydu (1998) captures the importance of this timing: “Choices in one period not only limit
future options, they may also precipitate later crises, structure available options, and shape the choices
made at those junctures” (353).
The sequence based at the center of my analysis traces “[…]a temporally ordered set of events
that takes place in a given context”, which in this case is Montes de María and Arauca (Falleti and
Mahoney 2015, 213). The first phase of this sequence (Historical Configuration) represents the
antecedent conditions, or “[…]impersonal factors such as the socio-economic conditions, class and
social alliances, diffuse cultural orientations, and the like” which ultimately shape and determine the
institutional outcome of interest (Capoccia 2016, 93).
Two critical junctures that provided exogenous shocks to these regions have been identified.
The first of these occurred with the emergence of armed non-state actors in these zones and their
subsequent attempt to assume control of territory and people, initiating the second phase of the
sequence (Insurgent Institutionalization). In recent years much research has been conducted on the
causal process where armed non-state actors emerge to govern particular communities (Mampilly
2011; Zukerman Daly 2016; Arjona 2014, 2015, 2016a), and as Paul Staniland (2014a) notes: “We need
to take history seriously to understand how armed groups emerge” (5). The tools provided by a
historical-institutional approach offer the necessary context to understand how these groups are
formed and the degree to which they consolidate control of people and territory in a given space.
43
The second critical juncture corresponds to the moment that counterinsurgent forces, whether
public or private, arrive and attempt to dislodge and displace these armed non-state actors from a
particular space once and for all (Period of Contestation). The interaction and confrontation between
various armed actors occurs in a localized setting where civilian agency also plays a critical role in the
territorial distribution of control and the production of violence (Kalyvas 2006). During this critical
juncture, the pre-existing structural constraints forged during the insurgent institutionalization phase
“[…]determine the range of options available to decision makers and may even predispose them to
choose a specific option” (Karl 1990, 7). Again, historical-institutionalism offers the best
methodological toolkit to unpack such a complex causal process where structure and agency collide
and interact to forge different outcomes in very violent and polarized settings, an evolving landscape
that David Kilcullen (2010) describes well: “[…]counterinsurgency is at heart an adaptation battle: a
struggle to rapidly develop and learn new techniques and apply them in a fast-moving, high-threat
environment, bringing them to bear before the enemy can evolve in response, and rapidly changing
them as the environment shifts” (2).
The primary method used to address the central research questions of this dissertation is
process tracing (Beach and Pedersen 2013; Falleti 2016; Bennett and Checkel 2015; Kittel and Kuehn
2013; Falleti and Mahoney 2015). George and Bennett (2005) describe process tracing as a
“[…]method [that] attempts to identify the intervening causal process – the causal chain and causal
mechanism – between an independent variable (or variables) and the outcome of the dependent
variable” (206). In the context of my project, process tracing implies a comparative historical analysis
of the primary sequence in Montes de María and Arauca in order to determine how state expansion
into insurgent controlled spaces engendered different outcomes. More specifically, my research design
and explanatory model are based around inductive process tracing, a method where “[…]the analyst
derives propositions and formulates sequences from empirical observations” (Falleti and Mahoney
44
2015, 229). Such a method is useful when examining complex causal processes such as insurgent
institutionalization and counterinsurgency because “[…]if important political outcomes depend not
on a few socioeconomic conditions but on a complex chain of strategic interaction, they cannot be
explained except by reference to that chain” (Hall 2003, 387). When analyzing such a sequence using
inductive process tracing, the goal is to unearth and develop the specific causal processes which
ultimately shape the outcome. Trampusch and Palier (2016) highlight the utility of this method for
unpacking sequences which consist of complex causal processes: “[…]inductive analysis of processes
does not merely consist of naïve observations of empirical events from which theoretical ideas are
derived, but rather forms a theoretically informed analysis (= decomposition) of processes that looks
for causal chains between the observed events” (445). In the context of my dissertation, inductive
process tracing required a laborious exploration of principally qualitative sources due to the
methodological difficulty posed by measuring embeddedness, a challenge highlighted by Durkheim
(1933): “[…]social solidarity is a wholly moral phenomenon which by itself is not amenable to exact
observation and especially not to measurement” (24). Only by observing, analyzing, and comparing
the effects of strong social solidarity through observable phenomena like civic participation,
mobilizations, and a lack of civilian willingness to collaborate with counterinsurgent forces was I able
to assess the level of insurgent-civilian embeddedness in Montes de María and Arauca, and even this
was made difficult by challenges which Jeffery Paige (1978) points out:
The sociological description and measurement of social movements are limited by the ambiguous boundaries of the phenomenon itself. Social movements seldom create the kind of repeated observable structures which make measurement convenient. Their institutional structure, tactics, and ideology change continually, their limits in time and space are ill defined, and their range of support is often difficult to determine. (86)
B. Research Design
My research design was heavily shaped by the demands imposed by the need to carry out
inductive process tracing in order to examine my case studies of interest in intimate detail. The
development of my explanatory model was guided by existing statistical data from a variety of
45
established think-tanks and government sources, the pre-existing secondary sources which focus on
these cases, an extensive archival ‘deep-dive’ of the media database housed by the Center for Popular
Research and Education (Centro de Investigación y Educación Popular – CINEP), and over one hundred
semi-structured elite and non-elite interviews conducted in Arauca, Montes de María, Cartagena,
Sincelejo, and Bogotá between January 2016 and December 2016. Those interviewed included regional
experts, academics, current and former government officials, members of the armed forces, leaders
and members of local civic organizations, human rights and peace activists, ex-combatants, indigenous
and Afro-Colombian community leaders, oil workers and union leaders, clerics, displaced persons and
other victims of the armed conflict. Depending on the interview participant, the questions centered
on the three aforementioned phases of the sequence of interest (historical configuration, insurgent
institutionalization, period of contestation) in the particular community in which they resided or
operated. More specifically, these focused on how insurgent control of the territorial, economic,
political, and civic spheres shifted and changed during these different phases. Between obtaining and
exhaustively analyzing all of the necessary field data to unpack the causal sequence in both Montes de
María and Arauca, I experienced considerable challenges, some more commonly experienced by social
scientists than others.
Firstly, Colombia provides an excellent national level case study to carry out both large-N
statistical studies as well as small-N comparative research on civil war and armed conflict due to the
fact that the country maintains a wealth of detailed records on these phenomena dating back decades.
The difficulty that emerges is that there is little consistency between many of the official and extra-
official sources of data in regards to the exact figures, a source of variation that is often caused by
disparities in data collection, classification, and built-in institutional biases. Furthermore, many of the
most comprehensive databases on all indicators of the armed conflict, social mobilization, and even
basic demographic information contain massive gaps for certain periods and places. Obviously, many
46
of these quantitative black holes owe much to the escalation and de-escalation of the armed conflict
in specific places at a particular moment in time, all of which made data collection impossible at that
juncture. These deficiencies do not disqualify the available quantitative data. Rather, they compel
researchers to compare and contrast different sources to examine the broader trends found between
them and to filter out those sources which deviate severely from the majority. The confirmation of
these trends is crucial to trace the conflict narrative in question. While this may have been unacceptable
for a more quantitative research project, my methodology is clearly qualitatively driven and as such
does not depend on descriptive statistics for anything beyond their ability to evaluate certain crucial
indicators during specific phases of the larger sequence.
Another difficulty related to data collection and content analysis was the paucity of established
scholarly research on my case studies. While greater research has been conducted on Montes de María
than Arauca, both remain less studied than other regions of Colombia such as Urabá, Magdalena
Medio, and La Macarena. As a result, I was forced to depend on unpublished historical accounts,
media sources, court documents, and more recent reports from official and unofficial sources to shape
my understanding of these places. Subsequently, I rely strongly on the field interviews I conducted -
particularly in the case of Arauca – for much of my empirical support. This is challenging insofar as
in many cases these particular testimonies have never been documented before and are therefore much
more difficult to verify except by triangulation with other local testimonies and other secondary
sources. Joe Yates (2004) asserts that “[…]very rarely are marginalized people provided with an
opportunity to tell their story and have their voices meaningfully heard in the research process” (1).
However, researchers conducting field work in high-risk environments also need to take into
consideration Lee Ann Fujii’s (2010) caveat that: “People forget some details and misremember others.
They rearrange chronologies, confuse sequences, and give greater weight to some moments over
others. In addition, institutions of all kinds, from prisons to schools, socialize people to construct the
47
past in certain ways” (232). In other words, social scientists should welcome every opportunity to talk
to those who have never been given a platform before, but by no means should they blindly accept
the veracity of field interviews without cross-verification or confirmation of crucial details. All of the
included information obtained from field interviews has been subjugated to a rigorous cross-
examination and only appears in the dissertation because it has been confirmed through triangulation
with other sources, established documentation, and/or the public record insofar as it exists in these
places.
C. Fieldwork
Conducting fieldwork as an outsider in a conflict or post-conflict setting provides a unique set
of challenges and risks (Fujii 2018; Kaplan 2017; Wood 2006). Furthermore, such high-risk field
research generally creates a plethora of ethical considerations for outside researchers, particularly in
instances where very clear power dynamics exist between the visiting scholar on one hand and local
habitants on the other (Lekha Sriram et al. 2009; Cronin-Furman and Lake 2018). As a North
American male, my presence and ability to conduct fieldwork in two rural war zones in Colombia
required an exhaustive ethics protocol and months of preparation in order to eventually visit the field
sites and to ensure my safety and that of my research participants. I owe a great debt to my institutional
affiliation with CINEP, one of Colombia’s oldest human rights and conflict research institutes based
in Bogotá, for sharing their networks with me and introducing me to individuals who conduct a variety
of functions on the ground in Montes de María and Arauca. Having an institutional sponsor that is
recognized and respected throughout Colombia provided me with a level of access to networks of
local researchers and community leaders that would have otherwise proven much more difficult to
establish. From these initial contacts with locals who were known on the ground, I was able to
successfully identify and locate a wide array of local participants who differed according to age, gender,
and vocation, but who shared a common experience of the armed conflict in these regions. Due to
48
these cross-sectional networks, I did not have to depend on a ‘snowball sampling’ method of
participant recruitment, which can be fraught with selection bias and other difficulties due to the very
real allegiances and prejudices that most people on the ground possess towards armed non-state
actors, political movements, and the national government. For example, community leaders with
greater sympathies towards the insurgent groups or the armed forces are more likely in turn to
associate with other people with similar sympathies and thus it is easy to get trapped in these invisible
cohorts without actually understanding or perceiving them.
Local contacts in rural settings such as Montes de María or Arauca are absolutely crucial for
outsiders. Whether the armed conflict is active or the post-conflict reconstruction is underway, people
who live in such spaces are highly attentive to and vigilant of outsiders given their extensive
experiences with competing armed actors, and as such will often report any irregular occurrences to
whatever group happens to maintain control of that community. In other words, there are eyes and
ears everywhere. Thus, when establishing contact with willing participants and arranging interviews,
one should always follow the parameters set by the interviewees due to their far superior
understanding of local customs, rules, and perceptions. Interestingly enough, most research
participants in the Araucan piedmont wanted to conduct their interviews in a public setting in front
of other people despite the fact that this sub-region continues to be highly controlled by the insurgent
groups. In the Araucan plains and Montes de María on the other hand, participants more often than
not sought to speak in private settings and for this reason I was frequently forced to travel significant
distances on barely passable rural roads to interview them in the comfort of their home villages.
Finally, the logic of positionality manifests itself across time and space dependent on the
subject and the context they are in. Carrying out fieldwork as an outsider in high-risk contexts such as
Arauca and Montes de María obviously came with its advantages and disadvantages. My position and
stature as an academic researcher at an elite North American university affiliated with a widely
49
respected Colombian human rights think-tank opened several doors that might have otherwise
remained shut. On the other hand, as a tall, light skinned North American male with red hair, I was
unable to either blend in or reduce my presence in an environment where such physical attributes
widely condemned me to being perceived as an American military advisor, a DEA agent, or a foreign
mercenary. Fortunately, I was able to offset this unwanted stigmatization by constantly being
accompanied by a Colombian research assistant. Over the course of numerous visits to Montes de
María and Arauca, I always hired a research assistant to help with crucial administrative duties related
to arranging, recording, and transcribing field interviews. Perhaps more importantly, these individuals
helped ‘soften up’ my personal image amongst locals in these zones while also offering a more efficient
means to socialize with local habitants in a manner typical to the Colombian countryside. Despite my
fluency in Spanish, in Colombia one has to socialize extensively in order to gain peoples’ trust before
requesting an interview with them, a skill that my different RAs all excelled in. In particular, the
presence of a Colombian research assistant not only made the research participants more relaxed, but
it also gave them enough confidence to converse about topics that otherwise would have been
extremely difficult for them to talk about with a foreign researcher.
D. Rationale for National and Sub-National Case Selection
Perhaps no other country provides more suitable case studies to explore this particular
research puzzle than Colombia. For over fifty years the Colombian civil war has produced any and
every negative externality associated with internal conflict as violence continues to be the medium
through which social, economic, and political disputes are resolved in this South American country.
Colombia also has a long tradition of armed insurrection, para-statal repression, and illicit criminal
networks all operating in the shadows of the legal bureaucratic authority (Palacios 2006; González
2014; Safford and Palacios 2002; Bejarano 2003). The extensive study of this topic has even led to its
own field of study in Colombia known as ‘violentology’ (violentología). The South American country
50
possesses the longest running leftist insurgency in the world today and as such various swaths of the
national territory have existed under alternative forms of governance for decades at a time,
administered by Marxist guerrillas and right-wing paramilitaries alike (González, Bolívar, and Vásquez
2003; Arjona 2016b; Gutiérrez 2019). Finally, the recent attempts by the Colombian state and its allies
to displace and expel insurgent groups from their regional strongholds throughout the country is only
the latest of successive efforts by the country’s political elites and military to defeat armed challenges
to its rule (Porch and Delgado 2010; Mills et al. 2016; Kline 2009; 2015).
Civil wars, historically and presently, have rarely been fought between the state on one side
and a unified insurgent front on the other. Most internal conflicts have been fought between a variety
of formal and informal forces, on behalf of both the state and the insurgents, all interacting and
colliding in a variety of forms depending on the local conditions (Kalyvas 2006; Staniland 2012;
Gutiérrez-Sanín 2019). During each successive armed conflict over the course of the 20 th century in
Colombia, virtually every attempt at counterinsurgency has included such public and private armed
groups. Philip Mauceri (2004) captures this symbiosis succinctly: “The state response to insurgent
groups in Colombia can best be characterized as ‘abdication and privatization,’ a process in which
state actors provide the legal framework, legitimacy, logistical support, and on occasion armaments to
private societal actors in order to combat insurgents” (154-155). Throughout the entirety of the
current armed conflict, various administrations have legalized the creation of paramilitary groups
depending on the immediate pressures posed by insurgent groups, whereas others oversaw robust
military efforts to stamp out armed non-state groups where they could. Many of the pre-eminent cases
found in the COIN literature focus on expeditionary counterinsurgency campaigns – military efforts
directed from outside powers and manned with foreign troops - related to decolonization and the
vagaries of the Cold War. Colombia on the other hand represents a rich opportunity to examine
“[…]what happens when a government engages in extremely long-term operations, with only limited
51
external support, against insurgents operating within its own territory” (Mills et al. 2016, 13). Colombia
thus represents a ‘home-turf counterinsurgency’ to specialists on the topic, even if it must be noted
that the United States’ military has played an integral if not indirect role in shaping counterinsurgency
doctrine in Colombia. With the implementation of Plan Colombia in 2000 and the initiation of
Democratic Security Policy in 2002, “[…]the Colombian military became larger, better equipped, and
more adequately trained”, all of which “[…]made it possible to attack FARC troop concentrations
and prevent the continuation of the insurgent groups’ transition to a war of movements” (Kline 2009,
45). Jorge Delgado (2015) emphasizes the crucial role of the U.S. government and military in the
formation and implementation of Democratic Security Policy:
Washington was eager to assist the {Government of Colombia} in the refinement of the main operational concepts of the {Democratic Security Policy}. The US Embassy in Bogotá insisted that it was necessary to ensure that ‘the Colombian strategy dovetails with the {United States government} effort’…The United States would assist through the framework of Plan Colombia to help the country to incorporate inter-agency coordination techniques efficiently to intervene in the areas with insurgent presence and promote economic development. (414)
For all of the reasons outlined above, Colombia serves as an excellent case study to explore
my principal research questions related to counterinsurgent efficiency and insurgent resilience. Within
Colombia the justification for my case selection and periodization is simple. Upon taking office in
2002, President Álvaro Uribe took full advantage of the military aid and assistance provided by the
United States under Plan Colombia and unleashed an unprecedented military offensive against armed
non-state actors (primarily leftist guerrillas) operating beyond the purview of the central government
in Bogotá (Isacson and Poe 2009). Serving two-terms consecutively until 2010, Uribe oversaw a
massive shift in the balance of power between the Colombian state and said actors while also
embarking upon a belated form of state-building in these spaces. The initial successes of Uribe’s first
term were obscured in his second term by scandals related to the means with which the president was
attempting to achieve his end goals. Despite continuing to apply pressure against the embattled
insurgent groups and overseeing several high profile targeted assassinations of insurgent commanders
52
during the latter half of his tenure in power, Uribe left office midway through 2010 leaving various
regions of the country which remained under insurgent control (Delgado 2015).
An excellent baseline for comparing the overall efficiency of Colombian state expansion
efforts during this period – and more importantly to explain the variation in outcomes found
throughout Colombia - can be found in two locations which served as laboratories for Uribe’s eventual
nationwide military and state expansion project: Montes de María and Arauca. At the onset of his first
term, Uribe launched numerous military initiatives by Decree 1837 to reclaim the most violent areas
of the country which were contested by guerrillas and paramilitaries alike. Two “zones of rehabilitation
and consolidation” (zonas de rehabilitación y consolidación) were established, the first constituting three
municipalities in the department of Arauca, and the other encompassing twenty-three municipalities
in the departments of Bolívar and Sucre centered on the region of Montes de María. In these areas,
the military operated with emergency powers ostensibly with the aim of re-establishing and
maintaining public order (Leal Buitrago 2003; Defensoría del Pueblo 2003). It warrants mention that
in April 2003, some eight months after the establishment of these two special zones in Montes de
María and northern Arauca, Colombia’s Constitutional Court declared this controversial strategy to
be illegal and ordered Uribe to scale-back the more coercive tactics employed by the Colombian
military in these regions (i.e. mass arrests, arbitrary detentions, suspension of habeus corpus, etc.). Even
though this initiative was short-lived, these two regions remained heavily militarized and were subject
to the various incarnations of the Colombian government’s ever-evolving state expansion campaign
found in other conflict zones throughout the country (Amnistía Internacional 2004).
Paul Staniland (2014a) argues that ‘[…]a strong explanation of a small number of cases rather
than a vague, lowest-common denominator explanation of a large number of cases’ provides greater
value than more expansive yet less detailed studies of violence in similar contexts” (11). Similarly, my
research prioritizes a strong explanation of my cases, following in the stead of recent works dedicated
53
to exploring sub-national variation in socio-political phenomena in order to examine “[…]important
outcomes that are difficult to detect with a national-level lens” (Giraudy, Moncada, and Snyder 2019,
5). These case studies also afford the benefits of both the most similar and most different methods of
case selection. The most similar method can be found in the contrast provided at the sub-departmental
and even sub-municipal level in the case of Arauca, as despite the shared contiguous territory, political
administration, and distribution of armed actors, the counterinsurgency outcomes varied between the
Araucan plains and the piedmont. The most different method can be seen in a comparison between
Montes de María and Arauca (Seawright and Gerring 2008). The case selection of these two regions
are justified by the following criteria: (a) both initially played host to strong contingents of leftist
insurgent groups, while also offering numerous illicit rent extraction opportunities such as coca
production/trafficking, kidnapping, extortion, and contraband; (b) both were subject to an AUC-
sponsored paramilitary incursion that was subsequently augmented by a military intervention
beginning with Decree 1837; and yet (c) the two demonstrate a drastic variation in conflict outcomes.
The focus of my analysis compares these regions in broad terms. However, in my theoretical
chapters I briefly examine this variation in outcomes using a mixed-methods approach which
incorporates statistical data and primary and secondary qualitative sources to assess the effect of state
expansion on three neighboring municipalities in both Arauca (Arauca municipality, Arauquita, Tame)
and Montes de María (San Jacinto, El Carmen de Bolívar, Ovejas). In these sub-sections, I specifically
focus on three municipalities in each region for purposes of parsimony and representation. Whereas
I conducted extensive fieldwork across five of the seven municipalities found in Arauca (Arauca
municipality, Arauquita, Saravena, Fortul, Tame) and five in Montes de María (San Jacinto, El Carmen
de Bolívar, Ovejas, San Onofre, Sincelejo), I reduce the focus to three municipalities in each region to
provide a better and clearer representation of the diversity of outcomes found at the sub-national
level. I selected Arauca municipality, Arauquita, and Tame because this provides an even balance
54
between the piedmont and the plains; the first is clearly located in the plains, the second in the
piedmont, and the third is divided between the two sub-regions. Whereas Tame was not technically
included in the zone of rehabilitation and consolidation, it was nevertheless heavily militarized at this
juncture and experienced the highest rates of violence in the region. Furthermore, Tame does not
share a border with Venezuela, a geographical consideration which provides an important analytical
contrast when studying the effects of insurgent resilience in such contexts. The three municipalities
selected in Montes de María were at the geographical center of the region and also happened to be
three of the most heavily impacted by the regional armed conflict.
E. A Brief History of Colombia
Following Colombia’s independence from Spain in 1810, the fledgling republic got off to
rocky start. Political elites from both the Liberal and Conservative parties struggled to compete
peacefully, while geographical obstacles and opportunistic regional elites made centralization difficult,
as “[…]some historians assert that Colombia was a country in permanent war during the 19th century”
(Comisión Valenciana de Verificación de Derechos Humanos 2005, 27). All violent conflicts during
this period “[…]occurred along the lines of the ‘hereditary [partisan] hatreds’ between Liberals and
Conservatives” (Daly 2014, 345), as from independence well into the mid-20th century “[…]party
identification was inculcated in offspring in a fashion resembling religious belief” (Oquist 1980, 78).
Colombia remained free of civil war for the first four decades of the 20th century, in no small
part due to founding of the National Military Academy (1907) and the Superior School of War (1909),
both of which marked the beginnings of professional armed forces in Colombia (Rodríguez 1993).
The political ideology and partisan alignment of the newly founded professional military conformed
strongly to the Conservative party, as many officers were recruited from the provincial middle class,
an organic receptacle of support for this political party (Dufort 2017). In 1930, the Liberals took
control of the government after an extended hiatus, introducing several important progressive reforms
55
which generated fervent Conservative opposition, and by extension political instability, factors which
brought boiling political and social tensions to the fore in the mid-to-late 1940s (Fluharty 1966; Tirado
Mejía 1981; Correa Peraza 2009).
After the Liberal loss in the 1946 presidential elections, both political parties faced growing
internal schisms between moderate and radical factions, exacerbating regional cleavages between
Liberals and Conservatives throughout the Colombian countryside during the first two years of
Mariano Ospina Pérez’s (1946-1950) term. On April 9th, 1948, the Liberal leader Jorge Eliécer Gaitan
was assassinated in downtown Bogotá, triggering el Bogotazo, a spontaneous series of riots which
engulfed the Colombian capital and left much of the city destroyed and hundreds dead in a matter of
hours (Bushnell 1993). In the aftermath of Gaitan’s assassination, the partisan-driven bloodshed
spread throughout the country, initiating a ten-year civil war known as la Violencia (Ortiz Sarmiento
1985; Oquist 1980; Henderson 1985).
Regional outposts which had historically been administered as personal fiefdoms by political
bosses (gamonales), quickly degenerated into armed conflict between hastily organized groups of armed
men purportedly representing one side or the other of the partisan impasse, yet many mobilized merely
to resolve longstanding local conflicts in their communities (Roldán 2002). The violence was chaotic
and unorganized for the most part, characterized by extreme sadism and the partisan cleansing of
entire sub-regions through massive forced displacement (Pécaut 1987). The partisan conflict
challenged peripheral power dynamics and agrarian social structures, as “[…]rural terror rearranged
social classes in the countryside and relations of leadership and power in the different regions”
(Sánchez and Meertens 2001, 17).
At the end of the carnage, la Violencia had “[…]wreaked havoc on 64 percent of the country’s
territory and left 250,000 dead” (Daly 2014, 333). In several regions, groups of Liberal guerrillas
formed to protect themselves and their communities from their Conservative rivals, in some cases
56
amounting to irregular regional armies which were mobilized well beyond the end of this conflict
(Villaneuva Martínez 2012). President Gustavo Rojas Pinilla, who had himself come to power in a
1953 military coup at the height of the conflict, calmed most if not all regions of the national territory
with an offer of amnesty and concessions. However these efforts were not enough to prevent political
elites from both the Liberal and Conservative parties from joining together and making common cause
to force his ouster in 1958 (Ayala Diago 1996; Saénz Rovner 2002; Karl 2017).
The following two decades saw the pacted return to democratic rule by both the Liberal and
Conservative parties in an agreement known as the National Front, in which they alternated four
consecutive presidents between 1958 to 1974 in an attempt to prevent another episode such as la
Violencia (Bejarano 2011). The first decade of the National Front governments saw some notable
institutional and policy reforms. In 1958, the national government approved Law 19, a legislative act
which created community action boards (juntas de acción comunal - JAC), democratic, communal
organizations established at the neighborhood and village level throughout Colombia in order to create
a channel of communication between these citizens and their locally elected officials. Oliver Kaplan
(2017) clarifies the historical formation and original purpose of the JACs: “The juntas as an
organizational form were not imposed by the state. Rather, the state created and encouraged a legally
recognized vehicle that communities could freely adopt – it attempted to institutionalize local councils.
Virtually any community could be organized into a junta and recognized by the government” (80).
Following this, the basis for the National Front’s attempt at land reform, Law 135, was passed in 1961.
Apart from land redistribution, the official agrarian policy sought to ease rural tensions by extending
support to the decimated peasantry in minifundia areas and colonization zones alike, providing greater
technical assistance to improve productivity, increasing incomes by establishing and encouraging
peasant cooperatives, and making better services available to the Colombian peasantry (Zamosc 1986).
The new law also witnessed the creation of the Colombian Institute for Agrarian Reform (Instituto
57
Colombiano de la Reforma Agraria - INCORA), the official state bureaucracy tasked with carrying out this
ambitious land reform campaign by distributing titles to landless peasants throughout the Colombian
countryside. However, from the beginning there was predictable opposition to the redistributive
project from radical Conservative politicians and rural landed elites. Marco Palacios (2006) describes
this: “From the outset, the government agency created to administer the reform (INCORA) was
submerged in legalistic quicksand that made expropriations almost impossible…By 1971 only around
1 percent of the land originally targeted in the reform had actually been expropriated” (182-3).
This controversial policy was complemented by the formation of the National Association of
Peasant Users (Asociación Nacional de Usuarios Campesinos - ANUC) in 1967, a national level organization
entrusted with matters relating to the adjudication and ownership of land (Fals Borda 1976). The
Liberal administration of Carlos Lleras Restrepo (1966-1970) wanted to provide an organizational
mechanism with which the Colombian peasantry could overcome any obstacles regional landholding
elites placed in their way: “On the one hand, the peasant economy would be reinforced in the regions
of minifundia and colonization, which would be helped by the direct intervention of ANUC in the
provision of state services. On the other hand, the landless peasants were offered help in areas of
latifundia, where ANUC’s pressure upon the landowners was bound to be crucial” (Zamosc 1986, 53-
4). Despite the terms of the National Front pact, popular pressure from regional landed elites against
the proposed reform generated a cleavage within ANUC between the Armenia line and the Sincelejo
line, with the former seeking to limit the extent of usable land parceled off to poorer peasants, while
the latter sought to deepen and widen these redistributive efforts. This impasse resulted in the Pact of
Chicoral, an agreement signed in 1973 between political and regional elites from both parties, large
landholders, and cattle ranchers, which withdrew support for the land claims of poor peasants while
attempting to redirect them towards the colonization of the most peripheral, unsettled regions of the
country in the Eastern Plains, the Amazon, and the Pacific Coast (Reyes Posada 1978; LeGrand 1989).
58
Safford and Palacios (2002) describe this dramatic shift in national agrarian policy: “The second half
of the twentieth century has witnessed dynamic colonization in eight zones, covering some 300,000
square kilometres of land – almost a quarter of the nation’s territory. It has been estimated that some
375,000 were engaged in colonization in 1964, 1.3 million in 1990. Over the past fifty years they have
cleared and settled some 3.5 million hectares” (311).
The exclusionary tendencies of the bipartisan agreement shut out many social and political
actors who might have otherwise sought peaceful means of contestation, while the Colombian military
increasingly sought to repress any extra-institutional forms of dissent. It was in this exact environment,
that the country’s first leftist insurgent groups emerged in the mid-to-late 1960s. The Revolutionary
Armed Forces of Colombia (Fuerzas Armadas Revolucionarias de Colombia - FARC) was established as
the armed wing of the Colombian Communist Party (Partido Comunista Colombiano – PCC) in 1964 by
a handful of peasant leaders who had survived a military campaign to expel them from their
“independent republic” in Tolima department (Rivera Cusicanqui 1987). The National Liberation
Army (Ejército de Liberación Nacional – ELN) was originally founded in the same year in San Vicente de
Chucurí in Santander department by a handful of idealistic young guerrillas who had been trained in
Cuba, radical university students, and a collection of Catholic priests inspired by the teachings of
Liberation Theology. Initially, the ELN embraced the foco theory of guerrilla warfare conceived by Che
Guevara in the late 1950s and early 1960s (Medina Gallego 1996).2
These insurgent groups were quite limited in the scope and scale of their operations until the
late 1970s and early 1980s. Many suffered near collapse when confronted with the overpowering force
of the Colombian military’s counterinsurgency operations during this period of time (González,
Bolívar, and Vázquez 2003). American military analyst Thomas Marks (2002) observes that this initial
2 While over the course of the 1960s, 1970s, and 1980s other smaller guerrilla groups and dissident factions would emerge, the FARC and the ELN were the most notable in terms of size and the capacity for action.
59
period of insurgent incubation did not concern the country’s political elites, only the armed forces:
“[…]following La Violencia, the insurgents remained largely out there, out of sight, out of mind,
patiently building an alternative society. No one much cared…As long as the guerrillas were
revolutionary homesteaders in areas no one else wanted, the government bothered with them only
when their actions forced a response. It was the job of the police and the military, went the logic, to
keep an eye on them” (4).
After a series of crucial tactical and strategic reforms, the FARC and the ELN saw their
fortunes improve. The FARC held its Sixth Conference in 1978, at which the insurgent organization
concluded that a nationwide expansion was necessary for its very survival as a viable revolutionary
project. Following this early period of growth, the guerrilla group convened at the Seventh Conference
in 1982, where it decided to double the number of active fronts, establish a presence closer to and
within medium sized urban centers, and to involve itself in the protection and taxation of the
burgeoning illegal drug industry (CNMH 2014a).3 The confluence of illegal macro-economic trends
with poorly planned resettlement policies tabled by the National Front governments created the
conditions necessary for the FARC to enter the cocaine industry in the late 1970s and 1980s (González
2014). The ELN on the other hand initially eschewed any form of participation in this illicit trade.
After convening a national level congress in 1983, the group’s leadership sought to imitate the success
of the Domingo Laín front in Arauca and expand into regions rich in hydrocarbons. By 1986, the
insurgent group “[…]had grown by over 500% and had become a national organization” (Gutiérrez
and González 2008, 31; Peñate 1998). Additionally, both insurgent groups continued and even
increased their use of kidnapping and extortion as a means of augmenting their revolutionary budgets.
3 It was also at this conference that the FARC adopted the addendum ‘peoples army’ (Ejército del Pueblo) to its name, thus becoming FARC-EP.
60
By the early 1980s a perfect storm was forming in Colombia. First and second-generation
guerrilla groups were expanding throughout the country and increasingly on the march in the
countryside, while the rapid growth of the international cocaine trade saw the South American nation
become the epicenter of this burgeoning illegal enterprise (Thoumi 1995). The main consortia fuelling
this industry used their newfound largesse to penetrate and corrupt domestic law enforcement, the
armed forces, and even local and national level politics. Simultaneously, the billions of illicitly gained
dollars returning to Colombia were laundered in a variety of legal businesses, and used to purchase
substantial tracts of land in rural communities (Bagley 1990; Molano 2004; Duncan 2014). The
territorial expansion of both the leftist insurgencies and nouveau riche drug traffickers created fresh
tensions in the countryside, as the former sought to extort the latter, along with other traditional
landed elites, often by kidnapping them and their relatives for ransom. Predictably, this practice
generated a violent backlash from the large landholders, new and old, who joined forces with drug
traffickers to sponsor nascent ‘self-defense’ forces to protect themselves and their properties from
further insurgent incursions (Ronderos 2014).
The Conservative government of Belisario Betancur (1982-1986) initiated peace negotiations
with the FARC in 1984, establishing a ceasefire between the guerrillas and the state. Simultaneously,
the FARC formed its political party, the Patriotic Union (Unión Patriótica - UP), to represent the
insurgent group in local and national level elections following the political opening of the 1980s. This
reform permitted the direct election of mayors and governors and paved the way for the eventual 1991
Constitution and the decentralization of the national political system as a whole (Archer and Chernick
1989; Eaton 2005). Unsurprisingly, the FARC peace negotiations eventually fell apart due to the
systematic violations of the ceasefire by the Colombian military, who in an alliance with nascent
paramilitary groups, unleashed a massive wave of violence against members and sympathizers of the
UP, a political cleanising which claimed thousands of lives in the mid-to-late 1980s (El Tiempo 2013).
61
Table 1. Insurgent Annual Finances in 1991 (USD)
Source of Revenue FARC ELN
Kidnapping $21.9m $21.3m
Mining $8.9m $28.1m
Extortion $8.9m $12.7m
Public Embezzlement N/A $9.6m
Narcotics $99m N/A
Other $7.4m $5.9
Total $146.1m $77.6m
Source: Author’s Calculations (Based on Semana 1992)4
Despite the worsening violence, the FARC and the ELN aggressively expanded over the
course of the 1990s and consolidated a strong presence throughout the length and width of the
national territory. Kidnappings increased dramatically, particularly with the introduction of “miracle
fishing” (pescas milagrosas), in which the insurgents would set up roadblocks on intercity highways and
forcibly abduct more affluent travellers for ransom. Apart from vastly improved armaments for the
rank and file troops, the guerrillas also began to deploy the use of improvised bombs made from gas
cylinders (cilindros) which would be launched at military and police installations throughout rural
Colombia, often causing enormous collateral damage, both physical and human, due to the notorious
inaccuracy of these homemade projectiles (CNMH 2014a). The FARC in particular began to mount
military offensives against the Colombian armed forces, handing serious defeats to the Colombian
military in highly coordinated operations in rural theatres such as Miraflores, El Billar, and Las Delicias
(Mills et al. 2016; Barrera 2017).
Faced with the growing influence of the FARC and ELN, two of the largest and strongest
paramilitary organizations, the Peasant Self-Defense Forces of Cordoba and Urabá (Autodefensas
Campesinas de Córdoba y Urabá - ACCU) and the the Peasant Self-Defense Forces of Magdalena Medio
4 This article was based on specific documents seized and retrieved from members of both the FARC and the ELN by the Colombian authorities over the course of 1991-1992 period. “In such a way it is not an exaggeration to say that in the past 20 months, the authorities have learned more about the FARC and the ELN’s finances than in the remainder of 30 years of the antisubversive struggle.” (Semana 1992)
62
(Autodefensas Campesinas de Magdalena Medio – ACMM) sought to take advantage of Law 356 passed in
1994 which legalized “vigilance cooperatives” known by their acronym, CONVIVIR, which were in
reality self-defense groups or paramilitaries.5 Despite the official repeal of this controversial decree in
1997, the Castaño brothers, who were the founders and leaders of the ACCU, had already established
a national-level umbrella paramilitary group, the United Self-Defense Forces of Colombia (Autodefensas
Unidas de Colombia - AUC) in 1996, and the nationwide expansion of the paramilitary project spread
like wildfire throughout the entire reach of the country (Romero 2003; Duncan 2006; Gutiérrez 2008,
2019).
Operating with the complicity of the Colombian armed forces, uninhibited by conventional
laws of war or modern conceptions of human rights, the paramilitary expansion across Colombia was
accompanied by a level of brutality and sadism which had not been seen since the darkest days of la
Violencia. The AUC was highly decentralized and wrought with internal rivalries and conflicts, yet all
paramilitary groups shared a similar repertoire of violence against their enemies. The paramilitaries
specialized in indiscriminate massacres, selective assassinations and disappearances, and the massive
forced displacement of civilians living in rural conflict zones, employing chainsaws, sledgehammers,
and even alligators (caimanes) to terrorize, murder, and dispose of their victims bodies in the most
publicly horrific way possible. The objective of this extreme violence ostensibly was
counterinsurgency, but in many regional and sub-regional cases the paramilitary project was actually
driven by landed elites seeking to preserve or re-obtain regional power, or by the paramilitary
commanders’ desire to control illegal economic opportunities related to extortion, drug trafficking,
and public embezzlement (González, Bolívar, and Vásquez 2003; Gutiérrez-Sanín and Barón 2005;
Duncan 2006). The objective of the paramilitaries was not simply the physical and symbolic
5 Vigilante groups or paramilitaries have a long and violent history in Colombia. Ironically enough, Law 48 was passed in 1968 during the administration of the progressive Carlos Lleras Restrepo, a decree which permitted the formation of these groups in rural areas in which guerrillas were known to operate in the general absence of the state (Richani 2002).
63
elimination of the perceived enemies of the state-paramilitary project. Rather, they sought the
complete eradication of any insurgent order that had existed prior. Effectively, paramilitary groups
were engaging in a socio-political cleansing of various rural theatres of war, sometimes so the state
could govern these territories, in other instances so they could govern themselves (Gutiérrez-Sanín
2019).
Levels of violence reached an all-time high at the turn of the century amidst the paramilitary
boom and increased guerrilla consolidation in their respective zones of influence. In an attempt to
combat the international perception that Colombia was on the verge of being a ‘failed state’,
Conservative President Andrés Pastrana Arango (1998-2002) oversaw the negotiation and passage of
a multi-billion dollar military aid package, Plan Colombia, with the United States in 1999. The robust
joint-security agreement, initially conceived and outlined as an antidrug strategy, suddenly evolved
“[…]into a counterterror and counterinsurgency strategy” after the terror attacks of September 11th,
2001 (Sweig 2002, 124). Between 2000 and 2007, Colombia received some $5.4 billion USD in
assistance, of which slightly more than 80% was earmarked for the security forces (Isacson and Poe
2009, 4). Simultaneously, Pastrana opened talks with the FARC’s leadership, initiating the El Caguán
peace negotiations and granting the leftist insurgents a demilitarized zone consisting of five
municipalities in southern Colombia, covering an area roughly the size of Switzerland (Bejarano and
Pizarro 2004). It quickly became clear that the FARC was not negotiating in good faith, and was
instead using the talks as a means to strengthen and consolidate its presence throughout the country
(Mills et al. 2016). The massive surge in paramilitary violence at this juncture in time was equally
problematic, as the Colombian military were either turning a blind eye to their excesses, or directly
collaborating with them in carrying out their worst atrocities (González, Bolívar, and Vázquez 2003).
Unsurprisingly, the peace talks fell apart in early 2002, as both Pastrana’s credibility and the potential
for a negotiated settlement to the armed conflict were left in tatters.
64
In the wake of the collapse of the Pastrana peace talks and amidst unprecedented levels of
violence, the Colombian electorate went to the polls in 2002 and voted in Álvaro Uribe Vélez, an
outsider candidate with a zero-tolerance policy towards the FARC. Upon taking office, President
Uribe harnessed the massive military aid provided by Plan Colombia and unleashed a hitherto
unprecedented counterinsurgency offensive christened ‘Democratic Security Policy’ (la Política de
Seguridad Democrática – DSP), ostensibly against all armed non-state actors operating beyond the
purview of the law. In reality, this campaign was primarily directed against the leftist insurgent groups
the FARC and the ELN. During this time “[…]the patterns of the past began to be broken and, finally,
the national government began to have effective control of the entire nation” (Kline 2009, 5).
Figure 4. Growth in Colombian Armed Forces - Number of Troops (1989-2015)
Year
Source: Barrera 2017 {Based on World Development Indicators – World Bank}
Under the auspices of DSP, Uribe sought to expand, overhaul, and modernize the Colombian
armed forces and police with the assistance of U.S. military advisors and their state of the art military
technology (Porch and Delgado 2010). Based on the oft-celebrated British counterinsurgency
playbook in colonial Malaya and Kenya, this policy sought the implementation of a three-step plan to
65
control, stabilize, and consolidate territory (clear, hold, build) which was ruled or administered by
armed non-state actors, by expanding the Colombian state’s military and bureaucratic presence into
these spaces while simultaneously combating and expelling the insurgents (Isacson and Poe 2009).
While the Colombian armed forces underwent a massive expansion, tripling in size between 1994 and
2014, it also began to depend less on conscription and more on the recruitment and training of a
professional class of soldier. The Colombian military underwent a much-needed restructuring during
the Pastrana administration, particularly in the area of military intelligence, which proved to be of
enormous importance once Uribe actively deployed his soldiers to every conflict zone throughout the
country. It warrants mention that this latter component, along with the crucial military aid and
hardware provided by the United States under the terms of Plan Colombia, was perhaps the most
valuable asset the North American power bestowed on the Colombian state (Porch and Delgado
2010).
Over the course of Uribe’s two terms in office (2002-2010), DSP underwent a progressive
evolution of different phases, which reflected the government’s attempt to refine and improve its state
expansion efforts into the various conflict zones found throughout the national territory.6 Upon taking
power in August of 2002, Uribe tested out his incipient counterinsurgency strategy in two specifically
appointed “zones of rehabilitation and consolidation” in the volatile conflict zones of Arauca and
Montes de María. The following year, this policy was expanded to include Cundinamarca, the Andean
department surrounding Bogotá, with the Colombian armed forces successfully expelling the FARC
from the area around the capital during Operation Liberty I (Operación Libertad I). In 2004, this
operation was emulated in the FARC’s historic stronghold in the nearby southern departments of
Meta and Caquetá in what was christened Plan Patriot (Plan Patriota) (Leal Buitrago 2006). The
6 Uribe was the first Colombian president to successfully serve two terms consecutively, having amended the constitution in order to do so. However in his attempt to amend it again to serve a third term, the Constitutional Court overruled the proposal, thus restricting his time in power.
66
contrasting success of Plan Patriot found between Cundinamarca, Meta, and Caquetá forced top
military officials to revise their tactics, and over the course of the following three year period (2005-
2007), the Colombian state’s counterinsurgency policy was reformulated according to the “Integrated
Action” doctrine, which sought to include both social and political components in order to better
complement the hitherto military focus of DSP. During Uribe’s second term, the president oversaw
the gradual implementation of the new counterinsurgency doctrine designed to strengthen and fortify
the state’s institutional presence in these confict zones, first with the creation of the Center for
Coordination of Integrated Action (CCAI) initiative in affected areas, followed by the unveiling of
Plan Consolidation (Plan Consolidación) in 2007. In response, the FARC adopted and implemented a
new military doctrine in 2008, Plan Rebirth (Plan Renacer), after years of military setback at the hands
of the Colombian armed forces (Mills et al. 2016). By 2009, PSD was again being re-branded with
‘Strategic Leap’ (Salto Estrategico), the final phase of Uribe’s counterinsurgency strategy before he
transferred power to his handpicked successor and former defense minister, Juan Manuel Santos, in
2010 (Porch and Delgado 2010; Isacson and Poe 2009).
Figure 5. Colombian Homicide Rates per 100 000 Persons (1990-2015)
Source: FIP 2016 (Based on Data from the National Police and DANE) * Period highlighted in pink represents Democratic Security Policy
67
The main successes of Uribe’s security policies, which both his most fervent supporters and
detractors alike will agree, were the dramatic reduction in levels of violence at the national level, the
securitization of most urban centers, and, perhaps most importantly, reclaiming control of the national
highway system. Prior to Uribe, inter-city transit had been extremely difficult due to guerrilla and
paramilitary activity on the country’s major highways (Delgado 2015). Between 2003 and 2006, the
Uribe administration negotiated the Agreement of Santa Fe de Ralito with the AUC, culminating in
the passage of Law 975, or the Justice and Peace Law (La Ley de Justicia y Paz) in 2005. This agreement
led to the demobilization of various paramilitary blocks throughout Colombia, an event which
highlighted the close historical ties, personal and professional, the president maintained with these
terms of Law 975: “The paramilitaries were granted additional benefits, such as a reduction in prison
terms for those convicted of crimes, allowing them to serve five to eight years in jail minus credits for
good behaviour and time spent in negotiations” (236). While the general consensus recognizes the
above mentioned achievements of Democratic Security Policy, the broader picture is not nearly as
clear. Isacson and Poe (2009) evaluate the efficacy of Uribe’s counterinsurgency campaign as neither
success nor failure, claiming instead that “[…]the results have been mixed” (4). Other analysts such
as US military instructor Douglas Porch (2012) assert that Democratic Security Policy “[…]is generally
lauded as evidence of the effectiveness of both counterinsurgency (COIN) doctrine and security
assistance” (243).
Figure 6. Armed Actions by Armed Actors in Colombia (1990-2011)
7 Hailing from a prominent cattle ranching family in Antioquia, Álvaro Uribe entered local politics after his father was murdered in the early 1980s, allegedly by the FARC. By the time he had risen to become governor of his home department, he was one of the key proponents of the CONVIVIR initiative and has consistently faced strong accusations of links with paramilitary groups throughout his political career. At the present, many of his former cabinet members and close family members are either imprisoned or awaiting trial for their connections to crimes committed by such “self-defense” groups.
While certain statistical indicators of conflict and public order do indicate that DSP succeeded
to a certain degree, Uribe’s second term in office experienced several scandals which brought into
question whether the ends did in fact justify the means. These scandals exposed the deep infiltration
of paramilitary groups in regional and national level politics (parapolítica), the systematic extrajudicial
killings of thousands of civilians by the armed forces in order to pass them off as guerrilla casualties
(false positives), as well as the illegal use of domestic security agencies to monitor and target members
of the opposition and civilians alike (Acemoglu, Robinson, and Santos 2013; Acemoglu et al. 2016).
However, during this same period a new strategic initiative to locate and assassinate insurgent
commanders was quite successful in taking down various High-Value Targets (HVTs) from the FARC
and the ELN, battlefield successes that sustained popular support for Uribe and his counterinsurgent
policy. 8 To date, Álvaro Uribe remains a powerful, if not polarizing figure in the Colombian body
politic, for the simple fact that he took the offensive to the guerrillas with unprecedented success.
While never able to defeat either the FARC or the ELN decisively on the battlefield, the Uribe
administration reduced their ranks and applied sufficient pressure to force these groups’ decimated
leadership to rethink their appetite for continuing the armed struggle. Whereas in 2001 the FARC and
8 In particular, during the second term of Uribe’s government, numerous high-ranking FARC commanders were killed in targeted bombings (Mono Jojoy, El Negro Acacio, Raul Reyes, Martín Caballero, etc.), a phenomenon which fuelled the public’s perception that the FARC was on its last legs.
69
ELN possessed more active members than at any other time in their respective histories (FARC – 20
000 guerrillas; ELN – 4500 guerrillas), by 2012 the FARC had been reduced to between 8000-9000
insurgents in arms, while the ELN counted some 1500 (CNMH 2014a).9 Perhaps most importantly,
Uribe’s counterinsurgency strategy achieved enough on the battlefield to make his successor Juan
Manuel Santos’s peace process a possibility instead of a remote aspiration.
Since 2010, the Colombian armed conflict has entered a phase which can be aptly described
as the “end game” of the protracted civil war, with a more restrained counterinsurgency focus, coupled
with the initiation of the Havana Peace Process in 2012 between the Santos administration and the
FARC. The ongoing implementation of the accord is proving to be exceptionally difficult, particularly
in light of the 2018 electoral victory of Uribe’s candidate, Iván Duque Márquez, who was opposed to
the Havana peace process, and has thus done everything within his power to ensure that the actual
implementation of the peace accord does not occur. In recent years, several worrisome trends have
emerged. Since 2016 there has been a massive rise in the assassinations of community leaders and
FARC ex-combatants throughout the country, a phenomenon which can be attributed to the
fluctuations in local power dynamics in the aftermath of the FARC’s demobilization. Second, more
FARC dissident factions (disidencias) have emerged recently and are returning, or continuing to operate,
in their former zones of influence. On August 29th, 2019, several top ranking FARC commanders
publicly repudiated the 2016 treaty and made a renewed call for arms purportedly from their refuge in
neighbouring Venezuela, thus throwing the future of the Havana Peace Accords into jeopardy. Of
equal concern, coca cultivation has exploded over the past couple of years and has now returned to
previous record levels, if not greater. Finally, the ongoing political and economic crises in Venezuela
9 Calculating the exact strength of clandestine insurgent groups such as the FARC and ELN is extraordinarily difficult, if not outright impossible. However, these figures are fairly consistent between diverging official and unofficial sources. These do not include militia members, who are more difficult to tally than formal guerrilla combatants.
70
has directly affected Colombia more than any other country in the region given the close proximity of
the two and the extremely poor bilateral relations between the Maduro and Duque governments.
71
Chapter Three. Pre-Existing Configuration
“The past is not simply “received” by the present. The present is “haunted” by the past and the past is modeled, invented, reinvented, and reconstructed by the present.”
- Jan Assmann, Moses the Egyptian (1997, 9)
The pre-existing configuration of Montes de María and Arauca played a deterministic role in
shaping the manner in which insurgent groups were able to embed themselves in these regions. Ana
Arjona (2016a) posits that “[…]social orders are not created in a vacuum” (63), an observation echoed
by Oliver Kaplan (2017), who notes: “Civil wars are not fought in social vacuums. They are fought in
social landscapes. These landscapes are often variable, with notable social differences from one town
or village to the next” (34). Historically, state presence has been exceptionally weak throughout the
Colombian periphery, and in many cases only select elite constituencies had access to the state and
were often tasked in its absence with administering local territory and people (LeGrand 1986;
González, Bolívar, and Vásquez 2003; González 2014). The lack of state presence in rural communities
outside of the Andean highlands meant that landed elites were responsible for establishing and
administering their own agrarian social orders, or in the absence of such a class, this responsibility fell
upon local peasants who often looked to communal organizations and civic leaders for leadership
In contrast to Jeffery M. Paige’s (1978) theory on peasant mobilization that suggests that
smallholding peasants are quiescent and thus unlikely to revolt, my research follows in the tradition
of Eric Wolf (1969) and James C. Scott (1976) by staking the opposing position. My theoretical model
holds that horizontal agrarian social structures are more conducive to insurgent embeddedness than
vertical structures due to the stronger level of communal cohesion between peasants. In the absence
of landed elites, horizontal rural hierarchies are more egalitarian and therefore peasant smallholders
72
are more likely to share similar immediate and long-term interests which in turn solidifies common
identities between them. Durkheim (1933) emphasizes the importance of such shared interests for
communal identity formation: “The totality of beliefs and sentiments common to average citizens of
the same society forms a determinate system which has its own life; one may call it the collective or
common conscience” (79). This is not to say that landless or land-poor peasants in vertical agrarian
structures do not share common interests or identities. Rather, in such highly stratified and divided
class structures they are substantially more dependent on large landholders to survive and therefore
less prone to challenge their authority. As Wolf (1969) points out: “A rebellion cannot start from a
situation of complete impotence” (290). In cases where they do possess sufficient group cohesion and
resources mobilization to challenge landed elites, they face an almost certain violent counter-reaction,
one which is often accompanied with support of the armed agents of the state.
Organizational endowments are inherently tied to the character of the pre-existing agrarian
social structure. Peasant hierarchies shape how different groups ultimately organize and mobilize to
resolve their collective action problems. I argue that horizontal agrarian social structures that emerge
in spaces where the state and landed elites are conspicuously absent will develop a variety of local
mechanisms for resolving collective actions problems. This multiplicity in turn engenders a multiplex
network of organizational endowments that are reflective of the dense local peasant social networks
from which they emerge, mechanisms that will ultimately determine their capacity for mobilization
(Tilly 1978, 2003). According to Oberschall, a “[…]viable network of communal relations can be the
foundation and breeding ground for the rapid growth of modern associational networks” (Oberschall
1973, 123). James Scott (1976) echoes the importance of these civi networks: “[C]ommunitarian
structures…have, due to their traditional solidarity, a greater capacity for collective action. For such
villages, it would seem, the organizational barriers to action are reduced simply because they have
73
recourse to an existing structure of local cooperation that has remained intact; their “little tradition”
is a ready made vehicle of action” (202).
Vertical agrarian social structures in contrast are unlikely to develop such a diverse array of
organizational endowments. These rural social hierarchies often are more prone to creating or
adopting monolithic organizational endowments that are centered around one principal civic
organization, or a handful of disconnected, competing organizations that are top-down structures. In
other instances, there will literally be a complete absence of organizational endowments found in such
vertical agrarian social structures. Joel Migdal (1974) attributes the organizational deficiencies found
in vertical agrarian social structures to the preferences of well entrenched landed elites:
Peasants under vigilant and strong lords had types of social organization which reflected the limitations the lord put on them. The most striking characteristic was the atomization of the community into individual households. In the extreme case of haciendas, for example, lords attempted to make themselves the suppliers of all peasant needs. The hacendado sought to direct as many of the peasants’ contacts as possible into dyadic relations with him, undercutting the need for dependence of the peasants on one another and thus undermining any possible basis of association that might challenge the lord. (42)
In the Colombian case, many of the relevant endowments which appeared during the period
in question were the result of top-down state policies that both created formal civic organizations at
the sub-national level in an attempt to improve local governance throughout the country (Eaton 2006;
Kaplan 2017; Vargas 2018). This provides an excellent opportunity for comparative analysis at the
sub-national level by examining the manner in which these civic reforms affected different
communities.
The rest of this chapter outlines the pre-existing configuration of Montes de María and Arauca
(plains and piedmont) in the territorial, economic, political, and civic spheres. The central focus is on
the long and short-term development of the agrarian social structures which emerged in these
disparate spaces, while also emphasizing the particular conditions found in these distinct zones prior
to the arrival of armed insurgents in the early 1980s. More specifically, the following sub-sections
examine the pre-existing land tenure patterns, the effects of the National Front reforms, the
74
agricultural modalities, the limits to state institutions or presence, social mobilizations, and the role of
emergent civic organizations such as ANUC and the community actions boards in both Montes de
María and Arauca prior to the arrival of armed non-state actors in the early 1980s.
A. Montes de María
Initially founded as an isolated refuge for runaway slaves during the early colonial era, the
Spanish authorities in nearby Cartagena eventually asserted control of Montes de María over the
course of the 18th century. Similar to elsewhere in Colombia and Latin America, the Spaniards
established the hacienda (latifundia) land tenure system in the region, reordering local society into a
highly stratified rural hierarchy which persisted well into the 20th century. Following independence,
Montes de María was ostensibly governed by political appointees based in Cartagena, yet landed elites,
most of whom supported the Colombian Liberal party, were the de facto source of governance in the
underdeveloped and forgotten region.
Table 2. Territorial, Political, Economic, and Civic Spheres in Montes de María (Pre-1980)
Territorial
Political
Economic
Civic
● Hacienda
(Latifundista) Land Tenure Pattern
• Appointed Departmental Governors (Bolívar, 1857; Sucre 1966)
• Subsistence Agriculture
• Tobacco Boom (1850s onwards)
● Early 20th century
Peasant Leagues ● ANUC (1960s, 1970s)
Source: Author’s Elaboration
It was only during the mid-19th century tobacco boom that Montes de María was finally
integrated into the national economy, serving as a major production enclave for Colombia’s main
export at that point in time. However, with the sudden rise of coffee as Colombia’s main source of
foreign exchange in the first half of the 20th century, the regional tobacco industry slowly began to
languish, and large landholders and landless peasants alike turned to cattle ranching and subsistence
agriculture to survive. The region can trace its first civic organizations back to the rural peasant leagues
75
founded during the first decades of the 20th century, but it was the creation of ANUC in the late 1960s
that saw Montes de María become one of the main focal points of peasant activism in Colombia over
the course of the 1970s.
i. The Territorial Sphere
The region of Montes de María, also known as the Serranía de San Jacinto, can be found roughly
between the respective capitals of the departments of Sucre (Sincelejo) and Bolívar (Cartagena) in the
northern Caribbean coast of Colombia. Characterized by low-lying hills and covered in jungle adjacent
to the sea, in geographic terms the only major distinction between the sub-regions of Montes de María
is between the coastal zone (zona costera), the high mountain (alta montaña) and the zone below (zona
baja), but all share very similar historical traits.10 The region is comprised of fifteen municipalities, eight
from Sucre department (San Onofre, Chalán, Colosó, Ovejas, Palmito, Los Palmitos, Morroa,
Toluviejo), and another seven from Bolívar department (Córdoba, Zambrano, El Carmen de Bolívar,
San Jacinto, San Juan Nepomuceno, El Guamo, Maria la Baja), which cover a geographic area of 6466
square kilometers between the Magdalena river and the Caribbean sea (PNUD 2010). Ironically
enough, despite being located between two extremely important bodies of water, the region of Montes
de María possesses no rivers itself, as local populations have always been dependent on various local
streams (arroyos) for themselves and their livestock.
The territory is extremely complex. The maze of rugged hills covered in lush vegetation range
rise as high as 810 meters above sea level (el Cerro Maco), although the average peak is closer to 400
meters (Porras 2014). The coastal zone is found in those municipalities adjacent to the Caribbean Sea,
namely San Onofre, Toluviejo, and Palmito, whereas the high mountain and zone below are harder to
articulate in such precise terms. The high mountain is the intermediate area found between Maria La
10 Interview 38, Sincelejo (2016).
76
Baja, San Onofre, San Jacinto, El Carmen de Bolívar, Ovejas, Chalán, and Colosó. Whereas the
remainder of the region can be tentatively characterized as the zone below, this title refers more to the
eastern and southern swathes of Montes de María, particularly in municipalities such as El Guamo,
San Juan Nepomuceno, Córdoba, Zambrano, Los Palmitos, and Morroa. These distinctions are of
some importance but not enough to affect theoretical assumptions given the homogenous character
of the region’s peasantry and geography.
Map 1. Montes de María
Source: Nicolás Herrera (2017)
Montes de María was first inhabited during the colonial era by runaway African slaves from
the nearby province of Cartagena, which was then the epicenter of the Spanish viceroyalty in northern
Colombia. These escapees eventually established clandestine communities (palenques) in Montes de
María, a region that “[…]offered the best strategic advantages” due to the abundance of “[…]thick
forests, swamps, and the mountains of María…topographic conditions which made it difficult for
Spanish troops to access the runaway territory” (Porras 2014, 338). The palenques encountered
77
something between mild resistance and general indifference from a sparse distribution of indigenous
tribes such as the Zenú, an ethnic grouping which had far greater presence to the southwest in what
is today the department of Córdoba (Falchetti 1995). Over the course of this period, the region also
witnessed the arrival of poor mestizos referred to as rochelas, who similar to their African descendant
counterparts, were also seeking greater refuge from the exclusive and repressive tendencies of the
Spanish colonial system (Fals Borda 1976).11
The Caribbean coast of Colombia was of enormous economic and geostrategic importance to
Spain, serving as a port of entry for the European colonization of not only Colombia, but also the rest
of Latin America.12 Initially, the isolated communities established in Montes de María were largely self-
sufficient and seemingly removed from this colonial framework, agricultural enclaves which thrived
on the illicit production of aguardiente, a potent sugar based liquor.13 The end of the encomienda model
of landownership and labor - a system of labor that effectively gave each successful conquistador the
right to the forced labor of all local native habitants who resided under their area of conquest - in the
early 18th century saw the emergence of the hacienda system. The new system differed from the old
insofar as former indigenous servants, African slaves, and poor mestizos alike were incorporated into
the ever-growing rural proletariat as “[…]different types of tenants, sharecroppers, and peons tied by
bondage debts” (Zamosc 1986, 9). Landed elites in Cartagena and Mompox expanded the hacienda
system and cattle ranching throughout the entirety of the Caribbean coast and by the dawn of the 19 th
century, Montes de María had finally been integrated into the rest of the colonial viceroyalty (Reyes
Posada 1978).
11 In Montes de María “individualism prevails, the nature of the montemariano is libertarian”. Interview 94, Sincelejo, 2016.
12 Interview 101, Sincelejo, 2018.
13 Interview 94, Sincelejo, 2016.
78
During the Wars of Independence, Montes de María was not spared from the death and
destruction that accompanied the violent conflict. Between 1815 and 1821, Ovejas, El Carmen de
Bolívar, and San Jacinto witnessed brutal encounters between the Royalist and Republican troops
(Támara 1997). With independence, the transfer of power from Spanish colonial elites to creole
oligarchs did not alter the social, economic, or political spheres in Montes de María in the slightest. If
anything, the first years of independence laid the foundation for what was to be the further
entrenchment of the hacienda system in Montes de María with the introduction of tobacco production
mid-century.14
The sudden conversion of Montes de María into an export-based enclave in the mid-19th
century put enormous tensions on the region’s peasantry by spurring the accelerated accumulation of
land by large landholders and foreign entrepreneurs alike.15 Following the War of a Thousand Days
(1899-1902), many mid-to-high high ranking military men who had participated in the conflict were
awarded large tracts of land in Montes de María for their service to the partisan cause, thus displacing
even more peasants from their smallholdings. The state of the national economy was in tatters after
the ruinous effects of this war and thus land may have been the only commodity worth anything in
real terms with which the political parties could incentivize their military leaders to mobilize and fight.
However, the land tenure pattern throughout the various villages and municipalities of the region was
hardly homogenous as land titling was a convoluted, complicated affair. Numerous rural communities
were comprised of peasant smallholders who worked extremely limited family plots in the general
absence of large landholders and survived through basic subsistence farming and small-scale animal
husbandry of chickens, hens, and pigs. For example, in the remote village of Las Lajas in the high
mountain of San Jacinto, the family plots were hereditary and measured between two and ten
14 Interview 101, Sincelejo, 2018.
15 Interview 34, El Carmen de Bolívar, 2016.
79
hectares.16 Residents from other villages in the region comprised of smallholding peasant families
recall that the average individual farm was larger that those found in Las Lajas, yet did not exceed fifty
to sixty hectares maximum.17 In one village in the zone below of San Jacinto, local peasant leaders
claim that prior to the arrival of armed non-state actors to the region, local land “[…]belonged to
everybody and was shared”.18
In other communities, the majority, if not all of the land belonged to a couple of prominent
landowning families who had established themselves well before the advent of the 20 th century. In
Bajo Grande, a small hamlet in the zone below of San Jacinto, most local peasants either possessed
minimal amounts of land themselves or were landless, and thus toiled for large landholders who
themselves occupied public lands.19 Initially an indigenous village in the colonial era, the hamlet of
Chengue in the high mountain of Ovejas was eventually settled by the Oviedo and the Mariño families,
who established massive haciendas encompassing the whole surrounding area. Other peasants who
settled there ended up with no land and no other choice but to work for these particular families.20 In
nearby El Salado, a tobacco dry-port in the zone below of El Carmen de Bolívar, the Cohen Mesa
family were one of the wealthiest families in the area, with some twenty different properties measuring
up to ninety hectares apiece for a grand total of 1800 hectares worth of property in the rural zone.
The Cohen family contracted all the necessary labor needed to maintain agricultural production on
their estates in a sharecropping system which was widely deployed throughout rural communities in
Montes de María (aparcería), while they devoted themselves to other endeavors such as cattle
ranching.21
16 Interview 77, San Jacinto, 2016.
17 Interview 86, El Carmen de Bolívar, 2016.
18 Interview 81, San Jacinto, 2016.
19 Interview 93, Bogotá D.C., 2016.
20 Interview 90, El Carmen de Bolívar, 2016.
21 Interview 85, El Carmen de Bolívar, 2016.
80
Thus, Montes de María by mid-20th century was a patchwork of multiple sub-regional land
tenure patterns in which the ownership of land varied markedly from one village to the next. The only
constant throughout the region is that the peasantry for the most part were land poor or landless and
survived due to subsistence agriculture and sharecropping schemes. Apart from the influx of
entrepreneurs and merchants from elsewhere in Colombia and beyond during the tobacco boom in
the 19th century, the composition of the regional peasantry remained largely static and homogenous,
with local habitants remaining within their native towns and villages their entire lives.22 A melting pot
of African, indigenous, and Spanish cultures, the region retained an extremely agricultural character
throughout the centuries, which is perhaps best evidenced by the long-standing practice of the minga,
a reciprocal labor exchange between neighbors in particular communities, who would work together
in a rotating cycle to perform certain labor intensive activities on their smallholdings to maximize
efficiency and protect their harvests (Pérez 2010).23
The mid-century carnage of la Violencia did not wreak havoc on Montes de María in the same
manner as it played out in nearby departments such as Antioquia, yet it did alter the composition of
the regional social hierarchy. During this period, numerous paisas fled their homes region due to
violence or the desire to improve their economic fortunes and many settled in Montes de María.24 In
Montes de María, the quick upward ascension of these migrants on the local hierarchy and their
particular ethos further complicated what was already a complex local social order. These migrants
were “focused more on entrepreneurship, building estates and acquiring private property, which
22 Interview 85, El Carmen de Bolívar, 2016; Interview 77, San Jacinto, 2016.
23 Interview 94, Sincelejo, 2016.
24 Paisas is the term used in Colombia to describe those hailing from the department of Antioquia, and the coffee axis (el eje cafetero) extending south along the western range of the Andes. They are distinguished in Colombia as: […]the nation’s sharpest businessmen and pragmatic technocrats, a region of aggressive colonizers who were also fiercely Catholic. A prolific lot…who gave rise to a society characterized by a sense of strong regional identity, large families, and small property holders” (Roldán 2002, 11).
81
clashed profoundly with local peasant, Afro-descendent, and indigenous traditions”, such as shared
work systems like the minga and customs of reciprocity.25
During the return to democratic rule under the National Front, Montes de María served as
one of the testing grounds for the mixture of agrarian reform policies of the 1960s, most notably Law
135 which created INCORA and gave peasants the right to occupy, work, and legally claim title to idle
land, even if it technically belonged to another landholder. Contrary to previous efforts at agrarian
reform such as Law 200 of 1936, the new legal regime established under Law 135 ensured that all
public lands (terrenos baldíos) were only able to legally be allocated by official state bureaucracies in the
form of INCORA and the National Agrarian Fund (Fondo Nacional Agrario), the former of which was
tasked with titling, the latter effectively serving as a land bank. Both of these entities under Law 135
were able to expropriate and redistribute unproductive lands which legally belonged to large
landholders and cattle ranchers.26
The role of INCORA in Montes de María produced mixed results, with some communities
receiving the necessary bureaucratic support in claiming title to idle lands which they had worked, in
many cases for generations, under the well established sharecropping system in the region. In Montes
de María, INCORA succeeded in procuring land from latifundistas and redistributing it to landless
peasants, with one-third to one-fourth of the available land in the region expropriated for the local
agrarian reform. For example, in neighboring Córdoba department, INCORA was only able to title
some ten thousand hectares of land for local peasants, a fraction of what the institute achieved in
Montes de María.27 Many communities previously dominated by large landholders suddenly
experienced dramatic shifts in the local social hierarchies because of the success of INCORA in
25 Interview 93, Bogotá D.C., 2016.
26 Interview 101, Sincelejo, 2018.
27 Interview 38, Sincelejo, 2016.
82
breaking up these large unproductive estates. One elderly peasant in the village of Palmar (Ovejas),
recalls that “in the beginning there were large landholders”. However, with the arrival of INCORA
“the lands were distributed in general to give ten hectares to each family”.28
In other communities, local attempts by INCORA to break up the established latifundia were
less successful. For example, in El Salado “some peasants obtained land because INCORA broke up
a few haciendas”. However, others “stayed without land and began to struggle for it”, further generating
“large conflicts between the landholders and the peasants which left many dead”.29 In contrast to other
regions in Colombia, the state-designated metric for creating a self-sufficient family held farm, the
Family Agricultural Unit (unidad agrícola familiar – UAF), was lower and was thereby considered
insufficient by many local peasants in Montes de María. In Chengue, INCORA purchased a stretch of
land measuring some 77 hectares, in which 49 families were settled, each receiving personal plots
measuring 1.6 hectares in total, an amount of land sufficient for erecting a familial domicile and little
else. One local leader from Chengue explains that for this reason the agrarian reform “didn’t change
anything between peasants and large landholders”, land continued to be denied to poorer villagers and
“they had to keep working on the haciendas”.30 Another problem was the limited number of recipients
who were able to receive title from INCORA. In the high mountain community of Las Lajas (San
Jacinto), many families held hereditary title to extremely small plots ranging from two to ten hectares
in size, whereas another eight families had been able to obtain titles to larger tracts of land measuring
up to twenty hectares by INCORA. These fortunate individuals were a minority however, representing
roughly ten percent of the families in Las Lajas.31
ii. The Economic Sphere
28 Interview 91, Ovejas, 2016.
29 Interview 85, El Carmen de Bolívar, 2016.
30 Interview 90, El Carmen de Bolívar, 2016.
31 Interview 77, San Jacinto, 2016.
83
Initially, the economic life of Montes de María was based on subsistence farming. However,
all of this changed with the sudden rise of the local tobacco industry in the mid-19th century. The
production of this crop was centered around the stretch of land which runs north from San Juan
Nepomuceno southwards through neighboring San Jacinto, El Carmen de Bolívar, Ovejas, Colosó,
Chalán, Los Palmitos and beyond. Tobacco production in Colombia in during the first half of the 19 th
century was principally concentrated in departments in the interior, namely Tolima and Santander,
with the state maintaining a monopoly on the production, marketing, and sale of the product.32 Montes
de María quickly became one of the most important producers of tobacco in all of Colombia. In 1857,
tobacco exports from El Carmen de Bolivar constituted a fifth of those produced in Ambalema
(Tolima), however, five years later the two produced practically the same amount of tobacco for export
in 1862. The following year, El Carmen de Bolívar led the entire country in tobacco exports and by
1888 the region produced more than the rest of Colombia combined (Posada Carbó 1998). The
expansion of the tobacco industry in Montes de María was centered in and around El Carmen de
Bolívar, which quickly became the unofficial capital of the region. Sincelejo served as the commercial
hub for the burgeoning industry, while the nearby town of Magangue, located on the Magdalena river,
became the collection center where all the tobacco was then sent upstream to foreign North American
and European markets.33
The increased demand for Colombian tobacco from European markets propelled both foreign
investment in the production of local tobacco and the migration of European entrepreneurs to the
region to monopolize the production and export of this particular cash crop. Newly established
commercial agencies would lend money to local peasants to plant tobacco, and in turn would purchase
the crops directly from them. This system was atomized and designed to encourage individual rather
32 Interview 94, Sincelejo, 2016.
33 Interview 38, Sincelejo, 2016.
84
than collective production, all in order to foster a greater worker dependency on the foreign buying
houses.34 Many of these foreign merchants would end up staying in Montes de María and quickly
converted their newfound economic success into extensive rural estates. An illustrative example of
this is the Cohen family, which came to be the dominant landholding family in the rural town of El
Salado in El Carmen de Bolívar. The family’s patriarch, a Jewish merchant from England, arrived in
El Carmen de Bolívar in the first half of the 19th century and quickly established himself as a leading
buyer for an English agency operating in the region.35
The tobacco industry remained the principal agricultural activity in Montes de María well into
the 20th century and was the driving force between the growth and quasi-modernization of the region.
By the turn of the century, Sincelejo, El Carmen de Bolívar, and Magangue had “[…]become the
biggest urban centers {in the region}. They grew 400% because they controlled the tobacco circuit in
the Caribbean”.36 The importance of tobacco was such that the chronic civil wars of 19 th century
Colombia did not affect production, as tobacco remained arguably the economic activity that sustained
the wartime economy and was thus spared from its destructive social and economic effects (Porras
2014). The sudden decline in regional tobacco production followed the end of the Second World War,
as the national economy had gradually shifted away from an export model based on tobacco to one
dependent on coffee, a crop whose production had surged in Colombia in the first half of the 20 th
century and peaked around the 1940s (Bergquist 1986; Bejarano 2011).37
The rise of tobacco in Montes de María brought sudden growth and stimulated the local
economy, however, when demand declined after la Violencia there was no agricultural activity of any
significance which could fill the void. Certain towns excelled in producing specific crops for domestic
34 Interview 101, Sincelejo, 2018.
35 Interview 85, El Carmen de Bolívar, 2016; Interview 94, Sincelejo, 2016.
36 Interview 94, Sincelejo, 2016.
37 Interview 38, Sincelejo, 2016.
85
consumption, such as avocado farming in Chengue (Ovejas), or yucca in San Rafael (Ovejas).
However, beyond these small production enclaves, the main two forms of agricultural activity which
continued in the face of the tobacco bust were subsistence farming and cattle ranching, the former
practiced by the peasantry, the latter by landed elites. As one local native of Ovejas puts it: “The
tobacco leaf was a curse, despite generating employment it also took the municipality backwards
because there was no diversification of crops, peasants were dependent on tobacco and if there wasn’t
any then they had nothing”.38 Another peasant leader from Las Palmas (San Jacinto) recalls that even
after the decline of tobacco that it remained the lifeblood of his community, employing some three
thousand families in the community. The crop enabled local peasants, as “each nuclear family was its
own company”, with tobacco “making them very prosperous and productive”.39 In El Salado (El
Carmen de Bolívar), tobacco remained a key crop throughout the 20th century. The village produced
between one and one and a half million kilos which were the “key sustenance which generated income
for families in El Salado”.40
Despite the apparent micro-level economic empowerment of peasant communities in Montes
de María, others contend that the tobacco boom and bust cycle in Montes de María enabled a deeper
entrenchment of an already very stratified class system with the arrival of foreign entrepreneurs who
were far better situated for local advancement, a process which led to the greater concentration of
land in fewer hands. According to Porras (2014):
The change was stimulating an active commerce, awakened, whose excesses began to be invested in the purchase and concentration of land, developing a latifundista logic which had characterized the province of Cartagena since colonial times, all of which entailed a break with the past, as until then the prevailing agrarian structure in Montes de María had been constituted exclusively by public lands, with public access for subaltern sectors comprised of the indigenous, Afro-descendants, freemen and peasants. As a consequence, new social conflicts arose, this time over access, tenancy, and use of the land. (347)
38 At its height, tobacco employed 1300 women and 400 men in the municipal capital of Ovejas alone. Presently, the industry employs some 130 women. Interview 92, Ovejas, 2016.
39 Interview 81, San Jacinto, 2016.
40 Interview 85, El Carmen de Bolívar, 2016.
86
iii. The Political Sphere
Similar to most of Colombia outside of the capital and other major urban centers, formal
political institutions carried little importance in peripheral regions for most of the 20 th century. In
regions like the Caribbean coast or the Eastern plains, political power was wielded by landed elites
who were beholden to one of the two traditional political parties and were largely left alone by political
elites in distant Bogotá (Palacios 2005; Robinson 2013). The only time when these regional balances
of power were contested and reconfigured was during the outbreak of national level conflicts such as
la Violencia or the current protracted civil war.
While various Colombian historians have claimed that la Violencia and its partisan bloodletting
left the northern Caribbean region relatively untouched, this is disputed by locals and other experts
embedded in the region who remember that the predominantly Liberal region experienced violent
repression at the hands of the highly politicized, Conservative police forces of the era, los chulavitas.
Long time regional peasant organizer and ANUC leader, Jesús ‘Chucho’ María Pérez (2010) recalls
this partisan repression in Montes de María: “As such they {local Conservatives} were subjugating all
of the Liberals, not only with arbitrary imprisonment, but also with lashes from the whips of the
political police of the time” (10). Landed elites used “the armed wing of the regional police to displace
peasants from their plots and in particular those known Liberals who possessed land”.41 These micro-
level cleavages in many cases preceded la Violencia by decades, if not longer, and would emerge yet
again later on in the 20th century with the intensification of the more recent armed conflict. A telling
example of this is the family-based group known as ‘los Rodríguez’ from the village of Macayepo in the
high mountain area of El Carmen de Bolívar:
All the guerrilla groups throughout history - from the era of la Violencia when the Liberal guerrillas settled in Montes de María, until the intervention of the present guerrillas - declared war on los Rodríguez, who were Conservatives and had ties to the the political police in the time of {Conservative president} Laureano
41 Interview 93, Bogotá D.C., 2016.
87
Gómez. This is how the people of Macayepo became stigmatized as cattle rustlers and criminals over the course of several decades, due to one family which was devoted to breaking the law. In the middle of the 20th century, there was a massacre in Macayepo where there were women abused and where their breasts were cut off, all because of this stigma and the confrontation between the Liberals and the Conservatives.42
Prior to the arrival of armed non-state actors to Montes de María, the level of state presence
in the region was extremely limited. While the municipal capitals possessed a small contingent of
national police, the rural villages in the coastal zone, the high mountain, and the zone below were
virtually abandoned to regulate and govern themselves. One local peasant in the high mountain of San
Jacinto remembers that “there were no state entities which responded to or would solve our
problems”, as in the event that somebody died “we would take care of it ourselves”.43 Another lifelong
resident of El Salado echoes this level of official negligence:
Social policy was one of abandonment. There were no doctors. Maybe a nurse would visit once a week. There were no roads. To enter El Salado from El Carmen de Bolívar people would take five to six hours, if it didn’t rain. They put a health center there later which didn’t function, as there was no medicine or experts. To transport goods, we had to move everything by mule. The municipal government never fulfilled their social policies, as the relationship between the administration and the population was based on abandonment and a failure to keep promises.44
Rural towns and villages were lucky to count on the presence of a police inspector, although
this does not appear to have been a common feature of these agrarian communities found throughout
the coastal zone, the high mountain, or the zone below.45 Quite simply, “there was a total
disconnection between the municipal capitals and the rural zones”, a dynamic which was made worse
by inclement weather such as torrential seasonal rains which made local dirt paths impassable.46
Despite the weak state presence, communities throughout Montes de María were generally peaceful
prior to the arrival of armed groups over the course of the 1980s. One local farmer in Las Lajas (San
42 Interview 94, Sincelejo, 2016.
43 Interview 78, San Jacinto, 2016.
44 Interview 85, El Carmen de Bolívar, 2016.
45 Interview 40, El Carmen de Bolívar, 2016.
46 Interview 83, El Carmen de Bolívar, 2016.
88
Jacinto) recalls that regardless of the isolation, “camaraderie, equality, and tranquility prevailed”.47
Throughout the patchwork of peasant communities across the region “everybody knew one another
and were like a family”, with people “lending one another a hand when they needed it”.48 According
to one fisherman from the idyllic coastal town of El Rincon del Mar (San Onofre):
It was paradise on earth. There weren’t problems of any kind. There was no electricity but everybody knew one another, and we used homemade lamps in order to see at night. The social fabric of the community was very strong. It was like one large family and everybody was brothers. There was no institutional presence however the village elders served as a kind of authority within the community. Also, local priests were like gods themselves.49
Even though the common recollection is that there was virtually no official state nor
institutional presence, this does not mean that formal politics did not exist in any shape or form in
Montes de María. In a pattern seen throughout the entire Caribbean coast, and in all of Colombia,
regional politics functioned along a well-entrenched pattern of clientelism. In rural dynamics such as
Montes de María and elsewhere in latifundia zones throughout the country, the landed elites’ political
power “[…]is based on the economic dependency of the peasants on the {boss} or large landholder”
(Reyes Posada 1978, 113). The logic of regional clientelist politics in Montes de María was simple, as
landed elites decided who their peons would vote for. A local expert explains this longstanding
regional practice:
Historically, people voted because the political capital of the region was based on land. What this entailed was that whoever had a lot of land not only possessed capital and the means of production, rather that those who worked their haciendas were also converted into votes. Before the commercialized clientelist dynamic such as vote buying, patronage dominated as everybody voted for whomever their boss wanted them to.50
Locals throughout the region recall that politicians and other candidates only recognized that
their communities existed during election campaigns. Political candidates were often linked to one of
47 Interview 77, San Jacinto, 2016.
48 Interview 78, San Jacinto, 2016; Interview 79, San Jacinto, 2016; Interview 81, San Jacinto, 2016; Interview 85, El Carmen de Bolívar, 2016.
49 Interview 45, San Onofre, 2016.
50 Interview 93, Bogotá D.C., 2016.
89
the two traditional parties, yet there was virtually no ideological distinction between them as they owed
their position to their affiliation with regional political clan bosses (caciques).51 These political ‘clans’
found throughout Montes de María and the larger Caribbean region were family-based and often local
politics was the family business. Alejandro Reyes Posada (1978) describes one such politician from
Ovejas and how economic power was converted into political power and vice versa: “He is the heir
to political and economic power. He moves fifty percent of the tobacco wealth in Ovejas. His political
lieutenants are the tobacco-buying agents, who also trade in supplies and advance money to tobacco
growers. They force the peasants they buy the crops from, as much as the factory workers who process
the tobacco, to vote” (128).
iv. The Civic Sphere
Over the course of the 20th century, the civic sphere in Montes de María became a highly
contested space where peasants attempted to redress long-standing regional inequalities about land
distribution with landed elites. The extreme concentration of land engendered the “proletarianization
of the peasantry which led to the creation of large peasant movements.”52 The first decades of the 20th
century witnessed the emergence of increased peasant activism in rural Colombia and this trend was
especially strong in Montes de María. For example, the first agrarian union in Colombia was founded
by tobacco workers in Colosó in 1913. The first recorded instance of peasant mobilization occurred
in the 1920s when a local latifundista fenced in an expanse of land that was communal ground for local
small holders and was their main source of water and firewood. The union members promptly took
down the fences and reclaimed the commons. This event constituted “an exercise in the recuperation
and demand for their communal land. These agrarian unions functioned in favor of access to land,
51 Interview 85, El Carmen de Bolívar, 2016.
52 Interview 93, Bogotá D.C., 2016.
90
indisputably along these lines and of course, as all unions do, to improve the labor dynamics for all
involved”.53
Various Liberal governments attempted land reform in the 1930s, most notably the creation
of Law 200 by the Alfonso López Pumarejo administration (1934-38). However, the application of
Law 200 ended up benefitting large landholders more than land-poor peasants in rural environs shaped
by highly unequal land ownership such as Montes de María. Landed elites were able to subvert,
manipulate, and co-opt local lawyers or judges into respecting their land titles, even if they were false
or established on a highly questionable basis. Despite the failure of the Liberal land reforms of the
time, peasant activists were not deterred from continuing to organize or mobilize in pursuit of their
common goals in Montes de María. Over the course of the 1940s and 1950s, even during the height
of la Violencia, a regional trade guild composed of tobacco growers and merchants alike, purchased
various expanses of land with the goal of giving small producers and cultivators access to their own
land.54
Early efforts at land reform by the National Front governments similarly failed as land
concentration intensified throughout the Caribbean coast, while thousands of sharecroppers and poor
rural tenants were forcibly evicted from the lands which they had worked for decades, if not
generations. In traditional latifundia regions such as Montes de María, this process was designed to
depopulate the available land, while increasing the number of cattle on these estates (Reyes Posada
1978). To make matters worse, the peasant capacity for collective action and resistance had declined
markedly over previous decades. The organizational legacy of the peasant leagues was in serious
decline, dropping from some 567 to some 89 active organizations by 1965, and given that half of these
original registered bodies were located in the Caribbean region, this reduction in peasant activism
53 Interview 101, Sincelejo, 2018.
54 Ibid.
91
presented an ominous sign for the future of the common peasant in regions such as Montes de María
(Zamosc 1986).
The Lleras Restrepo administration sought to give the peasantry the necessary tools they
needed to challenge these entrenched interests through the creation of ANUC, representing an
attempt to break the regional impasse imposed on the country by regional landed elites so that the
Colombian peasantry could assert their democratic rights. Rivera Cusicanqui (1987) describes ANUC
as a means “[…]to enable the organized peasantry to overcome its desperate poverty, demand and
obtain concessions from the State and the dominant classes, acquire and exercise its right to citizenship
and become a factor for democratization and political stability” (62). Through design and reach,
ANUC was designed to provide the entire Colombian peasantry with the means to unite in common
cause against the large landholding classes, regardless of geographical distance and regional
differences. A former ANUC leader from Montes de María summarizes the importance of achieving
this goal in the region: “All of these disperse communities united for the first time to work along the
same general lines to struggle for the general welfare of rural habitants” (Pérez 2010, 17).
The structure of ANUC was such that it gave the organization substantial influence from the
village level to the national stage. At the most local level, there was a local committee in every village
(corregimiento/vereda), which would serve as a pool of recruits from the ‘users associations’ (asociación de
usuarios) as they were called, to the association at the municipal level. The municipal association had a
board which was democratically elected, and these associations provided and received support from
the departmental association, which possessed at least one member from each municipal association.
Beyond this, ANUC possessed a congress known as the National Assembly, which was comprised of
all of the members of the departmental associations’ boards. Finally, there was a national level board,
composed of one representative from each departmental association, from which five members were
elected to the national executive committee to represent ANUC at the national level (Zamosc 1986).
92
In no other region did ANUC experience such a sharp rise and subsequent fall in later years
as Montes de María. When Sucre separated from western Bolívar in 1966 to become a department,
President Lleras appointed Apolinar Díaz Callejas, a progressive Liberal lawyer from Colosó and
assistant director of INCORA, as the first governor of the department. Under the departmental
administration of Díaz Callejas, Sucre became the first department to organize and convene a
departmental association of ANUC in 1968. The first few years of ANUC saw massive growth, with
some 600 000 members registered nationwide by March of 1968, a level of mobilization which was
sustained over the following years. By October, 1971, the organization possessed 989 306 registered
members, 28 departmental associations, 634 municipal associations, and some 13 983 trained leaders
(Zamosc 1986, 57). Mobilizing peasants at the sub-departmental level was not always easy, as ‘Chucho’
Pérez (2010) remembers this process in Montes de María and elsewhere in Sucre: “[…]the campaign
opened with a seminary to train leaders and operatives in the department. The enrollment of members
encountered various inconveniences, well there was a lot of fear on part of peasants who were scared
of being evicted from their small sharecropping or rental plots” (27).
When Carlos Lleras Restrepo’s term ended in 1970, the final National Front administration
led by Conservative Misael Pastrana Borrero inherited his ambitious ‘bourgeois reformist’ agenda.
However, the prevailing circumstances were such that the new administration and ANUC were
heading towards an imminent showdown. Peasant leaders from the organization complained that
other official state bureaucracies and entities related to agriculture and development had blacklisted
ANUC members from joining. Meanwhile, a wave of land occupations by peasant activists on the
Caribbean coast terrified regional landed elites and their political allies, a trend which generated fresh
problems because quite simply, ANUC was perceived to have gone “[…]beyond the state’s control”
(Reyes Posada 1978, 151). By the 1972 national level congress, ANUC split into two factions with
those more moderate peasant leaders who favored maintaining ties with the Colombian government
93
convening in the city of Armenia, while the majority of ANUC leaders at the departmental and
municipal levels who sought a complete organizational autonomy from the state sent their
representatives to an alternative congress in Sincelejo. The two currents, the Armenia line and the
Sincelejo line as they were called, would become even more polarized the following year when
President Pastrana sided with the moderate faction and large landholders nationwide with the
bipartisan Pact of Chicoral, which sought to “[…]impose a business model of integrated rural
development instead of the redistributive model of agrarian reform” (Porras n.d., 3).55
Despite this massive betrayal by the Colombian government, ANUC was in reality becoming
exactly what its members wanted it to be. Free from constraints imposed by political elites in Bogotá,
and armed with a nationwide peasant organization with local reach throughout the length and width
of the national territory, ANUC “[…]quickly became a pressure group instrumental in revealing the
total inability of the State to fulfill the targets of agricultural reform and its weakness in the face of the
landowners’ counter-attack which had quickly gained the support of splinter groups of the ruling
classes and their spokesmen inside the traditional parties.” (Rivera Cusicanqui 1987, 75). At this
juncture, ANUC began to deploy a new tactic which would cause massive upheaval in rural
communities throughout the country during the 1970s and 1980s: land occupations. And nowhere
else in Colombia were these more common or effective than on the Caribbean coast. Between 1970
and 1978, some fifty percent of all recorded land occupations in Colombia occurred on the Caribbean
coast, with almost a fifth of these, some 199 total, taking place in Sucre department alone (Zamosc
1986, 75). The rash of peasant evictions in the late 1960s coincided with the establishment and
consolidation of ANUC in Sucre, thus creating the conditions for increased peasant activism in the
region. One local activist recalls that in this department, those invading and occupying the land
55 Interview 93, Bogotá D.C., 2016.
94
“[…]were practically jurists”, as they “[…]knew the laws inside out” (Escobar 1982, 56). The
organization behind occupying idle haciendas “[…]was much better in Sucre”, as large landholders
were so frightened by the rise of peasant squatters that in 1971 they offered “[…]to hand over 10
percent of their lands for agrarian reform purposes” (Zamosc 1986, 79).
One of the regional focal points of this proliferation of land occupations was in Montes de
María, with the most affected areas being municipalities such as Ovejas, Colosó, Chalán, El Carmen
de Bolívar, San Jacinto, and Maria La Baja.56 The local municipal councils of ANUC were strong
proponents of the Sincelejo line and were the main protagonists behind the massive increase in peasant
occupations of large haciendas, a phenomenon which forced many landed elites to sell part or all of
their holdings to INCORA so that they could be parceled off to land poor peasants. A regional expert
recalls how these activists carried out these mobilizations:
They chose the objective with anticipation. The common characteristics that these lands had is that they belonged to exclusively latifundista families. The García Badell family from Corozal had more than 20 000 hectares of land in Ovejas, and these cattle ranchers had broken their contracts with sharecroppers and many of them knew the land perfectly and after they identified the properties, they went with their families and raised poorly improvised domiciles by cutting wood, cane stalks, and palms leaves, raising these structures and assembling a camping ground. These campsites were frequently destroyed by police and they would leave the property, but a few days, weeks, or months later they would return and reassemble the site, all of which generated a tense situation and conflict until INCORA intervened and declared the land a public good, a process which people called ‘Incorizing’ the land.57
This tactic proved to be a particularly effective modality of protest in Montes de María, as it
required between fifty to a couple of hundred participants who would invade large estates, construct
hastily built settlements on the occupied territory, and then slowly begin to work the land themselves.
The police had little recourse as “[…]they lacked the sufficient force to intervene in all of the different
invasions; it was impossible to dislodge all of the peasants” (Pérez 2010, 81). Clearly, such a strategy
was temporarily successful insofar as the proliferation of land occupations in different locations
56 Interview 94, Sincelejo, 2016.
57 Interview 101, Sincelejo, 2018.
95
impeded a massive coordinated response by landed elites and the state. However, participation in the
occupations began to wane considerably in the late 1970s and early 1980s once local peasant leaders
and peasants themselves began to be targeted violently in retaliation. While this form of social
mobilization continued to be employed well into the 1980s, others were used less frequently by
ANUC, as their repertoire of protest was expanded to include road and highway closures and other
occupations such as the ‘Seizure of Sincelejo’ in 1985, however these proved to be unsuccessful in
achieving the organization’s ultimate aims (Escobar 1998). All in all, the golden age of peasant activism
during this period saw some 140 000 hectares of land redistributed to landless peasants in the larger
region, representing close to 25% of the available land in the entire region (Porras 2014).
The surge in land occupations elicited two reactions from large landholders in Colombia at
this juncture: either they felt that resistance was futile and sold parts of their estates to INCORA so
they could be parceled off to poorer peasants, or they deployed local police and their own nascent
paramilitary groups to evict peasant squatters from their properties. The latter option became
increasingly common over the course of the decade, as “the implementation of the counter agrarian
reform with the Pact of Chicoral permitted Liberal and Conservative elites alike to cast aside
INCORA’s {land} titles with proto-paramilitaries”.58 Local cattle ranchers and latifundistas who were
loth to cede any part of their estates to peasant activists, would often deploy a combination of local
judicial ploys and private violence to impede the success of the occupation (Reyes Posada 1978, 165-
166). While ANUC had to contend at the regional and national levels with other similar peasant
organizations founded by the Catholic Church (FANAL), or the Colombian Communist Party
(FENSUAGRO), in Montes de María “FANAL and FENSUAGRO were minimal expressions of
rural civil society in Sucre”.59 The level of peasant collective action in Montes de María was remarkably
58 Interview 93, Bogotá D.C., 2016.
59 Interview 94, Sincelejo, 2016.
96
high during this period, and the level of coordination between associations from the village to the
national level were extremely efficient. ANUC members would collectively finance their leaders’ trips
to other locations in the region, or the country, thus creating a shared responsibility and a communal
investment in the outcome. One human rights lawyer from Sincelejo who grew up inside of the
organization explains the importance of this:
There was a very sophisticated system, protected by ethical commitments which permitted these peasant user committees at the village, town, or municipal levels to survive…a very strong type of organizational structure was needed to be able to function in an effective manner from bottom to top and from top to bottom to penetrate those spaces and recuperate land. Women and children had their function inside of this process, as men were the principal target to be captured and jailed so women fulfilled the vital role in the recuperation of land because they were in charge of preparing food for the men and they also got involved in the process, but when the situation got complicated with the police or the army. Women and children were those who shielded peasant leaders who were carrying out the protest or land occupation.60
The split between the Sincelejo and Armenia factions in the mid-1970s ended up reducing
overall membership in ANUC from 800 000 to 310 000, or a 61% decline in participation (Bagley and
Botero Zea 1978). To make matters worse, the local leadership of ANUC who formed the Sincelejo
line splintered into three different factions at this precise juncture due to ideological squabbles over
the preferred repertoire of action and mobilization, thereby weakening the organization and further
eroding regional peasant support for the cause (Zamosc 1986). ANUC hitherto had to contend with
FANAL and FENSUAGRO, but in Montes de María and beyond ANUC had remained the most
popular and the most effective peasant organization in terms of scope and ability to achieve results
on the ground. However, with the growing repression of landed elites and state agents against peasant
activists, the organization’s regional and national level leadership began to question what was the best
course of action for its membership. One faction called for continued autonomy from the state and
for ANUC to remain a peasant organization configured along the lines of a civic organization, while
another sought to convert ANUC into an autonomous political actor which would participate
60 Interview 101, Sincelejo, 2018.
97
democratically in local, regional, and even perhaps national elections. The third course of action was
for ANUC to combine all forms of struggle, violent and non-violent, in its confrontation with the
large landholding class and the state.61 These strategies were irreconcilable, as the success of the early
1970s proved to be the high point for peasant activism in Montes de María.
During this period, there were few other civic organizations which maintained a meaningful
presence in Montes de María. Community action boards (JACs) were formed sporadically throughout
the region yet they fulfilled a distinct function from that of ANUC. In those villages where a JAC had
actually been formed, the local junta president frequently served as the leader of the local ANUC
committee as well.62 Some villages and towns were fortunate enough in the sense that their JAC
actually performed its function and was able to deliver basic goods and services to the community. In
Las Palmas (San Jacinto), the JAC is remembered as “a vital organization for the development of the
village”, particularly in regards to local roads, health, and education.63 Similarly, the JAC “was a
management body and interlocutor between the community and the government to obtain basic
services and protect fundamental rights” in El Salado, as the junta was responsible for the village’s first
electric plant, high school, and local aqueduct.64 It appears that the JACs were far more common and
effective in Ovejas in comparison with the rest of the region.65 For example, in Chengue the village’s
JAC went door to door to collect familial quotas for the construction of a local electric plant, while
the entity also collaborated with the JACs in neighboring town to maintain the local roads connecting
61 Interview 94, Sincelejo, 2016.
62 Interview 85, El Carmen de Bolívar, 2016; Interview 93, Bogotá D.C., 2016.
Cordillera.74 The first arrivals were mule drivers, as “[…]only men entered the zone and when they
were settled and had made the land livable, they sent for their families”.75 The task of settling the land
required an intensive process of deforestation which in turn spawned the lucrative yet illicit cross-
border timber trade with Venezuela.76 The actual process of travelling to the piedmont and settling
public lands differed, although the most common route was the path leading from Pamplona (Norte
de Santander), to Cubará (Boyacá), which finally descended into Saravena. Quite frequently, groups
of families from the same towns and villages in departments such as Santander and Norte de Santander
made this journey together, with many establishing small hamlets composed of peasants from the
same locations in the Andean highlands.77 Travelling migrants first stopped in Tunebia (Boyacá) in
order to present the necessary paperwork to the regional office of the Caja Agraria so they could
legalize their claims. One such woman from Cáchira (Norte de Santander) who made this trek with
her young children remembers her colonization experience vividly:
Once we arrived at La Primavera and saw the land my husband had bought, there was no house or farm so we had to stay with my brother-in-law for a bit until we could build a wood and earth shack. We began to work clearing the forest by hand with only the aid of axes and machetes and in those early days we helped my brother-in-law with his cows so we could at least give milk to our small children. Our neighbors also helped out with food when there wasn’t any, as everybody helped one another out like that. Little by little we improved our situation and finally built a better house with greater comforts.78
The first wave of colonization was directed by the Caja Agraria and INCORA under ‘Project
Arauca One’ (Proyecto Arauca Uno). This phase of semi-planned migration saw roughly some 5000
families arrive in the piedmont over the course of the 1960s, settling principally in Saravena and Fortul
(El Espectador 2014). The standard unit of land which the Caja Agraria and later INCORA would
parcel off was fifty hectares of land, the UAF, a plot measuring three hundred meters by one hundred
74 In 1970 Saravena legally became independent of Tame, and similarly Fortul became a municipality in 1990. According to a local historian in Tame, “[…]the center of colonization was centered in Saravena and little by little it expanded towards Fortul, Arauquita, and Tame.” Interview 71, Tame, 2016.
Source: Sistema de Información de Desarrollo Rural - SIDER
79 Interview 23, Fortul, 2016.
80 Interview 71, Tame, 2016.
81 Interview 23, Fortul, 2016.
82 This lack of clarity in land titling and formal ownership was more strongly felt in Arauquita and Tame than in Saravena and Fortul, reflective of the chronology of directed and spontaneous colonization in the sub-region.
108
Every municipality in the piedmont has a particular mix of migrant populations, yet broadly
speaking “the population of the piedmont is multicultural with people from Santander, Boyacá, Valle,
Cudinamarca, Antioquia, and Tolima”.83 Saravena alone was settled by colonos from thirteen different
departments.84 Arauquita possesses a greater influence of Venezuelans and Afro-Colombians from
departments such as Cauca and Antioquia, as they were brought to the region during the height of the
timber trade.85 Tame is the only municipality in the entire department which possesses mountain,
piedmont, and plains, and thus is settled in certain part by creole llaneros, while the piedmont portion
of this municipality was colonized by migrants from all over Colombia. Due to proximity, Tame has
been heavily influenced by migrants from neighboring Boyacá, a department which for much of the
20th century maintained the distinction as being the poorest in the entire country (Rausch 2015). A
majority of migrants in the piedmont can trace their origins back to Norte de Santander, Boyacá, and
Santander (Rucinque 1972).
For the first decades of the colonization period, there was surprisingly little contact between
communities in the piedmont and in the plains, save northern Tame and western Arauca municipality,
both of which served as places where the two sub-regions finally met.86 The sudden influx of poor
peasants from the Andean departments such as Boyacá and Santander into the previously uninhabited
jungles of the piedmont dramatically altered the future balance of power in the department of Arauca,
as by the late 1970s “[…]the balance of population in the department had reversed, with almost three-
fourths of its 78,000-odd residents living in the piedmont.” (Carroll 2011, 181). These changes
predictably brought to the fore some latent tensions between the two populations, but the peasant
83 Interview 50, Arauca municipality, 2016.
84 Interview 67, Saravena, 2016.
85 Interview 98, Fortul, 2018.
86 This was largely the result of the lack of roads connecting the two sub-regions in any shape or form. (Interview 71, Tame, 2016)
109
settlers, despite sharing the same nationality as the llaneros, were never assimilated or fully accepted by
their counterparts in the plains.87 Unsurprisingly, the main flashpoint between colonos and llaneros in
Arauca was over land. As the colonization movement expanded beyond the focal point of Saravena
well into Fortul, Arauquita, and Tame, many of these peasant migrants found that the public lands
they had hoped to settle were already claimed by other recent arrivals, and therefore they pushed into
existing haciendas and other idle properties, effectively occupying these large estates. This was
particularly true in the part of Tame where the plains and the piedmont meet, as in many cases the
migrants effectively invaded and divided entire landed estates amongst themselves.88
Whereas indigenous groups saw their traditional way of life completely disrupted by the
encroachment of European migrants during the colonial period and the destruction of the Wars of
Independence, over the course of the 20th century the most frequent flashpoints between settlers and
natives in Arauca were the theft of cattle and other agricultural crops, a practice which invited violent
reprisals from rural homesteaders as time wore on (Rausch 1993). The colonization of the piedmont
claimed the last remaining space in the department which indigenous tribes could claim for themselves,
and while nominally better than their peers from the plains, relations between the new piedmont
arrivals in the 1950s and 1960s were also prone to periodic outbreaks of violence due to ongoing cattle
rustling committed by groups such as the Guahibo. After one such incident in Arauquita, the local
police inspector “backed by many people from the town, sent his men to the Island of Reinera and
the area near Lipa, and there they burned down the native settlements” (Pérez Bareño 2015, 39).
Another local in Tame recalls when various llaneros from this municipality poisoned between ten and
fifteen natives in retaliation for their incursion on local cattle ranches,89 while another sixteen were
87 Interview 67, Saravena, 2016.
88 Interview 73, Tame, 2016.
89 Interview 71, Tame, 2016.
110
notoriously ambushed and killed at La Rubiera in Arauca municipality in 1968 (Pérez Bareño 2015,
40). By the 1960s and 1970s, most indigenous groups were completely marginalized by the
colonization of both the plains and the piedmont, concentrated in a handful of impoverished
communities throughout Arauca.
ii. The Economic Sphere
From the time that Spanish settlers and missionaries settled in the plains straddling what today
is the Orinoco watershed between Colombian and Venezuela until the present, cattle ranching has
been the staple of all local economic life in the plains. In other regions of Colombia such as the
Caribbean coast or the upper Magdalena valley, cattle ranching was selected and practiced extensively
by land holding rural elites because ‘‘[…]the monopoly of available land was the only way to control
labor” and this in turn generated political power and territorial control for these particular individuals
(Reyes Posada 1978, 4).
The early Spanish settlers and missionaries brought the agricultural traditions of Andalusia and
Extremadura to the Americas by introducing the cattle ranch (el hato) to the plains due to the fact that
the climate patterns and the quality of soil and grasses prevented any other viable agricultural activity
save for the raising of livestock and the occasional harvest of plaintains, yucca, and corn.90 Cattle
ranchers depended on accessibility to massive stretches of land, without fences or impediments, in
order to sustain and raise their herds, and thus grazed them freely in the thousands of hectares which
surrounded their hato up to an extended radius (Otto and Anderson 1986). Their cattle were branded
or marked in order to keep track of which belonged to what hato. However, the concept of private
property was complicated, as formal title did not exist and ownership was very loosely established on
understandings between neighboring cattle ranchers and the general consensus that cattle were free to
90 Interview 69, Tame, 2016.
111
roam as far as they pleased. A man’s social and economic status was entirely dependent of the amount
of cattle under his control, as “the elites were clearly defined and were composed of the hato owners,
large cattle ranchers, and land holders, who measure their wealth in cattle and in the amount of
hectares they possess”.91 In the absence of formal title: “A man claimed informally all the land
occupied by his cattle, and holdings of 2,500 square kilometers {250 000 hectares} were not
uncommon”. Only those ranches with more than 2,000 head of cattle were referred to as hatos and the
owners of these “[…]wielded most of the political and economic power” (Loy 1978, 507).
The introduction of the Spanish hato and the religious missions engendered an extremely
vertical social structure in the Eastern Plains of Colombia. While there was a very particular division
of labor established on the hato, the key roles were clearly defined by the owners themselves (ganaderos
or patrones), the administrators of their estates (mayordomos), and then the laborers (peones) who carried
out a variety of specialized functions on the hatos. On the ruling class and the expectations of their
subordinates, Rausch (2015) observes that: “The owners, at the top of this social pyramid, generally
lived in the towns. They formed a class apart, expecting complete submission from their workers in
this semifeudal arrangement” (6). Frequently, the peons and mounted llaneros “[…]lived outside the
the towns, had no access to schools or health facilities, realistic opportunities to gain title to land on
which they may have lived for many years” (28). Despite the dismantling of the hacienda system
established by the religious missions, private individuals re-asserted control over the trade in the 19th
century. However, what cattle ranching had recovered was severely limited by a restricted demand
from the Andean highlands, a trend influenced by the availability of better quality livestock from
nearby savannas in Cundinamarca and Boyacá, or even the Caribbean coast. There was also a
concerted effort by coastal ranchers and merchants to prevent their competitors in the plains from
91 Interview 48, Bogotá D.C., 2016.
112
expanding exports to markets within and outside of the country, a practice which had continued from
the colonial period (Loy 1981, 161).
The social and economic backwardness of cattle ranching in the Eastern Plains persisted in an
era when the trade experienced a pronounced boom in Colombia in the latter half of the 20 th century
(Van Ausdal 2009). Despite efforts by the authorities to legalize a system of land title and ownership,
local ranchers and their peons made little to no attempt to register their claims, nor were they
interested in the improved techniques adopted by their competitors elsewhere in the country,
innovations such as barbwire fences and improved African grasses for their cattle. With no intention
of applying for legal ownership, nor modernizing their estates and ranching practices, an anti-progress,
quasi-nomadic form of cattle ranching took hold in the region. The hatos which were not legally
constituted in the first place were frequently “[…]passed along from father to son without any
semblance of legality or documentation and anyone who sought to challenge this would either be
overwhelmed with lawsuits or armed confrontations” (Rausch 1993, 109). Isolated from internal
markets for their cattle, hato owners depended on both the legal and illegal trade of their livestock with
neighboring Venezuela via the Arauca and Orinoco rivers.
The practice of raising cattle in the plains remained largely consistent until certain events in the
latter half of the 20th century forced a belated, partial modernization of the industry. Following the
turbulence of la Violencia, particularly in places like Meta, Casanare, and Arauca, coupled with the
attempts of the early National Front governments to resolve the agrarian issue, the arrival of thousands
of landless migrants from the interior dramatically changed the regional dynamic. The sudden
introduction of thousands of peasants into a territory where an extreme concentration of land existed
in a climate where land title and ownership was extremely ambiguous predictably generated problems
for the hato owners and cattle ranchers. In a bid to quell the land invasions and lay claim to what they
believed to be theirs, local landed elites and ranchers belatedly erected barbwire fences to demarcate
113
their properties. Often these disputes between large cattle ranchers and peasant squatters would be
resolved by the local police inspector. Once the borders of their claim was somewhat established, they
would get approval from the police inspector before passing the documents over to the municipal
office of INCORA for legalization.92 They were able to do so despite the fact that “to erect fences
has always come with a high cost. Those who closed their properties first were precisely the rich or
those who brought a lot of cattle to take advantage of the pasture and grasses”.93 Perhaps nowhere
else was the threat of colonization felt more acutely than in Tame. Divided between plains and
piedmont, this particular municipality had always been characterized by large hatos.94 The mass arrival
of migrants in the 1960s and 1970s led to the fragmentation of numerous large estates by squatter
peasants. Others hastily erected fences in the hope of stemming future invasions (Pérez Bareño 2015,
61).
The effect of the introduction of barbwire fences on hatos in the plains compounded and
reinforced the extreme concentration of land, while also bringing to the fore the subtle enmity between
the owners and their peons, and between competing landowners themselves. The quasi-feudal social
relations that were characteristic of the plains from the colonial era well into the 20 th century have
persisted until the present. Notwithstanding improvements in the local quality of life and a modest
modernization of the form of cattle ranching practiced in the region, plains habitants continue
“[…]living as their ancestors lived” in highly vertical agrarian social structures with and large (and
mostly absentee) landholders on the top, and peons on the bottom (Loy 1981, 167).95
The state-driven colonization experiment in the Araucan piedmont lasted until the early 1970s
when national-level economic priorities shifted and support for aspiring homesteaders in colonization
92 Interview 100, Tame, 2016.
93 Interview 23, Fortul, 2016.
94 Interview 69, Tame, 2016.
95 Interview 74, Tame, 2016.
114
zones across the Colombian periphery began to dry up. However, this did not dissuade other migrants
from pouring into the sub-region to stake their own claim and settle small plots of land, with or
without formal title from the authorities. Over the course of these decades, a very unique model of
peasant society emerged in the piedmont of Arauca, one which was complemented by a distinct
economic modality geared towards a smallholding system of agricultural production.
The conditions of early piedmont society were harsh. One local leader from Gaitan (Arauquita)
recalls that even though the zone was calm and safe in his childhood, the “poverty was extreme”. To
compound this economic misery, the isolation of the piedmont was equally as severe as there were no
access roads, a lack of communication which “didn’t allow trade to exist” in the sub-region.96 Another
account from Saravena highlights this same lack of connection between villages within the sub-region
and beyond. Those who were sick were forced to travel downriver at considerable cost to Arauca City
for medical attention. What schools existed initially were in extremely precarious condition, with desks
made from bricks and crude wooden slabs. Piedmont residents depended on Coleman lamps for light,
although numerous peasants improvised and made their own appliances because electricity was not
made available until 1989, and even then it was only operational for two hours a day.97 These immense
hardships fostered a high level of self-reliance and collaboration amongst peasant settlers. Everybody
was expected to contribute in some shape or form, local exchanges were often based on a system of
barter and peasants “did not live under the logic of the market”.98 Their diet consisted of bush meat
(carne de monte), fish (which were caught in abundance in the extensive local fluvial system), and
whatever crops remained after harvest.99 There was a great deal of camaraderie and sharing between
96 Interview 62, Arauquita, 2016.
97 Interview 67, Saravena, 2016.
98 Interview 65, Saravena, 2016.
99 Interview 96, Bogotá D.C., 2016.
115
migrant peasants in these communities, yet the rough conditions predictably led to intense
confrontations when disputes over property boundaries arose.100
In the piedmont, a great familiarity existed between neighbors as “everybody watched each
others back”. Close interpersonal social relations served a strategic purpose in “generating protection
for those very same people in the sub-region”.101 The social fabric was extremely strong and was
described as “a deep rooted brotherhood” between local habitants where “everybody ate or nobody
did”.102 In contrast to the neighboring plains, the limits placed on peasants by this smallholding system
created both a closer intimacy with their neighbors, and less freedom at the same time.103 Neighbors
were expected to contribute to communal activites such as the maintenance and upkeep of common
spaces in their communities, while also helping to transport residents who were either ill or pregnant
to the nearest doctor or midwife.104 This high level of collective action was partly born out of necessity,
as the first authority to register a presence in the piedmont was local clergy, followed by the emergence
of local community leaders who disseminated lessons and little booklets to peasant settlers regarding
basic health advice such as malaria prevention, instructions on home construction, and even how to
effectively castrate livestock.105
The economic modality which emerged and developed in the sub-region was largely shaped
by the origins of the peasant settlers, the distribution of land as directed and overseen under INCORA,
and the quality of the soil. Unlike other sub-regions found in the Eastern Plains, the land tenure pattern
of the piedmont is one which is remarkably equitable, as quite simply “there are no elites, it is a
100 “There was always violence, even when the guerrilla groups weren’t present. Sometime people wouldn’t respect others
land. At that point, the community would solve these disputes amongst themselves, regardless of who was at fault.” Interview 63, Arauquita, 2016.
services, and education, as everybody’s children, peasants and workers alike, study in the same
schools”.112
iii. The Political Sphere
Notwithstanding the waves of Venezuelan and European migrants to the plains, the cultural
identity of the sub-region remained fairly homogenous and was strongly tied to its partisan and
political allegiance to the Liberal Party. With every national-level civil war, the loyalty of the plains
elites to the Liberal cause grew even stronger, as one prominent Liberal general of the 19 th century,
General Vargas Santos, resided in Tame and organized an army in Venezuela just prior to the War of
a Thousand Days in 1895 and ended up invading and seizing control of Arauca City during the final
years of the 19th century (Rausch 1993, 179). As the partisan-driven violence and repression increased,
so did the resolve and partisan commitment of the Liberal fighters, a phenomenon which would be
passed from one generation to the next in the region over the course of the 20th century.
The outbreak of the partisan civil war throughout Colombia in the late 1940s had a strong and
lasting impact on the Eastern Plains region. Arauca was no exception as Guadalupe Salcedo was
arguably the strongest Liberal rebel commander in the region, if not the entire country, over the course
of the conflict. Born in Los Chorros, Tame, to a Venezuelan father and indigenous mother, Salcedo
was raised in Arauquita. The assassination of Jorge Eliécer Gaitan in April, 1948, brought ominous
consequences to Liberal bastions like Arauca. The local response was swift and violent in places such
as Tame, as a local politician recalls:
Some Conservatives here who were happy organized a dance in Las Queseras….The Liberals were disgusted and organized a group of boys to pacify those people so that ‘they wouldn’t dance’, haha, and they killed them all, hehe, a goddaughter of mine was killed. They killed her so that she wouldn’t dance, hehe, and everything started from there, the dislike and the hate. There were about eight deaths. They killed them with knives and machetes. (Pérez Bareño 2015, 44)
112 Interview 65, Saravena, 2016.
118
The wave of Conservative oppression followed. In the Eastern plains, those publicly identified
as Liberal supporters lost their jobs and were intimidated, while peasants were stripped of their lands
and often killed.113 Large landholders from Boyacá “[…]took advantage of situation to recoup lands
they had lost after López Pumarejo’s semi-land reform” (Rausch 2015, 34). Between the Colombian
military and the Conservative police, they targeted and killed everything that they associated with the
Liberal party in Tame (Pérez Bareño 2015).
Unwilling to idly sit back and accept the government’s numerous abuses, a combination of
local farmhands, cattle rustlers, and erstwhile victims of Conservative repression found common cause
against the Colombian government and began to organize and fight back. The first groups arose in
Tame, numbering around six hundred men armed with only sticks and clubs. However, these fighters
had the advantage of knowing the terrain very well. They were all very capable of swimming and
crossing the various local rivers in contrast to the army’s soldiers and the police. Furthermore, these
rebels hailed from local communities and “[…]the people loved the guerrillas more than the Army
because they were Liberals” (Pérez Bareño 2015, 47). Once organized, these rebel groups consolidated
their control of Liberal zones of influence and would launch frequent attacks on Conservative targets
wherever they were to be found.
These groups benefitted from the partisan cleansing of other regions, as numerous Liberal
refugees from Conservative strongholds such as Boyacá sought protection in numbers and ended
joining rebel groups in the Eastern Plains. Unlike other Liberal and Conservative insurgent factions
found elsewhere in Colombia during this chaotic time, the rebel army led by Guadalupe Salcedo was
quite disciplined and refrained from arbitrary violence against civilians, a tendency which reflected the
insurgent commander’s personal dislike for unjustified coercion. This in turn guaranteed the support
113 One such way of identifying Liberal supporters was through their partisan identity cards, a tactic which persists to this day in order to ostracize those who were born in a certain place, and who are stigmatized as such along these lines.
119
of the local civilian population in Arauca. As Jane Rausch (2015) writes: “Key to rebel success was
their ability to mobilize the civilian population in favor of their movement. By consolidating their
authority and power structures, they increased control” (43). A local military commander conceded
that the guerrillas had both the sympathy and complicity of the majority of local residents, who in
exchange for their support received security and order as Salcedo’s soldiers imposed their own legal
codes, courts, and even a system of taxation. Over the course of la Violencia, the Liberal guerrillas in
the Eastern Plains “[…]created enclaves that were effective centers of permanent refuge and often
insurmountable barriers to the imposition of government authority” (Rausch 2015, 49).
Following the sudden coup launched by General Rojas Pinilla in 1953, the new military
president was able to convince a Liberal general who had family in Arauca to go to the region and
persuade Guadelupe Salcedo and his followers to lay down their arms in 1953.114 Recognizing the need
for more inclusive policy proposals in the Eastern Plains, Rojas Pinilla sought to calm the region in
aftermath of the amnesty by offering free land to settlers and improved access to credit. The seeds of
the eventual colonization of the piedmont can be traced back to this period, as the Rojas Pinilla
administration “wanted to give work to all of the guerrillas, while also expanding agricultural frontier,
a process which evolved over the period from 1955 until 1980”.115 While helping to bring down the
Conservative government and serving as a key impetus for the agrarian policies proposed and
established in the 1950s and 1960s, the Liberal guerrillas ultimately failed to achieve any long term
reform in the region, as “[…]the true winners were the large landowners, who with the support of the
military would retain their control over the region” (Rausch 2015, 53).
114 By this point in time, the conflict had reached something of an impasse because “[…]the government couldn’t
dominate the guerrillas and that is natural; you know how guerrillas are difficult to defeat because they operate in their own environment.” (Pérez Bareño 2015, 38)
115 Interview 71, Tame, 2016.
120
For the majority of Arauca’s post-independence history, the territory was classified at various
stages as either a comisaria (commisary) or an intendancia (intendancy) before finally becoming a
department in 1991. Following la Violencia, local governance in the region was practically non-existent,
as the entire intendancy of Arauca was officially administered from the city of Sogamoso in eastern
Boyacá.116 While intendencies were officially administered by executively appointed intendants, the
Araucan political elite was thoroughly dominated by a group of plains cattle ranching elites who were
fervent supporters of the Liberal party. Between this ruling group, there were two dominant factions
vying for control of the intendancy’s budgets: the Villabonistas and the Latorristas.117. These elites
exercised power through clientelist linkages with core constituencies primarily found in Arauca
municipality and Tame.
The faction led by Alfonso Latorre Gómez was more dominant from 1966 onwards. Latorre
was the intendancy’s representative in the mid-1960s before subsequently serving as the territory’s
senator from 1966 until 1990, although he was the veritable powerbroker of Arauca prior to the arrival
of the insurgent groups and the political opening of the mid-1980s. Latorre was responsible for
approving all appointed intendants - or those charged with administering the intendancy - and thus
controlled all public jobs and budgets in the department as a result (Carroll 2011). The Araucan
powerbroker maintained tight control over the region’s politics at the expense of the piedmont, a
distribution of power which “was problematic and less inclusive” due to the direction of public
budgets towards sustaining Latorre’s clientelist networks rather than investing in the development of
communities in both the plains and the piedmont.118 Formal political institutions in Arauca remained
dominated by plains landowning elites until the political decentralization of the mid 1980s, leaving the
piedmont ignored and marginalized (Carroll 2011).
iv. The Civic Sphere
The national and regional effects of the National Front reforms in the late 1950s and 1960s
did not impact the Araucan plains in the same manner as they were felt in other regions of Colombia.
Until the discovery of oil in the early 1980s - and arguably until the present - the Araucan plains’
traditional social order continued to be defined by cattle ranching and a vertical social hierarchy.
ANUC did not emerge as a central motor of local peasant life, perhaps because of the general lack of
non-cattle related agricultural activity in the sub-region. Community action boards were established in
every village and neighborhood in the plains, as mandated by law, yet these JACs were not a crucial
component to local development or administration in these communities. One native of Puerto
Rondón comments that in his municipality “the creation of the JACs created a mechanism for co-
existence and respect, however, relations are more focused towards official state entities.”119 The
policies of the National Front made little contribution to the Araucan plains in terms of social
development and the traditional llanero social order persisted throughout the 20th century.120 Peasant
collective action was not a necessity in the plains for several reasons. Historically, relations between
peasants who resided in the same villages or worked on the same hato were characterized by a
reciprocity which manifested itself in various ways. Locals frequently engaged in a system of barter
(trueque), or would simply mutually share what scarce crops they produced on their land, as these
practices were sufficient to sustain these communities and hence local social organizations were not
necessary.121
119 Interview 53, Arauca municipality, 2016.
120 Interview 57, Arauca municipality, 2016.
121 Interview 59, Arauquita, 2016.
122
This is not to say that peasant collective action (in the form of social organizations and mass
mobilizations) was not evident in the plains during this period of time. However, due to the fact that
Arauca municipality was the political capital and administrative center of the department, locals, in
traditional clientelist fashion, would frequently look to official government entities for handouts,
employment opportunities, and for the local government to resolve their problems, whatever they
may have been.122 As previously mentioned, Arauca municipality and the other plains municipalities
(Tame to much lesser degree) were historically more integrated into neighboring Venezuela by land
and river connections than they were with the Colombian interior, and therefore local peasants “had
access to more goods and they had less difficulty getting by” as the majority of their needs were met
by the more highly developed neighboring country.123 Similarly, the dearth of formal and informal
organization in the plains can be attributed to the extreme under-population of rural areas and the
sparse distribution of people, a dynamic which owes much to the historic influence of cattle ranching.
Participation in social organizations is weak “due to the large estates and distances between people
which mean that they aren’t in contact frequently, there is little organization in the countryside because
everything is concentrated in the towns”.124 A veteran union organizer from the piedmont who works
throughout the department highlights all of the above reasons in explaining the historic and present
lack of collective action found in the plains:
In the plains, the population is less active because there is not much criteria for unity. It {the plains} is where all the political, administrative, and financial power lies, which creates a scavenger culture: “What can I get from the government?” These interests divide the population. Another factor to take into account is that the plains are underpopulated which generates less contact between neighbors. From one farm to the next can take a day travelling on horseback in terms of distance.125
122 Interview 95, Bogotá D.C., 2016.
123 Interview 65, Saravena, 2016.
124 Interview 67, Saravena, 2016.
125 Interview 50, Arauca municipality, 2016.
123
In stark contrast, the policies of the National Front governments were strongly felt throughout
the piedmont during the colonization period. INCORA took over where the Caja Agraria left off in
1962 as the main bureaucratic body responsible for overseeing the directed colonization of the
Araucan piedmont, approving and regulating the formal titling of small, evenly distributed plots of
land throughout these regions. During this decade “the entire social fabric was given life by the state”
in colonization zones such as the piedmont.126 After settling down in the zone “the migrants started
to organize, beginning with the Agrarian Social Organization (Organización Social Agraria)”, a peasant
movement which principally sought “state presence in this abandoned region with the goal of
guaranteeing their fundamental rights”.127 The history of INCORA in the piedmont is credited by
many longstanding peasant migrants in the sub-region as the driving force behind the directed
colonization and establishing a model peasant economy by preventing the emergence of any large
landholders.128 As such, it is remembered “with great affection and love by rural habitants” in the four
piedmont municipalities for the various functions it carried out in the early development and
settlement of these communities.
Apart from the adjudication and titling of land claims, INCORA also mapped out and cleared
the rough access roads which opened up unsettled lands to new arrivals from the interior, while also
providing credits to purchase materials to these homesteaders to clear their claims and erect
rudimentary domiciles.129 Under INCORA’s guidance, all newly arriving peasants to the piedmont
were enrolled in the sub-region’s biggest agricultural cooperative, the Agricultural Cooperative of
Sarare (Cooperativa Agropecuaria del Sarare - COAGROSARARE), while the institute also actively
126 Interview 95, Bogotá D.C., 2016.
127 Interview 48, Bogotá D.C., 2016.
128 Interview 49, Arauca municipality, 2016.
129 Interview 62, Arauquita, 2016.
124
fostered participation in ANUC when it was established in the piedmont later on that decade.130
Various other projects such as the development of the famous local aqueduct system in Saravena
occurred under INCORA’s watch, however the state bureaucracy’s influence began to wane in the
early 1970s before becoming virtually defunct in the sub-region. According to a Fortul-based activist,
“corruption and political opportunism killed INCORA”, as others recalled that administrators from
outside the sub-region began to embezzle funds or use the organization for their own ends. The end
result of this corruption was the lack of funds to provide infrastructure connecting the sub-region to
the interior, and thus leading to its decline in Arauca.131
In the piedmont, ANUC was highly active and one of the most integral social organizations
in regards to local peasant participation, although “it was strongest in Saravena”.132 The municipal
chapter of ANUC in this municipality formed the frontline of the opposition to the increased
inefficiency of INCORA in the piedmont, and in response dramatically increased its confrontation
the Colombian state through mass mobilizations beginning with the first civic strike launched in
Saravena in 1972. Many of the ANUC leaders who emerged in the piedmont at this juncture in time
had previous experience organizing in the Santanders, or possessed some form of formal education,
and therefore were quite effective in their newfound capacity as community leaders in a colonization
zone.133 Migrants from both Santander and Norte de Santander had a massive influence on the
colonization process in the piedmont due to the fact the region was destroyed following more than a
century of constant civil wars, leaving a highly unequal land tenure pattern circumscribed by an
unproductive agricultural economy (Gutiérrez de Pineda 1968). The region’s historic resistance to
130 Interview 95, Bogotá D.C., 2016.
131 Interview 23, Fortul, 2016.
132 It is important to take into account the fact that “ANUC was an association for articulating claims, one which made
demands but did not manage resources.” Interview 74, Tame, 2016.
133 One local public servant in Tame cites the origins of peasants in the piedmont as the reason why locals participated so actively in social organizations such as ANUC and the JACs: “They were settlers from much more structured cities like Bucaramanga which had a greater degree of organization.” Interview 69, Tame, 2016.
125
abusive forms of authority was complemented by a propensity for social organization, as many of the
first trade unions and popular social movements originated in the region in the first decades of the
20th century (Van Isschot 2015; Gill 2016). Confirming the regional legacy of this migration to Arauca,
a local religious figure observes that the piedmont “has been enormously influenced by the Santanders,
its population is characterized by being very active and quite fond of organizing”.134
Complementing the increased activity of ANUC over the course of the 1970s was the
emergence and consolidation of the community action boards as the most important local
administrative and deliberative civic organization in the piedmont.135 Throughout every community in
the sub-region, the JACs “have been the base for development and the mechanism to achieve results
from the state”, serving each village and town as a kind of “mini-state or government”.136 Although
the law that founded the JACs dates back to the late 1950s during the earliest years of the National
Front governments, these local councils only began to appear in the piedmont in the mid-to-late 1970s
at the encouragement of progressive local Catholic clergy.137 These councils are credited with having
fostered the development of rural communities throughout the piedmont, from the construction and
outfitting of local schools to establishing and improving hand-built roads and bridges, one of the key
challenges inhibiting basic communication between villages in the sub-region.138 Due to the quasi-legal
character of the JACs and the dense webs of these organizations at the village, municipal, and
departmental levels, the JACs historically served as a means to further expand and fortify peasant
134 Interview 49, Arauca municipality, 2016.
135 Perhaps the best description of the JACs come from a local teacher and union organizer from Tame. These organizations “represent the meeting point between civil society and institutionality. They have never administered state resources, rather they are tasked with managing and conceiving development projects for the development of these communities.” Interview 74, Tame, 2016.
138 One former JAC president in rural Arauquita recalls neighboring villages joining together led by their respective juntas in order “to repair the roads and bridges which connected them”. Interview 63, Arauquita, 2016.
126
social networks in the piedmont by “generating conscientious work throughout the community”.139
The juntas created what one prominent local activist terms “the new village” (la nueva vereda), a place
which “has schools for children and for communal meetings”, as the JAC is a body “which dialogues
with all of the state institutions which can provide benefits for these village residents”.140
Apart from ANUC and the JACs, a variety of different kinds of social organizations emerged
over the 1960s and 1970s in the piedmont. With the exception of COAGROSARARE which was
founded in 1963, a slew of other agricultural cooperatives, trade guilds (gremios), unions, and other
voluntary associations were formed during this period of time. These differed from the above insofar
as they were not created by the impetus of the state or local authorities, rather they were the product
of efforts at the local level by peasant organizers. This grassroots organizational boom saw common
peasant guilds formed for practically every distinct agricultural activity practiced in the piedmont
including cattle ranching, yucca, cacao, coffee, and plantains, while also others were constituted to
represent specific common vocations such as fishermen, transporters, teachers and students.141 In the
piedmont, the dense network of organizational endowments was characterized by a high level of
interconnectedness between them. For example, the same peasant leaders who served in ANUC would
also double as the president of their local JACs, while their constituents in these rural communities all
belonged to agricultural cooperatives and guilds (Gutiérrez n.d., 36). With the rapid decline of
INCORA and the sudden rise of peasant activism in the early 1970s, relations between the piedmont
population and the Colombian state worsened dramatically in a very short period of time, eventually
progressing into an armed confrontation as the decade wore on.142
141 In a trend that remains true to this day “participation in the guilds was extremely high, everybody belonged to at least
one.” Interview 67, Saravena, 2016.
142 Interview 95, Bogotá D.C., 2016.
127
There is strong evidence to suggest that the piedmont capacity for peasant collective action
was born of necessity due to the extreme inadequacy local settlers felt in term of official government
presence. A second-generation homesteader from Botalón (Tame) remarks that their very survival
depended on organizing: “In front of such a difficult situation, it was necessary to create cooperatives
to survive and colonize the zone. Various peasants would get together and work one person’s land,
later somebody else’s, and they rotated like this”.143 A community leader from Arauquita remembers
a similar level of selflessness and collaboration amongst local settlers in his community: “Before the
arrival of the oil industry, if a local community lacked a bridge or anything like that, the same people
would resolve it. If it was a bridge then they would build something provisional using wooden slabs
and there would be a beautiful competition between groups of local residents to see who could
contribute more”.144
One particular way in which peasant collective action manifested itself in public and
demonstrable manner was the emergence of mass mobilizations in the 1970s. Towards the beginning
of this decade, the radicalization of many peasant leaders coincided with the sudden rupture between
piedmont society and INCORA, and by extension the Colombian state.145 This confrontation
culminated in the first civic strike of its kind in the Eastern Plains, and arguably all of Colombia, during
this period of time. In March 1972, numerous peasant leaders throughout Saravena convened a
majority of the municipality’s peasant settlers in the municipal capital for thirteen days to protest the
lack of infrastructure and other forms of assistance to migrant communities. An estimated three
thousand peasants occupied the main square of Saravena with the support of peasant leaders and local
143 Interview 76, Bogotá D.C., 2016.
144 Interview 59, Arauquita, 2016.
145 Whereas INCORA was initially successful in breaking up and assigning titles for smallholders in the piedmont, it failed when it came to providing the other necessities of a fledgling society: “Later, in order to build roads the only way to make the state realize our necessities was to mobilize strikes and marches. Later on, these peasant mobilizations were organized in order to bring electricity to the department.” Interview 53, Arauca municipality, 2016.
128
clergy alike, motivated by the fact that the population was completely isolated from the interior and
from one another, with a lack of access roads connecting even the municipal capitals of Saravena,
Arauquita, and Arauca municipality (Zamosc 1986, 93).146 This strike “represented the profound
tensions existing between communities organized by ANUC and the JACs” and the state, a situation
which was increasingly characterized by “enormous mutual mistrust”.147 The national government
eventually sent emissaries to restore calm to the piedmont by signing a pact with local notables in
order to meet a series of demands, which were quickly forgotten altogether by the official authorities.148
Following the first civic strike in 1972, several others followed over the course of the decade.
In a pattern which would become common in the years to come, the failure of the Colombian state
to properly address the peasant grievances expressed during the 1972 civic strike led to a fresh strike
three years later. Centered again in Saravena, the same core of peasant leaders this time convened
some seven thousand peasants to paralyze the town and the block the rudimentary unpaved road
leading to nearby Cubará for twenty-five days. That same year, Tame experienced its first civic
mobilization of this kind. Despite being llanero in character, the capital of this municipality was
sufficiently influenced by the colonization of the piedmont that in March 1975 virtually the entire
town mobilized to occupy Tame’s airport to protest the lack of teachers in the local high school.149
The shared sacrifice and risks that the civic strikes required of peasant settlers in the sub-region had a
fortifying effect on these individuals and their broader communities. Not only did these mobilizations
serve as a forum or means to make their demands, it also “created a political culture”, in which “the
148 One key demand was the construction of a bridge over the Bojabá River near the border of Saravena and Cubará in
neighboring Boyacá department. It was only constructed ten years later after another massive civic strike in the piedmont. (Interview 66, Saravena, 2016)
149 Interview 72, Tame, 2016.
129
state was seen as an enemy because it repressed the peasants and didn’t take their interests into
account.”150
The historic abandonment of the department and the residual tensions that this extreme
isolation caused generated far greater problems in the piedmont than it did in the plains. Until the
belated construction of roads and other vital infrastructure in the department in the 1980s, Arauca
remained as disconnected as it had been one hundred years earlier (Rausch 1993, 124). One local
community leader in Arauquita remembers that in that period “there was no state presence
whatsoever” in the piedmont.151 Another peasant activist from the northern Tame comments on the
struggles her family endured in her infancy: “There was no power, roads, not even schools. The great
struggle that my parents had was to obtain a school”.152 The municipal capitals and towns were little
better. Tame was afflicted by a “total abandonment by the state”, an isolation which meant that the
streets were unpaved, there were no roads connecting the town to any other settlement, as “everything
was transported in and out by planes which had survived the Second World War”.153
Before the belated modernization made possible by the advent of oil rents in the mid-1980s,
the department was extremely underdeveloped. According to one regional historian, the education
system was abysmal, with one-fourth of eligible students attending high school, forty percent were
fortunate enough to attend elementary school of any kind, while neither pre-school nor post-
secondary education existed in the broader department. To top it all off, there was “no electricity,
running water, or sewage” in Arauca.154 A rural homesteader in Fortul recalls having to give birth by
herself in extremely precarious conditions: “I went into labor late at night and my husband was away
150 Interview 95, Bogotá D.C., 2016.
151 Interview 63, Arauquita, 2016.
152 Interview 75, Bogotá D.C., 2016.
153 Interview 72, Tame, 2016.
154 Interview 71, Tame, 2016.
130
working in Venezuela. I had only my children with me at our farm and it was too risky to send my
eldest to fetch the nearest neighbors in the dark because I was afraid some animal would eat her, so I
ended up giving birth alone on the dirt floor and severing the umbilical cord myself”.155
The perceived abandonment by locals was not necessarily reflected in the actual presence the
Colombian state did maintain in Arauca. In many piedmont communities prior to the arrival of
insurgent groups, there was a basic presence of state sanctioned authorities. One lifelong resident of
Arauquita remembers that “before the arrival of the subversive groups, there was a court, a police
inspector, and a mayor, all of whom mediated local problems…more than anything these related to
property boundaries”.156 While the Colombian military did maintain small detachments in the
department to maintain the appearance that the national government was protecting the sovereignty
of the Colombian side of the border in the Eastern Plains, the reality is that the main unit of the armed
forces responsible for maintaining security and order in the region was a cavalry group, the “Guides
of Casanare” which were stationed in Yopal (capital of Casanare), a few days by horseback to the
piedmont.157
Those areas of the department which suffered from a comparable isolation due to the
‘absence’ of the Colombian state, yet which had access to greater luxuries were generally found next
to the Venezuelan border. Towns such as Arauca City and Arauquita were just as disconnected from
the interior of their own country as rural areas of the plains and the piedmont. However they
“benefitted greatly because they shared a border, they benefitted from Venezuelan oil, and they could
trade with them”.158 The existence of historic border trade and commerce between the two countries
ensured that “products of the first necessity” were made available courtesy of Venezuelan merchants,
155 Interview 96, Bogotá D.C., 2016.
156 Interview 59, Arauquita, 2016.
157 Interview 47, Bogotá D.C., 2016.
158 Interview 59, Arauquita, 2016.
131
the local antenna broadcast was similarly Venezuelan, as the better quality of Venezuelan highways in
the states of Apure and Táchira meant that Colombian could travel from Arauca to Cúcuta (Norte de
Santander) in a fraction of the time it would take on extremely precarious access roads in Colombia.159
As Arauca lurched into the final years of its pre-modern period it was characterized by a
strange, contradictory dynamic in which local peasants sought greater state presence and all of the
perceived benefits that came along with this, while concurrently harboring “a total mistrust” towards
the very same state (Pérez Bareño 2015, 59). The establishment of social organizations such as the
JACs in every village and neighborhood of Arauca endowed the organizational infrastructure to 734
communities in the department.160 This hypothetically gave each community in Arauca a mechanism
with which it could dialogue with official state entities and interlocutors to promote its own claims
and development. These efforts coupled with the surge in social mobilizations over this period helped
give a greater organizational dynamic to the department. However, as the former political boss Octavio
Sarmiento laments, it was a case of too little too late:
A pure sentiment has prevailed that the state and its institutions are worthless, citizens keep turning their backs to the criteria of the state and its institutions, few care if the state addresses {their problems} or not, the population feels marginal to the decisions of the country’s center. The insularity has discriminated against us because it has served as a shackle which prevents us from walking with the necessary pace towards progress. (Pérez Bareño 2015, 64)
C. Primary Cleavages in Montes de María and Arauca
The primary cleavages which emerged in Montes de María and Arauca in the latter half of the
20th century were distinct to these agrarian social structures found in these regions, yet they were both
driven by the desire of rural actors to construct and protect the optimal social order for their particular
constituency, whether landed elites or peasants. The Colombian state’s presence was conspicuously
insufficient, if not completely absent, in both of these regions. What formal political institutions did
exist were co-opted by regional landed elites, and these political bosses paid little if any attention to
growing peasant demands in Montes de María or Arauca. The regional economies were dependent
entirely on agriculture, and peasants were effectively denied an opportunity at social advancement due
to excessive land concentration (in the case of Montes de María), and the lack of basic infrastructure
(in the case of the piedmont). In order to address these cleavages, peasant leaders in both regions used
civic organizations created by National Front governments to organize and mobilize their
constituencies to engage in diverging forms of collective action to achieve their goals vis-à-vis their
opponents.
Settled in the colonial era, the rural social dynamic in Montes de María became highly stratified
over the years due to implementation of the hacienda system and the eventual rise of tobacco
production following independence. The unequal distribution of land emerged as the primary cleavage
in the region over the course of the 20th century, with the entrenched vertical agrarian social structure
finding itself challenged by the top-down reforms of the National Front governments in the 1960s,
policies which fuelled an escalating cycle of contention between regional landed elites and land-poor
peasants during the 1970s. Zamosc (1986) articulates the zero-sum nature of this social conflict which
intensified in regions characterized by the latifundia system:
[T]he massive eviction of peasants and the acceleration of landlord enclosures wherever the definition of property was unclear would become clear expressions of the primacy of the conflict over land in the areas of latifundia. The landless emerged, therefore, as the main sector of the peasantry, and the possibility of an independent peasant economy clearly depended upon the result of their impending struggle to break the power of the landowners, destroy large-scale property, and establish themselves as free peasants. (29)
Similar to Montes de María, the Araucan plains were the product of hundreds of years of
colonial and post-colonial settlement, a process which witnessed the slow emergence of a vertical
llanero social order in Arauca municipality, Cravo Norte, Puerto Rondón, and southeastern Tame.
However, there has historically been little political or social conflict in the Araucan plains outside of
the bipartisan bloodletting manifested during la Violencia. In the absence of any such class conflicts,
133
the centuries old enmity between creole descendants and indigenous groups emerged as the only
visible cleavage during the same period, albeit one that had dwindled in importance towards the end
of the 20th century. The lack of social conflict or a primary cleavage in the Araucan plains owes much
to the dependence of local habitants on large landholders and political elites and their patronage
networks which sufficed to meet peasant needs and demands. As Jessica Price (2019) points out:
“Where clientelist networks predominate, people are unlikely to develop independent protest capacity
and clientelist bosses are unlikely to lead protests” (408).
In contrast to Montes de María and the Araucan plains, the Araucan piedmont was carved out
of the inhospitable jungle by impoverished peasants from the interior of the country who had little to
lose following the turbulence of la Violencia. These migrants established a diverse society in the pocket
of territory stretching from the foothills of the Andes eastward through northern Tame, Fortul,
Saravena, and Arauquita. As the sub-region became increasingly settled over the course of the 1960s
and 1970s, robust horizontal peasant social networks mobilized for the greater state provision of
public goods and services, which the successive governments responded to with violent repression
and false promises. The inadequate state presence found in the piedmont during the colonization
period became the driving factor behind the various civic strikes which occurred prior to the arrival
of the insurgent groups. The fact that the land question had effectively been resolved in the piedmont
meant that peasants were wholly focused on other grievances related to the lack of public goods and
infrastructure, a reality which left these migrant communities completely isolated from the interior of
the country. According to Zamosc (1986), this cleavage was characteristic of colonization zones such
as the piedmont:
In this sense, the most important demands were those connected with roads, communications, and all the infrastructure that was needed to bring the products to the market, the prices of the pioneering crops, the basic credits, the procedures of land titling, and the introduction of services to support the peasant economy and to make the extremely difficult conditions of life more bearable. (28)
134
Chapter Four. Insurgent Institutionalization
“People who live at subsistence level want first things to be put first. They are not particularly interested in freedom of religion, freedom of the press, free enterprise as we understand it, or the secret ballot. Their needs are more basic: land,
tools, fertilizers, something better than rags for their children, houses to replace their shacks, freedom from police oppression, medical attention, primary schools. Those who have known only poverty have begun to wonder why they should continue to wait passively for improvements. They see – and not always through Red-tinted glasses – examples of peoples who have changed the structure of their societies, and they ask, ‘What have we to lose?’ When a great many
people begin to ask themselves this question, a revolutionary guerrilla situation is incipient.”
- Mao Zedong, On Guerrilla Warfare (2005, 6)
Whereas the previous chapter outlines the pre-existing agrarian social structures which
emerged over time in Montes de María and Arauca, this chapter unpacks the process of insurgent
institutionalization in order to better identify and analyze the constellations of local anchors or barriers
that enabled or constrained insurgent strategies to consolidate control of territory, people, economic
resources, political institutions, and civil society in these disparate zones. While local anchors and
barriers are of enormous importance for armed non-state actors who are seeking to appropriate the
primary cleavage, these do not solely determine the degree to which an armed non-state actor is able
to embed itself in a given space. The role of agency is also of enormous importance, particularly after
critical junctures, as both insurgent groups and civilian communities need to decide upon a fixed
strategy of how they are going to manage their interactions with one another. As Ana Arjona (2017)
points out:
Armed groups combine different strategies to penetrate local communities, take over control over their territory, and establish different forms of rule. The use of violence, political mobilization, selective incentives, and governance – by which I mean the creation of new institutions and the provision of public goods – is essential in these quests. (764)
At such a crucial moment, insurgent leaders need to pay serious consideration to the primary
cleavage affecting these peasant societies, while also taking into account the pre-existing collective
action mechanisms used to address their grievances and the extent to which they can impose
themselves on such communities before experiencing collective resistance. If armed groups are able
135
to appropriate the primary cleavage - or assume peasant grievances and mobilize effectively in support
of them - this allows them to better embed themselves in the human terrain and to establish a
comprehensive system of nested governance. The co-optation of key functions in the territorial,
economic, political, and civic spheres synchronizes and binds together the interests of armed non-
state actors with those of the civilian population, a process which legitimates the insurgents’ use of
selective violence against recalcitrant local figureheads of authority and civilians alike. Podder (2013)
argues that: “Unlike groups that are predatory, abusive, and disruptive for peace, certain {armed non-
state actors} offer alternatives to a weak and inefficient government as the legitimate representative
of minority grievances” (19). Upon doing so, armed groups become interlocutors between local
populations and the state, which during periods of heightened conflict and extreme polarization can
serve as a “pilgrimage” between individuals sharing a common grievance in the same geographic space,
a binding process that further commits insurgents and civilians alike to the same emergent non-state
social order (Anderson 2006; Rubin 2019; Levitsky and Way 2012; Ling 2006). These networks serve
as the basis for successful insurgent mobilization and consolidation (Viterna 2006; Parkinson 2013;
Staniland 2014). As Oberschall (1973) notes: “Rapid mobilization does not occur through recruitment
of large numbers of isolated and solitary individuals. It occurs as a result of recruiting blocs of people
who are already highly organized and participants” (125).
Some armed actors face considerable difficulties in appropriating local grievances and
embedding themselves in civilian populations, thereby lacking the necessary organizational structures
to regulate these populations. As Eric Mosinger (2018) notes, “[…]groups that fail to establish
stationary institutions turn to predation instead” (65). Such groups typically resort to provisional
predation when their attempts to insert themselves into peasant communities are met with barriers in
the form of communal resistance (e.g. a lack of local buy-in to their project), or when local conditions
provide little to no anchors for such groups to embed themselves with. In the case of the former,
136
communities are often resistant to non-state rule when these groups maintain clearly different short
and long-term goals than local civilian communities, and when their imposition within pre-existing
organizations and institutions negatively affects their efficacy to mediate local collective action
problems and grievances (Arjona 2016b; Kaplan 2017). However, the pre-existing agrarian social
structures in many underdeveloped and atomized rural communities simply do not provide any
substantial anchors or barriers to armed non-stop actors (Pearlman 2020). Mkandawire (2002)
highlights this dynamic:
[T]he grievances that rebel movements claim to seek to address are often not salient in local political situations. There is no landlord from whom to free the masses or upon whose surpluses guerrillas can survive. Guerrillas cannot offer an immediate end to predation by local potentates, since such predation hardly exists; nor can they liberate peasants from the heavy exaction of national government. (199)
This chapter examines the second key phase - insurgent institutionalization - of the broader
sequence in question, focusing on the arrival, integration, and consolidation of leftist insurgent groups
in Montes de María and Arauca. Substantially shorter than the previous period examined in Chapter
Three, this timeframe nevertheless is crucial to understanding the context of the larger research puzzle
at the heart of this dissertation. The breakdown and inability of the National Front governments to
rectify the longstanding structural dynamics which had fostered chronic political infighting and rural
upheaval throughout Colombia since independence finally came to a head with the emergence and
consolidation of leftist insurgent movements founded in the 1960s.
Whereas previous civil wars in Colombian history were primarily driven by the extreme
competition between the Liberal and Conservative parties, a zero-sum political dynamic which gave
little other recourse to political impasse except for armed violence, the emergence of the current civil
war differed insofar as it was born not from partisan loyalties, but from the longstanding unresolved
struggle for access to and control of land throughout Colombia (Safford and Palacios 2002; Palacios
2005). During this period, various distinct actors have attempted to assert control of land and people
137
in order to promote, establish, and protect a particular agrarian social structure, often through the co-
optation of the available organizational endowments in these particular spaces. Agrarian conflicts in
Colombia vary in character but are invariably tied to the land question, representing struggles between
armed actors who are appropriating local cleavages as their own on behalf on a specific constituency
(or constituencies) in an attempt to implement or protect their optimal agrarian model. According to
Dario Fajardo (2015), these agrarian conflicts:
[…]have been phenomena associated with the frequently violent usurpations of peasant and indigenous land and territory, misappropriations of the country’s public lands, private impositions of rent and other charges for access to this land, in no small occasion with the support of state agents, as well as land occupations on behalf of peasants without land or with minimal access to it, of properties established in an irregular manner. (3)
The principal insurgent groups in Colombia only truly began to develop a military capacity
capable of confronting the Colombian state over the course of the 1980s and 1990s, a period of time
which coincided with the meteoric rise of the cocaine trade and the decentralization of the country’s
political system between 1986 and 1991. While the former helped finance this insurgent expansion,
particularly the growth of the FARC, the latter provided a new arena for disparate armed-non state
actors to compete with the state and one another (Echandía 2006; López 2010). During this period
the municipal political institutions became highly coveted by armed non-state actors from the late
1980s onwards as their “[…]administrative decentralization, combined with their weakness in matters
of justice and public security, permit the illegal groups to profit from municipal income” (Rubio 2005,
108). Political power at the sub-national level became a mechanism for competing actors to fortify
their ranks and to gain an advantage over their rivals.
A. Montes de María In Montes de María, the pitched battles between ANUC activist and regional landed elites in
the 1970s inspired the formation of a handful of small insurgent groups in the early 1980s. The
escalation of this agrarian conflict between smaller insurgent groups and proto-paramilitary
organizations was finally punctuated by the full scale arrival of the FARC in the early 1990s, and by
138
1994 this insurgent group had asserted territorial control over virtually the entire region. The FARC
sustained its operations in Montes de María primarily through the extortion and kidnapping of large
landholders, as well as becoming involved in drug trafficking throughout the region. Despite the
political decentralization between 1986 and 1991, the insurgent groups did not intervene much
influence in local politics, limiting their actions to enforcing abstinence during election and on rare
occasions making clandestine arrangements with municipal officials. Finally, the FARC’s attempt to
subvert and co-opt civic organizations, namely ANUC, was met with considerable resistance from
local community leaders, ultimately leading to a pronounced and length decline in regional civil society.
i. The Territorial Sphere
In the early-to-mid 1980s, numerous small leftist guerrilla groups began to appear in rural
communities in Montes de María, as the agrarian struggles of the 1960s and 1970s “seduced armed
groups to the region with the perception that maybe they would be able to find social support to install
an insurgent project with the possibility of success in Montes de María”.161 The first of these was the
Movimiento Izquierdista Revolucionario-Patria Libre (MIR-PL), followed by the near simultaneous arrival of
the Partido Revolucionario de los Trabajadores (PRT), the Ejército Popular de Liberación (EPL), and the ELN.
Many of the insurgent leaders from MIR-PL and the PRT had been actively involved with ANUC and
the regional surge in peasant activism in the 1970s, and thus continued to support the ongoing land
occupations in the region. In contrast to the ELN and the EPL (and later the FARC), the MIR-PL
and PRT were largely locally led and manned. Native to Montes de María, they possessed pre-existing
ties with these communities, coupled with an intimate knowledge of the terrain. However, these early
insurgent groups were extremely small and therefore lacked the military capacity to carry out larger
attacks against their adversaries. Whatever security they offered to the regional peasant leadership was
161 Interview 38, Sincelejo, 2016.
139
offset by the increased stigmatization they brought to these very same communities (Restrepo and
Rodríguez 2000).162
The EPL first arrived during this heady period and attempted to replicate the success with
which the group had implanted itself in the nearby region of Urabá and the south of Córdoba
department.163 The ELN arrived in Montes de María the mid-to-late 1980s, establishing the Frente Jaime
Bateman Cayón in the high mountains of San Jacinto, El Carmen de Bolívar, and Ovejas.164 Although
the group’s presence in the region was relatively modest as was that of the EPL, the ELN began to
kidnap and extort local cattle ranchers and hacienda owners to finance its regional operations (Duica
2013). Furthermore, the ELN strengthened its position in Montes de María by merging with the MIR-
PL in 1987. That same year the ELN attacked the oil complex at Coveñas, causing substantial damage
and demonstrating the group’s ability to target energy multinationals throughout the entire country
(Currea-Lugo 2014). The ELN suffered a setback around this time when the former MIR-PL leaders
who had quickly become disillusioned with the merger, split from the group by forming the dissident
Corriente de Renovación Socialista (CRS) in the early 1990s, only to demobilize shortly after in 1994.165
Proto-paramilitary groups in Montes de María gained greater traction as the confrontations
between peasant activists and landed elites intensified in the late 1970s and early 1980s with the
emergence of small insurgent groups. Many of the rural bourgeoisie finally caved in to their threats by
selling their properties to INCORA to be distributed amongst landless peasants, while other large
landholders steadfastly refused to give in to the insurgents’ demands and in turn were promptly
attacked for their recalcitrance (Verdad Abierta 2010). Some of these were longstanding patrons of
specific rural communities, while others were drug traffickers who purchased large tracts of land in
162 Interview 101, Sincelejo, 2016.
163 Interview 38, Sincelejo, 2016; Interview 77, San Jacinto, 2016.
164 Interview 84, El Carmen de Bolívar, 2016.
165 Interview 36, Ovejas, 2016.
140
the coastal zone and the zone below close to the Magdalena river. These initial insurgent assaults on
the regional landed elites caused the latter to “stigmatize the peasantry and they began to intimidate
and kill members of the community”. By the mid-to-late 1980s, there was a proliferation of these
proto-paramilitary groups throughout the region.166 One native from El Salado describes the rise of
these groups in the region: “In the 1980s, they began to form clans such as los Méndez sponsored by
the Cohens. They allied themselves with common delinquents to defend their land and this
phenomenon marked the beginning of paramilitarism in El Salado. These alliances were created
because the powerful families were afraid that the peasants were going to turn against them”.167 The
rise of these self-defense groups coincided with a dramatic increase in forced displacement and
selective assassinations in Montes de María, a phenomenon that received the explicit support of what
local military and police attachments did exist in the region (Verdad Abierta 2010).
Instead of directly engaging one another, insurgent and paramilitary groups in Montes de
María frequently targeted civilians perceived to be allied with one side or the other. The conflict
between the early insurgent groups and the paramilitary organizations only worsened when the FARC
entered the region in the early 1990s, as the various paramilitary factions proved too weak to impede
the guerrilla hegemon’s massive incursion into the zone (Verdad Abierta 2010). Compared to other
regions of Colombia, the FARC arrived in Montes de María in a rather belated yet abrupt fashion in
the period between 1989 and 1991 (CNMH 2009). The manner in which the 35th front (Antonio José
de Sucre front) and 37th front (Benkos Biohó front) arrived and settled in Montes de María varies
between communities found the coastal region, the high mountain, and the zone below. Local
community leaders from the pre-dominantly Afro-Colombian population in the municipality of San
Onofre remember the FARC arriving during this period in a very subtle fashion “with the goal of
166 Interview 80, San Jacinto, 2016.
167 Interview 85, El Carmen de Bolívar, 2016.
141
penetrating society and recruiting members”, although their arrival coincided with a series of selective
killings of local peasants and community leaders.168 Similarly, the 37th front arrived in the zone below
“between 1989 and 1990 and started committing selective killings” and after the FARC “didn’t have
to hide from anybody because it had complete control of the territory”.169
The demobilization of other small insurgent groups (PRT, EPL, CRS) in Montes de Maria
between 1991 and 1994 left their former zones of influence wide open for the 35 th and 37th fronts to
move into.170 Many demobilizing insurgents from these smaller groups ended up joining the FARC
and the ELN, while others became active in local civil society and politics.171 The FARC’s strategy for
maximizing its territorial control was to establish a massive network of rural camps throughout the
region, utilizing small towns and villages primarily as a source of food, information, and as transit
corridors.172 In El Salado, after the 37th front first arrived it slowly turned the rural community into
“a strategic corridor” surrounded with a series of camps located within fifteen kilometers of the actual
village.173 Elsewhere in towns such as Las Palmas (San Jacinto), Chengue (Ovejas), Macayepo (El
Carmen de Bolivar), and Las Lajas (San Jacinto), local peasant habitants similarly recall that the
insurgents circulated between a series of isolated camps in the hilly foliage, transiting through their
rural villages with varying degrees of frequency.174
Under the command of Martín Caballero, the FARC aggressively took over Montes de Maria
in the early 1990s. A veteran of other guerrilla campaigns elsewhere in Colombia, Caballero had been
tasked with the creation and growth of the Bloque Caribe of the FARC on the Caribbean coast. The
168 Interview 43, San Onofre, 2016.
169 Interview 81, San Jacinto, 2016; Interview 36, Ovejas, 2016.
170 Interview 101, Sincelejo, 2016.
171 Interview 78, San Jacinto, 2016.
172 Interview 87, El Carmen de Bolivar, 2016.
173 Interview 85, El Carmen de Bolivar, 2016.
174 Interview 81, San Jacinto, 2016; Interview 30, Ovejas, 2016; Interview 84, El Carmen de Bolivar, 2016; Interview 77, San Jacinto, 2016.
142
FARC quickly occupied the territory left behind by the smaller demobilizing groups, while carving out
a vast area of influence in the heart of the larger region, particularly in the municipalities of Los
Palmitos, Ovejas, El Carmen, San Juan Nepomuceno, and San Jacinto, pushing the ELN to more
peripheral areas in the process (Duica 2013). The 35th front controlled municipalities located in Sucre
department such as San Onofre, Chalán, Los Palmitos, Morroa, Colosó, San Pedro, Ovejas, Tolviejo,
Buenavista, Galeras, Sincé, El Roble, Betulia, and San Benito Abad, while the 37 th front settled in
municipalities located in Bolívar department such as El Carmen de Bolívar, San Jacinto, María la Baja,
San Juan Nepomuceno, El Guamo, Mahates, Calamar, Zambrano, and Córdoba (MPE 2007a; MPE
2007b). The FARC’s regional presence also spilled over into nearby cities such as Sincelejo, Cartagena,
and Barranquilla.175 Martín Caballero brought numerous family members and old friends he trusted to
the region and assigned them to key positions in his insurgent block. Whereas most of the FARC’s
commanders in the zone hailed from outside the region, the rank and file combatants were largely
from the Caribbean coast and the south of Bolívar (El Tiempo 2007b; Araujo 2008).
The rise of the FARC’s 35th and 37th fronts as the armed non-state hegemon began in 1991
and ended roughly in 1994, when the FARC began to assert its regional dominance over peasant
communities, official state agents, landed elites (and their private ‘self-defense’ groups). At this time,
the FARC’s local structures finally took form. The 35th front was divided into three sub-units, while
the 37th front was configured into four. The 35th front was estimated to possess between 200 and 250
formal combatants, and these in turn were divided into three companies: the Carmenza Beltrán
company in Morroa, Colosó, Ovejas, Tolúviejo, San Onofre, Corozal, Chalán, and Los Palmitos; the
Robinson Jiménez company in the plains of Sucre just southeast of Montes de María, in Betulia, Sincé,
Buenavista, and Galeras; and finally the Policarpa Salavarrieta company which operated with the 37th
175 Interview 97, Bogotá D.C., 2017.
143
front between the departments of Bolívar and Sucre (MPE 2007b). The 37 th front possessed
somewhere in excess of 250 formal combatants, which were spread between four sub-units: the Pedro
Góngora Chamorro mobile company; the Che Guevara company; the Palenque company in northeast
El Carmen de Bolívar, Zambrano, and Córdoba; and the Caribbean Special Forces Unit which moved
between El Carmen de Bolívar, San Jacinto, María la Baja, San Juan Nepomuceno, El Guamo,
Mahates, Calamar, Zambrano, and Córdoba (MPE 2007c). By the mid-1990s, the FARC had also
infiltrated the larger towns and municipal capitals of the region with a network of militia members,
who operated as civilians and thus were undetectable to most people in the region.176 The FARC
militias played a crucial role as informants to the guerrillas, passing along vital information about
“where the army and the police were moving and identifying who were paramilitaries”, while these
individuals also “were in charge of obtaining supplies for the insurgents”.177
The FARC’s incursion into Montes de María signified a new social order for the region’s
inhabitants, as the group’s sudden presence came with a variety of new regulations and sanctions for
those who violated them. The 35th and 37th fronts seized control of the main highways, the rural access
roads, and even the remote footpaths in the most hidden communities, and as such would often force
peasants to ask permission to move from one place to the next.178 Once the FARC settled in Montes
de María, the modest presence of the National Police and the Armed forces that did exist in the region
quickly vacated the zone (Verdad Abierta 2010). Over time, the insurgent group tightened its control
over civilian communities to the extent that peasants eventually were forced to ask the FARC
permission to slaughter their own livestock (CNMH 2009). The insurgents “imposed a military
discipline on the population” in their zones of influence, as the group was “very violent and arbitrary
176 Interview 42, San Onofre, 2016.
177 Interview 87, El Carmen de Bolívar, 2016.
178 Interview 78, San Jacinto, 2016.
144
as a mechanism to obtain greater control” over the region.179 The FARC depended on selective
assassinations to keep peasant communities in line, yet unlike the paramilitary groups which operated
in the region, the insurgents did not practice massacres (Verdad Abierta 2014).
Between 1994 and 1996, the FARC began to escalate the level of violent attacks directed
against the Colombian armed forces and the National Police, while also increasingly attacking non-
combatants and their properties, both of which constitute violations of International Humanitarian
Law. As demonstrated in Figure 12, the 35th front and the 37th front began to violently exercise its
authority in the region from 1994 onwards. More than half of these military actions and violations of
IHL occurred in the three neighboring municipalities - El Carmen de Bolívar, San Jacinto, and Ovejas
- which served as the historic crux of Montes de María (CNMH 2009). The military and the National
Police possessed a modest presence in the region, yet the FARC sought to completely eradicate the
Colombian state’s authority in order to supplant it with its own. The insurgent group’s strategy focused
on “[…]hitting the armed forces while at the same eradicating state institutions in the local sphere, all
of which implicated incursions into the more populated towns to destroy the police stations and other
public bureaucratic installations, especially mayoral buildings, the political institutional centers of
power” (CNMH 2009, 234). An example of the FARC’s usage of “seizures” (tomas) of towns during
this period occurred in the small rural municipal capital of Chalán. In December 1995, some sixty
fighters from the 35th front assaulted the town’s police installation, yet were repelled by the official
authorities. Undeterred, the FARC again attacked the police station three months later, this time
arming a donkey with large amounts of explosives and remotely detonating the farm animal in front
of the police station. Subsequently, an estimated eighty insurgents continued the assault on the eleven
police officers holed up in the remains of the station with rockets, grenades, and other explosives until
179 Interview 84, El Carmen de Bolívar, 2016.
145
those surviving police officers surrendered only to be promptly executed and incinerated in the main
square (El Tiempo 1996b).
The uptick in military actions committed by the FARC in Montes de María coincided with a
similar increase in violations of IHL, most notably against landed elites who failed to meet regular
extortion payments (vacunas) and ransom demands, and also against peasants who resisted their newly
established hegemony or broke their social regulations (CNMH 2009, 234). Certain occupations
became extremely dangerous, as the increase of FARC manned checkpoints and roadblocks leading
in and out rural communities led to a rash of lethal violence against transport drivers who provoked
suspicion due to their unsupervised mobility in and around the region (Verdad Abierta 2014). As the
FARC became the de facto authority in Montes de María, the insurgent organization began to recruit
minors, both voluntarily and forcibly, some as young as twelve years of age. However, in instances of
forced recruitment, any resistance was met with a lethal punishment by the FARC, a situation which
led many families to leave their rural communities for the safety of nearby towns and cities (Verdad
Abierta 2014).
Whereas the FARC ostensibly entered Montes de María to protect and promote the rights of
the region’s historically disenfranchised peasantry, the 35th and 37th fronts committed brutal acts of
violence against lcoals peasants in Montes de María, while simultaneously protecting those large
landholders who paid monthly vacunas in exchange for their protection (Pérez 2010). Apart from the
FARC’s increased violence against the state, landed elites, and peasants in Montes de María, the
group’s intrusion into the lives of the region’s habitants became much more pronounced from 1994
onwards, stringently regulating crime, alcohol and drug consumption, sexual behavior, personal
appearance, and independent mobility.180 The FARC implemented curfews in many rural areas and on
180 Interview 87, El Carmen de Bolívar, 2016; Interview 30, Ovejas, 2016.
146
the regional highways during this time, while also enforcing a strict edict against any contact with the
military or police under the pain of death. A former journalist from El Carmen de Bolívar recalls a
telling anecdote about how seriously this rule was enforced: “A teacher in the village of Ojito Seco left
her oldest son in the house and the army arrived and told him to sell them a hen. That night, the
guerrillas entered the village and they killed him for having sold the hen to the soldiers. The teacher
was a local nurse as well and she had to treat the guerrillas and still they killed her son”.181
From the 1994 onwards, the regional conflict in Montes de María entered an escalating spiral
downwards, from which it would not escape until well over a decade later. The municipal capitals also
became increasingly unstable, as in San Jacinto “police would hide in their stations at 5pm because at
that hour the guerrillas would come out on patrol throughout the town”, operating armed and
uniformed in plain view of everyone, the insurgents even “sat and drank beer in the main plaza”.182
While all armed actors were increasing their violent output, it was directed “[…]more against civilians
than between one another”. With the escalation of violence, both the FARC and the proto-
paramilitary groups “[…]were bringing reinforcements and growing” (Verdad Abierta 2010). The
Colombian authorities, most notably the armed forces, were noticeable mostly for their absence and
inaction against these armed groups as they quickly converted Montes de María into one of the most
hotly contested warzones in Colombia in the latter half of the 1990s.
The Colombian armed forces were present in the region, albeit primarily in the form of the
Naval Armada’s 1st Brigade of the Marine Infantry, which had larger installments near the Caribbean
sea and the Magdalena river, as according to Colombian law and the Navy’s martial code, these forces
were effectively prohibited from extending their military operations outside of a restricted radius from
where they were based near water, a limitation which allowed both insurgent groups and nascent
181 Interview 83, El Carmen de Bolívar, 2016.
182 Interview 82, El Carmen de Bolívar, 2016.
147
paramilitary organizations to expand and consolidate throughout the heart of the region (CNMH
2009, 197). The modest military presence was concentrated in the larger towns, municipal capitals,
and throughout a patchwork of certain bases, most of which had been belatedly created in 1986 by
the then-governor in order to combat the guerrillas. By the mid-1990s, the Colombian armed forces’
presence in the region was most notable in the Navy’s 5th Rifleman Battalion and the Rafael Núñez
School of Carbineers located in Corazal (Sucre), a base in Malagana (Mahates), another in Sincelejo,
and a training base near Coveñas, which was tasked with protecting the tail-end of the Caño Limón-
Coveñas oil pipeline. From the base in Malagana, the armed forces would carry out patrols in María
la Baja, el Guamo, San Jacinto, and San Juan de Nepomuceno, while only randomly venturing into El
Carmen de Bolívar, whereas from the installation in Sincelejo, the Colombian Navy patrolled San
Onofre and other municipalities southwards. Unsurprisingly, these insufficient regional forces lacked
both the “logistical support and means to participate in battle”, all of which generated negative results
for the military when it attempted to confront the insurgents during this time.183
ii. The Economic Sphere
In the case of Montes de María, insurgent groups did not find a bounty of economic
endowments as they did elsewhere in Colombia. According to one regional human right lawyer: “Of
all the fronts in Colombia, those in Montes de María were amongst the poorest”.184 The insurgents
were often forced to resort to coercive predation against civilian communities as in order to procure
basic necessities such as food and water, fostering “relations that were mainly economic” between the
two groups.185 One peasant from the high mountain village of Las Lajas (San Jacinto) explains the
effects of this increasingly one-way relationship: “There existed a great distrust on our behalf towards
183 Interview 97, Bogotá D.C., 2017.
184 Interview 38, Sincelejo, 2016.
185 Interview 82, San Jacinto, 2016.
148
the insurgent groups. These groups never helped us organize to fight for our social demands. Instead
of taking care of us, they lived off of and began to extort us. Relations got really bad because one day’s
wages for a peasant was not enough for both his family and the insurgents. This ended up creating
hatred”.186
In accordance with the FARC’s internal policy that each front and block needed to be self-
sufficient by generating their own revenues, the 35th and 37th fronts quickly developed and expanded
local rackets to sustain their regional operations. The insurgent group supported itself economically
by extorting and kidnapping large landholders and cattle ranchers, eventually turning the region into
a major hub for the latter activity.187 In an effort to raise funds for their operations in the region, these
FARC fronts carried out a series of selective assassinations against landowners who did not meet their
demands, while dramatically increasing the number of kidnappings for ransom and destroying the
properties of those unwilling or unable to meet their extortion payments. The insurgent group
specifically targeted many of the most prominent families in the region, many of whom had ties to or
control of their own proto-paramilitary groups such as Enilse ‘La Gata’ López in Magangue and the
Cohen family of El Salado.188 The FARC would use rural communities as its “operational centers”,
according to one displaced peasant from rural Ovejas: “The guerrillas would leave their stolen cars
and also do kidnapping exchanges there, all of which produced a powerful stigmatization against the
community”.189
Whereas the FARC initially targeted recalcitrant landed elites for kidnapping for ransom, the
group eventually expanded its operations to the local highway system traversing the region in which
insurgents would arbitrarily abduct travelers they deemed sufficiently affluent for ransom at the
186 Interview 77, San Jacinto, 2016.
187 Interview 43, San Onofre, 2016; Interview 35, Ovejas, 2016.
insurgent group tolerated zero dissent or other challenges to its authority in the zone, and as a result
local participation in civic organizations declined markedly during this time (Verdad Abierta 2014).
A clear example of the FARC’s effect on regional civil society can be seen in how local
participation in the JACs evaporated under the insurgent group’s rule. Those community action boards
which did exist fell victim to increased repression from the proto-paramilitary groups and the FARC
alike.211 In one village, Chengue (Ovejas), the JAC was quite important to the local community but
when the FARC arrived nobody wanted to participate anymore for fear of stigmatization and the junta
eventually disappeared.212 Another human rights lawyer confirms this account of what happened in
Chengue:
When the FARC arrived, they used the JAC as its principal interlocutor and they sent political messages through them. After the authorities would arrive and they took issue with the JAC because of its contact with the insurgents, all of which brought consequences and nobody wanted to belong to this organization because it had become a forced interlocutor for the insurgents. With the FARC’s arrival, the role of the community action board was affected, nobody wanted to participate, and as a result, all the progress, the activities, and the functions it had achieved were undone.213
Similar to Chengue, local peasants in Montes de María simply stopped participating in the civic life of
their villages in response to increased insurgent encroachment into their daily lives.214 Quite simply,
armed actors in Montes de María during this period “[…]by way of assassinations and threats,
destroyed ANUC, the JACs, the unions, and the human rights organizations” (PNUD 2010, 20).
B. Arauca The FARC and the ELN registered their first belligerent actions in the Araucan piedmont in
1980, setting the tone for a rapid territorial takeover of the entire sub-region. After a period of
consolidation, both groups sent fighters to lay claim the vast expanses of the Araucan plains in 1986.
211 Interview 28, El Carmen de Bolívar, 2016; Interview 38, Sincelejo, 2016.
212 Interview 90, El Carmen de Bolívar, 2016.
213 Interview 94, Sincelejo, 2016.
214 Interview 85, El Carmen de Bolívar, 2016.
157
By the late 1980s, Arauca was completely under insurgent control. Whereas, the ELN sustained its
operations by extorting the oil multinationals which appeared in Arauca in the early-to-mid 1980s, the
FARC depended more on kidnapping and extortion until the group began to develop a small-scale
coca economy in rural piedmont communities over the course of the 1990s. Both groups similarly
benefitted by capturing municipal and departmental budgets swollen with oil royalties from the mid-
1980s onwards, a feat which they achieved through their skilled deployment of “armed clientelism”,
or the co-optation of political institutions through a combination of electoral strategy and coercion.
By the new millennium, the FARC and the ELN, the latter in particular, controlled all political offices
in the department. This level of political control would have been difficult had in not been for the
insurgent groups’ ability to “co-opt the civic sphere”, directly infiltrating and regulating the vast
networks of civic organizations found throughout the piedmont.
i. The Territorial Sphere
The FARC was the first guerrilla group to launch an attack in the Araucan piedmont, when the
groups violently seized Fortul in March 1980, killing some ten police officers.215 The following year,
this insurgent front was christened the Guadalupe Salcedo front (10th front) of the FARC (El Tiempo
1998b). Some piedmont locals suggest that the 10th front was initially comprised of rebels who had
mobilized during la Violencia under the command of Guadalupe Salcedo himself and had subsequently
been in hibernation while still retaining their basic organizational structure in their struggle against
cattle rustlers in the Araucan plains (Velasco 2016).216 Others claim, more plausibly, that the FARC
arrived in Arauca during the guerrilla organization’s expansion to other colonization zones in
Magdalena Medio and Urabá during the same period of time, although at the request of local members
the guerrilla group’s ‘Bolivarian militias’ in villages and towns. By the end of the decade, the FARC’s
structures in Arauca, combined with those in neighboring Casanare, functioned as a form of ‘mini-
bloque’ inside of the Eastern Block of the FARC, all of which was under the command of Germán
‘Grannobles’ Briceño Suárez, the brother of Jorge ‘Mono Jojoy’ Briceño Suárez, the commander of
the Eastern Block and one of the most powerful members of the entire insurgent organization
(ACNUR 2007).
The emergence of the ELN in the Araucan piedmont in 1980 can be traced to two key events.
First, in the aftermath of the near-decimation of the ELN by the Colombian military at Anorí in the
early 1970s, the remnants of the insurgent group relocated and reorganized in Arauca in the years
following this defeat (Peñate 1998). Second, various peasant leaders - many of whom were originally
from Santander - who were active in local peasant civil society, faced increased violent persecution at
the hands of the Colombian military and police as the decade wore on (El Espectador 2014).227 These
particular peasant leaders felt that their only recourse was to arm themselves and their communities
and in the process made contact with mutual acquaintances in Santander who were longstanding
members of the ELN (Medina 1996; Plazas Días 2017). A nascent guerrilla cell in the piedmont
formed in Alto San Joaquín (Saravena) in 1978 and after two years of preparation some twenty
guerrillas promptly attacked a police outpost in Betoyes (Tame) in September 1980, killing four
officers and forcing the eight survivors to surrender their weapons. The Domingo Laín front of the
ELN was born (El Espectador 2014).228
227 The most recognized founders of the ELN in Arauca are Raymundo Cruz, his farmhand Atilano (surname unknown),
Efraín Pabón Pabón, and Daniel (surname unknown). All four were heavily involved in ANUC and other social organizations such as COAGROSARARE in the piedmont, and were amongst the leading organizers of the 1972 and 1975 civic strikes in Saravena (Interview 49, Arauca municipality, 2016). Pabón and Atilano hailed from Boyacá, Cruz from Cundinamarca, while Daniel was originally from Santander.
228 The ELN in the piedmont initially wanted to name the front after Guadalupe Salcedo but the FARC had already claimed that name, so they settled on Domingo Laín Sáenz, a Spanish priest who had joined the ELN in the 1960s only to fall in combat some years later.
162
Over the following years, the ELN became firmly entrenched in piedmont society. The
Domingo Laín front “[…]was a guerrilla {group} of family members, friends, acquaintances,
neighbors, all of which allowed them to grow easily” (El Espectador 2014). This was not difficult for
the first generation leadership of the front as its’ founders were peasant leaders in ANUC and the
JACs who “decided to become guerrillas due to the abandonment of the state”. Furthermore, the
majority of peasant leaders in Sarvena and Fortul “either joined them or began to support their
project”.229 The ELN in Arauca deployed a strategy of “co-optation, subordination, and submission
of local power structures such as social organizations” (FIP 2015, 23). Thus, the founding members
of the Domingo Laín front already possessed substantial local networks and a strong capacity for
communal organization. Atilano and Daniel assumed command of the ELN in the piedmont after
Efraín Pabón was killed by the Colombian military in an operation in Santander, and the insurgent
front quickly consolidated control of many rural communities throughout the zone by establishing a
rigid code of conduct which was applicable to all and disobeyed by none (El Espectador 2014). As
one native from Saravena puts it, “there was room for debate but at the same time if one didn’t comply
with their law then they were going to die”.230
The severity of this social regulation worked. The social codes imposed by the ELN existed
in both towns and rural villages alike. One social leader remembers the extremity of some of their
prejudices: “Notices appeared where they advised that they were going to do a ‘cleansing’ of different
groups such as women who had relations with military men, and lesbians and homosexuals”.231 Other
prohibitions included long hair for men, excessive piercings, prostitution, consuming or dealing drugs,
thievery, and talking to a police officer or soldier. The punishments for breaking any of these ranged
229 Interview 95, Bogotá D.C., 2016.
230 Interview 48, Bogotá D.C., 2016.
231 Interview 66, Saravena, 2016.
163
from clearing rural fields with machetes to execution.232 The ELN grew rapidly in the piedmont as
between its arrival in 1980 and 1982, the Domingo Laín front expanded from some 25 initial
combatants to 150 members assigned to commisions (comisiones), or local units of twelve to fifteen
fighters stationed in Saravena, Arauquita, Fortul, and Tame. The quick growth of the insurgent front
can be attributed to the fact that “[…]it was a guerrilla {group} of family members, friends,
acquaintances, neighbors, all of which allowed them to grow easily” (El Espectador 2014).
In contrast to the FARC during this time, the ELN preferred a more direct confrontation with
its opponents. The Domingo Laín front established the Simacota Company in 1986, a highly trained
mobile unit with commanders schooled in Cuba and Vietnam, founded with the express purpose of
carrying out larger, more logistically difficult operations against the Colombian military and National
Police. Similar to the FARC, the ELN also expanded into plains communities in Tame, Puerto
Rondón, Cravo Norte, and Arauca municipality during the mid-to-late 1980s, although it appears the
FARC was stronger in this sub-region.233 The leadership of the Domingo Laín front changed hands
around this juncture, with Armel ‘El Chino’ Robles Cermeño taking control of the ELN in Arauca. A
teacher and union activist by trade, under “El Chino” the ELN tightened its control over the piedmont
as violence against recalcitrant local figureheads of authority such as judges, civil servants, and political
activists increased (Velasco 2016). For example, the highest ranking Catholic priest in the department,
Bishop Jesús Emilio Jaramillo, publicly criticized the ELN, and in response three insurgents
kidnapped, tortured, and murdered Jaramillo in rural Fortul in early October 1989 (Semana 1989). The
ELN claimed responsibility while simultaneously accusing the Bishop of collaborating with the
military and nascent paramilitary groups, and also of embezzling funds that were earmarked for local
During the first decade of insurgent consolidation in Arauca, the ELN expanded its formal
units, albeit in much smaller and more fluid commissions spread throughout the piedmont. Key to
the success of the Domingo Laín front was the development of hundred, if not thousands, of militia
members (milicianos) in the broader department. Whereas the ELN and FARC alike operated in plain
view in rural villages in Arauca, armed and uniformed, they functioned more clandestinely in larger
towns and municipal capitals. In these settings, they maintained control through the deployment of
militia members, who were trained cadres in service of the ELN who dressed and operated as civilians
in urban settings. As a lifetime resident of Saravena observes: “It is difficult to identify exactly who is
a militia member and who isn’t. They are invisible”.234 They were often armed and would carry out
small attacks or other acts of violence for the guerrilla organization, and perhaps more importantly
they would gather intelligence, send and receive messages between guerrilla units and civilians, and
perform other crucial tasks for the armed group.235 The Domingo Laín front aggressively recruited
local youth in the piedmont and established a network of indoctrination and training camps for this
purpose in the sub-region.236 Apparently in certain villages, attendance for local adolescents was
mandatory. The ELN also differed from the FARC in regards to the terms of enlistment, as the former
offered a particular form of military service which was a five-year commitment to the military wing of
the organization, after which one was free to leave the armed struggle, albeit with certain political and
social commitments remaining.237
Until the 1990s, the ELN in Arauca depended primarily on the Domingo Laín front, a
fearsome military unit which was supported by the specialized Simacota Company, while two fronts
234 Interview 66, Saravena, 2016.
235 Interview 65, Saravena, 2016.
236 The local school system served as fertile ground for insurgent recruitment. For example, Armel ‘El Chino’ Robles
was reportedly a teacher of Gustavo ‘Pablito’ Giraldo in the 1980s in a piedmont school. The student eventually replaced the teacher as the commander of the ELN in Arauca.
237 “The principal incentive to perform such a service was the lack of opportunities, especially academic and employment. It represented a good option to obtain power and money.” Interview 49, Arauca municipality, 2016.
165
in nearby Casanare (José David Suárez front and Los Libertadores front), one in northern Boyacá
(Adonay Ardila front), and another in the south of Norte de Santander (Efraín Pabón front), all were
available for support if need be. The ELN on the other hand relied on the Domingo Laín front to
serve as the focal point of its presence in the entire northeastern region of Colombia. During the
1990s, a second specialized unit, the Capitán Pomarés company, was established to augment the
Simacota company and to serve in the border area between Casanare and Arauca. The remainder of
the ELN’s presence in the department was divided between nine other commissions and vast
networks of militia members which maintained a presence in every municipality in Arauca (ACNUR
2007). In 1996, the ELN’s Central Committee formed the Eastern War Front (Frente de Guerra Oriental
– FGO), which consisted of all of the ELN insurgent structures operating in Arauca, the José David
Suárez front, the Efraín Pabón front, and the Adonay Ardila front (Fundación Ideas para la Paz 2015).
Mapping out the division of space between the FARC and the ELN in Arauca is a complicated
task. There were certain communities and even municipalities in which one group maintained
hegemony, yet the majority were very closely occupied and contested by both the Domingo Laín front
and the 10th front throughout the department. For example, the village of Pueblo Nuevo is literally
divided in half between members and sympathizers beholden to both groups.238 The only municipality
where one group maintains a near complete hegemony is the ELN in Saravena. The ELN also
maintains a greater presence in the piedmont corridor adjacent to the Eastern Andes which can be
found in western Fortul and Tame, as well as the border region with Venezuela along the Arauca River
in both Saravena and Arauquita. The FARC on the other hand has maintained a greater presence in
rural Arauquita, eastern Fortul, and certain parts of northeastern Tame.239 For the most part, the FARC
were more dominant than the ELN in the plains, a fact which can be explained by the greater presence
of the department’s rudimentary transportation infrastructure shows, the provision of fundamental
public goods such as justice, security, and order were of paramount importance for both the FARC
and the ELN throughout every corner of Arauca. A women’s rights leader from Fortul describes this
dynamic: “In that era, the laws that ruled our society were created by the people and by the
guerrillas”.246 The FARC and the ELN succeeded in establishing themselves as the de facto arbiter of
justice in Arauca, a feat which these groups were able to achieve due to the insufficient deployment
of state manpower and the willingness of the insurgents to physically eliminate any rival claimants to
power without hesitation.247 Efforts to legally investigate and prosecute insurgents accused of
committing a variety of crimes always led nowhere, as the guerrillas forced countless judges to
renounce their positions and leave their posts in places such as Saravena, Fortul, Tame, and Arauquita
during the early 1990s. Failure to comply was accompanied by a certain death penalty. Indicative of
the insurgents’ success, some 218 of the 248 cases pending in the department in 1991 for serious
offenses such as homicide, kidnapping, and extortion were against known members of the FARC and
the ELN, while the remainder implicated members of incipient paramilitary structures in Arauca. Very
few, if any, of these cases were ever even investigated (El Tiempo 1991).
The insurgent justice which these organizations administered was quite severe and draconian,
however, it was also “much quicker” than that provided by the state, which in comparison was “very
slow”.248 Any perceived threats to their authority were persecuted and expelled from the department
or killed outright, as this treatment extended beyond the National Police and Colombian armed forces
to religious clergy.249 A disturbing practice emerged where women who greeted, conversed with, or
246 Interview 19, Saravena, 2016.
247 A former mayor of Arauquita claims that prior to oil “there were only five police officers for the entire municipality” and the insurgents were responsible for resolving issues related to “infidelities, the sale of property, everything” as the “state was not present”. Interview 13, Arauquita, 2016.
248 Interview 50, Arauca municipality, 2016.
249 Interview 56, Arauca municipality, 2016.
168
even attempted to date men from either the police or the military were killed by the insurgent groups.
Over the period 1994 to 1995, there were between one to two dozen adolescent girls who were
executed by the ELN in Saravena alone, solely “[…]for being girlfriends or friends with police officers
and soldiers” (El Tiempo 1995b). A former mayor of Tame explains this macabre phenomenon:
“Women became military targets when they dated somebody from the military, as the insurgents
believed they could be turned into informants. In Tame, there was a guerrilla called Fidel Gallo who
they nicknamed the ‘ladykiller’ (el mata viejas) because he killed any woman who looked at a soldier or
a cop”.250
The extremely strict social code maintained by the insurgents in Arauca, historically and
presently, “had to do with security”.251 The FARC and the ELN provided security and order for the
communities under their control by deploying a highly developed system of populational control.252
People from outside the region faced enormous difficulty entering peasant communities in the
piedmont without being properly vetted by the insurgents first.253 The FARC and the ELN settled
disputes between family and neighbors, adjudicated weddings, issued work permits, appointed
teachers in local schools, and managed the local distribution and usage of land in the department.254
Of enormous importance to their control of the piedmont, the FARC and the ELN heavily intervened
in the local real estate market, effectively regulating all local land purchases and only allowing those
who were related or known to long-established local peasants to buy small plots in the sub-region.255
In places such as Tame where numerous settlers had obtained and developed their smallholdings by
250 Interview 73, Tame, 2016.
251 A peasant in Fortul highlights this apparent contradiction: “I never understood protection to be honest, because here nobody can slip up because the insurgents will kill them at once.” Interview 22, Fortul, 2016.
means of squatting, these groups offered crucial support and protection to peasant migrants without
formal title.256
Recruitment was another area in which the FARC and the ELN competed. The two groups
sought to expand their ranks in the department by recruiting locally, principally from the rural, agrarian
communities in which they had consolidated a strong presence. Their target group was adolescents of
both genders, although it was not uncommon for the FARC and the ELN to attempt to recruit youths
as young as twelve or thirteen years old.257 In piedmont communities, the process of local recruitment
changed from voluntary to quasi-obligatory over the course of the 1990s. A piedmont native from
Tame recalls this:
Almost all of the members of the guerrillas were young people from the community, including in many cases my friends’ parents and my classmates. In the early 1990s, people enlisted because they wanted to, due to household needs, as there were no possibilities to go and study, no purpose, nor were their parents nor anybody else for that matter telling them they had to go and study a vocation. The insurgents would come and listen to them and give them a shoulder to cry on, all the while offering them money and other luxuries that one could only receive if they went that route. By the late 1990s, it became more of an obligation to go with them. According to the guerrillas, we had to accompany them, we had to defend the country, we had to give mandatory service to defend the land and all of that.258
The Gaviría (1990-1994) and Samper (1994-1998) administrations struggled to confront the
increased threat posed by the rise of the FARC and the ELN during the 1990s. As Omar Gutiérrez
(2010) notes: “Between 1992 and 1996, the same guerrilla fronts {in Arauca} labored intensely to
expand territorially and socially (a product of the terms set out at their congresses and conferences)”
(12). The Domingo Laín front and all of the other ELN structures in the department consolidated
considerable control over Araucan society. Various sources recall how the ELN became so belligerent
towards the military and the police that “[…]the police had to confine themselves in their main
barracks between 1995 and 2000”, as the group’s harassment of the authorities “[…]were routine
256 Interview 100, Tame, 2016.
257 Interview 96, Bogotá D.C., 2016.
258 Interview 99, Tame, 2016.
170
affairs in the central plazas of Saravena, Arauquita, and Fortul” (El Espectador 2014).259 Even the
wives of police and military men “became military targets of the guerrillas, they killed them on the
highways in front of others to sow terror”. The plains sub-region was not exempt from the
intensification of insurgent activity against the state, as “[…]between 1996 and 2000 there was a
scourge of violence in Puerto Rondón and Cravo Norte”.260
During the second half of the 1990s the ELN increased its attacks on oil infrastructure in
northern Arauquita and Saravena. The FARC simultaneously began to engage in similar attacks while
also increasing its level of belligerency with the Colombian authorities to such an extent that by 1998
and onwards the group surpassed the ELN as the most active guerrilla outfit in the department
(Gómez Rivas 2016). In April of this year, insurgents from the FARC’s 10th and 45th fronts coordinated
a series of simultaneous attacks against both the armed forces and the police in four different
municipalities in the plains and the piedmont (Cravo Norte, Puerto Rondón, Saravena, Cubará) at
exactly the same time, leaving one civilian dead, ten policemen injured, while the guerrillas also stole
two ambulances, destroyed eight buildings, and blew up a section of the oil pipeline (El Tiempo
1997a). The conflict continued to produce civilian casualties on all sides, as in 1999 the FARC, under
direct orders from ‘Grannobles’, kidnapped and murdered three American activists, all of whom were
working for the protection of U’wa rights in their ancestral homeland in northwestern Saravena and
Cubará.
This period of time also marked greater confrontation on part of the guerrillas with authorities
across the border in neighboring Venezuela. The Venezuelan military and National Guard periodically
abused Colombians who crossed the border with any frequency. As the FARC and the ELN
strengthened their grip on the department in the 1990s, they began to make greater incursions into
259 Interview 69, Tame, 2016.
260 Interview 53, Arauca municipality, 2016.
171
Apure and beyond, often with lethal results. In March 1995, the ELN attacked a military post in
Cararabo (Apure), leaving some eight Venezuelan soldiers dead. After this attack and the ensuing
political fallout between the Venezuelan and Colombian governments, the Colombian armed forces
deployed over one hundred extra soldiers to patrol the Arauca river and established two new bases on
the same waterway in an attempt to curb future incursions by either of the insurgent groups into
foreign territory (El Tiempo 1996c). Over the course of the next two years, the ELN structures based
in Arauca carried out eight further attacks inside of Venezuela and two separate executions of civilians,
increasing the death toll to twenty-four, nineteen of whom were members of the military or National
guardsmen (El Tiempo 1997b).
The election of Hugo Chávez in 1998 would quickly lead to a dramatic change in official policy
towards the Colombian armed conflict in Caracas, as the new president quickly adopted a policy of
neutrality. The previously assembled border commissions were suspended as Venezuelan border states
such as Apure, Táchira, and Zulia suddenly became rearguards for both the FARC and the ELN
(Malamud 2004).261 The ELN consolidated control of the Venezuelan side of the border stretching
from Saravena to Arauca City, whereas the FARC ended up settling further inside the interior of Apure
state.262 The new border policy permitted both insurgent groups a resting place which the Colombian
authorities could not enter, a staging ground for attacks against targets inside of Colombia, and a
further source of revenues, as they quickly divided up the available rackets with the Venezuelan
National Guard. One particular industry which had historically predominated the border region –
contraband - proliferated under the new regime, as when “Chávez devalued the Venezuelan currency
we began to buy everything in Venezuela, especially gasoline and even cattle. All of this was possible
261 Interview 59, Arauquita, 2016.
262 Interview 51, Arauca municipality, 2016.
172
because the Venezuelan National Guard is very corrupt, and the guerrillas make things worse by
smuggling in large amounts of contraband and demanding taxes”.263
During the late 1990s, both formal and informal pro-state forces attempted to combat the
insurgent hegemony in Arauca. The first appearance of paramilitary groups in Arauca date back to
1983, a period when small paramilitary comprised of soldiers from nearby military detachments and
common criminals from the region threatened and eliminated peasant activists who led the civic strikes
throughout the department. Over the course of 1988 and 1989, a group called Grupo Cívico Armado de
Arauca (Cruciagar) appeared in Arauca municipality and was apparently responsible for a variety of
gruesome murders committed against peasant leaders from the piedmont (Carroll 2011, 211).264 In late
1998, another paramilitary group named ‘Cooperativa El Corral’ appeared in Arauca municipality under
the controversial terms of the CONVIVIR policy established during the Samper administration. The
group was stationed in front of the 18th Brigade of the Colombian military’s base in Arauca City and
began engaging in selective assassinations of trade unionists, peasant leaders, and other perceived
‘guerrilla sympathizers’, while also threatening and displacing particular communities which were
located on land which the multinationals wished to develop for further oil exploration (Carroll 2011,
241).265 The paramilitary group’s lifespan in the department was cut short due to revelations from a
prominent local journalist, Efraín Varela, that the group had been imported into Arauca from a
paramilitary haven in Santander department and the ensuing public outcry led to the dismantling of
El Corral.266
263 Interview 10, Arauca municipality, 2016.
264 Interview 21, Saravena, 2016.
265 Interview 8, Arauca municipality, 2016.
266 After Varela’s revelations went public in February 1998, a massive civic strike against the paramilitary incursion was convened throughout the entire department, a mobilization which helped force the national government to apply sufficient pressure in order to shut the paramilitary group down.
173
The worsening violence of the departmental armed conflict was evident in the increasingly
rash and irresponsible conduct of all the armed actors present in Arauca. In December 1998, the
Colombian armed forces carried out the targeted bombing of Santo Domingo, a small hamlet in
northeastern Tame where the army had been engaged in skirmishes with FARC a few days prior.267
The aftermath of the strike left seventeen civilians dead, six of which were young girls between four
and fourteen years of age, with some twenty-one wounded, some permanently disabled. One of the
support planes was manned by American contractors working for Occidental in support of the
Colombian Air Force, a fact which further fuelled the public backlash to the brazen carelessness of
the military operation (Semana 2009a). In the final years of the 1990s, the Colombian military and the
National Police did try, albeit unsuccessfully, to counter the alarming growth of insurgent activity in
Arauca. For example, the Colombian army deployed an astounding two thousand extra soldiers to the
municipality of Tame alone to combat both the FARC and the ELN during Operation Nemesis
(Operación Némesis) in June 1999. Despite the overwhelming level of force sent to the municipality, the
net result was some five ELN members killed in action (El Tiempo 1999).
A key limitation facing the Colombian government’s attempts to assume control of the
department was the inequitable allocation of military recourses. The priority of the armed forces in
the region was the protection of the Caño Limón oilfield and pipeline, a complicated task that
demanded the lion’s share of the state’s focus, thus leaving other rural communities in the plains and
the piedmont at the mercy of the FARC and the ELN. A former General of the 18th Brigade describes
these challenges: “The army’s fundamental problem was when the first oil wells appeared, the
government gave priority to the protection of the oil industry, the pipelines, and the oil production
stations. The little army presence that there was in the zone was devoted more than anything to protect
267 Interview 6, Arauca municipality, 2016.
174
the oil resources than to say attack the terrorist groups”.268 During the worsening climate in the early
2000s, the Pastrana administration reinforced the pipeline’s protection, while perhaps more
importantly creating a support structure from the District Attorney (fiscalía), which was housed inside
of the 18th Brigade’s military base. This new unit accompanied anti-guerrilla units throughout the
department and in less than one year had captured fifty-six people purportedly involved in attacks on
oil infrastructure, a stark contrast from the two people who had been captured in the fifteen-year span
prior (Semana 2002).
Those troop deployments to towns and villages which were firmly under the control of the
FARC and the ELN were often quite ineffective due to the reluctance of the detachments to actually
operate in hostile territory and the limited timeframes in which they actually committed to such
operations. One peasant remembers when she lived in Puerto Nidia (Fortul) how once “they tried to
take control but they only lasted four months. It was absurd! The police and the soldiers were terrified
because there were so few of them”.269 In response to the state’s aggressions, the FARC and the ELN
explicitly forbid any form of contact with any representative of official authority in the department, a
transgression that was punishable by death. A woman’s rights leader from the piedmont recalls the
level of polarization: “The tension of the war was terrible here. Such was the level of mistrust that
even barbers were killed because they had cut the hair of soldiers and police”.270 Affirming this extreme
level of polarization and stigmatization, a religious leader in Arauquita comments:
Speaking with a soldier was reason enough to be killed. Also, any favor or job done for them was certain death. That’s why so many laundry women and shoeblacks were killed. You also couldn’t sell the soldiers or police anything. For that reason, people preferred to keep away from the soldiers, but at the same time they began to see the civilian population as accomplices of the guerrillas, as we didn’t inform them about the attacks that the insurgents had prepared for them. However, people were caught between a rock and a hard place. If they talked the guerrillas would kill them. It is in this context that the military became full of resentment and began to stigmatize civilians as guerrillas.271
268 Interview 47, Bogotá D.C., 2016.
269 Interview 22, Fortul, 2016.
270 Interview 19, Saravena, 2016.
271 Interview 59, Arauquita, 2016.
175
ii. The Economic Sphere
Until the confirmation of vast oil deposits in northeastern Arauquita, Colombia had been a
net importer of petroleum. This all changed with the discovery of the Caño Limón oil field in 1982,
as “[…]the discovery of two billion barrels of oil deposits below the soil of the department of Arauca
allowed Colombia to become an important oil exporting country” (Sarmiento 2015, 11). However, the
extraction and transportation of this fossil fuel required the construction of a terrestrial oil pipeline
through Arauquita and Saravena, then onwards through the departments of Boyacá, Norte de
Santander, Cesar, Magdalena, Bolívar, and Sucre to the Caribbean port Coveñas. Ecopetrol partnered
with Occidental Petroleum, a multinational based out of California in a consortium called the Cravo
Norte Association (Asociación Cravo Norte), and by 1985 the Caño Limón – Coveñas pipeline was
complete and Arauca began to produce oil for export. Arauca quickly converted into one of the
wealthiest departments in Colombia as the massive windfall from oil rents finally provided the means
for the modernization of the department in the form of basic infrastructure and public goods and
services.
Yet, oil wealth also brought many problems to the region such as environmental degradation,
increased militarization, the escalation of the local armed conflict, and endemic corruption. The
establishment of a large oil production complex dramatically altered Arauca’s diverse fluvial network
as entire rivers, fresh water ponds, and lagoons dried up or were contaminated with the run-off
produced by production at Caño Limón.272 A local human rights worker describes the disruptive
effects of oil exploration on rural communities in Arauca: “The oilmen who began to carry out the
seismic, mineral, and energy exploration were the first displacers of migrants and indigenous people
who resided in and around the Caño Limón complex. These multinationals had the state’s complete
287 The FARC and ELN were unable to prevent the establishment of Caño Limón due to their limited presence or military capacity, yet the insurgents were able to stop Occidental from developing an oil field discovered in Caricare (Tame) in 1983. Simply put, “[…]in this field the guerrillas didn’t let them {Oxy} extract.” Interview 8, Arauca municipality, 2016.
179
departmental government in Arauca City.288 The discovery of oil in Arauca exacerbated the primary
cleavage affecting civilian communities, while providing armed insurgent groups with additional
economic resources for their consolidation and expansion in the zone. A former army general who
served in the region for decades describes the effect of oil on the nascent insurgent project in Arauca:
“The violence and lack of security began to appear when it became known that there was oil in Arauca.
I attribute the need of these terrorist guerrilla groups to strengthen their finances in order to grow as
the generating cause of this violence”.289
The Domingo Laín front initially supported itself through kidnappings and the extortion of
large cattle ranchers and landholders on both sides of the Colombian-Venezuelan border.290 However,
the ELN’s well documented 1984 kidnapping of four foreign oil engineers employed by a German
contracting firm, Mannesmann AG, not only gave the insurgent group the financial resources to
expand, but more importantly it provided a blueprint for future revenue extraction in the department
and beyond (Velasco 2016; Carroll 2011). With the completion of the pipeline and the beginning of
massive oil production in Arauca in the period 1985-1986, the ELN started to attack critical oil
infrastructure and continued to kidnap personnel working for the oil multinationals and the
contractors who maintained their operations in exchange for financial remuneration. The ELN’s
subsequent penetration of local politics was such that they were both able to siphon off a considerable
percentage of oil revenues through municipal budgets and the extremely lucrative practice of extorting
local contractors who were hired to build infrastructure and perform public services in the larger
department.291
288 Interview 59, Arauquita, 2016.
289 Interview 47, Bogotá D.C., 2016.
290 Ejército de Liberación Nacional, “Crónica del surgimiento del Frente Domingo Laín”, January 14th, 2007.
http://www.cedema.org/ver.php?id=1734.
291 Interview 50, Arauca municipality, 2016.
180
The ELN’s monopoly on this racket came under threat from the FARC in the 1990s. The
FARC substantially increased its attacks on oil infrastructure during this period in an attempt to co-
opt and control all of the potential revenues in the department from coca production to extortion of
the oil multinationals. For example, when the price of oil rose in September, 2000, the departmental
budget received an additional $223 million USD. The FARC leadership in Arauca told the then-
governor Gustavo Carmelo Castellanos that he needed to deliver half of that sum to the insurgent
group, or they “[…]would blow up the pipeline and there wouldn’t be money for anybody”. Upon
becoming governor in January 2001, Gallardo was given a month to hand over the FARC’s share and
upon neglecting to do so, the FARC attacked the pipeline with explosives ten times in one day (Semana
2002).
The rise of the coca trade in Arauca can trace its beginning to the late 1980s, when the FARC
gradually started providing coca seeds to peasant farmers who lived in areas where this particular
group maintained a very strong presence such as central and southern Arauquita. A peasant leader
from rural Arauquita recalls that: “Coca was one of the strongest products of our village’s economy.
The crop began approximately in 1987 and the boom occurred around 2000”. 292 Another farmer from
the municipal capital of Tame places this date around the same time, stating that “coca plants began
to be seen in 1988” in rural communities in the piedmont section of his municipality.293 Over the
course of the 1980s, the FARC had already consolidated itself firmly in Arauca, establishing
clandestine laboratories, landing strips for small aircraft, and smuggling routes for precursor chemical
needed to process cocaine throughout the department. Nearby departments in the plains and beyond
(e.g. Meta, Guaviare) which produced large amounts of coca would supply laboratories and transport
points in the piedmont region of neighboring Casanare where the merchandise would then be moved
292 Interview 62, Arauquita, 2016.
293 Interview 72, Tame, 2016.
181
on to Arauca before being exported to external markets.294 The increased traffic of narcotics through
the department compelled local commanders of the 10th front of the FARC to introduce the crop to
peasant small holders in rural communities under their control.295 Over time, more and more peasants
began cultivating coca in the piedmont. However, the insurgents maintained a strict system of
regulation over land usage and only allowed peasant farmers to grow one hectare of coca for every
four that they planted for plaintains, yucca, cacao, and other subsistence crops in an effort to mask
the illicit activity.296
In sharp contrast to the FARC, the ELN prohibited any of its members from participating
directly or indirectly in the coca trade on ideological grounds. This explains why coca cultivation was
minimal to non-existent in peasant communities which were almost completely dominated by the
ELN such as Saravena.297 The growing coca trade in the piedmont attracted new arrivals from outside
the region due to the substantially better wages agricultural workers (jornaleros) could earn working as
coca harvesters (raspachines). The influx of agricultural workers generated greater demand for
restaurants, temporary housing, and bars, and directly financed a micro-level consumption boom in
these particular communities.298 Despite this influx of new capital, traditional agriculture suffered as
the production of coca “left local farms without labor” due to the exponentially greater wages offered
to harvest coca in compared to traditional crops.299 Villages such as El Oasis, Aguachica, Pueblo
Nuevo, Filipinas, Panamá de Arauca, and Bocas del Ele witnessed dramatic changes during this time
due to the massive influx of young men from other regions working as raspachines, a process which
294 Interview 12, Arauquita, 2016.
295 It is important to clarify that the FARC did not force anyone to plant coca in the piedmont, the decision was taken by peasants who sought to increase their earnings. Interview 99, Tame, 2016.
304 The first fiscal year that oil royalties became available to the intendency’s government in 1986-1987, Arauca’s budget increased twenty fold (Carroll, 2011).
184
management at the municipal and departmental level by fielding and supporting political candidates
and parties in local elections beginning in the mid 1980s and early 1990s.305 Whereas the FARC was
first to enter the political sphere in the piedmont, first through the communist PCC and later with the
UP in 1985, the ELN broke from its historic policy of electoral abstentionism in 1988 when it began
to support a political faction known as the Saravena Liberals.306 This group of politicians were fearful
of the rise of the UP in the piedmont, and offered the Domingo Laín front access to municipal budgets
if the ELN would in turn use its control of local JACs in the piedmont to deliver votes towards their
candidates (Peñate 1991). The strategy worked swimmingly, as the UP and the Saravena Liberals
virtually controlled the department’s municipal councils and mayoralties until the early 2000.307
By the early 1990s, such was the dominance of insurgent influence in municipal and
departmental politics that “between 1986 and 2002 there was a kind of co-government between the
ELN, the FARC, and the Liberal party which implicated a constant agreement of budgetary plans with
the insurgents”.308 Local peasants were expected to vote for whichever candidate their local JAC
presidents decreed, as it was clear that these were the preferred options of whichever insurgent group
maintained greater influence over their communities, whether in the plains or the piedmont.309 It came
as no surprise to anybody in the department when the departmental senator, Elías Matus Torres, was
arrested in Bogotá in the company of the commander of the Domingo Laín front, Armel ‘El Chino’
Robles Cermeño, in October 2000 (Semana 2002). The conversion of Arauca from an intendancy to
a formal department in 1991 marked the first formal gubernatorial elections for the territory, as the
ELN was able to mobilize enough votes in the piedmont to successfully elect the Liberal candidate,
mobilization capacity of the peasantry as anchors to embed themselves into these communities.
Whereas the FARC arrived at the invitation of the local Colombian Communist Party members, the
ELN emerged organically from the regional civic leadership, and thus already intimately understood
and shared the same enmity towards the state as the rest of the piedmont peasantry. For both nascent
insurgent fronts, appropriating the primary cleavage in support of the peasant cause proved a seamless
task given the prevailing context. The Colombian state was increasingly perceived as the problem
rather than the solution throughout the piedmont and civilians threw their full support behind the
insurgent project, a decision that Wolf (1969) describes as a strategic move embraced by peasants to
guarantee their livelihood in precarious contexts:
Perhaps it is precisely when the peasant can no longer rely on his accustomed institutional context to reduce his risks, but when alternative institutions are either too chaotic or too restrictive to guarantee a viable commitment to new ways, that the psychological, economic, social and political tensions all mount toward peasant rebellion and involvement in revolution. (xiv-xv)
The success with which the FARC and the ELN appropriated the primary cleavage in the
piedmont is evidenced by the concessions gained from the civic strikes over the course of the 1980s.
Previous mobilizations had failed in the sense that the state failed to deliver on its promises, yet with
insurgent backing these strikes began to bear fruit in the form of bridges, roads, electricity, schools,
and other public goods and services in the piedmont (Carroll 2011). By synchronizing group goals to
those of the broader peasantry, the FARC and the ELN became wedded to civilian communities in
the piedmont, providing these groups with a crucial base of support that gave them the ability to
supplant the state in the sub-region. The benefits of expansive hard and soft support meant that the
FARC and the ELN “[…]could acquire almost everything they needed from the populace,
progressively attenuating government authority and creating “counter-institutions” to provide what
the government could or would not” (Long 2006, 22).
204
Neither the insurgents nor the civilian population wanted the sub-region to remain isolated
and cut off from the rest of Colombia. Rather, they like other revolutionary challengers before them,
sought “[…]to enter the polity” (Tilly 1978, 54). The FARC and the ELN achieved the control of
territory and people within short order of arriving in the piedmont, as the insurgent groups
“[…]became the arbiter of conflicts between local settlers, and brought with it arbitrary practices,
firing squads, and inflexible and draconian codes of conduct for combatants as much as for the civilian
population which is still maintained to this day” (El Espectador 2014). The insurgent groups in the
piedmont quickly became interlocutors between peasant communities and the central government by
co-opting the civic sphere and the abundance of organizational endowments to serve as both a local
governance mechanism and a means of contesting the state, all of which enabled the gradual takeover
of local political institutions and productive activities.
The insurgent groups encroached further and further into the political and economic spheres
following the construction of the Caño-Limón oil complex and pipeline. This coincided with the
decentralization of Colombia’s political institutions in the mid-1980s, a reform that “[…]provided
armed groups with the opportunity to develop their activities and establish much better terms with
those municipal governments than with national institutions” (Boesten 2013, 247). The insurgent
social order in the piedmont strengthened considerably due to the FARC and the ELN’s ability to co-
opt municipal and departmental political institutions, a feat they achieved largely through their
deployment of local JACs in local elections to appropriate and invest the surge in oil revenues
generated from Caño Limón into local infrastructure and development projects demanded by peasants
during countless civic mobilizations (Peñate 1998).361 By addressing local grievances as their own,
insurgent groups in the Araucan piedmont gained legitimacy and expanded the reservoir for hard and
soft support amongst the civilian population. Arjona (2017) breaks down the importance of
appropriating the primary cleavage:
By giving the community what it lacks, the group gives locals a reason to form positive beliefs about its involvement in local affairs. Locals are more likely to cooperate with a group that they believe works for their common good by both obeying and offering voluntary support. Second, by ruling over both public and private life, the armed group becomes a very powerful local actor. Because of its capacity to decide on many local affairs and bring about change, locals may want to be on good terms with the group. Those willing to have access to power, for example, are more likely to cooperate. And finally, as the armed group gains more sympathizers in the locality, others have incentives to cooperate in order to obtain the approval and recognition of their fellow residents. (767)
The success of the insurgent groups in controlling every aspect of piedmont society,
particularly in forcing concessions from oil multinationals and departmental political institutions, only
served to further exacerbate their confrontation with the state. The highly skewed distribution of oil
rents which favored the plains over the piedmont generated a widespread perception of relative
deprivation, as piedmont habitants “saw that the oil companies were taking all of the wealth and
weren’t contributing to local society”.362 The rise in military repression against peasant activists fuelled
the widespread perception that the state priority in Arauca was the protection of the oil industry, a
belief that ended up “creating greater loyalty to the armed groups”.363 Both the FARC and the ELN
took advantage of this climate by encouraging the pre-existing polarization between local peasant
communities and the state. This was a key component of the ELN’s revised ‘Popular Power’ (Poder
Popular) strategy devised in the 1980s, which viewed peasant civic organizations as the key to their own
social bases (UC-ELN 1990). Similarly, the FARC’s 10th front, in accordance with the expansive
agenda established at the insurgent group’s seventh conference in 1982, sought to cultivate greater
social interaction with peasant communities under their control by encouraging the creation of
‘solidarity nuclei’ (nucleos de solidaridad) in order to co-opt local organizations. The benefits for the
362 Interview 59, Arauquita, 2016.
363 Interview 65, Arauca municipality, 2016.
206
insurgent groups were numerous, as each passing abuse led to the delegitimation of the state as the
veritable authority in the eyes of the sub-region’s habitants.364
The FARC and the ELN’s appropriation of the primary cleavage and their continued ability
to sustain it, even after many of the initial grievances related to lack of state investment were addressed
in the 1980s and 1990s, was crucial to keeping piedmont civilians bound to the insurgent order.365
Despite the hard fought gains made by the piedmont insurgents and civil society, the increased state
repression against these mobilizations sustained the peasant-state cleavage by shifting the focus away
from demands centered on public goods and services to the defense of human rights (Carroll 2011).
Participation in the mobilizations was expected of everyone in these communities, regardless of
gender, age, ethnicity, or religion, as a peasant activist from rural Tame describes the immense
sacrifices made by common people in the face of frequent state violence: “All the nearby communities
joined together and fought for their rights. It wasn’t easy because people shed their blood, lost their
freedom and lives, but despite this they helped to develop our communities. When people went to the
strikes, they were willing to fight and give their lives.”366 Sustained polarization strengthened insurgent-
peasant identities by providing a common adversary, a fact accentuated by Gruber and Pospisil (2015):
“Identity is a crucial factor in any conflict setting because it differentiates ‘us’ from ‘them’, thus shaping
and being shaped by the conflict at the same time” (231).
This cycle of escalation only served to draw the insurgent groups and the piedmont population
closer together. By the 1990s, the FARC and the ELN had established a robust insurgent order, one
that was firmly entrenched in the piedmont. A newspaper article from this period captures this
insurgent hegemony:
They are the real power, the para-state. They settle conjugal and family conflicts, they select the teachers and through armed pressure the towns have electricity. People aren’t afraid as they are accustomed to living with
364 Interview 95, Bogotá D.C., 2016.
365 Interview 47, Bogotá D.C., 2016.
366 Interview 75, Bogotá D.C., 2016.
207
the guerrillas. The para-state of Arauca is a territory prohibited to everybody, the Army, the Police, the departmental and national government, the Liberals and the Conservatives, the corporations…for everything that doesn’t have the blessing of the guerrillas. There are kilometers and kilometers of highways under the control of armed subversion. On the highway that leads from Arauca’s capital to the {village} of Panamá de Arauca, there is an enormous sign written with white letters on a green background that says: Vehicular transit is forbidden from 7pm to 5am. Simón Bolívar Guerrilla Coordinating Board. (El Tiempo 1992)
The insurgent groups’ ability to insert themselves between the piedmont population and the state,
serving as an armed interlocutor by administering and regulating all formal and informal processes in
the territorial, economic, political, and civic spheres gave the FARC and the ELN complete
populational control over civilians. The state’s violent yet ineffective attempts to break these peasant-
insurgent linkages over the two decades preceding Democratic Security Policy only served to fortify
this level of control, a group dynamic which Durkheim (1961) suggests is the product of violent
contention:
When a society is going through circumstances which sadden, perplex or irritate it, it exercises a pressure over its members, to make them bear witness, by significant acts, to their sorrow, perplexity or anger. It imposes upon them the duty of weeping, groaning or inflicting wounds upon themselves or others, for those collective manifestations, and the moral communion which they show and strengthen, restore to the group the energy which circumstances threaten to take away from it, and thus they enable it to become settled. (459)
Of equal importance, when people are united by a common cause, identity, or adversary (or
any combination of the three) and they endure sustained levels of highly polarized confrontation, it
becomes more difficult over time for individual members or groups to break away from their specific
group due to sunk costs and extreme group pressures. For piedmont residents during this period,
neutrality was not a viable option for civilians while defection was an offense punishable by death.
This escalation of revolutionary commitment seen in the piedmont is well described by Levitsky and
Way (2013) in path dependent terms when they analyze the resiliency of revolutionary party structures
elsewhere in the world:
[I]ntense polarization sharpens “us-them” distinctions, strengthening within-group ties and fostering perceptions of a “linked fate” among party cadres. Where cadres have participated in prolonged violent struggle, they are more likely to view party membership in “moral” terms, and to frame choices about cooperation or defection in terms of loyalty rather than a simple material calculus. The polarization generated by revolutionary wars often persists into the post-revolutionary era, effectively “trapping” potential defectors
208
within the ruling party. When the opposition can be credibly linked to historic enemy and when abandoning the ruling party is viewed as disloyalty or even treason, the cost of defection will be high. (9)
By the time the counterinsurgent expansion into the department began with the 2001
paramilitary incursion and the ascendance of Álvaro Uribe the following year, relations between the
civilian population and the authorities “were non-existent” in Arauca, largely due to the strict and
highly effective prohibition maintained and enforced by not only the FARC and ELN, but also by the
entirety of the piedmont civilian population. According to a prominent trade unionist and activist
from the piedmont: “If the insurgency has done anything in regards to the civilian population, it has
been to train us to distance ourselves from the armed forces because we have been classified as being
the enemy’s collaborators. For this, the same community has been hurt by the military as they have
attacked us”.367
367 Interview 50, Arauca municipality, 2016.
209
Chapter Five. Period of Contestation in Montes de María
“The first method, labeled "tightening the encirclement" (or "tightening the noose"), was to be used when the
encircled area was small and the enemy weak and consisted of a simultaneous advance around the entire perimeter. The second, or "hammer and anvil," technique involved an advance by only a portion of the
encircling forces, while the remaining elements waited for the guerrillas to be driven upon their defensive positions, which were often established along some barrier or obstacle. The third approach consisted of
sending one or more forces into the encircled area, splitting it into two or more smaller pockets, which were then reduced piecemeal. Finally, the fourth tactic, which was to be used when the guerrillas had established a strong fortified position, employed a powerful assault force to overrun the main guerrilla bastion. Once this
had been achieved, the encircling forces would advance to mop up the remaining resistance.”
- Andrew J. Birtle, U.S. Army counterinsurgency and contingency operations doctrine, 1942-1976 (2006, 140)
“But when there is no king to conquer, no capital to seize, no organized army to overthrow, and when there are no celebrated strongholds to capture, and no great centres of population to occupy, the objective is not so easy to select. It is then that the regular troops are forced to resort to cattle lifting and village burning and that
the war assumes an aspect which may shock the humanitarian. “In planning a war against an uncivilized nation who has, perhaps, no capital,” says Lord Wolseley, “your first object should be the capture of
whatever they prize most, and the destruction or deprivation of/which will probably bring the war most rapidly to a conclusion.” This goes to the root of the whole matter. If the enemy cannot be touched in his
patriotism or his honour, he can be touched through his pocket.”
- Colonel C.E. Callwell, Small Wars - Their Principles and Practice (1906, 40)
Amidst a backdrop of skyrocketing violence and intense polarization, the Colombian state and
its paramilitary allies began to push back against insurgent groups during the late 1990s and early
2000s, forcing many sub-national armed units of the FARC and the ELN into retreat. The period of
contestation began with three simultaneous events: the nationwide paramilitary expansion beginning
in 1997, the increased role of the United States in the funding and training of the Colombian military
from 1999 onwards, and the ascendance of Álvaro Uribe Vélez in 2002.
The preceding period of insurgent institutionalization throughout Colombia generated a
massive stigmatization against civilians living in rural zones controlled by such groups. The widespread
perception amongst regional landed elites, paramilitary groups, the armed forces, and the central
210
government in Bogotá was that virtually everybody living in these territories provided some form of
hard or soft support to the insurgent groups. This perception overlooked the fact that disobedience
or non-cooperation with insurgent rules and directives in such spaces was punishable by a severe if
not lethal sanction. Civilians living under insurgent rule thus found themselves doubly condemned, as
they needed to obey and respect insurgent social codes, yet in doing so they were branded as
sympathizers and supporters of the guerrilla groups by the state authorities. Eric Fichtl (2004)
describes the pervasiveness of this stigmatization at the time:
One recurrent theme in accusations of collaboration is “guilt by association,” though in Colombia’s boundless war, most victims are never afforded a chance to prove their guilt or innocence. In this climate, if one person is accused of collaboration with an armed faction, his or her entire family is often considered suspect. This principle has frequently been applied to the entire populations of towns, especially by the state security forces in reference to towns and villages in guerrilla-dominated areas.
Historically, the Colombian military has been tasked with directly confronting insurgent
groups, yet regional commanders were all too often quite content to remain in their bases while the
insurgents controlled the countryside due to the lack of pressure from above (Porch and Delgado
2010). The key function of paramilitary groups on the other hand has been “[…]to develop repressive
operations against the civilian population and to implement a model of counterinsurgent warfare in
which paramilitaries have proven to be an effective instrument”, an objective achieved through the
“[…]physical elimination, disappearance, and forced displacement of community leaders and their
social bases” (González et al. 2003, 61).
In the context of the current civil war, the nationwide paramilitary expansion under the AUC
occurred years before the negotiation of Plan Colombia or the implementation of Democratic Security
Policy. There is an obvious need to distinguish between the formal armed forces and other state-
parallel paramilitary groups, but in light of the fact that both share a common enemy, a relatively
similar cause, and collaborated in a variety of ways, my model holds that the period of contestation
begins when the first counterinsurgent groups inserts itself into insurgent controlled space in an
211
attempt to combat its rivals. As seen from these two cases, paramilitary groups and the military do not
necessarily expand into such spaces simultaneously and hence this periodization needs to be clarified
as such.
Counterinsurgent forces did not merely arrive and converge in insurgent controlled zones such
as Montes de María and Arauca by way of historical accident. Rather, paramilitary bosses and top
military officials targeted regions that were heavily stigmatized as guerrilla havens. Upon first entering
these territories with the goal of dismantling the established insurgent social order, the prevailing
paramilitary strategy was driven by an oversimplified perception which made no distinction between
combatants and non-combatants. The majority of paramilitary blocs throughout Colombia avoided
direct militarily engagement with insurgent groups when they could. Rather, their function was to
“cleanse” insurgent spaces, or to eliminate the social bases of support for these groups found in the
countryside by way of massacres, selective assassination, and especially forced displacement
(González, Vásquez, and Bolívar 2003; Gutiérrez-Sanín 2019).
More so than the Colombian military at this juncture in time, this paramilitary expansion
stemmed the guerrilla growth which had begun in the early 1980s, as this particular state-paramilitary
symbiosis “[…]proved to be unprecedentedly efficacious in cutting the guerrillas’ access to the
population by attacking the guerrillas’ social base through land dispossession, threats, homicides and
massacres” (Dufort 2017, 341). The paramilitary strategy of “draining the sea” was practiced widely
throughout the country, making Colombia one of the world’s largest producers of internally displaced
persons during the period of contestation (CODHES 2010). There is substantial variation between
the manner in which disparate paramilitary blocs functioned and how they forged relations with
official state agents and regional landed elites alike, yet there was a playbook for how they entered
insurgent controlled spaces and used violence. Throughout Colombia, formal military patrols would
deploy in order to clear a specific zone of locally based insurgent units, and paramilitary forces would
212
arrive in the immediate aftermath to conduct brutal cleansing operations against the civilian
population. Their extra-lethal repertoires of violence were intentional and designed to destroy the
social fabric of peasant communities in insurgent controlled zones, a strategy explained by Alejandro
Reyes Posada et al. (2010):
Indiscriminate massacres serve as a catalyst to scare the population which does not identify with the paramilitaries and to organize new bases of support in their favor with those persons who don’t flee. The massacre was thus a military technique to select adversaries or friends where civilians displaced by the terror exercised by the “liberators” implicitly confessed their adhesion to the guerrillas, and as a consequence, justified a posteriori their victimization and the theft of their land. (103)
In most cases, these particular communities were targeted collectively for merely having
shared the same proximate space as insurgent groups. Having cleansed these spaces of perceived
sources of insurgent support, “[…]the paramilitaries brought in their own cadres: the latter’s role was
to ‘protect’ the population against guerrilla influence” (Rojas 2009, 228). Paramilitary blocs thus served
“[…]as a rearguard that consolidates - with an antisubversive program - the territories taken by the
army” (Gutiérrez Sanin and Barón 2005, 6). These vacated properties would then be recycled among
paramilitary sponsors, supporters, or by the fighters themselves in a reconfiguration of the pre-existing
regional land tenure pattern (Human Rights Watch 1998).
Figure 10. Forced Displacement in Montes de María and Arauca
Source: Author’s Elaboration (Data from Observatorio del Programa Presidencial de Derechos Humanos y Derecho Internacional Humanitario de la Vicepresidencia de la República)
213
The importance of paramilitary groups to counterinsurgency outcomes in Colombia cannot
be understated, as their efficacy owed much to the amount of local buy-in that these groups obtained
from landed elites and peasants in contested spaces. State-parallel militias such as those found in
Colombia possess advantages that state-manipulated militias do not. Distant and recent history is full
of state-manipulated militias which defected to the insurgents once they received their weapons and
training (Galula 1964; Tucker 2015). In contrast, autonomous paramilitary groups are frequently
mobilized organically by constituencies disaffected with insurgent predation and can draw on local
networks and resources for the counterinsurgent cause.
The successful negotiation of Plan Colombia in 1999 coincided with the AUC expansion
throughout Colombia, and quickly turned the South American country into one of the largest
recipients of U.S. military aid in the world. The dramatic infusion of military hardware and expertise
sought to modernize the Colombian state’s capacity to combat drug cultivation and trafficking first
and foremost, an undertaking which largely focused on the modernization and professionalization of
the country’s underdeveloped military (Porch and Delgado 2015). With the election of Álvaro Uribe
Velez in mid-2002, the Colombian military was finally unleashed against leftist insurgent groups under
the new president’s Democratic Security Policy in August of the same year. Within his first month in
office Uribe selected two “zones of rehabilitation and consolidation”, or heavily contested guerrilla
zones, to test out his new counterinsurgency strategy: Montes de María and Arauca (Delgado 2015;
Leal Buitrago 2006; Kline 2009). Over the course of his first term in office, the Colombian military
expanded its presence into every sub-national conflict zone in an attempt to confront and expel leftist
insurgent groups from these spaces. The efficacy of the Colombian state’s counterinsurgency efforts
in rural war zones was heavily shaped by the preceding efforts by paramilitary blocs and their local
allies to dismantle insurgent organizational infrastructure and their networks of civilian support.
214
The following two chapters chapter break down the period of contestation in Montes de María
and Arauca, and warrants mention that each region possesses different temporal parameters. While
the period of contestation in both Montes de María and Arauca first began with the paramilitary
incursion, the arrival of the distinct AUC units in these two regions occurred at different junctures.
Furthermore, the success of the joint state-paramilitary project in Montes de María led to the expulsion
of insurgent groups from the zone by late 2007, bringing a long overdue peace to the region. In
contrast, Arauca remained a heavily contested zone throughout Uribe’s two terms in office. By the
time he finally left office in 2010, the FARC and the ELN still controlled the department. In these
two chapters I analyze the disparate effects of state expansion into these regions, how these
paramilitary incursions into Arauca and Montes de María unfolded, the subsequent expansion of
conventional forces into their territory beginning in late 2002, how insurgent groups were affected by
and how they responded to these enormous challenges to their established rule, before analyzing the
final counterinsurgency outcomes.
A. Evaluating Counterinsurgency Outcomes in Montes de María
The Colombian state massively expanded its presence into Montes de María and Arauca in
2002 with the intention of expelling the guerrillas once and for all from their strongholds throughout
Colombia. The counterinsurgent strategy based on clearing, holding, and consolidating territory that
had previously been controlled by armed insurgents proved largely successful in Montes de María, yet
a few key municipalities – El Carmen de Bolívar and Ovejas in particular – remained highly contested
until 2006. A brief analysis of municipal homicide rates for every one hundred thousand residents in
Montes de María, a useful metric to ascertain the stability (or lack thereof) of a given location, reveals
the efficacy of Democratic Security Policy in reducing violence in this region. As Figure 11
demonstrates, the homicide rates in San Jacinto, El Carmen de Bolívar, and Ovejas experienced
215
fluctuations due to the ebb and flow of the local conflict, yet all declined markedly over the same
period of time, thus indicating a notable improvement in security.
Figure 11. Homicide Rate per 100 000 Residents in Montes de María (1990-2013)
Source: Author’s Elaboration (Data from Observatorio del Programa Presidencial de Derechos Humanos y Derecho Internacional Humanitario de la Vicepresidencia de la República)
The difficulty considering the homicide metric in isolation is that it does not capture non-
lethal forms of violence deployed by armed actors such as forced displacement, kidnapping, and
disappearances. As such, homicide rates, while helpful, need to be augmented by other indicators of
armed conflict. It is possible to develop an accurate idea of the level of control exercised between
competing actors by comparing and triangulating qualitative accounts from local residents, secondary
analyses conducted by Colombian research institutes, and other quantitative sources of data on violent
actions initiated (e.g. armed actions and confrontations) committed at the municipal level by the
Colombian military, insurgent groups, and paramilitary organizations. Unlike homicides, armed
actions require a higher level of logistical coordination and collective action from specific actors and
therefore represent not only the presence of competing actors, but also the intensity of the armed
conflict in a given place at a particular moment in time.
216
Figure 12. Armed Actions in Montes de María (1998-2013)
Source: Author’s Elaboration (Data from Observatorio del Programa Presidencial de Derechos Humanos y Derecho Internacional Humanitario de la Vicepresidencia de la República)
According to Figure 12, El Carmen de Bolívar and Ovejas witnessed a rise in these attacks
between 2002 and 2006, only for these to fall off dramatically afterwards, whereas San Jacinto
maintained comparatively low levels of armed confrontation during the same period. These descriptive
statistics largely concur with a study conducted by the Conflict Analysis Resource Center (Centro de
Recursos para el Analisis de Conflictos – CERAC), a Colombian think tank specializing in the study of the
domestic armed conflict and other violence. CERAC’s typology examines how the armed conflict
affected each and every municipality in Colombia from 2000 until 2012. While two of the three
municipalities in question in Montes de María were deemed to have been “strongly affected” by a
“high intensity” level of armed confrontation between opposing actors during this period, the conflict
was classified as ‘interrupted’ in Montes de María.
Table 8. State of Conflict and Outcome per Municipality in Montes de María (2000-2012)
Municipality State of Conflict Intensity Result
San Jacinto Interrupted Conflict Low Intensity Lighly Affected & Interrupted
El Carmen de Bolívar Interrupted Conflict High Intensity Strongly Affected & Interrupted
Ovejas Interrupted Conflict High Intensity Strongly Affected & Interrupted
217
Source: CERAC Typology by Municipality of the Armed Conflict
The success of the state expansion campaign in Montes de María was very visible within the
region and beyond. A local community leader in El Carmen de Bolívar remembered that prior to
Uribe’s first term, “the situation and public order was terrible, there was a lot of fear. The guerrillas
painted their graffiti on houses and made incursions into the municipal capital. People didn’t leave
their houses after 8pm. There were threatening flyers, and the highway was closed after 6pm and
nobody could travel anywhere”.368 A community leader from San Onofre similarly recalled that travel
before Uribe was extremely difficult, if not impossible due to the insurgent and paramilitary control
of the highways. But from 2004 onwards, “that all changed. One could travel at night to Cartagena or
Sincelejo at night without any fear”.369 A peasant farmer in the high mountain of San Jacinto similarly
credited Democratic Security Policy for improving security in his rural community completely, stating
emphatically that: “Uribe’s government was the best. Thanks to him we could recuperate our land in
our village after having been displaced. Like that, law and order finally arrived as four thousand men
entered the zone to combat the guerrillas. Uribe said in his first term that he was going to save Montes
de María and he did”.370 Most everybody interviewed recognized the targeted assassination of Martín
Caballero as signaling the death knell of the regional insurgencies. Former Colonel Colón described
this event in simple terms: “After October 24th, 2007, not a single living guerrilla remained in Montes
de María. Not a single FARC, ELN, or ERP militant”.371
B. Counterinsurgent Victory in Montes de María
The arrival and territorial consolidation of both the FARC and the ELN in the late 1980s and
early 1990s caused great panic amongst landed elites in Montes de María. As a result, large landholders
368 Interview 28, El Carmen de Bolívar, 2016.
369 Interview 42, San Onofre, 2016.
370 Interview 78, San Jacinto, 2016.
371 Interview 97, Bogotá D.C., 2016.
218
and cattle ranchers convened in Sucre with regional political bosses and paramilitary leaders from
neighboring Córdoba to form the Bloque Héroes de los Montes de María (BHMM) in 1997. The incursion
of this paramilitary group into the heart of the region began in 1997 with a string of massacres in rural
communities. Based in the coastal zone and the zone below adjacent to the Magdalena River, the
paramilitaries operated in tandem with the 1st Marine Infantry Brigade of the Colombian Navy and
conducted incursions into supposed guerrilla strongholds to rid the insurgents of their social bases of
support. The methods employed were some of the most brutal ever in the context of the Colombian
conflict, as between 1997 and 2002 Montes de María became the region with one of the highest levels
of forced displacement and massacres in the country. The paramilitary strategy was based on making
periodic incursions into guerrilla-controlled territory in the high mountain and the zone below in order
to conduct selective killings, forced displacement, and massacres, all with the aim of destabilizing their
opponents. The end result of this campaign was that by the time Álvaro Uribe came to power in 2002
and converted the region into one of his “zones of rehabilitation”, Montes de María had arguably
become Colombia’s most violent killing field.372
Figure 13. Insurgent Trajectory and Counterinsurgency Outcome in Montes de María
Source: Author’s Elaboration
The state expansion campaign in Montes de María succeeded in severing local communities
from the insurgent groups, eventually expelling the insurgents from the region within five years of its
implementation. Beginning in 2004, the local attachment of the Colombian Navy began to directly
372 Interview 26, Cartagena (2016).
219
confront the BHMM, a strategic decision that both forced the paramilitaries to demobilize the
following year in 2005, while also generating local buy-in from many civilians who had been brutally
victimized by this particular group. The ELN was not able to resist and retreated from the region in
2003-2004, while the FARC’s fronts initially withstood the onslaught, but were increasingly
surrounded and cut off as the war ground on. Initially, the FARC succeeded in avoiding a direct
confrontation with the Colombian military, although the increased level of rebel desertions gave
regional military commanders the intelligence they needed to locate and strategically bomb these
insurgent refuges. Finally, in late 2007 both the 37th and the 35th fronts virtually collapsed with the
targeted assassination of local commander Martín Caballero, a feat which saw all surviving members
either disperse elsewhere, or surrender to local authorities.373 With a superior numerical advantage and
a vastly improved technological capacity, the Colombian armed forces made good use of intelligence
and collaboration which came from a local peasantry that had been extremely traumatized by the
paramilitary incursion while simultaneously feeling betrayed by insurgents.
i. The Paramilitary Incursion
The BHMM owes its formation to the confluence of local, regional, and national level
paramilitary trends. By the early-to-mid 1990s, the entrance of the FARC had greatly weakened most
of these first generation proto-paramilitary groups, and as a result, regional landed elites formed their
own CONVIVIR organizations during the Samper administration in a futile attempt to stem the
guerrilla tide (Verdad Abierta 2010). Despite these efforts, the FARC had firmly consolidated itself in
the region by 1996 and was committing acts of violence against large landholders with far greater
frequency. The AUC had sponsored the formation of two powerful blocks to the west and east of
Montes de María - Bloque Córdoba and Bloque Norte – and the Castaño brothers then centered their
373 Interview 87, El Carmen de Bolívar, 2016.
220
attention on the formation of the BHMM in 1997. According to Verdad Abierta (2010):
“Paramilitarism {in Montes de María} was born in 1997 at a meeting on the farm, Las Canarias, owned
by the ex-governor Miguel Nule Amín in the rural area of Sincelejo…It was convened to seal a
counter-guerrilla alliance between a hundred farmers and politicians with some paramilitary bosses
who came from the neighboring department of Córdoba”.
The BHMM was able to quickly assemble hundreds of soldiers by incorporating the remnants
of the proto-paramilitary clans and the regional Convivir apparatus into its fold, while also recruiting
local peasants, insurgent desertors, army guides, and common criminals. The BHMM’s leadership
represented the rank-and-file’s makeup, as all except the top-ranking commander, Edward ‘Diego
Vecino’ Cobos Téllez, were born and raised in Montes de María. His three top sub-commanders –
Rodrigo ‘Cadena’ Mercado Peluffo, Uber ‘Juancho Dique’ Bánquez, and Román Zabala – were all
native sons of the region. It warrants mention that whereas Diego Vecino was the top ranked
commander, he was more the ‘political’ head of the BHMM while Cadena served as the military
commander for the paramilitary group (Tribunal Superior de Bogotá Sala de Justicia y Paz 2017). As
a young man, Cadena began was a member of the proto-paramilitary group, Los Rodríguez, in his home
village of Macayepo, before eventually rising to assume a leadership role within the new organization
(Verdad Abierta 2010).
The BHMM was divided into four fronts, all ostensibly under the central command of Diego
Vecino, who in turn had the full support of regional politicians, paramilitary bosses from neighboring
Córdoba and Magdalena departments, and other powerful figures. These fronts were broken down
into sub-regional spheres of influence, including the Gulf of Morrosquillo (coastal zone), Canal de
Dique (coastal zone/Cartagena), the Plains of Bolívar and Sucre (zone below), and La Mojana (south
of the region). Each front served as a quasi-autonomous group in their zone of influence, staffed with
young men largely recruited from Montes de María or other communities on Colombia’s Caribbean
221
coast.374 Many recruits were insurgent defectors, who after abandoning the FARC and seeking refuge
in local military installations, would then be passed along to the BHMM.375 Young men and women
from the region, some of whom were minors, were motivated by various considerations ranging from
vengeance to economic enticements, while others were simply tricked or coerced into enlisting.376
According to Sarah Zukerman Daly’s (2016) index, the BHMM was a locally configured group, as
upon demobilization in July 2005, some 58% of the 598 demobilized fighters were from Montes de
María, while 67% of the block’s troops remained in the region after disbanding (106-107).
The BHMM first arrived, settled, and established a network of bases in those municipalities
adjacent to the Magdalena River and the coastal zone, sub-regions which boasted a substantial
presence of large landholders, cattle ranchers, and drug traffickers.377 Not only did these municipalities
boast organic allies and sponsors, but the FARC’s presence in these particular communities was not
as consolidated as it was in either the high mountain or the zone below, providing the BHMM with a
launching pad to make incursions, effectively creating a cordon around the heart of the larger region.378
The BHMM also maintained a considerable presence in municipal capitals such as Córdoba Tetón,
San Juan Nepomuceno, El Guamo, San Onofre, El Carmen de Bolívar and San Jacinto, as the armed
group were also present in the closest metropolitian areas, Sincelejo, Barranquilla, and Cartagena.379
The paramilitaries made inroads into these more densely populated communities by first arriving
dressed as civilians in order to cultivate local informants so that they could determine who was
supporting the insurgent groups with information or supplies.380 Having identified these potential
374 Interview 39, San Jacinto, 2016; Interview 34, El Carmen de Bolívar, 2016;
375 Interview 30, El Carmen de Bolívar, 2016.
376 Interview 33, Ovejas, 2016.
377 Interview 82, El Carmen de Bolívar, 2016.
378 Interview 88, Sincelejo, 2016.
379 Interview 87, El Carmen de Bolívar, 2016.
380 Interview 42, San Onofre, 2016.
222
targets, the paramilitaries then entered these towns in unmarked vans and would either assassinate
these targets on the spot, or forcibly abduct and make them disappear permanently.381
Figure 14. Conflict Events Committed by Paramilitary Groups in Montes de María (1988-2007)
Source: Author’s Elaboration (CNMH 2009, 238) Paramilitary incursions into the high mountain and the zone below were conducted in a
familiar pattern where the Colombian military would enter and clear a particular area first, and upon
departing would then be replaced by a paramilitary patrol that would brutalize the community. In
other instances, various paramilitary detachments from different fronts would arrive simultaneously
in a targeted community from every direction, overwhelming any insurgent or civilian resistance that
was offered.382 After torturing and killing those who had previously been singled out by informants as
insurgent collaborators, the paramilitary troops would frequently appropriate all the livestock they
could find before leaving.383 En route to the specified destination, the paramilitaries would often kill
whomever they came across, either to prevent them from warning others of their approach, or to
simply demonstrate their power and authority. As one community leader from Macayepo recalls:
“Various times they would come up the mountain to attack guerrilla camps and upon descending again
381 Interview 82, El Carmen de Bolívar, 2016.
382 Interview 82, El Carmen de Bolívar, 2016.
383 Interview 77, San Jacinto, 2016.
223
would kill everything they found”.384 These ‘caravans of death’ were a common tactic of the BHMM,
with the entry and escape routes leading to and away from the various massacre sites littered with
selective assassinations.385 Another common tactic of the BHMM during these incursions into the high
mountain and zone below was to establish temporary roadblocks and checkpoints on highly travelled
rural access roads, where they would detain, interrogate, and kill certain travellers who were deemed
suspicious by the counterinsurgent forces.386Many civilians were targeted merely for transiting
frequently between the urban and rural areas. For example, in January 2000, over a dozen young men
who sold snacks informally on the buses to and from El Carmen de Bolívar were detained, abducted,
and disappeared by the BHMM (Verdad Abierta 2014).387
From its arrival, the BHMM violently targeted what remained of Montes de María’s civic
leadership, killing Guillermo Montero Carpio, the regional president of ANUC, in broad daylight in
Sincelejo on June 11th, 1997 (El Tiempo 1997). During this period, ANUC and the regional peasant
movements “became completely invisible” due to the escalating violence, a fact which was reflected
by the almost total decline in civic mobilizations.388 Other civic leaders and political activists from the
left were systematically hunted down by BHMM hitmen in Montes de María. The armed group not
only carried out extensive assassination attempts against former municipal mayors and council
members within the region, but also extended its scope to urban centers as far away as Barranquilla,
killing several former militants and leftist activists who had taken refuge there (Corporación Nuevo
Arco Iris 2014).389 The practice of killing community leaders sowed greater fear and broke down
communal trust between peasants living within these communities because locals understood that
384 Interview 84, El Carmen de Bolívar, 2016.
385 Interview 29, El Carmen de Bolívar, 2016.
386 Ibid.
387 Interview 42, San Onofre, 2016.
388 Interview 101, Sincelejo, 2016.
389 Interview 93, Bogotá D.C., 2016.
224
somebody amongst them was providing information to the BHMM. One night in San Rafael (Ovejas),
the local BHMM detachment arrived in the dead of night to eliminate a notable local civic leader.
Instead of searching for him at his family residence, the assassins went directly to his mistress’s home,
where the intended target was in fact staying that particular night. Following his assassination, there
was “mutual distrust between neighbors who had once trusted each other with everything”.390
In addition to selective assassination, the BHMM also generated massive levels of forced
displacement in Montes de María. Driven by the counterinsurgent ethos of “draining the sea to kill
the fish”, the BHMM forcibly vacated rural zones of civilian populations in order to deny the FARC
and other smaller insurgent groups the vital support networks on which they depended to operate.
Through a brutal deployment of selective and indiscriminate violence, the paramilitary group quickly
converted Montes de María into one of Colombia’s highest regional producers of internal displaced
persons. In the case of Bajo Grande, most habitants were forcibly displaced in 1996, and when many
tried to return a few years later, this provoked the paramilitaries to burn the entire village to the ground
in an act of reprisal.391 Often, this pattern of forced displacement occurred in communities which had
previously experienced land conflicts between large landholders and their proto-paramilitary groups
on one side, and peasant smallholders on the other, providing a mechanism for landed elites to assert
their interests at the expense of the peasantry.392
However, the sudden increase in paramilitary massacres was arguably the key factor in
emptying the countryside in Montes de María. From the time of its arrival, the BHMM would commit
the majority of these massacres on its own, but when the armed group needed assistance from more
established blocks elsewhere on the coast, it could depend on their support. This was exemplified by
390 Interview 36, Ovejas, 2016.
391 Interview 80, San Jacinto, 2016.
392 Interview 81, San Jacinto, 2016.
225
a series of extremely violent massacres committed in the rural village of El Salado located in the zone
below of El Carmen de Bolívar between 1997 and 2000. By the 1990s, El Salado had become one of
the largest flashpoints in the broader region, with fierce conflict ensuing between the FARC and the
proto-paramilitary group, Los Méndez, which was backed by the Cohens, the town’s most wealthy
landowning family. In 1995, Santander Cohen was under siege from the insurgents and the head of
the local Battalion of Marine Infantry came to his rescue. However, the FARC ambushed the rescue
squad as it left the village, killing Cohen, the commander, and twenty-eight other soldiers (Semana
2020). El Salado permanently became stigmatized as a guerrilla village in the eyes of the Colombian
military and paramilitary fighters in the region.
On February 16th, 2000, three disparate groups of paramilitaries from the BHMM, Bloque
Norte, and Bloque Córdoba, three hundred men in total, descended upon El Salado from every road
leading to the village.393 Led by insurgent deserters, the different paramilitary troops proceeded
towards El Salado on foot, confiscating all livestock they found and killing any civilian they
encountered. The local FARC detachment quickly found out about the impending advance and set
out to confront the various paramilitary factions descending upon El Salado. After realizing how
outnumbered they were however, the insurgents retreated rapidly and warned locals who remained in
the village center to flee immediately (CNMH 2009). By the morning of February 18 th, the
paramilitaries arrived in El Salado in full force. One survivor describes what ensued:
They got everybody together in front of the church, identified themselves as the AUC, and they told us that they were under orders to burn the town to the ground and kill everybody. They told us that we were guerrilla auxiliaries. The first five people killed were singled out by informants, but after that it turned into a game of chance: they forced us to say a number between one and twenty and if you said the wrong number, they would kill you. If people cried for their family members, they were also killed. Fear consumed all of us as the paramilitaries stayed for three days after the massacre. In this time, they raped women. All of these deeds were accompanied by music as they took all the instruments out of the House of Music and they celebrated every death accordingly and showered themselves in rum and beer. The most astonishing thing is that from the 16th until the 20th, the paramilitaries were inside of El Salado controlling and terrorizing people and we weren’t allowed to cry, we weren’t allowed to bury the victims, and despite how close the Malagana military base was, two hours by car, and the one in Corazal, an hour and fifteen minutes by car, the proper authorities
393 Interview 40, El Carmen de Bolívar, 2016.
226
never mobilized, neither the Marine Infantry or the National Army, to help us in our time of need. The worst was about fifteen minutes to half an hour after the paramilitaries left, the armed forces arrived via the exact same road they had left on. The community saw how some of the soldiers who entered had actually been at the massacre as paramilitary combatants.394
By the time the various paramilitary factions had left the village, sixty residents had been brutally
murdered, some by dismemberment and impalement. In the aftermath of the second massacre in three
years, El Salado and its outlying areas became a virtual ghost town due to the massive exodus of local
residents (CNMH 2009).
Similar to paramilitary blocks elsewhere in Colombia, the BHMM frequently deployed sexual
violence against men, women, and children in Montes de María. Rape would often be used by
paramilitary forces during its incursions into communities to commit selective assassinations, forcibly
displacement, and massacres. During the massacre in El Salado, many women were forced to strip
naked and dance in front of their husbands, some of whom were subsequently abused sexually by
paramilitary fighters. One young woman who was pregnant was impaled for allegedly being a guerrilla
informant, while another sixteen-year old named Nayibis Contreras was hung from a tree and stabbed
with bayonets for supposedly dating a guerrilla commander stationed nearby (Verdad Abierta 2009;
Semana 2008b). In contrast to leftist insurgent groups which maintained strict prohibitions against
sexual violence towards civilians, the BHMM did not possess such regulations against this type of
predatory behavior. One mid-ranking paramilitary commander, Marco ‘El Oso’ Tulio Pérez, who
presided over the rural coastal community of La Libertad in San Onofre, was a notorious culprit of
such crimes. He would reportedly rape underage girls in front of their families and abduct others and
keep them against their will for days (Verdad Abierta 2008).395
While the BHMM principally sought to vacate the countryside of any potential social networks
of support for the regional insurgent groups, the paramilitary block also directly clashed various times
394 Ibid.
395 Interview 45, San Onofre, 2016.
227
with the FARC, the ELN, and even the Revolutionary People’s Army (Ejército Revolucionario del Pueblo
– ERP).396 Between 2000 and 2002, there were eighteen registered confrontations between the BHMM
and insurgent groups, mainly the FARC, with notable battles being fought in rural communities
throughout the region. These confrontations produced dozens of combatant deaths, while also
forcibly displacing numerous civilians from their communities in the process, and were based more
on these various armed actors attempting to assert “strategic control over vital {geographic} points”,
in particular the local highway system and natural corridors, than any determined effort to eradicate
their opponent entirely (Echandía 2006: 203; Cantor 2018). Following the beginning of the AUC’s
nationwide ceasfire under the terms of its negotiations with the Uribe administration in December
2002, the BHMM and the FARC avoided entering one another’s territorial strongholds in an unofficial
pact of non-aggression (Porch 2012).
The entrance and consolidation of the BHMM in Montes de María occurred with the overt
and explicit support of the regional authorities, namely the 1st Marine Infantry Brigade of the
Colombian Navy and the National Police. According to one peasant leader in the high mountain of
San Jacinto, “the military never touched the paras, they would pass right by them and let them continue
on their way”.397 Similar to other regions of Colombia, the military and their paramilitary alies would
refer to one another as “cousins” (primos), and one such benefit provided by such an alliance was that
casualties inflicted upon the insurgent groups by the BHMM were attributed and claimed by the 1st
Brigade.398 In other instances, it has been established that members of formal state apparatuses
disguised themselves as paramilitaries in order to participate in massacres, as some individuals
396 The ERP was a breakaway faction of the ELN in Montes de María that formed in 1996 and inhabited a few rural spaces in the high mountain before demobilizing in early 2007.
397 Interview 77, San Jacinto, 2016.
398 Interview 89, El Carmen de Bolívar, 2016.
228
effectively moonlighted for the armed group when they were not on duty.399 Many of those formal
agents who did not directly participate in paramilitary operations would nevertheless provide the
BHMM with classified intelligence to carry out selective killings (Verdad Abierta 2009). Such was the
level of intelligence sharing that local civilians “weren’t able to denounce the paramilitary abuses
against the community because they would find out immediately”.400 Those who did attempt to report
such crimes committed by the BHMM would often be intercepted, abducted, and killed immediately
after leaving the police station or military base where they had registered their complaint.401
Even though coca was neither cultivated nor processed in the region, Montes de María
nevertheless possessed various potential points of departure to export this illegal merchandise abroad,
as somewhere between thirty and thirty-five tons of cocaine were exported annually from the Gulf of
Morrosquillo between 2002 and 2004 (Verdad Abierta 2010).402 BHMM commanders organized
shipments through different coastal routes, serving as middlemen between producers in nearby coca
regions and foreign clients who arrived in the Gulf of Morrosquillo to purchase their narcotic
shipments (Verdad Abierta 2009). While much of the BHMM’s revenues were derived from drug
trafficking, the armed group were also actively involved in extortion schemes of large cattle ranchers,
local merchants and transportation workers, and the lucrative practice of embezzling public funds
from departmental and municipal budgets and public contracts (CNMH 2012; Verdad Abierta
2010).403
In contrast to the insurgent practice of largely abstaining from municipal and departmental-
level politics, the BHMM was founded with the overt support of regional political elites and this
399 Interview 79, San Jacinto, 2016; Interview 80, San Jacinto, 2016.
400 Interview 42, San Onofre, 2016.
401 Interview 43, San Onofre, 2016.
402 Interview 97, Bogotá D.C., 2016.
403 Interview 46, San Onofre, 2016; Interview 42, San Onofre, 2016; Interview 36, Ovejas, 2016.
229
remained the case throughout the armed group’s tenure in Montes de María.404 This relationship was
based upon electoral needs and the effective deployment of violence. After the paramilitaries
consolidated their presence in Montes de María, the armed group quickly eliminated any elected
official or aspiring candidate who opposed their project.405 In order for its preferred candidates to win,
the BHMM used widespread voter intimidation tactics and vote buying to ensure the desired outcome,
as in one telling example only one candidate ran in the 2003 mayoral elections in San Onofre. Quite
simply, no other candidates were allowed to participate.406 Reflecting a nationwide trend, the BHMM
did its part to bring out the vote for Álvaro Uribe’s 2002 presidential campaign by publicly threatening
to kill anybody who did not vote for him, and by also casting false votes using identification cards of
people who were living outside of Colombia, or those who were long deceased.407
The region’s political elites were directly involved in the planning of paramilitary operations
and other assorted acts of violence. Political barons who held seats in the National Congress such as
Miguel Nule Amín and Álvaro García Romero were directly involved in the planning and execution
of the paramilitary massacre in Macayepo, yet perhaps no other politician in Montes de María blurred
the line between elected representative and paramilitary fighter more than Salvador Arana. In his
capacity as governor of Sucre between 2001 and 2003, Arana – who would later serve as Álvaro Uribe’s
ambassador to Chile - ordered the assassination in March 2003 of the then-mayor of El Roble,
Eudaldo León Díaz Salgado, because Díaz had publicly denounced Arana as corrupt. The governor
was also directly implicated along with Cadena in the previous killing of Yolanda Paternina, who had
been investigating the Chengue massacre (Semana 2008). One individual who served as Arana’s driver
during this period, Jairo Antonio Castillo Peralta, later came forward and testified to something that
404 Interview 42, San Onofre, 2016; Interview 46, San Onofre, 2016.
405 Interview 46, San Onofre, 2016.
406 Interview 89, El Carmen de Bolívar, 2016; Interview 35, Ovejas, 2016; Interview 43, San Onofre, 2016.
407Interview 45, San Onofre, 2016; Interview 46, San Onofre, 2016.
230
was an “open secret’” in the region, namely “that the real paramilitary bosses in the region were for
many years the politicians” (Verdad Abierta 2008). Over the course of the BHMM’s tenure in the
region, the para-political alliance completely dominated politics from municipal councils to the
election of departmental governors, and national-level congressmen and senators. The extent of the
parapolítica scandal in Montes de María was extensive, with some thirty-five politicians indicted and
tried for their links to the BHMM, eight of whom were former mayors, seven ex-councilmen, a former
departmental legislator, three former governors, three former congressmen, three active congressmen,
and three sitting senators elected for the 2006-2010 term, and two mayors and five councilmen who
were elected to office in 2007 (Verdad Abierta 2009).
The BHMM’s relations with civilians can best be described as predatory and repressive, as the
group instantly surpassed the FARC in terms of its cruelty and willingness to arbitrarily use violence
under any and every circumstance imaginable. As one native of El Carmen de Bolívar puts it: “At least
with the guerrillas you could negotiate, whereas with the arrival of the paramilitaries everything got
worse, the situation completely deteriorated”.408 One consequence of this was the loss or
decomposition of the ‘social fabric’ (el tejido social), as neighbors ceased to trust or confide in one
another due to the fact that certain individuals opted to collaborate with the paramilitary block.409 In
El Rincon del Mar, where Cadena forced townspeople to regularly attend his meetings, locals obliged
out of fear, because as one fisherman put it “our own people sold us out”.410 A similar binding
mechanism forced civilians in other communities under the BHMM’s control to wittingly become
informants. In the municipal capitals of Zambrano, El Carmen de Bolívar, San Juan Nepomuceno,
San Jacinto, Córdoba Tetón, and El Guamo, locals would automatically report the presence of
408 Interview 34, El Carmen de Bolívar, 2016.
409 Interview 32, El Carmen de Bolívar, 2016.
410 Interview 45, San Onofre, 2016.
231
outsiders in the town to the nearest paramilitary detachment for fear of being reported on by some
other resident for not doing so (Verdad Abierta 2011).
With such a system of populational control firmly established in areas under its occupation,
the BHMM enacted a strict and coercive regulatory regime of civilian communities throughout Montes
de María. All social and familial conflicts were adjudicated and resolved by the local paramilitary
commander. One former JAC president in rural San Onofre describes this dynamic: “If you had a
problem with your wife, it wouldn’t be resolved by the Police Inspector. The paras would sort it out
with weapons and they would put you to work cleaning for a week, or whatever else they felt like”.411
All local community meetings were explicitly prohibited, and numerous locals from across the region
remember that the paramilitaries established a curfew that was rigidly enforced, as from 6pm onwards
people were not even allowed to sit in front of their homes.412 Apart from these restrictions, the
BHMM frequently expropriated civilians of their personal property and the paramilitary bloc had no
particular interest in local economic development or sustaining a basic standard of living for those
civilians living under their mandate. Local agriculture stagnated due to the inability of those living in
towns and even villages to travel short distances to tend to their plots in rural areas, all of which
“generated poverty because the paras changed our livelihoods”.413 A displaced farmer from rural San
Jacinto affirms these dramatic economic changes: “Economically it was impossible to live in San
Jacinto. There was no source of employment whatsoever. The community didn’t dare try to take our
products outside the municipality. Consequently, this created a local dynamic of barter”.414
ii. State Expansion
411 Interview 46, San Onofre, 2016.
412 Interview 89, El Carmen de Bolívar, 2016; Interview 32, San Jacinto, 2016.
413 Interview 44, San Onofre, 2016.
414 Interview 41, San Jacinto, 2016.
232
In the end, the BHMM succeeded in emptying out the countryside in Montes de María, with
close to 120 000 people fleeing their rural homes for the relative safety of nearby municipal capitals
and cities (PODEC 2011). Verdad Abierta (2010) describes the effect of this displacement: “The
paramilitary expansion destroyed everything except for the guerrillas, which was supposedly its
principal objective…It destroyed the poorest families, leaving hundreds of widows and orphans with
their souls and estates in ruins. It dispossessed peasants of their land and crushed what was left of
local leadership. It asphyxiated any political renewal that was starting to come to life”. However, the
BHMM’s extensive dirty war throughout the region paved the way for Uribe’s eventual success in
expelling the insurgent groups from Montes de María. Initially, the region was identified and targeted
by the Uribe administration due to the widespread stigmatization that it was an insurgent haven. One
resident in El Carmen de Bolívar recalls that the state’s authorities claimed that in the region “of every
three men, two are guerrillas”.415 One former journalist from El Carmen de Bolívar affirms that “there
were a lot of guerrillas settled in the zone”, and that Uribe was driven by the strong electoral support
he received in Montes de María to deliver on the election promises he made “to counterattack the
insurgent groups in the region”.416 Many others highlight the fact that Uribe selected Montes de María
to be a test site for his security policy because he along with the military high command “decided to
create a strategy to show results in a zone where they could be achieved”.417
In reality, much of the massive forced displacement caused by the paramilitary incursion in
the late 1990s and early 2000s was driven by regional landed elites to carry out a counter-agrarian
reform to violently take back land won by peasant activists over the course of the 1970s and 1980s
(Duica 2013). When the state expanded its presence into Montes de María, instead of redressing these
415 Interview 31, El Carmen de Bolívar, 2016.
416 Interview 83, El Carmen de Bolívar, 2016.
417 Interview 36, Ovejas, 2016.
233
extensive land seizures, it sided with those responsible for the dispossessions by “forcing local
peasants to leave the countryside and sell their lands at a loss”.418 A doctor from El Carmen de Bolívar
says that with Democratic Security Policy, “there was more security to protect the land than the actual
community itself”.419 According to one peasant leader from Chengue (Ovejas): “It was a strategy to
demonstrate results at the international level to attract an injection of capital while enabling the
entrance of entrepreneurs with the goal of buying land. The objective responded more to the clearance
of the zone so these businessmen could enter and cultivate mega-crops like African palm and teak”.420
Ex-General Rafael Colón of the Marine Infantry’s 1st Brigade during this period deconstructs Uribe’s
intervention in Montes de María: “Uribe’s biggest challenge was to show the country that the
government was present throughout the territory and that the FARC was not going to maintain its
military capacity in places where kidnappings, illegal checkpoints, and other crimes were daily news.
There were also very strong political interests and many wealthy cattle ranchers around the Gulf of
Morrosquillo. It was a strategic zone with political and social conditions that needed to be attended
to”.421
Thus within Álvaro Uribe’s first month in office, the military’s powers were expanded
dramatically at the expense of the civil liberties and freedoms of the region’s habitants with the
introduction of a variety of new restrictions.422 During Uribe’s first year in office, his government
outlined the Integrated Strategy for Highway Security (Estrategia Integral de Seguridad en Carreteras), which
sought to counteract any threats posed by armed non-state actors or criminal groups to Colombia’s
highway system (República de Colombia 2003). In Montes de María, the most notable change was the
418 Interview 92, Ovejas, 2016.
419 Interview 34, El Carmen de Bolívar, 2016.
420 Interview 30, Ovejas, 2016.
421 Interview 97, Bogotá D.C., 2016.
422 Interview 35, Ovejas, 2016.
234
state’s instant takeover of the major highways traversing the region (Troncal del Caribe, Troncal del
Occidente) by establishing a series of heavily armed military checkpoints. Prior, travel throughout the
region was impossible due to the FARC’s ubiquitous presence on the main highways and rural roads
alike (Porch 2012). After the official authorities successfully reclaimed control of these roads, the
regional military and police placed intense restrictions against nocturnal travel, implementing a curfew
from 6pm in the evening until 6am in the morning, a restriction that would only be lifted four years
later in 2006.423 The expansion of these military checkpoints not only served to regulate traffic and
assert control over the regional highway system, but also as a means of collecting intelligence and
cutting off the insurgent groups’ access to food and other supplies. This was done through rigorous
controls placed on the amount of food, medicines, batteries, cigarettes, and portable radios that rural
inhabitants were allowed to take from urban centers back to their small towns and villages.424 At the
final checkpoints controlled by the Colombian government, soldiers maintained detailed records of
how much each family consumed every week, and strictly regulated the amount of goods and
consumables each passing civilian was transporting to their communities.425 If an individual was
carrying more than their weekly allowance of goods, then the surplus would be confiscated and
returned only the following week.426
According to a former general, prior to Plan Colombia: “Montes de María was the land of
nobody. Not the navy, not the army, not the police. Some bases were set up there but they didn't do
anything”.427 Existing military installations were located in Corazal, Malagana, Sincelejo, and Coveñas,
while the National Police maintained an ineffective presence in most municipal capitals. Following the
423 Interview 29, El Carmen de Bolívar, 2016; Interview 88, Sincelejo, 2016.
424 Interview 29, El Carmen de Bolívar, 2016.
425 Interview 40, El Carmen de Bolívar, 2016.
426 Interview 77, San Jacinto, 2016.
427 Interview 101, Bogotá D.C., 2016.
235
implementation of Plan Colombia and later Democratic Security policy, patrol bases and smaller
installations were installed in San Jacinto, San Juan Nepomuceno, el Guamo, Córdoba Tetón, La
Cansona (El Carmen de Bolívar), Chalán, Ovejas, and San Onofre. Apart from these, the Colombian
armed forces channeled greater resources into patrolling, and by extension, controlling the Magdalena
River and the Dique Canal, while also increasing security of the Caño Limón-Coveñas oil pipeline and
two communication towers located at Cerro Maco and Cerro la Pita. This increase “allowed for more
profound operations and permanent patrols which meant greater control of the territory”.428 Whereas
in 2000, the region could only count on one thousand police and military personnel, by 2004 this had
risen dramatically to eight thousand armed authorities (Cantor 2018).
As was the case of the zone of rehabilitation and consolidation in Arauca, the Constitutional
Court ostensibly prohibited the extension of the security initiative in Montes de María in early 2003
after only eights months of implementation. In reality, however, the militarization of the zone
continued throughout the remainder of Uribe’s time in office (PODEC 2011). The expansion of the
armed forces coincided with the entrance of various specialized judicial bodies to develop and
prosecute cases against suspected insurgent collaborators, most notably the District Attorney’s office
(Fiscalía General de la Nación - FGN), the Central Command of Judicial Police and Intelligence (Dirección
Central de Policía Judicial e Inteligencia – DIJIN), and a newly created Elite Group unit (Grupo Élite) whose
responsibility was to track down and detain suspected insurgent auxiliaries429 The implementation of
Democratic Security policy in Montes de María also saw the creation of the informant network (red de
cooperantes) in civilian communities to prevent and prosecute potential criminal acts, and the
development of the rewards program (programa de recompensas) to pay informants for information which
prevents “[…]terrorist attacks or the capture of members of illegally armed groups” (República de
428 Interview 97, Bogotá D.C., 2016.
429 Interview 36, Ovejas, 2016; Interview 41, San Jacinto, 2016.
236
Colombia 2003). Under this system, the potential rewards which informants could claim were based
on the military’s valuation of the targets themselves. The third measure was the renewed efforts of the
national government to induce and incentivize members of armed groups to turn themselves in and
demobilize voluntarily, as per the conditions outlined in the Presidential Decree 1228 of 2003.
Based largely in rural communities, the network met with mixed results as in some cases
informants were given military training and were converted into anonymous military guides called
‘caratapadas’ due to their faces being covered in order to shield their identities. They would in turn
accompany military patrols to rural villages during raids and point out purported insurgent
collaborators to their handlers.430 In numerous cases, peasant community leaders ended up becoming
informants themselves, a fact which devastated what remained of regional civil society during this
period due to these individuals’ expansive knowledge of local dynamics.431 A local leader from El
Salado echoes this, stating “the first thing we lost was trust in other people and in the institutions in
our own community. That’s how the social fabric deteriorated so badly”.432 Similarly, a community
leader from rural San Onofre cites the informant network as having “generated distrust within the
community, between friends and family. It produced a decomposition of the social fabric”.433
Numerous others claim the informants network was successful in their particular communities due to
the financial incentives, although many used the program to settle personal scores or to simply provide
misleading information about innocent people in order to collect the reward.434 While the network
“was an excellent strategy in itself”, it ultimately suffered from “a lack of quality control of information
to check whether it was true or false”.435 From 2002 onwards, there was a marked increase in the
430 Interview 84, El Carmen de Bolívar, 2016; Interview 35, Ovejas, 2016.
431 Interview 43, San Onofre, 2016.
432 Interview 40, El Carmen de Bolívar, 2016.
433 Interview 46, San Onofre, 2016.
434 Interview 92, Ovejas, 2016.
435 Interview 97, Bogotá D.C., 2016.
237
number of arbitrary detentions and mass arrests in Montes de María. According to official assessments,
some 328 people were captured in twenty-six separate mass arrests orchestrated by the district
attorney’s office between 2003 and 2004 in Bolívar department, of which 231 were eventually released.
Most of these individuals were arrested and detained based on the word of paid informants. Tellingly,
a majority of the accused “[…]exercised leadership and the defense of rights or fulfilled tasks which
guaranteed the subsistence and continued presence of peasants in rural zones difficult to access” (El
Espectador 2017).
While the success of the informant network remains in question, two other intelligence-
gathering initiatives proved instrumental to the Colombian military’s eventual counterinsurgent
success in Montes de María. The first of these was tied into the promotion of the desertion and
demobilization initiative promoted by the national government throughout the country.436 Despite the
serious risks and challenges associated with insurgent desertion, an act punishable by death within all
of the active insurgent groups in the region, many rank and file and even middle ranking commanders
attempted to escape. Upon surrendering to the official authorities, they were detained for up to one
month in the nearest battalion where they were extensively debriefed by military intelligence officials
(Araujo 2008).437 The second initiative was not necessarily designed as an intelligence-gathering
strategy, although it clearly became one during its implementation throughout the region. Although
numerous displaced peasant communities attempted to organize their collective returns (retornos) to
their abandoned villages and towns prior to 2002, these attempts became far more frequent under the
Uribe administration. Previous attempts at a large-scale return were almost always thwarted by the
BHMM and the FARC alike. However, the armed forces began to accompany these displaced peasant
populations back to their communities in order to protect them from any potential attacks from armed
436 Ibid.
437 Interview 87, El Carmen de Bolívar, 2016.
238
non-state actors. Following the massacre and mass displacement in Macayepo, that village was virtually
abandoned. However, most peasants found the living conditions in Sincelejo more intolerable and
sought to return. They were only able to do so by collaborating with the Colombian military to evict
the insurgent groups from their community. One peasant leader from this village describes this
process:
When Uribe became president, his government began to look for displaced peasants from Macayepo in Sincelejo and he gave us the option to return under one fundamental condition: we had to help combat the guerrillas. The Armada provided total security for us and they began to transport us in helicopters to return to our land. In light of the help provided by the government, the return began in 2004 and people from Macayepo decided to arm ourselves to confront the guerrillas, as we had a large amount of resentment and hate for them. The government paid three million pesos for every guerrilla killed in action. This prompted people from Macayepo to turn into guides because many members of the Armada hired us to tell them where the guerrillas were camped and to show them all the different routes and corridors. To serve as guides, we had to camouflage ourselves and they trained us militarily, even teaching us how to detect anti-personal mines. If we didn’t collaborate with the government, they threatened to not provide any security for the process of our return. In this context, Macayepo’s people weren’t able to leave alone to go anywhere, nor to work. We always had to be with soldiers who protected us because we were completely stigmatized as being paramilitaries and the insurgents wanted to kill us.438
The regional armed conflict changed markedly over the course of 2003 as the Armada’s
command structure experienced an important rotation. In early 2003, Admiral Guillermo Barrera
relocated his headquarters from Cartagena to the municipal capital of San Jacinto at the orders of
President Uribe. Almost simultaneously, Colonel Rafael Colón assumed control of the 1st Brigade.
Both men set out to learn from the failures of the zone of rehabilitation and consolidation initiative.
Despite the infusion of new military resources under Plan Colombia and Uribe’s Democratic Security
policy, the 1st Brigade of the Marine Infantry achieved minimal operational success against leftist
insurgent groups prior to the shakeup of the command structure. The ascendance of Barrera and
Colón in 2003 saw the two military commanders dramatically altering the focus of the Armada’s
operations by channeling the newfound military might of the 1st Brigade against both the BHMM and
the FARC with equal ferocity. Whereas from 1997 until 2003, the regional armed forces overtly
collaborated with the BHMM when conducting operations in Montes de María, from 2003 onwards
438 Interview 84, El Carmen de Bolívar, 2016.
239
the regional military command began to disrupt the paramilitary bloc’s economic networks, while
actively trying to capture and prosecute high-ranking BHMM commanders such as Marco ‘El Oso’
Tulio Pérez, who was arrested and jailed in 2003. The very visual manner in which the 1st Brigade
combated the regional paramilitary structures generated an unprecedented amount of civilian goodwill
for the military in Montes de María (Porch 2012).439 As a result of the nationwide demobilization of
the AUC under the Justice and Peace Law (La Ley de Justicia y Paz), the BHMM formally demobilized
in a ceremony in María La Baja in July of 2005.
The successful persecution of the BHMM by Colón and his troops ultimately proved
instrumental in the Armada’s subsequent battle against the remaining leftist insurgent groups in the
region. The paramilitary demobilization prompted more peasants to return to their rural communities,
albeit under the protection of the armed forces.440 From October 2003 onwards, Colonel Colón had
been waging what he called ‘the battle of Aromeras’ against the FARC’s 37 th front in an extremely
inhospitable rural zone of El Carmen de Bolívar, Zambrano, San Jacinto, and San Juan Nepomuceno
(Araujo 2008). The increased military pressure directed against the FARC’s 37th front in the Aromeras
culminated in Operation Neptune (Operación Neptuno) in June of 2004, waged by the Neptune Task
Force (Fuerza de Tarea Neptuno) with 13 counterguerrilla and intelligence organisms and a Special
Forces battallion (GILES) attempted to dismantle the Che Guevara company. However, the
insurgents had vacated the zone prior in anticipation of the operation, demonstrating the FARC’s
highly developed intelligence network in the region while underscoring the difficulty the armed forces
maintained in supplying troops with basic supplies and water in these rural expanses (Porch 2012).
439 Interview 45, San Onofre, 2016; Interview 84, El Carmen de Bolívar, 2016.
440 Interview 28, El Carmen de Bolívar, 2016.
240
According to one displaced peasant from San Jacinto, 2004 marked a breaking point from that
moment onwards “the guerrillas began to withdraw due to the high military investment”.441 The FARC
had been reduced from 500 armed combatants to 220 in a matter of years, with the remainder being
killed, captured, or demobilized, while kidnappings dropped from 131 in 2002 to 30 some two years
later (Semana 2004). Apart from the increased number of displaced peasant returning to their rural
communities, there were also other signs of greater civilian collaboration with the state authorities at
this juncture. Similar to the usage of ‘caratapadas’ in Macayepo, the Colombian Marines increasingly
used expert peasant guides (zorro-solos) to track insurgent marches and patrols (Araujo 2008). The
creation of the Joint Command of the Caribbean (Comando Conjunto del Caribe) integrated the command
of all the branches of the armed forces in the region in late 2004, providing long overdue assistance
from the Colombian army and air force to the Naval Armada’s 1st Marine Brigade, while offering
improved weaponry such as high caliber morters and mine resistent armoured vehicles. The Command
was also plugged into the Naval Intelligence Network of the Caribbean (Red de Intelligencia Naval del
Caribe – RINCA), an organ which had substantially upgraded its human intelligence, signal intelligence,
and counterintelligence after 1998. In short time, counterintelligence operations uncovered networks
of FARC infiltrators in local military structures, with some eighteen soldiers arrested as a result (Porch
and Delgado 2010; Porch 2012).
Over the course of 2005, the Colombian air force bombed southern Aromeras relentlessly
during Operation Fortress (Operación Fortaleza), while the armed forces succeeded in seizing control of
valuable geostrategic points, while also tightening the perimeter around the 37th front and installing
fixed patrol bases at principal elevations. In June, Operation Omega picked up where Operation
Fortress left off, as the Joint Command of the Caribbean carried out some 383 military actions until
441 Interview 41, San Jacinto, 2016.
241
December. Through a constant deployment of helicopter incursions, the regional military command
succeeded in assuming control of key spaces in the high mountain and the zone below, while also
destroying Martín Caballero’s principal camp which was located near Aceituno. The Command’s
strategy produced several key captures and desertions of mid-ranking commanders, all the while
tightening the noose around the Che Guevara company which protected Caballero (Araujo 2008).
The increase in insurgent desertions from 2003 onwards was crucial in improving the
Colombian armed forces intelligence capacity in Montes de María, particularly in light of the mixed
results of the informant network in the zone. Upon surrendering, insurgent deserters provided detailed
information to the official authorities regarding guerrilla structures, personnel, and the locations of
their base camps; in some cases they would even serve as guides to track down insurgent groups in
rural zones (El Tiempo 2007a). The intense control of food entering rural communities directly
affected the insurgents’ diet and morale, while the dramatic surge in military operations forced
insurgent patrols to constantly be on the move in order to avoid the incessant helicopter assaults on
their camps (Araujo 2008). One former member of the 37th front who escaped and demobilized in
2004 claims that by that point “we couldn’t eat well let alone sleep, all of which brought an intense
exhaustion”.442 From 2003 onwards, RINCA began intensely filtering and better deploying the human
intelligence it obtained, establishing a rubric known as the ‘Five Rings’ (los cinco anillos), where
informants were compartmentalized according to their relation and proximity to Martín Caballero
(Araujo 2008). The 1st Brigade began to aggressively intercept and disrupt the insurgent supply chain
from nearby cities to the guerrilla camps, thus eventually forcing guerrilla combatants themselves to
go into nearby towns to procure supplies where they would then be captured, or in some cases take
advantage of the opportunity to turn themselves in to the authorities. The improved local intelligence
442 Interview 87, El Carmen de Bolívar, 2016.
242
was also crucial insofar as it allowed the Naval Armada to develop operations to locate and claim
control of insurgent water sources scattered throughout the region, a vital resource for all actors given
the lack of rivers in Montes de María (Araujo 2008).
(Due to the increased pressure following Operation Omega) Caballero did not realize that
military intelligence officers had infiltrated the FARC’s ranks, while others had flipped Caballero’s key
contact in Medellín who frequently visited her commander in his rural camps (Las Dos Orillas 2014).
Based on the intelligence gleaned from these sources, the Joint Command of the Caribbean planned
an audacious mission, Operation Paris Lineage (Operación Linaje París), which was designed to both
rescue high value hostages and to neutralize Caballero simultaneously on New Year’s Eve of 2006.
During the surprise attack on Cabellero’s camp, six of his fighters were killed, including his girlfriend
and one of his sons; the insurgents also lost numerous armaments and other valuable supplies (Araujo
2008). In the first four months of 2007, there were substantial insurgent desertions from all armed
groups in Montes de María, many of them high ranking commanders, with 49 from the FARC, 30
from the ERP, and 11 from the ELN (El Tiempo 2007d).
Between February and July of 2007, the Joint Command had successfully captured one of
Caballero’s cousins, two of his sons, and one of his nieces outside of Montes de María and from these
RINCA had obtained vital intelligence on the FARC commander’s movements (El Tiempo 2007a;
Semana 2007a). Over the course of Operation Alcatraz, the naval intelligence organ used both
insurgent deserters and double agents to great effect. As Porch (2012) highlights, “The net result was
that guerrillas or milicianos who were killed, captured, or defected, disappeared with irreplaceable
knowledge of logistical and revenue sources” (265). With supply lines severely disrupted, the FARC
could no longer produce the requisite number of homemade mines needed to slow the advance of
Marine foot patrols, the patrols incessantly hunted Caballero and his personal security ring supported
by a fleet of helicopter gunships and artillery manned by the armed forces from strategic summits
243
throughout the zone. Over the course of the latter half of 2007, the FARC had become so desperate
due to the escalation of pressure and the dire conditions that they were fighting amongst each other
over chocolate bars.443 On October 24th, a military infiltrator in Caballero’s camp phoned in his exact
coordinates using GPS, and the Joint Force of Dissuasive Action (Fuerza Conjunta de Acción Disuasiva -
FUCAD), a newer integrated force created one year prior, was dispatched. Half an hour later an Air
Force bomber had delivered a precision strike to the camp in the rural hamlet of Aceituno (El Carmen
de Bolívar). As a result of Operation Aromo, Martín Caballero and 19 other FARC fighters were killed,
while another three wounded were quickly captured after the assault (El Tiempo 2007a; Porch 2012,
266).
With the decapitation of the FARC in Montes de María, the armed forces “achieved a
complete recuperation of territory”, as from this pivotal moment onwards “the FARC retreated and
ceased to have a presence in the zone”.444 Most insurgents still loyal to the guerrilla organization were
forced to regroup in the south of Bolívar and the Bajo Cauca region, yet some isolated fighters
remained and tried to reactivate the regional insurgency, attempts that were all thwarted by the
Colombian military. Other more astute fighters turned themselves in and demobilized (Caracol 2008a;
El Espectador 2008). While most rural inhabitants assert that the insurgent groups quickly disappeared
from their communities and were not seen again after the success of Operation Aromo, the Colombian
authorities continued to kill and capture the remnants of the FARC and the ELN in the zone until
2010, as after this juncture it appears that the 35th and 37th front relocated permanently to northeastern
Antioquia (El Espectador 2012).
Clearly the Colombian military deserves much credit for ending the armed conflict in Montes
de María, but many of the military’s more controversial methods for combatting armed non-state
actors are still remembered unfavorably by the region’s civilian population. Apart from the Naval
Armada’s clear collusion with the BHMM and the proto-paramilitary groups which preceded it,
another controversial practice employed by the 1st Brigade in the region was the generation of “false
positives” (CNMH 2009; Caracol 2009). A more common form of mistreatment perpetuated by the
armed forces throughout Montes de María with alarming frequency was the widespread stigmatization
of civilians as guerrilla auxiliaries.445 This negative behavior manifested itself in a variety of ways. In
the aftermath of the horrendous massacre in Chengue, a Marine patrol arrived on the scene the soldiers
“beat, robbed, and kicked people still in the village”.446 In another instance, a peasant recalls how an
army unit arrived in his village of Las Lajas (San Jacinto) and, after beating his son, taped a flare on
his back and told him it was a bomb while simultaneously mocking him.447 In a serious violation of
International Humanitarian Law, the armed forces refused in many instances to let international
organizations such as Doctors Without Borders and the Red Cross enter rural communities because
they thought they would aid wounded insurgents.448
The Naval Armada stigmatized and persecuted what remained of civil society in Montes de
María from 2002 onwards.449 It was only in 2003 that civic organizations began to re-appear in the
region after a prolonged hiatus due to the increased returns of displaced peasants to their rural
communities. This coincided with the creation of the Foundation Development and Peace Network
of Montes de María (Fundación Red Desarrollo y Paz de los Montes de María), an initiative spearheaded by
the United Nations Development Program (UNDP), the Mennonite Church, and the regional
diocese.450 Simultaneously, the Naval Armada began to attempt to replace and fulfill many of the
445 Interview 79, San Jacinto, 2016.
446 Interview 90, El Carmen de Bolívar, 2016.
447 Iterview 77, San Jacinto, 2016.
448 Interview 97, Bogotá D.C., 2016.
449 Interview 26, Cartagena, 2016.
450 Interview 29, El Carmen de Bolívar, 2016.
245
functions previously carried out by civic organizations in Montes de María in order to co-opt the civic
sphere and other public spaces. This conformed with one of the central goals of Democratic Security
Policy, which was for the Colombian government to work “[…]with every sector of civil society” to
promote security, including “[…]the academic sector, the private sector, with national and foreign
human rights non-government organizations, with local civic associations, and with the Church”
(República de Colombia 2003, 17). According to Esquivia Ballestas and Gerlach (2009): “[…]the
concentrated military presence in the region undermines the authority of civilian leaders and results in
the militarization of society. The military is smothering civil society by directing sports programs,
health centers, and schools and controlling the public space” (297).
Apart from its functions related to security, counterinsurgency, and public order, the military
began to link local farmers in the region with business and tourism interests in Cartegena and beyond
to “stimulate development and incentivize solidarity” by way of providing security to local
communities (Semana 2004). The Colombian government unveiled social programs throughout
Montes de María such as Families in Action (Familias en Acción), a cash transfer initiative to subsidize
poorer families based on how many children they had per household (Porch 2012). Subsequent
research has demonstrated how Familias en Acción functioned as an effective electoral tool for Uribe’s
government, one which simultaneously deepening clientelistic relations of dependency between
civilian communities and the Colombian state (Barrios González 2011; Velasco 2016). Cantor (2018)
elaborates how social programs such as these influenced local community organizations in Montes de
María:
[T]he Armed Forces sought to co-opt local social organisations into a policy of both favouring their presence and acquiring information on the guerrilla. Officers intervened directly in community meetings and social assistance programs were established to win over sectors of the community. Money was offered to youths to act as informers. Those individuals who opposed the Military presence were publicly signaled as ‘suspicious’ – i.e. suspected of guerrilla links – and numerous leaders were arrested and detained on this basis, causing several families to re-displace. On at least two separate occasions, paramilitaries entered the town with the connivance of soldiers and threatened individuals. The community became ‘intimidated and
246
divided’, with sectors that benefited from the social programs arguing that if the Military left the guerrilla would now attack them. (409-410)
By the time Martín Caballero had been killed in late 2007, regional civil society had experienced
a complete rebirth in Montes de María. Prior to the demobilization of the BHMM and the defeat of
the FARC, local civil society slowly but surely started to reappear, although under a totally different
guise. Fearful of past stigmatization, many younger relatives of former ANUC leaders took the helm
and organized victims rights groups and civic associations throughout Montes de María, albeit with
completely different names.451 Most mobilizations at the present are based in the municipal capitals
given the dramatic transformation of the rural landholding pattern in the region. Following the
massive forced displacement and the virtual evacuation of the countryside in the region, paisa
entrepreneurs and agribusinesses alike entered Montes de María following the defeat of the FARC
and bought massive tracts of land from displaced peasants at prices substantially below market value.452
Thus, the once highly active peasant civil society has now been transformed into an urban
phenomenon which has focused its demands on lack of social services and inadequate infrastructure,
and perhaps more interestingly has conformed to regional clientelist traditions in terms of their
cohesion (or lack thereof) and motives. Coinciding with the nationwide implementation of Plan
Consolidation, regional civil society became dependent on external sources for funding, thereby
causing an intense competition between community leaders for resources from these outside bodies.453
iii. The Insurgent Response
Following the FARC’s territorial consolidation in the mid-1990s, the insurgent group entered
the new millennium by dramatically escalating its armed confrontation with the Colombian state, a
dynamic which only worsened with the arrival of Álvaro Uribe to the presidency in 2002. Prior to
451 Interview 93, Bogotá D.C., 2016.
452 Interview 26, Cartagena, 2016.
453 Interview 97, Bogotá D.C., 2016; Interview 90, El Carmen de Bolívar, 2016.
247
2002, the FARC had increased its incursions into municipal capitals, kidnappings, and attacks on a
variety of state targets, most notable police stations, mayoral offices, and even against politicians and
aspiring candidates in Montes de María (El Tiempo 1997d; El Tiempo 1997e; El Tiempo 1998).
Following the implementation of Democratic Security Policy in 2002, the main insurgent groups in
Colombia all adapted and changed their military strategies from movement-based warfare to classic
guerrilla warfare. This was particularly noticeable in Montes de María, as former General Rafael Colón
describes: “The insurgents became more agile and divided themselves into smaller groups. They also
began to use anti-personnel mines, attacked helicopters lending support to troops, and their terrorist
actions grew”.454 As can be inferred from Figure 22, the FARC was the most active insurgent group
in Montes de María from 1994 until 2007, although the group’s military capacity was severely curtailed
from 2004 onwards by the Colombian military’s efforts. The ELN increased its level of bellicosity
between 1999 and 2001; however after this brief period the Jaime Bateman Cayón front never again
produced the same level of engagement.455
Figure 15. Conflict Events per Insurgent Group in Montes de María (1988-2007)
Source: Author’s Elaboration (CNMH 2009, 233)
454 Interview 97, Bogotá D.C., 2016.
455 Interview 30, Ovejas, 2016.
248
The FARC’s strategy during this period heavily depended on the extensive deployment of
improvised landmines to prevent Marine patrols from accessing the remote network of insurgent
camps in places like Aromeras, which was so heavily mined that is was referred to by the insurgents
as ‘Cambodia’ (Camboya) (Araujo 2008). The entrances to each camp were carefully protected with
these improvised devices, while these installations were not accessible by rural roads or even footpaths,
as the guerrilla fighters would literally hack a clearing through the jungle to reach them. Given the
scarcity of water in the region, especially during the dry season, rural water sources such as wells and
creeks were of enormous importance to both the military and the insurgent groups, and became a
critical flashpoint between these actors. The FARC’s fronts increasingly targeted military patrols at
these water sources by planting numerous mines in and around them in anticipation of a Marine
incursion. From 2004 onwards, the success of the Joint Command in confiscating the necessary
material to produce landmines forced the insurgents to improvise, and they thereby began to rig
“Vietnamese traps”, or camouflaged pits with a depth of two meters that were lined with sharpened
wooden stakes smeared with excrement or acid, leading to several military casualties due to the fatal
injuries and infections they produced (El Tiempo 2004).
The deaths and injuries caused by these tactics dealt a serious psychological blow to the
military’s morale, they also however led Martín Caballero to overestimate his own organization’s
military capacity (Porch 2012). By avoiding direct confrontation and skillfully deploying these
landmines, the FARC was able to stave off the military’s growing offensive until 2005, when the Joint
Command of the Caribbean’s strategy began to asphyxiate the insurgents, not only of the materials
they needed to produce landmines, but also of other basic necessities such as food and medicines.
Despite the noose being tightened around the insurgent group’s neck, the FARC was still capable of
launching lethal attacks against the regional military command, as evidenced by an ambush prepared
and carried out by the newly integrated 35th and 37th fronts in June, 2006, an operation which left nine
249
soldiers dead as they were patrolling the El Carmen de Bolívar-Zambrano highway (Araujo 2008).456
The FARC also increasingly resorted to attacks on vital infrastructure, namely electrical towers. Over
the course of the armed conflict, Martín Caballero’s group is credited with blowing up some fifty of
these, seriously disrupting the regional electrical grid in the process (Semana 2007a).
Two advantages maintained by the FARC during this period was the inhospitable terrain of
rural Montes de María, and the group’s impressive internal discipline. From 2005 onwards, Caballero
and his personal phalanx of protection provided by the 37th front’s Che Guevara commission
principally operated in the Aromeras. This zone was described by a Colombian journalist as “[…]a
completely unpopulated zone, a desert where water is practically non-existent. There are small oases,
but many were mined by the guerrillas to prevent the soldiers from supplying themselves. The
temperature exceeds 35 degrees Celsius. It is a little hell” (Semana 2007b). This was a seemingly perfect
environment for a skilled guerilla commander to operate, albeit one which would eventually be turned
against the insurgents. From mid-2005 onwards, the insurgent fronts had effectively reconfigured and
were deployed in different groups, as “[…]the troops were close but the guerrillas didn’t give up.
Caballero organized them in commandos of six units that silently spied on soldiers and mined the
terrain surrounding them” (Araujo 2008, 266). Porch (2012) highlights how Caballero and his troops
staved off being killed or captured for as long as they did:
A master of insurgency, Caballero offered no significant target to attack, only atomized assailants often dressed as civilians who could only be picked off with concise intelligence by small groups of special operators capable of moving after dark – precisely those capabilities that the Colmar lacked in 2004…Caballero kept his forces constantly circulating among well defended campsites that contained assembly and training areas, kitchens, caletas (raised, covered sleeping platforms), and water. Guerrillas were trained to hide their tracks and move in irregular patterns to discourage pursuit, and to remain aware of escape routes and rendezvous points if forced to disperse. (262)
456 The FARC’s two fronts in Montes de María merged into one from the beginning of 2006 onwards due to the dramatic reduction in troop size, and as a strategic gambit. Martín Caballero remained the top FARC commander in the region, with the 35th front’s commander ‘Manuel’ taking second in command (Araujo 2008, 298).
250
The internal discipline imposed by Caballero and all other commanders was severe. In contrast
to military patrols, which notoriously left tracks wherever they went, the insurgents carefully avoided
leaving any trace of their movements. During intense military operations, insurgents would operate in
virtual silence, whether preparing and cooking food or cutting firewood, they were extremely careful
not to produce enough smoke as to give away the location of their camps. The use of radios and cell
phones to communicate was strictly prohibited. Water tanks were carefully transported by mules to
the FARC’s remote camps, with the insurgents disguised as common mule herders (muleros), and in
the absence of these animals, ordinary fighters were used to carry these receptacles (Araujo 2008). In
order to avoid detection at military checkpoints, the insurgents and their collaborators needed to find
deceptive means of smuggling these supplies into rural communities, often moving foodstuffs around
in improvised garbage carts (zorras) to avoid detection.457 Messages sent by Caballero, whether to
subordinates in the region or his superiors elsewhere in Colombia, were dispatched by a trusted group
of confidants, most of whom were actually direct relatives of the guerrilla commander (El Tiempo
2007c). During the 2006 national elections, the FARC commander dispatched several of his more
trusted fighters to nearby municipal capitals in order to gather information on troop buildups in these
more heavily populated areas (Araujo 2008). However, as the pressure mounted in 2004 and 2005,
increasing numbers of insurgents took advantage of these supply and reconnaissance missions to
abandon their revolutionary commitment by turning themselves in to the authorities.
The most common factors behind these defections were the ever-worsening conditions. Even
high and mid-ranking commanders deserted during this period due to the severe conditions and
discipline demanded by Martín Caballero. Many of the deserters were young women who ended up
pregnant or with young children as a result of passing romantic encounters with young men who were
457 Interview 88, Sincelejo, 2016.
251
similarly enlisted in the insurgent ranks, and fearing for the future of their children turned themselves
in at the first opportunity presented itself (El Tiempo 2007e). The FARC had long regulated sexual
relations between its fighters, in particular to avoid unwanted pregnancies, and as such pregnant
females were punished for not instantly reporting a potential conception to their commanders. If
possible, these young women would be forced to undergo abortions, or in the case that the pregnancy
was too advanced, then they would be required to give up their children for adoption after delivering
them (Araujo 2008).
Many of rank and file recruits in Montes de María were undereducated or illiterate adolescents
as young as fourteen years of age, who were taken from their peasant communities and quickly
indoctrinated by their guerrilla superiors (El Tiempo 2007e). Apart from the ease with which these
impressionable peasant recruits could be molded by insurgent commanders, another benefit to the
group was the fact that these fighters felt powerless to escape as their superiors knew who their families
were and where they lived. Similar to how they treated infiltrators within the group, the FARC
maintained a zero tolerance towards deserters. Insurgent assassins were frequently deployed to
municipal capitals and beyond to locate and assassinate fleeing members of the group. In the event
that fighters successfully turned themselves in and demobilized, their family members were often tried
and sanctioned in their stead, thus dissuading potential defectors from abandoning the cause (Araujo
2008). This threat only compounded the extremely difficult terrain and the FARC’s vast network of
insurgents, militia members, and sympathetic peasants throughout the rural expanses of Montes de
María. Nonetheless, despite these daunting circumstances, hundreds of fighters ended up deserting
during the tail end of the regional armed conflict.
Insurgent relations with civilians became more coercive and repressive during the period of
contestation between the BHMM, the Colombian armed forces, and the myriad of insurgent groups
which remained in Montes de María. When regional civil society began to resurface during the early
252
years of Democratic Security Policy, it was restricted to municipal capitals and returning peasant
communities, both of which received substantially greater protection from the Colombian military. In
zones where the FARC held sway, however, the insurgent group threatened various displaced
communities from returning due to their perception that these peasants were now beholden, and by
extension, more loyal to the Colombian state than to the insurgent social order. Similarly, the FARC
prohibited international organizations such as the Red Cross and UN agencies from entering rural
communities where they held sway from 2004 onwards, as peasants in these zones were forbidden
from participating in any externally administered social projects (Cantor 2018). While the FARC’s 37th
and 35th front continued to retain the temporary loyalty of some peasants by purchasing their livestock,
in other cases their fighters began to break a cardinal rule of the larger organization by appropriating
livestock belonging to civilians in the zone under spurious circumstances (Semana 2007b). From 2005
onwards the regional insurgent groups became increasingly cut off from their remaining sources of
income from kidnapping and extortion, and as a result were far more predatory towards those civilians
which remained in communities under their control than at any other point in the past.458 This
economic pressure generated widespread resentment amongst peasant communities who had long
tolerated the insurgent groups at their own expense.459
C. The Outcome The period of contestation in Montes de María lasted roughly from the BHMM incursion in
1997 until the complete defeat of the regional insurgencies in late 2007 with the death of Martín
Caballero. The joint state-paramilitary expansion into Montes de María significantly reconfigured the
territorial distribution between all armed actors in the department. Firstly, the BHMM virtually
expelled all the insurgent groups from those municipalities adjacent to the Caribbean Sea and the
458 Interview 77, San Jacinto, 2016.
459 Interview 36, Ovejas, 2016.
253
Magdalena River, boxing the rebels into the high mountain and the zone below. The Colombian armed
forces expanded their presence substantially throughout the region by seizing complete control of the
principal municipal capitals and the major highways. The insurgents in response, retreated and fortified
their presence in rural communities in the high mountain and the zone below. Solicited by municipal
and departmental level political elites in both Bolívar and Sucre, the BHMM successfully seized control
of the majority of municipal and departmental level political institutions in Montes de María during
this period. Paramilitary influence in regional politics endured well beyond the demobilization of the
BHMM in 2005. The BHMM incursion also succeeded in wresting control of valuable drug trafficking
routes between the south of Bolívar department and the Gulf of Morrosquillo from the FARC, while
the state’s control of the region’s main highways deprived the insurgents of lucrative extortive and
kidnapping opportunities that had once been abundant.
Table 9. Armed Actor Control in Montes de María (1997-2007)
Territorial
Political
Economic
Civic
• Counterinsurgent control of municipal capitals and highways (2002-2007)
• Contested control of rural villages (1997-2007)
• Counterinsurgent control of departmental and municipal politics (1997-2007)
• Insurgent violence against politicians and disrupted elections (1997-2007)
• Insurgent control of extortion, kidnapping (1997-2004)
• Counterinsurgent control of extortion, drug smuggling, public embezzlement (1997-2005)
• Insurgent/Counterinsurgent repression of civic organizations and mobilizations (1997-2003)
• Counterinsurgent control of civic organizations and mobilizations (2004-2007)
Source: Author’s Elaboration
From 2005 onwards, the Colombian military’s efforts to encircle and asphyxiate the insurgents
began to produce results, as insurgent supply chains were severely disrupted and the shortages of basic
items negatively impacted their capacity to continue to produce highly effective landmines, while also
prompting greater insurgent defections due to weakened morale. Finally, the paramilitary incursion
into the region decimated what remained of civil society in Montes de María, forcing civic leaders and
their organizations into hiding. However, this changed in 2003 when a rash of new civic organizations
254
emerged in the region with the support of international organizations, promoting a variety of peace
building and human rights based initiatives which were tolerated and even passively supported by the
regional authorities, thus bringing the civic sphere under the protection of the Colombian state.
These separate yet interconnected counterinsurgent processes fostered the conditions for state
victory in Montes de María. The BHMM incursion succeeded in physically disconnecting civilian
populations from the insurgent groups, while also destroying whatever hard and soft support the
FARC maintained among civilian populations. As Table 10 demonstrates, there was a meaningful
demographic shift between rural and urban populations before and after the BHMM incursion into
El Carmen de Bolívar, the municipality most severely impacted by the armed conflict in Montes de
María. This municipality became increasingly urbanized over this period of time, a social phenomenon
which can be attributed to the massive forced displacement carried out by the BHMM in rural
communities throughout the municipality and the larger region (Porras 2014). Displaced peasants
tended to flee to the nearest municipal capital or city, a fact which is reflected in the immense informal
slums established by IDPs in the outskirts of El Carmen de Bolívar, San Jacinto, Sincelejo, and
Cartagena (Cantor 2018).460
Table 10. Urban and Rural Population of El Carmen de Bolívar (1993-2005)
Municipality Urban Population % of Population Rural Population % of Population Total
El Carmen (1993) 38.289 58% 27.711 42% 66.000
El Carmen (2005) 49.434 73% 18.529 27% 67.963
Source: DANE (1993, 2005)
Between 1997 and 2004, numerous villages in Montes de María that had long been stigmatized
as bastions of insurgent support suffered brutal massacres at the hands of the BHMM which were
designed to spur direct indiscriminate displacement, a wave of violence directed at civilian
communities which the FARC proved unable or unwilling to prevent. In Chengue, a BHMM incursion
failure to protect their purported social bases from paramilitary abuses engendered a deep hatred and
a widespread desire for revenge against the insurgents (Semana 2004).471 Others grew tired of the
instability of the armed conflict and the inability to live on their land in peace. For one rural habitant
in El Carmen de Bolívar: “People were tired of living with these groups and resented them. That’s
why they ended up collaborating with the state”.472 A former ERP fighter similarly cites exhaustion as
a key factor in driving a wedge between insurgents and civilians: “The civilian population grew tired
of the guerrillas because they threatened them and on occasion robbed things from them too. As a
result they started to collaborate with the government by informing in order to get rid of them”.473
The high level of stigmatization of peasants in Montes de María similarly caused problems for
civilians caught between the insurgent groups and their opponents, as “[t]he onus was on locals to
prove that they did not even passively support the insurgency”, a task which could only be proved by
collaborating with counterinsurgent forces (Porch 2012, 185). Within these regions, many peasants
perceived themselves as neutral, communities “stuck between these actors who committed abuses
against them”.474 However, from the outside “powerful stigmas were used against entire towns and
villages and everybody was branded as a guerrilla”.475 The switch in civilian collaboration only really
began to occur when Democratic Security Policy was established, as civilians were encouraged “first
by the military encirclement and then the increase in troops” and only then people “began to
collaborate and contribute in order to defeat them”.476 It became increasingly clear as more and more
civilians began to collaborate with the Colombian military that the perception that “[…]the guerrillas
flourished in Montes de María only because the people supported them” was in fact “a myth” (Porch
471 Interview 36, Ovejas, 2016.
472 Interview 84, El Carmen de Bolívar, 2016.
473 Interview 88, Sincelejo, 2016.
474 Interview 85, El Carmen de Bolívar, 2016.
475 Interview 89, El Carmen de Bolívar, 2016.
476 Interview 92, Ovejas, 2016.
258
2012, 253). While the counterinsurgents did seek active collaboration and other forms of hard support,
they also prevented many who had previously offered soft support to the insurgents to cease doing
so, thereby denying these groups of this vital source of assistance. Evans (2014) highlights the
importance of such strategic shifts in civilian support: “{Counterinsurgency} is not simply about
breaking the will of the insurgents to endure conflict but convincing the community from which this
highly localized emerges that the population’s interests are better served by acquiescence to – and,
ideally, support of {counterinsurgent forces}” (268).
As can be expected in an evolving context during civil war, the primary cleavage is susceptible
to change as the circumstances continue to develop and transform. In such contexts, previously
hardened attitudes and grievances give way to more primordial concerns, as Galula (1964) notes:
“[…]the population’s attitude in the middle stage of the war is dictated not so much by the relative
popularity and merits of the opponents as by the more primitive concern for safety” (14). In Montes
de María, the primary cleavage changed somewhat. Although it was still influenced by peasant access
to land, it shifted from the question of ownership to one of occupation and return. Victims of
paramilitary displacement desired more than anything else to return to their homes. Virtually all
accepted that the only way for this to happen was the expulsion of the insurgent groups, and the only
clear-cut way for this to be achieved was by supporting the counterinsurgents, or at a minimum,
ceasing to obey the insurgents. In such conflict dynamics where civilians are not intimately bound to
robust insurgent social orders, most individuals prioritize “their survival and that of their children”
over everything else (Evans 2014, 269). Montes de María proved no different as the reconfiguration
of the primary cleavage ended up uniting most civilians, regardless of their socio-economic position.
For displaced peasants living in the region’s municipal capitals, the Colombian military ended up being
259
the best provider of protection, particularly after the armed forces began to persecute the BHMM as
vigorously as they confronted the FARC and other insurgent groups from 2003 onwards.477
Perhaps the best example of how the Colombian military bound civilians in Montes de María
to the counterinsurgent cause was through the accompanied returns to previously abandoned villages.
Certain brave groups of individuals attempted returns prior to 2003, but it was only really when the
Marine Infantry and other state authorities actively began accompanying these returns that displaced
peasants were able to permanently re-settle their communities. Ironically, the peasant returns ate away
at what remained of the insurgent groups’ civilian support, while shrinking the physical space available
for them to operate in. The efficacy of the returns varied depending on the village, yet this strategy
was successful for the counterinsurgent cause insofar as it forced civilian returnees to depend on the
Colombian military and the National Police for their security, and by extension their livelihood, thus
further driving a wedge between the peasantry and the insurgent groups. The most successful of these
returns occurred in Chinulito, where, similar to Macayepo, the National Police and a group of peasant
soldiers “[…]set geographical limits around the town within the confines of which it ‘guarantees
security’. The Public Force has told community members that they should ask for permission to go
outside these limits – beyond which most of the Chinulito campesinos’ lands lie – and troops will
accompany them” (Cantor 2018, 398).
Rather than being a case of the Colombian state “winning hearts and minds”, the returns were
largely driven by necessity and coercion. The peasant returnees grudgingly accepted the military’s
accompaniment, but complained about the abusive tactics employed by the counterinsurgent forces.
In rural villages like El Salado, Las Palmas, and Pijiguay, the military willfully ignored judicial rulings
from national-level courts and directly co-opted local civic organizations and their funding, detained
477 Interview 41, San Jacinto, 2016.
260
and jailed dissident civic leaders, damaged civilian property, and violently threatened returnees who
would not serve as informers (Cantor 2018, 409-410). One former peasant leader from El Salado
comments that: “Even after having returned, the military continued to engage in selective
assassinations and arbitrary detentions against the community”.478 The FARC’s coercive attempt to
recover control of territory and people left behind by the BHMM demobilization saw an increase in
insurgent violence against purported counterinsurgent collaborators, a wave of selective violence
which only served to further drive civilians into the military’s arms given that it was the only option
available to them (Araujo 2008).
The massive displacement of the regional peasantry gave the Colombian military an
opportunity to increase its control over territory and people at the expense of the FARC, a shift in the
balance of power that Nagl (2002) argues is devastating to insurgent fortunes: “Once the local and
regular armed units are cut off from their sources of supply, personnel, and, most importantly,
intelligence, they wither on the vine or are easily coerced to surrender or destroyed by the security
forces with the aid of the local populace” (28). Perhaps more importantly, the sudden shift in military’s
relations with the BHMM under Colonel Colon and Admiral Barrera “generated trust and credibility
in the Armada”, while displacing the paramilitaries from their strongholds, co-opting the spaces they
controlled, and encircling the FARC in the broader region.479 As Kilcullen (2010) notes: “Like any
opponent in any war, an insurgent enemy needs to be pinned against an immovable object and “fixed”
in order to be destroyed” (9). Of equal importance, Long (2006) suggests that “[…]successful COIN
operations appear easier in isolated battlefields”, an observation that events bore out in Montes de
478 Interview 40, El Carmen de Bolívar, 2016.
479 Interview 84, El Carmen de Bolívar, 2016; Interview 94, Bogotá D.C., 2016; Interview 97, Bogotá D.C., 2016.
261
María because of the Colombian military’s ability to deny the insurgent groups an effective rearguard
by cutting off their mobility corridors (8).480
This territorial isolation further weakened the FARC as the group “[…]was increasingly
reduced to the unpopulated region, arid and mountainous, far from urban centers and from their
historic social base” (Semana 2007b). From 2004 onwards, the encirclement of the FARC tightened
around the “[…]isolated and resource starved” Aromeras, as Porch (2012) notes: “For Caballero, the
Aromeras became a prison with the Colmar guarding the flank of the Magdalena and denying him
population centers” (268). The region’s freshwater sources were limited to a series of creeks and
streams which evaporated during the dry season, and the BHMM and the armed forces had long cut
the insurgents off from the Magdalena river, leaving them extremely vulnerable during extended
droughts (Araujo 2008). The military “slowly began seizing control of rural water sources”, and the
Armada’s control of the main highways and access roads similarly “countered the flow of food
supplies, arms, and munitions to the guerrillas, while hampering drug trafficking in the region”.481 A
former FARC fighter recalls that: “We had to smuggle small amounts of food through different means
because the controls were rigorous”, while another peasant from rural San Jacinto remembers that
during this period “the military were in the village permanently and the guerrillas were starving to
death”.482 Towards the end of the regional conflict, a high-ranking FARC commander deserted and
turned himself in “after three days in the hills without eating, yet he had three million pesos in his
pocket”.483
The worsening conditions and increased military pressure against the insurgents in Montes de
María from 2004 onwards led to increased guerrilla desertions. John Nagl (2002) suggests that
480 Interview 38, Sincelejo, 2016.
481 Interview 101, Sincelejo, 2016.
482 Interview 87, El Carmen de Bolívar, 2016; Interview 77, San Jacinto, 2016.
483 Interview 38, Sincelejo, 2016.
262
desertions are valuable to a counterinsurgent cause: “In a battle against insurgents, persuading fighters
to surrender and provide information on their comrades is much more effective than killing them”
(93). Years of provisional predation in rural communities generated modest insurgent populational
control, and the FARC in Montes de María did not possess sufficient levels of internal cohesion and
commitment to commit its fighters to high levels of risk and adversity. As such, “[t]his made them
vulnerable to desertion, facilitated by the proximity of towns like Carmen and incentivized by the
expansion and publicizing of a demobilization program” (Porch 2012, 268). The tightening
encirclement of insurgents in the Aromeras, coupled with growing desertions and the shortages of
munitions and other components needed to prepare land mines both weakened morale and the
FARC’s capacity to organize sufficient defensive military measures.484 Increasingly, insurgent deserters
provided more and more valuable information to Colombian military intelligence, leading to the
captures of several of Martín Caballero’s family and inner circle, ultimately enabling the Joint
Command of the Caribbean to locate and decapitate the FARC leadership during Operation Alcatraz
in late 2007 (Araujo 2008; El Tiempo 2007a).485 After the death of Caballero, the remaining insurgent
troops promptly fled the region for good, representing a substantial insurgent defeat in the broader
Colombian armed conflict.
484 Interview 77, San Jacinto, 2016.
485 Interview 87, El Carmen de Bolívar, 2016.
263
Chapter Six. Period of Contestation in Arauca
Sonny: That's what it comes down to, availability. The people in this neighborhood who are on my side, who see me every day, they feel safe, because they know I'm close. And that gives them more reason to love me. But the people who want to do otherwise, they think twice, because they know I'm close. That gives them
more reason to fear me.
Colagero: Is it better to be loved or feared?
Sonny: That's a good question. It's nice to be both but it's very difficult. But if I had my choice, I'd rather be feared. Fear lasts longer than love. Friendships that are bought with money mean nothing. You see how it is
around here, I make a joke --everybody laughs. I know I'm funny, but not that funny. It's fear that keeps them loyal to me. But, the trick is not to be hated. That's why I treat my men good, but not too good. I give them too much they don't need me. I give them just enough where they need me, but they don't hate me.
- A Bronx Tale (1993)
A. Evaluating Counterinsurgency Outcomes in Arauca
The counterinsurgent strategy that was largely successful in Montes de María met with more
limited success in Arauca. There, any illusion of progress was based upon the initial achievement of
the Colombian army and the paramilitary Bloque Vencedores de Arauca (BVA) in reclaiming key spaces
in the plains, namely in Tame and Arauca municipality. In contrast, the state’s expansion into the
piedmont neither reduced violence, nor weakened the insurgent groups’ control over territory and
people living outside of the municipal capitals. As Figure 16 demonstrates, Arauquita and Tame were
marred by extremely high homicide rates between 2002 and 2010, a level of violence which one would
expect to find in a fiercely contested micro-level war zone. In contrast, the homicide rates in Arauca
municipality declined noticeably over the same period of time, representing a notable improvement in
security. Arauquita and Tame experienced extremely high incidences of armed actions and
confrontations between 2002 and 2010, similar to their respective homicide rates per every hundred
thousand residents. Arauca municipality maintained a comparatively low level of armed confrontation
during the same period. Again, these descriptive statistics confirm the variation of outcomes between
264
and within these cases, with Democratic Security Policy being successful in Montes de María, partially
successful in the Araucan plains, and largely unsuccessful in the Araucan piedmont.
Figure 16. Homicide Rate per 100.000 Persons in Arauca (1990-2013)
Source: Author’s Elaboration (Data from Observatorio del Programa Presidencial de Derechos Humanos y Derecho Internacional Humanitario de la Vicepresidencia de la República)
Of equal importance, all three of the municipalities in question in Arauca were deemed to have
been “strongly affected” by a “high intensity” level of armed confrontation between opposing actors
during this period, and the conflict was “persistent” or “permanent” in Arauca. These classifications
conform for the most part with my findings, with the notable difference being the characterization of
Arauca municipality, and by extension the plains sub-region, as a zone which continued to be a high
intensity conflict zone after Uribe’s tenure in power. As this chapter demonstrates, the guerrillas lacked
any comprehensive presence within Arauca municipality after 2005. However, they still supposedly
maintained active militia members in urban centers and rural towns of geostrategic value. According
to my sources, any violent actions committed by either the FARC or the ELN in Arauca municipality
265
from 2005 onwards were the result of incursions made by these groups into and around the capital
from neighboring Arauquita or Venezuela.486
Figure 17. Armed Actions and Confrontations in Arauca (1998-2013)
Source: Author’s Elaboration (Data from Observatorio del Programa Presidencial de Derechos Humanos y Derecho Internacional Humanitario de la Vicepresidencia de la República)
In the plains, virtually every interview participant recalled how the insurgent groups were
quickly displaced from the sub-region when the BVA arrived in 2001. A former victim from rural
Arauca municipality remarked that: “When the BVA arrived they came with the military and between
the two they finished the guerrillas off, there were only five to ten guerrillas and the paras and the
military had hundreds and therefore they couldn’t confront them”.487 According to a local journalist
in Arauca City, prior to 2002 the departmental capital was dangerous due to frequent shootings,
bombings, and kidnappings. After Uribe, “there was greater tranquility, the army strengthened, there
were a lot more enemy captures and less massacres”.488 Despite the initial gains of the counterinsurgent
efforts in the plains, various participants acknowledged that the insurgents returned to their
486 Interview 47, Bogotá D.C., 2016.
487 Interview 57, Arauca municipality, 2016.
488 Interview 4, Arauca municipality, 2016.
266
communities following the demobilization of the BVA in late 2005. One local from Cravo Norte
noted that before Uribe “the guerrillas were seen clearly throughout the plains”, but after “they weren’t
seen in uniforms or with arms, they operated in another way”.489
Table 11. State of Conflict and Outcome per Municipality in Arauca (2000-2012)
Municipality State of Conflict Intensity Result
Arauca Municipality Permanent Conflict High Intensity Strongly Affected & Persistent
Arauquita Permanent Conflict High Intensity Strongly Affected & Persistent
Tame Permanent Conflict High Intensity Strongly Affected & Persistent
Source: CERAC Typology by Municipality of the Armed Conflict
By contrast, most if not all interview participants in the piedmont roundly rejected the idea
that Democratic Security Policy had achieved its objectives in the sub-region. One civil servant from
Saravena remarked: “If Uribe’s strategy had been successful, the FARC and the ELN would have
ceased to have a presence in Fortul, Arauquita, Tame and Saravena, but that wasn’t the case”.490
Another peasant leader from rural Arauquita claims that: “During Uribe’s time the guerrillas
maintained the same presence in the village. When Uribe was campaigning he promised that in six
months the situation and security were going to get better, but that was a lie. On the contrary it got
worse”.491 In Fortul, a local business owner commented that while there was an increase in police and
military presence, “the guerrillas never lost social or territorial control in general terms”.492 A Tame
native was slightly more optimistic, opining that the state expansion “slightly reduced the scourge that
the insurgents had created, but in the end the strategy failed to remove the guerrillas”.493
B. Short-term COIN Success in the Plains, Insurgent Resilience in the Piedmont
the Venezuelan border. The paramilitaries took them to a nearby interrogation and torture chamber
manned by the BVA. After accusing the men of being members of the ELN, the commander killed
two of the brothers in front of the others. He then proceeded to force one of the surviving brothers
to fight one of his foot soldiers, and when his subordinate began to lose the physical contest, Martín
executed the defenseless victim. The paramilitaries proceeded to hang the fourth and final brother
from a nearby tree, before finally descending upon the three other detained men with knives and
machetes. After killing them, Martín and his men carved ‘ELN’ in their chests, before finally
dispensing of their corpses by the side of the unpaved highway leading to the municipal capital for all
to see (Verdad Abierta 2009).504
The BVA frequently targeted politicians, community leaders, JAC presidents, journalists,
teachers, businessmen, and municipal civil servants. Upon taking over Arauca municipality, the
paramilitary block unleashed a wave of selective killings against any public critic of previous
paramilitary efforts in the department, most notably gunning down the well-known journalist, Efraín
Varela, in broad daylight on July 28th, 2002.505 Apart from more conventional arms, the BVA’s methods
also included the use of chainsaws, while also incorporating wild animals into their repertoire, as the
group employed venomous snakes, attack dogs, and even caimans, a breed of alligator specific to
tropical environments in South America, to torture and kill its victims (El Tiempo 2012a).506 The BVA
further distinguished itself from its insurgent rivals through sexual violence. The paramilitary group
committed systematic violations of this nature against both men and women in Arauca during the four
years that it operated in the larger department. One survivor from Caracol recalls a disturbing case of
such violence:
The paras killed my godfather - the only homosexual in our town and the JAC president - along with his eleven year-old stepson whom he was responsible for. They killed them both for fun. At first, they attempted
to force my godfather to rape his stepson and when he resisted, the paramilitaries raped and killed both of them. They strangled the child and after they hung them both so it looked like a suicide.507
Given the armed group’s composition principally consisted of young males from outside the region,
combined with the top-down ethos of pillage and plunder which was embodied by the bloc’s
commanding structure, it became common practice for paramilitary patrols to kidnap young women
from civilian communities and take them back to their rural bases where they would then physically
and sexually abuse them before taking their lives (Verdad Abierta 2012b).508
The BVA planned, coordinated, and carried out their expansion, consolidation, and military
operations closely in concert with the Colombian military and National Police detachments
throughout the department. Paramilitary foot soldiers operated numerous roadblocks and checkpoints
on major highways, often for months at a time. Those civilians travelling through the plains who were
stopped and deemed suspicious by the BVA patrolmen due to the origin of their identification card
or some other arbitrary indicator, were taken out of their vehicles and executed by the side of the
road.509 Purportedly, in the municipal capitals of Tame and Cravo Norte, the “paramilitaries and the
police would conduct patrols side by side”.510 Further contributing to local confusion was the fact that
local detachments of the Colombian armed forces would often disguise themselves as paramilitaries
in order to carry out illegal operations.511 Whereas the state authorities needed the paramilitaries to
carry out illegal tasks which they could not be publicly seen engaging in, the BVA needed “the
information that the army provided them with in order to expand and advance to new territory”.512
As many civilians found out the hard way, reporting official military misconduct or paramilitary crimes
to the proper authorities all but guaranteed that the BVA would find out, almost always resulting in
the deaths of those well-intentioned citizens.513 During this time, numerous selective assassinations
and massacres were conducted effectively as joint operations, with the official authorities detaining
and profiling civilians first, after which they would depart and the paramilitaries would arrive
instantaneously and kill those suspected of being insurgent collaborators.514 One teacher in Tame
remembers a particularly illustrative example of the state’s collusion with paramilitary fighters:
There was a woman walking in the main square who two paras mistook for somebody else they wanted to eliminate. One of them jumped off their motorcycle, grabbed her by the hair and put a gun to her head, before realizing that she wasn’t the intended target. They took off and the woman, who was quite shaken up, went to the main police station in order to report what had just happened. When she entered the police commander’s office, he was sitting there relaxing with the two men who had almost just killed her. She fled Tame the following day.515
Despite the very public way in which the BVA and state agents interacted in Arauca, the
former was still responsible for funding its own operations and predictably the block was heavily
involved in the departmental drug trade from the moment it entered the zone.516 The BVA arrived in
Arauca at precisely the same moment as the FARC-led coca boom in the piedmont, and the BVA and
the insurgent groups clashed most frequently in places along the piedmont-plains border where this
crop was cultivated in large quantities.517 Aside from seizing important cocaine production and
transportation routes in the plains, the BVA also levied heavy protection payments against all local
merchants and cattle ranchers to subsidize its income from the drug trade. However, the BVA was
excessively predatory compared to its insurgent competitors in this regard, as the armed group’s
exorbitant demands of rural landholders and urban merchants alike led many to ruin. In many cases,
Figure 20. Forced Displacement in Arauca (1990-2013)
Source: Author’s Elaboration (Data from Observatorio del Programa Presidencial de Derechos Humanos y Derecho Internacional Humanitario de la Vicepresidencia de la República)
Even though the BVA had established a well-developed network of bases throughout the
Araucan plains, the paramilitary block’s numerous attempts to make inroads into the piedmont met
with little success. Whereas neither the FARC nor the ELN were able to offer much resistance to the
paramilitary incursion in the plains, the insurgents ferociously rebuffed their counterinsurgent
opponents when they attempted to enter the piedmont. The dense surveillance networks long
established by the insurgent groups throughout the sub-region helped them detect any potential
attempt by counterinsurgent forces to infiltrate these communities, overtly or covertly, further binding
insurgents and civilians together in the defense of their territory.528 The FARC and ELN cooperated
to establish a security cordon along the piedmont-plains border in Tame and Arauquita and waged
several intense battles with the BVA in towns located on this divide such as La Cabuya, Betoyes, and
Corocito (El Espectador 2014).529 Unable to enter the piedmont north of Tame, the BVA resorted to
three-quarters of the total U.S. military aid allocated to the department (USGAO 2005). In very little
time, the security at Arauca’s main oil producing facility was upgraded and it “[…]became a highly
technological center of protection and strict control of everything that moved near its perimeter” (El
Espectador 2014). The sudden technological advantages given to the Colombian military in Arauca
over its insurgent rivals were immense. According to the former commander of the 18th Brigade:
A fundamental factor of this technological benefit had two aspects: one, in intelligence with the electronic means to intercept communications and all of that, and two, the capacity to engage in combat at night. The guerrillas already knew how the army acted. They knew that helicopters and airplanes weren’t able to fly after 6pm, so they carried out all their actions at night - incursions, the movements of large columns of fighters, attacks on towns - because they knew that the army wouldn’t react because it didn’t have the means to. Night combat technology was acquired, night vision equipment so helicopters and planes could fly at night and it completely changed the balance of combat. Attacks against the oil pipeline were reduced because we in Arauca had the support of a plane crewed by intelligence agents who watched over the pipeline at night, who would patrol the pipeline and look where there were suspicious people and they would inform troop patrols by radio in real time. These would take off in a helicopter and arrive at the site in question in a few minutes. Our reactive capacity increased significantly.545
The troops responsible for reacting to potential attacks on oil infrastructure on short notice generally
belonged to the Mobile Brigade No. 5, all of which were trained by the U.S. military advisors
dispatched to Arauca. Arriving in January 2003, some sixty-eight U.S. Special Forces instructors
divided between the main battalions in Arauca City (thirty advisors) and Saravena (thirty-eight
advisors) taught counterinsurgency tactics and psychological warfare operations to some 1600
Colombian army regulars by 2005 (USGAO 2005; Defensoría del Pueblo 2003).
The sharp rise in attacks on the Caño Limón-Coveñas oil pipeline in 2001 prompted regional
military commanders to examine ways in which they could combat these acts of sabotage. For
example, from the opening of the pipeline in 1985 to 2001, only three people had ever been prosecuted
for some eight hundred attacks in Arauca. The former commander of the 18th Brigade reflects that
“for the most part these occurred because there was total impunity, none of these terrorist acts, the
deaths, the losses, absolutely nothing had been investigated by the judicial authorities”.546 Other
545 Interview 47, Bogotá D.C., 2016.
546 Interview 47, Bogotá D.C., 2016.
285
sources explore the source of this impunity, highlighting the fact that local prosecutors were from
Arauca and “like the rest of the population, they were either scared of or co-opted by the guerrillas”
(Semana 2003). As a result, the Attorney General’s Support Structure (Estructura de Apoyo – EDA) was
formed, a group of district attorneys dispatched from Bogotá who stayed on base with the army and
accompanied the troops when they conducted military operations throughout Arauca. With the
military intelligence gained during these operations, the Attorney General’s office and the EDA began
to prepare and prosecute cases against the responsible parties on a hitherto unprecedented scale.547
From the onset of Uribe’s first term in office, the Colombian government simultaneously
promoted the recuperation of regional political institutions and in order to achieve this objective in
Arauca, the Colombian government and military would need to find a way to break and permanently
disrupt the FARC and the ELN’s longstanding hegemony of departmental politics.548 As a former
elected representative from Tame points out, during this period “the ELN had absolute control of the
governorship”.549 After a team of legal investigators from the District Attorney’s office arrived in
Arauca in mid-2001, they uncovered the vast infiltration of the insurgent groups into municipal and
departmental politics (Garay Salamanca et al. 2017). Over the following two years, this specialized unit
of investigators and the regional military developed ‘Operation Dignity’ (Operación Dignidad), which
quickly led to the judicial disqualification in 2002 of the then-governor, Hector Gallardo, a vacancy
which Uribe quickly filled through a series of political appointees, some of whom were ex-military
officers, until the following gubernatorial elections in late 2003. The political turbulence of this period
was punctuated by the assassination of the mayor of Puerto Rondón, and the successive resignations
547 Ibid.
548 Interview 47, Bogotá D.C., 2016.
549 Interview 72, Tame, 2016.
286
of the mayors of Saravena, Arauquita, and Fortul that same year due to death threats from various
armed actors (Gómez Rivas 2016).
The departmental political reconfiguration was achieved in part by the extreme violence
unleashed against the traditional political class in Arauca and the Colombian state’s ability to judicially
interfere in the electoral proceedings and to harass and persecute those not aligned with Uribe’s project
into quiescence or exile. During the 2003 gubernatorial elections, it appeared that Helmer Muñoz, a
well-known piedmont-based priest and critic of Uribe, might win the governor’s seat over Julio Acosta
Bernal. Alarmed, the national government intervened in the elections to alter the outcome “[…]by
arresting just five days before the election, virtually all the major contenders for governor and Arauca
{municipality} executive that were not Acosta allies or Acosta himself, accusing them of links to the
guerrillas” (Carroll 2011, 247). As seen with the case of Hector Gallardo, Uribe frequently exercised
his right to make appointments to vacant posts during this period, invariably filling these vacancies
with military appointees.550 According to one high-ranking oil union leader at the time, “between 2002
and 2004 the armed forces governed Arauca more than the mayors did”.551 In 2006, Uribe won the
support of five of Arauca’s seven municipalities in that year’s presidential elections (Arauca
municipality, Cravo, Norte, Puerto Rondón, Tame, Saravena), an electoral shift that was mirrored by
the success of those parties supportive of his government, most notable the Radical Change (Cambio
Radical) and the Citizen Convergence (Convergencia Ciudadana) parties (Gómez Rivas 2016).
The increased protection afforded to oil infrastructure in Arauca represented a greater resolve
on the part of the national government to sustain departmental oil production at a time when oil prices
were rising fortuitously, having doubled between 1998 and 2001 and continuing to climb.
Furthermore, new exploration concessions had been granted to foreign energy multinationals
550 Interview 6, Arauca municipality, 2016.
551 Interview 8, Arauca municipality, 2016.
287
throughout the department, and the ability to develop these potential reserves depended entirely on
massive improvements in local security conditions (Carroll 2011).552 Under Uribe, this prerogative
became even more clearly defined as Occidental Petroleum’s Caño Limón concession was extended
in perpetuity in 2004.553 The imperative to protect Arauca’s oil fields and pipelines was partly driven
by the fiscal demands of the national government, but Uribe also sought to deny the insurgent groups
continued access to these extortive revenues. In a bid to further weaken their economic networks,
Uribe stripped control of Arauca’s oil revenues from the governor and local mayoralties in early 2003,
assigning the National Commission of Royalties this responsibility instead (El Tiempo, 2003c).
This controversial move was not without basis, as according the Anti-Corruption unit of the
Attorney General’s office, of the $1.3 billion USD that Arauca received in oil royalties between 1988
and 2003, at least $390 million of these were directly funneled to the ELN and the FARC in Arauca,
while the former group purportedly exercised control over the remaining funds by way of regulating
departmental budgets and the commission of contracts (Semana 2003; Duque Daza 2017). In an effort
to cut the armed groups off completely from collecting their quotas from public contracts, Uribe
assigned the construction of the $21 million USD Tame-Arauca highway to the Colombian army’s
Engineer Battalion, stating emphatically that “[…]we aren’t going to allow one peso more to be taken
by the corrupt or to {get into the hands of} the guerrillas” (Semana 2003). However, according to one
civil servant in the region “the money was handed over to the army and the highway ended up costing
triple the projected amount and was never fully completed by the army”.554
Another key source of illicit finance for Arauca’s insurgent groups, mainly the FARC, was the
cultivation of coca and drug trafficking. By the new millennium Arauca had become a relatively
The efficacy with which the BVA entered and established control of the Araucan plains was
the result of both the FARC and the ELN’s weak embeddedness in the sub-region, coupled with the
brutality of the counterinsurgent tactics employed. Similar to Montes de María, the insurgent groups
in the plains had intensified their persecution of large landholders and cattle ranchers over the course
of the 1990s, all of which fueled the stigmatization of local peasants as guerrilla collaborators by landed
elites and the Colombian military, further exacerbating an already highly polarized social dynamic in
the process. This stigmatization of civilian communities in the Araucan plains resembled Tilly’s (1999)
observation on extreme nationalism, insofar as rural elites were driven by “[…]claims to prior control
over a state, hence the exclusion of others from that priority. It authorizes agents of the nation to
subordinate, segregate, stigmatize, expel, or even exterminate others in the nation’s name” (172).
Already pre-disposed to the extreme paramilitary ethos of “re-stating the state”, landed elites threw
their considerable support behind a veritable paramilitary project when the AUC presented this
opportunity to them in the early 2000s, yet many of them ended up victims of the BVA (Gutiérrez
and Barón 2005).
In an unfortunate twist of irony, the stigmatization of the region’s peasantry as a common
enemy by landed elites and the Colombian military overshadowed a contrasting reality, one in which
local peasants had seen little to no actual gains under their predatory insurgent guardians. Upon their
arrival in plains communities, the FARC and the ELN found multiple barriers and few anchors to
develop a robust insurgent social order. The society comprised a highly atomized and dispersed
peasantry with little organizational capacity that was also heavily dependent on the massive cattle
ranching estates for their very livelihoods. Similarly problematic was the fact that the primary cleavage
between creole settlers and indigenous groups had lost most of its importance over the latter half of
the 20th century and thus there was no widespread peasant grievance to appropriate. Additionally, due
309
to the overwhelming dominance of the insurgents in co-opting all departmental and municipal political
posts from the late 1980s onwards, plains based insurgent structures were spared from having to bring
out the vote in peasant communities under their control. Faced with these conditions, the FARC and
the ELN quickly resorted to provisional predation upon their arrival in the plains, prioritizing
territorial mobility and economic resources over governance, security, and order. Thus, when these
groups were confronted by the BVA and the Colombian military, they unsurprisingly retreated to
Venezuela or the piedmont, leaving their former civilian charges to pay the costs of having tolerated
rebel rule for so many years.
Table 12. Armed Actor Control in the Araucan Plains (2001-2010)
Territorial
Political
Economic
Civic
• Counterinsurgent control of rural villages, highways, and municipal capitals (2001-2005)
• Minimal insurgent control of rural villages (2006-2010)
• Insurgent control in Apure (2001-2010)
• Counterinsurgent control of departmental and municipal politics (2001-2009)
• Counterinsurgent violence against leftwing politicians (2001-2005)
• Counterinsurgent control of contraband, coca, and extortions (2001-2005)
• FARC control of contraband, coca, and extortions (2005-2010)
• Counterinsurgent repression of civic organizations and mobilizations (2001-2005)
Source: Author’s Elaboration
As Timothy Wickham-Crowley (2015) highlights, the most important public good provided
by any aspiring non-state ruler is “[…]to protect the populace from both internal and external
violence” (65). The long-term viability of armed groups to rule communities is heavily dependent on
their ability to protect their social bases from external predation (Kalyvas 2006; Mampilly 2011;
Metelits 2011). When the insurgent groups - principally the FARC - returned to occupy the plains in
the aftermath of the BVA’s demobilization in order to assume control of local informal economic
activities (e.g. extortion, drug trafficking, and contraband), they did so in a substantially less visible
and conspicuous manner than before due to the lack of any meaningful civilian support for their
renewed presence. Similar to Montes de María, the case of the plains demonstrates how the efficacy
310
of the paramilitary dirty war created the necessary conditions on the ground for the short and long-
term success of counterinsurgent outcomes in these sub-national theatres of war (Gutiérrez-Sanín
2019). Even though the BVA expelled the FARC and the ELN from the plains, state expansion
ultimately proved unable to permanently remove the insurgent groups in the long-term, thereby
producing this particular counterinsurgency outcome (state advantage-insurgent disadvantage). This
result owes more to the spillover effects from the piedmont insurgencies and the short-termism of
the Colombian military’s strategy in the plains, yet still raises several questions about the overall
efficacy of Democratic Security Policy in the Araucan plains.
When the period of contestation began in the piedmont, both the FARC and the ELN oversaw
a complex and exceptionally robust system of nested governance throughout the sub-region.
However, this strength did not necessarily mean that the entire civilian population cared for,
supported, or appreciated the presence of the FARC and ELN in their day-to-day lives. Rather, the
pervasiveness of the insurgent rule locked civilians into their social order, regardless of their personal
beliefs or sentiments towards the dual insurgencies. According to a former civil servant from Tame,
peasant-insurgent relations during this period were characterized by alignment, displacement, and
mortality, representing different logics based on coercion, affection, and/or material gain:
The guerrillas embedded themselves deeply in piedmont society and a relationship was formed based on submission, collaboration, and fear. The population has always had a degree of familiarity with these groups because they have family members, friends, and acquaintances who belong to them. The guerrillas did good things such as prohibiting the use of dynamite for fishing and they also legalized marriages. But they also created a lot of suffering, they became instruments to kill people.631
These relations oscillated between “domination and cooperation”, yet were “determined more
by armed pressure than by the conscience. Civilians are motivated by an instinct to survive”.632 Of
equal importance, peasant communities in the piedmont possessed “greater ties to the insurgent
groups because everybody had friends and family members who belonged to them”.633 The FARC and
the ELN “usurped the state’s functions” and for this reason “the population has applauded the
subversive groups”. The guerrillas regulated and structured local communities and “supported their
petitions and demands while the state promoted prison and exile”.634
The wholesale insurgent co-optation of the civic sphere enabled these groups to capture
political power in the department, regulate order in local communities, while also protecting the
peasant smallholder system from external predation (Peñate 1998; Duque Daza 2017; Larratt-Smith
2020). The FARC and the ELN’s control of community actions boards, peasant cooperatives, trade
guilds, and other civic organizations provided excellent monitoring and enforcement mechanisms,
yielding broad networks of local information. Virtually all of these organizations were infiltrated to
the extent that the insurgents “essentially knew everything that occurred inside of these groups”, and
were thereby able to regulate their activities. Crucially, insurgent control of local production, trade,
and commerce engendered a particular dynamic where peasants could either support the insurgent
project and share in the benefits – whether providing intelligence or voting for a certain candidate -
or merely remain obedient and receive nothing.635 As one religious leader in Arauquita describes this:
“To be neutral was difficult. It was to opt for a very poor life and one without many opportunities.
The guerrillas banded together and co-opted everything and the population was forced to sympathize
in some form or another to eat or to work”.636 In binding civilian livelihoods to support for the
insurgent social order, the FARC and the ELN guaranteed that the local peasantry was going to be
633 Interview 49, Arauca municipality, 2016.
634 Interview 50, Arauca municipality, 2016.
635 Interview 49, Arauca municipality, 2016.
636 Interview 59, Arauquita, 2016.
312
stigmatized by the state, as local civilians “had to bow their heads to the insurgents because they
depended on them”.637
Contrary to the paramilitary incursions in Montes de María and the Araucan plains, which
created very favorable conditions for the Colombian military to pry civilian support away from the
established insurgent social order, the BVA enjoyed no such success in the piedmont. The insurgent
commitment to protecting the piedmont, reinforced by the wholesale support of the sub-region’s
civilian population for repelling the paramilitary advance, not only generated widespread goodwill and
gratitude towards the insurgent groups, but also highlighted the extreme brutality of the BVA and thus
drove civilians further into the insurgents’ arms (El Espectador 2014).638 According to a peasant leader
from rural Tame:
One in a certain way was thankful to god that the ELN and the FARC were there and they didn’t let the paras enter at that time because they committed so many massacres. If you were the mother of somebody in the guerrilla groups then they would kill you. The paras would kill your father, mother, brothers, everybody. At that time, people preferred that the guerrillas were there.639
In light of the well-known fact that the Colombian military collaborated intimately with the
BVA, relations between civilian communities and the armed forces reached a new low, as a human
rights defender in Tame puts it: “People felt unprotected with them. They never acted and only
stigmatized us”.640 If anything, the collaboration between the BVA and the Colombian Army’s 18 th
brigade only exacerbated the pre-existing cleavage, with the insurgent groups and the piedmont
peasantry on one side, and the Colombian state on the other. In Tame, the municipality most affected
by the regional armed conflict and also the only one divided evenly between the two sub-regions, the
insurgent security cordon was largely effective at protecting rural civilian communities in the piedmont
from being forcibly displaced by counterinsurgent forces as happened in the plains. As Table 13
There were certainly some civilians who perceived state expansion as a potential opportunity
to disobey or defect from the insurgent social order, but both insurgent groups were highly effective
at deterring any potential collaboration or support for the Colombian state during this time. According
to one peasant leader from northern Tame: “Some people started to collaborate with the army, but
the guerrillas bribed a commander to give them information about who was informing. This
commander sent them a file with their names and phone numbers and all those who had called were
killed”.643 In another instance, the FARC killed a husband and wife in rural Tame because their adult
son had enlisted in the Colombian army (El Tiempo 2004c). Women suspected of interacting with
members of the military or the police were only spared on the condition that they become informants
for the insurgent groups on the threat of having their children forcibly recruited into their ranks
(CODHES 2008). Rural communities that did not sufficiently resist counterinsurgent incursions were
accused of being paramilitary collaborators and forcibly displaced elsewhere by the insurgent groups
(El Tiempo 2004d). In a telling example, fourteen detectives who arrived in the (then BVA controlled)
municipal capital of Tame to assume investigative functions in early 2004 were forced to sleep on the
floor because no transport or moving company would dare deliver beds and other equipment to their
headquarters out of fear of becoming an insurgent military target (El Tiempo 2004e).
Table 14. Armed Actor Control in the Araucan Piedmont (2001-2010)
Territorial
Political
Economic
Civic
• Contested control of municipal capitals, rural villages, and highways (2002-2010)
• Insurgent control in Apure (2001-2010)
• Counterinsurgent control of departmental politics (2002-2009)
• ELN control of municipal politics
• ELN control of contraband, contracts, extortions (2001-2010)
• FARC control of coca, extortions (2001-2010)
• ELN control of civic organizations and mobilizations
Source: Author’s Elaboration
643 Interview 75, Bogotá D.C., 2016.
315
Various initiatives implemented under Democratic Security Policy, such as the informants
network, peasant soldiers, and ‘Soldier for a day’, sought to break these ties “by directly involving
people in the not only in the use of force, but by implicating that everybody had to collaborate which
only worsened the existing level of polarization”.644 Counterinsurgent forces wilfully or ignorantly
ignored the fact that obedience was the minimum requirement for civilian communities living under
insurgent control and conflated this soft support for active collaboration, a failure which ironically
only further bound insurgents and civilians together. According to a peasant leader from a rural
piedmont community, given the high frequency of insurgent-civilian interactions, “anybody could be
accused of being an insurgent collaborator for having sold these groups a chicken or for giving them
lunch, all of which we were directly or indirectly forced to do”.645 The pervasiveness of nested
governance in the piedmont gave civilians no option but to be allied with them, as one migrant settler
in rural Fortul elaborates:
In this place the insurgents classify you. You are for one side or the other. I don’t consider myself part of either group, but because I arrived here thanks to my brother and he is classified as supportive of the FARC, then so am I and that’s how it is. Here, you have to be classified, and they are the ones who decide and they don’t ask you anything. I don’t know the ideology of either group, nor do I share it, and look at me.646
Pre-insurgent peasant grievances against an unresponsive state were addressed with the sudden
modernization of the department in the 1980s and 1990s, yet the continued stigmatization of
piedmont civilians as insurgents helped the FARC and the ELN to re-frame the primary cleavage in
existentialist terms with the state as a repressive enemy which sought to undo peasant gains by
destroying and rebuilding piedmont society in favor of elite interests. Such a zero-sum contest helped
these groups sustain the primary cleavage well into the 21st century. As Lipset and Rokkan (1967) point
out, the more polarized such socio-political conflicts become, the more they are “[…]no longer over
644 Interview 25, Tame, 2016.
645 Interview 62, Arauquita, 2016.
646 Interview 22, Fortul, 2016.
316
specific gains or losses but over conceptions of moral right and over the interpretation of history and
human destiny” (11). Galula (1964) similarly highlights how primary cleavages driving insurgencies
can change over time: “All wars are theoretically fought for a political purpose, although in some cases
the final political outcome differs greatly from the one intended initially” (8).
The failure of the counterinsurgents to sufficiently weaken insurgent populational control in
the piedmont was compounded by several other strategic and operational errors, which only further
fuelled negative civic-military relations in the sub-region. Faced with two powerful insurgencies that
permeated every sphere of Araucan society and were strongly emdedded in rural communities
throughout the piedmont, the counterinsurgent forces could not depend on sheer coercion or force
alone if they were to dislodge and expel these groups from the zone. As Huntington (1968) observes:
“Numbers, weapons, and strategy all count in war, but major deficiencies in any one of those may still
be counterbalanced by superior cohesion and discipline” (23). The Colombian army initially faced
enormous challenges locating active insurgents themselves, and instead attempted unsuccessfully to
dismantle their social bases by prosecuting Araucan civil society through a series of mass arrests,
unlawful detentions, and selective assassinations.647 The minimal benefits accrued from this strategy
were far outweighed by the renewed vigor with which piedmont civic organizations mobilized against
the counterinsurgent cause in the department. The civic sphere became another battleground on which
the state was ill prepared to fight, let alone win, against the insurgent groups (Semana 2003). Given
that most of these civic leaders detained or killed “were not part of formal insurgent structures”, the
groups themselves “were not really affected by this strategy”, as civil society itself saw the emergence
of new leaders to replace those prosecuted in this war and did not cease mobilizing.648
and 1990s are in serious disrepair and are frequently dynamited by the insurgent groups during periods
of intensified confrontation with the Colombian state.
The departmental economy has suffered from the exhaustion of key oil wells at Caño-Limón
which are slated to dry up around 2022. Apart from oil production, the department has never been
able to develop any alternative productive industry that can generate sufficient income or employment
for the department to subsist on. Small producers in the piedmont and cattle ranchers in the plains
similarly face enormous challenges developing internal markets to generate revenues and stimulate the
departmental economy, leaving many to sell their products as contraband in neighboring Venezuela
when the binational exchange rate is advantageous to such cross-border trade.698 Similarly, various
rural farmers complained of the sudden disappearance of markets in Casanare or Meta for their
livestock, a change which they attribute to the Uribe family’s monopolization of pig raising throughout
the plains during his presidency.699 One local human rights defender in Arauca municipality evaluates
Uribe’s counterinsurgent strategy on the local quality of life in bleak terms: “The only thing that Uribe
did was to militarize the zone. There is no new employment, no improvement in the quality of life,
nor are there universities or possibilities to keep studying. There are young men who don’t have any
option but to join the guerrillas”.700
Civil society in Arauca remains strong and diverse, a surprising feat given the decades of armed
conflict that have torn away at the social fabric. At the present, various civic organizations have been
able to claim a modest degree of autonomy vis-à-vis the insurgent groups but only to the extent that
they do not collaborate with the state or exceed the parameters established by these armed actors.701
Until the FARC’s demobilization in 2017, both insurgent groups continued to maintain absolute
698 Interview 10, Arauca municipality, 2016.
699 Interview 62, Arauquita, 2016.
700 Interview 2, Arauca municipality, 2016.
701 Interview 49, Arauca municipality, 2016.
331
control over the JACs in the piedmont. Following the demobilization of the BVA and the return of
many displaced peasants to rural communities, the community action boards experienced a resurgence
in plains communities where they had long remained inactive due to the paramilitary incursion and
low levels of communal participation.702 In a twist of irony, many residents in both the piedmont and
the plains claim that due to the demobilization of the FARC, local JACs have lost much of their
impetus due to a decline in local participation. Quite simply, when these civic organizations were
supported and regulated by the insurgents, civilians possessed far greater motivation to comply with
their dictums and to contribute both the time and resources to sustain them.703
D. En fin: Potential Theoretical Contributions and Policy Implications
My dissertation clearly offers a great deal in regards to the study of insurgent formation, non-
state governance, and counterinsurgency, but the theoretical model provided also has potential to
contribute to explanations about other socio-political phenomena such as organized and unorganized
criminal dynamics, authoritarian and illiberal regimes, and more generally how embeddedness is
formed in precarious contexts. Perhaps most importantly, this research demonstrates the utility of
such embeddedness in fostering outcomes that are generally regarded by contemporary statesmen and
military commanders as negative and counterproductive. Social embeddedness is required on a
massive scale to foster the necessary collective action required by rebellions and revolutions alike
(Popkin 1979; Lichbach 1994; Petersen 2001; Wood 2003). However, social embeddedness in
contentious spaces can also serve as an inherently positive force, whether due to its ability to mitigate
interethnic violence (Varshney 2002; Belloni 2001), or because of its capacity to help foster conditions
for peace between warring actors in civil wars (Guelke 2003; Paffenholz and Spurk 2006; Farrington
2008; Orjuela 2003). Conversely, recent works grounded in the study of armed conflict and violence
702 Interview 72, Tame, 2016.
703 Interview 99, Tame, 2016.
332
have noted that the role of embeddedness may often intersect with that played by armed non-state
actors in theatres of conflict, and often for ends which are neither peaceful nor pluralistic (Belloni
2008; Arjona 2016a; Staniland 2014; Zukerman Daly 2016). This dissertation contributes to the latter
canon of research by establishing a dialogue between scholarship which focuses on the importance of
informal institutions (Helmke and Levitsky 2004; Bratton 2007; Grzymala-Busse 2010), with works
that study the formation of local social orders during times of conflict and the agency of civilians in
this complicated process (Mampilly 2011; Staniland 2012; Kaplan 2013; Arjona 2009; 2014; 2014b).
My theory on the historical processes that generate social embeddness between peasants in
rural communities, peasants and insurgent groups, and peasants and counterinsurgent forces, may also
be deployed to understand socio-political phenomena where binding group dynamics affect individual
agency, often superceding decisions or option which are in fact in a person’s best interests.
Environments where organized and unorganized criminal dynamics reign, whether the Camorra in
Naples or the Rolling Sixties Crips in South Central Los Angeles, demonstrate similar conditions
where the state and non-state source of authority overlap, yet most civilians who reside in these spaces
are beholden to the social order and the accompanying code of conduct implemented and enforced
by the latter, regardless of the former’s ability to intervene and prosecute inviduals when they please.
Many civilians may not appreciate or even particularly like these non-state groups, yet they understand
that defection to and collaboration with the state almost always comes with high personal costs that
transcend violent sanction, with familial and communal ostracization and forced exit often awaiting
those who are not killed outright for their transgressions. Similar situational ethics can be found in
authoritarian and illiberal regimes and may go a long way to explaining the resilience of such political
orders and the absence of widespread social upheaval against repressive and deeply unpopular regimes.
Apart from the obvious sanctions for challenging, opposing, or not supporting an autocratic ruler,
group pressures may also dictate individual support, acquiescence, general indifference and apathy in
333
such dynamics. The theoretical implications of this may go a long way to explaining how these regimes
survive by binding the minimum necessary amount of people to the established order through either
positive or negative inducements, an avenue of inquiry that is particularly relevant in the current global
context where democratic regression is in flux the world over.
Finally, this dissertation suggests that when there is an alternative to military counterinsurgency
- whether a negotiated settlement, power sharing agreements, political decentralization, or
bribery/appeasement - that these are all more desirable than the escalation of an armed conflict in
which the biggest loser will be the civilian population caught between the competing actors. As this
dissertation demonstrates, strategies which involve highly questionable and outright illegal tactics can
often be the most effective, thus incentivizing counterinsurgent strategists and political leaders with a
shaky moral compass to reframe and repackage such measures in a seemingly humane manner which
conforms to existing International Humanitarian Law. If these other options do prove unavailable,
then the strategies selected, developed, and deployed should involve the advice or input from
stakeholders on the ground, or if this is impossible, then from individuals who do possess sufficient
knowledge about the antecedent conditions of the space in question to make sound and ethical policy
recommendations. As Evans (2014) notes, “[…]the social assumptions underlying counter-insurgency
doctrine demand substantive engagement from social scientists and sociologists in particular. Absent
this engagement, stakeholders find themselves in the odd position of having a theory of warfare
focused on social behavior that is practically untouched by those who make the study of social
behavior their life’s work” (257).
334
Works Cited Abbs, Luke, Govinda Clayton, and Andrew Thomson. 2020. The Ties That Bind: Ethnicity, Pro-
government Militia, and the Dynamics of Violence in Civil War. Journal of Conflict Resolution 64, 5: 903–932.
Acemoglu, Daron, James A. Robinson, and Rafael J. Santos. 2013. The Monopoly of Violence: Evidence from Colombia. Journal of the European Economic Association 11, 1: 5–44.
Acemoglu, Daron, Leopoldo Fergusson, James A. Robinson, Dario Romero, and Juan F. Vargas. 2016. The Perils of High-Powered Incentives: Evidence from Colombia's False Positives. NBER Working Paper No. 22617.
Albertus, Michael, and Oliver Kaplan. 2013. Land Reform as a Counterinsurgency Policy: Evidence from Colombia. The Journal of Conflict Resolution 57, 2: 198-231.
Alley, April Longley. 2013. Tracking the ‘Arab Spring’: Yemen Changes Everything…And Nothing. Journal of Democracy 24, 4: 74–85.
Alto Comisionado de las Naciones Unidas para los Refugiados (ACNUR). 2007. Diagnóstico Departamental Arauca.
Aliyev, Huseyn. 2016. Strong militias, weak states and armed violence: Towards a theory of ‘state-parallel’ paramilitaries. Security Dialogue 47, 6: 498–516.
Alvarez, Alex. 2006. Militias and Genocide. War Crimes, Genocide, and Crimes against Humanity 2: 1-33. Amnesty International. 2004. Colombia: Laboratory of War: Repression and Violence. Amnesty
International UK. ———. 2005. Colombia: The Paramilitaries in Medellín: Demobilization or Legalization?. Amnesty
International UK. Anderson, Benedict. 2006. Imagined Communities: Reflections on the Origin and Spread of Nationalism.
London: Verso. Araujo, Fernando. 2008. El Trapecista. Madrid: Planeta. Archer, Ronald P., and Marc Chernick. 1989. El presidente frente a las instituciones nacionales. In La
democracia en blanco y negro: Colombia en los años Ochenta, edited by Patricia Vasquez de Urrutia. Bogotá: Ediciones Uniandes and CEREC.
Arjona, Ana, Nelson Kasfir, and Zachariah Mampilly. 2015. Rebel Governance in Civil War. Cambridge: Cambridge University Press.
Arjona, Ana. Wartime Institutions: A Research Agenda. Journal of Conflict Resolution 58, 8: 1360-1389. ———. 2015. Civilian Resistance to Rebel Governance. In Arjona et al. 2015. 1–20. ———. 2016b. Rebelocracy: Social Order in the Colombian Civil War. Cambridge: Cambridge University
Press. ———. 2016b. Institutions, Civilian Resistance, and Wartime Social Order: A Process-driven Natural
Experiment in the Colombian Civil War. Latin American Politics and Society 58, 3: 99–122. ———. 2017. Civilian Cooperation and Non-Cooperation with Non-State Armed Groups: The
Centrality of Obedience and Resistance. Small Wars & Insurgencies, 28, 4-5: 755-778. Aron, Raymond. 1950. Social Structure and the Ruling Class: Part 1. The British Journal of Sociology 1, 1:
1–16. Arrighi, Giovanni, Terrence K. Hopkins, and Immanuel Wallerstein. 1989. Antisystemic Movements. New
York: Verso. Ayala Diago, César Augusto. Resistencia y oposición al establecimiento del Frente Nacional. Bogotá: UNAL. Bagley, Bruce Michael, and Fernando Botero Zea. 1978. Organizaciones campesinas contemporáneas
en Colombia: Un estudio de caso de las Asociación Nacional de Usuarios Campesinos. Estudios Rurales Latinoamericanos 1,1.
335
———. 1990. Colombia y la Guerra contra la Droga. Economía Colombiana 226‑227. Balcells, Laia. 2017. Rivalry and Revenge: The Politics of Violence During Civil War. Cambridge: Cambridge
University Press. Ballvé, Teo. 2013. Grassroots Masquerades: Development, Paramilitaries, and Land Laundering in
Colombia. Geoforum 50: 62–75. ———. 2020. The Frontier Effect: State Formation and Violence in Colombia. Ithaca: Cornell
University Press. Banerjee, Abhijit, and Lakshmi Iyer. 2005. History, Institutions, and Economic Performance: The
Legacy of Colonial Land Tenure Systems in India. The American Economic Review 95, 4: 1190-1213. Barbosa Vargas, Julián Eduardo. 2015. Configuración diferenciada de las autodefensas campesinas de
Córdoba y Urabá en el Urabá: norte de Urabá, eje bananero, sur del Urabá antioqueño y Urabá chocoano. Análisis Político 28, 84: 39–57.
Barrera, Víctor. 2017. The Dark Side of Success. The Consequences of Military Modernization and Security Policies in Colombia. Unpublished Manuscript.
Barrios González, M. 2011. Clientelismo y Familias en Acción: una mirada desde lo local. Revista Opera 11, 11: 147-164.
Bayer, R., & Rupert, M. C. 2004. Effects of Civil Wars on International Trade, 1950-92. Journal of Peace Research 41, 6: 699–713.
Beach, Derek, and Rasmus Brun Pedersen. 2013. Process-Tracing Methods: Foundations and Guidelines. Ann Arbor: University of Michigan Press.
Beckert, Jens, and Matias Dewey. 2017. The Architecture of Illegal Markets: Towards an Economic Sociology of Illegality in the Economy. Oxford: Oxford University Press.
Bejarano, Ana María, and Eduardo Pizarro. 2004. Colombia: The Partial Collapse of the State and the Emergence of Aspiring State-Makers. In States-Within-States: Incipient Political Entities in the Post-Cold War Era, ed. by Paul Kingston and I. Spears. New York: Palgrave MacMillan. 99-118.
Bejarano, A.M. 2003. Protracted Conflict, Multiple Protagonists, and Staggered Negotiations: Colombia, 1982 – 2002. Canadian Journal of Latin American and Caribbean Studies 28, 55-56: 223 – 247.
———. 2011. Precarious Democracies. Understanding Regime Stability and Change in Colombia and Venezuela. Notre Dame: University of Notre Dame Press.
Belloni, Roberto. 2001. Civil Society and Peacebuilding in Bosnia and Herzegovina. Journal of Peace Research 38, 2: 163-180.
———. 2008. Civil society in war-to-democracy transitions. In From War to Democracy: Dilemmas of Peacebuilding, edited by Jarstad, Anna K. and Timothy D., Sisk. Cambridge: Cambridge University Press. 182–210.
Bennett Andrew, and Jeffrey Checkel (eds). 2014. Process Tracing: From Metaphor to Analytic Tool. Cambridge: Cambridge University Press.
Bergquist, Charles. 1986. Coffee and Conflict in Colombia, 1886-1910. Durham: Duke University Press. Berman, Eli, Jacob N. Shapiro, and Joseph H. Felter. 2011. Can Hearts and Minds be Bought? The
Economics of Counterinsurgency in Iraq. Journal of Political Economy 119, 4: 766–819. Berman, Eli, and Aila M. Matanock. 2015. The Empiricists' Insurgency. NBER Working Papers.
National Bureau of Economic Research. Besley, Timothy, and Marta Reynal-Querol. 2014. The Legacy of Historical Conflict: Evidence from
Africa. American Political Science Review 108, 2: 319–336. Bhavnani, Ravi, Dan Miodownik, and Hyun Jin Choi. 2011. Violence and Control in Civil Conflict:
Israel, the West Bank, and Gaza. Comparative Politics 44, 1: 61-80.
336
Black, Robert E., Saul S. Morris, and Jennifer Bryce. 2003. Where and Why Are 10 Million Children Dying Every Year? The Lancet 361, 9376: 2226-34.
Boesten, Jan. 2014. The Generalization of Particularized Trust. Paramilitarism and Structures of Trust in Colombia. Colombia Internacional 81: 237-265.
Böhmelt, T., C. Dworschak, U. Pilster, and J. Walterskirchen. 2019. A Cross-National Analysis of Forced Population Resettlement in Counterinsurgency Campaigns. Peace Economics, Peace Science and Public Policy 26, 1.
Boone, Catherine. 2003. Political Topographies of the African State: Territorial Authority and Institutional Choice. Cambridge: Cambridge University Press.
———. 2014. Property and Political Order in Africa: Land Rights and the Structure of Politics. Cambridge: Cambridge University Press.
Bratton, Michael. 2007. Formal Versus Informal Institutions in Africa. Journal of Democracy 18, 3: 96-110. Brewer Norman, Susan. 2010. Agrarian Class Structure and Guerilla Violence in Civil War: The Case of the FARC in Colombia. Conference paper. Brubaker, Rogers, and David D. Laitin. 1998. Ethnic and Nationalist Violence. Annual Review of Sociology 24, 1: 423-452. Burns, Ken, and Lynn Novick (Directors). 2016. The Vietnam War [TV Series]. Arlington: Public
Broadcasting Service. Bushnell, David. 1993. The Making of Modern Colombia: A Nation In Spite of Itself. Berkeley: University of
California. Byman, Daniel. 2016. ‘Death Solves All Problems’: The Authoritarian Model of Counterinsurgency.
Journal of Strategic Studies 39:1: 62-93. Callwell, C.E. 1906. Small Wars - Their Principles and Practice. London: Harrison and Sons. Campbell, Andrea. 2012. Policy makes mass politics. Annual Review of Political Science 15: 333-351. Cantor, David. 2018. Appendix 2: Montes de María Region: Bolívar and Sucre Departments.
In Appendix 2: Montes de María Region: Bolívar and Sucre Departments. Leiden: Brill. Capoccia, Giovanni, and R.D. Kelemen. 2007. The Study of Critical Junctures: Theory, Narrative,
and Counterfactuals in Historical Institutionalism. World Politics 59, 3: 341–369. Capoccia, Giovanni. 2016. When Do Institutions “Bite”? Historical Institutionalism and the Politics
of Institutional Change. Comparative Political Studies 49, 8: 1095–1127. Caracol Radio. 2008. Desvertebrado el frente 35 de las FARC en Sucre. May 18.
https://caracol.com.co/radio/2008/05/18/nacional/1211105280_598353.html ———. 2009. A juicio 11 infantes de marina involucrados en un caso de "falsos positivos". October
12. https://caracol.com.co/radio/2009/12/10/judicial/1260447960_922628.html Carroll, Leah. 2011. Violent Democratization: Social Movements, Elites, and Politics in Colombia’s Rural War
Zones, 1984–2008. Notre Dame: University of Notre Dame Press. Cederman, Lars-Erik, Simon Hug, Andreas Schädel, and Julian Wucherpfennig. 2015. Territorial
Autonomy in the Shadow of Conflict: Too Little, Too Late?—ERRATUM. American Political Science Review 109, 3: 635–635.
Centro de Recursos para el Análisis de Conflictos (CERAC). n.d. Análisis de Conflictos y Violencia Política Tipología por municipios del conflicto armado. http://www.cerac.org.co/es/l%C3%ADneas-de-investigaci%C3%B3n/analisis-conflicto/tipologia-por-municipios-del-conflicto-armado.html
Centro Nacional de la Memoria Histórica (CNMH). 2009. La Masacre de El Salado: Esa Guerra No Era Nuestra. Bogotá: CNRR.
———. 2012. Justicia y Paz: Los silencios y los olvidos de la verdad. Bogotá: CNRR. ———. 2014a. Guerrilla y población civil trayectoria de las FARC 1949-2013. Bogotá: CNRR.
337
———. 2014b. Recordar para reparar: Las masacres de Matal de Flor Amarillo y Corocito en Arauca. Bogotá: CNRR.
———. 2015. Cruzando la frontera: memorias del éxodo hacia Venezuela: el caso del río Arauca. Bogotá: CNRR. Centro de Investigación y Educación Popular (CINEP). Dynamics of Armed Conflict Database. Cigar, Norman. 2014. Tribal Militias: An Effective Tool to Counter Al-Qaida and Its Affiliates. Strategic
Studies Institute and U.S. Army War College Press. Consultoría para los Derechos Humanos y el Desplazamiento (CODHES). 2008. Arauca: Dilemas de
guerra, desafios humanitarios. Bogotá: Ediciones Ántropos. ———. 2010. Número de personas desplazadas por departamento de llegada. Bogotá: Ediciones
Ántropos. Colletta, Nat, and Michelle Cullen. 2000. Violent conflict and the transformation of social capital :
Lessons from Cambodia, Rwanda, Guatemala, and Somalia. Washington, D.C.: The World Bank.
Collier, Ruth Berins, and David Collier. 1991. Shaping the Political Arena: Critical Junctures, the Labor Movement, and Regime Dynamics in Latin America. Princeton: Princeton University Press.
Collier, Paul, and Anke Hoeffler. 1998. On economic causes of civil war. Oxford Economic Papers 50, 4: 563–573.
———. 2004. Greed and grievance in civil war. Oxford Economic Papers 56, 4: 563–595. Comisión Valenciana de Verificación de Derechos Humanos (CVVDH). 2008. Colombia: Rompiendo
el silencio. Coppola, Francis Ford (Director). 1979. Apocalypse Now [Movie]. San Francisco: Omni Zoetrope. Corporación Nuevo Arco Iris. 2014. Genocidio del Partido Revolucionario de los Trabajadores (P.R.T), en los
Montes de María, después del Acuerdo de Paz en 1991. https://www.arcoiris.com.co/2014/07/genocidio-del-partido-revolucionario-de-los-trabajadores-p-r-t-en-los-montes-de-maria-despues-del-acuerdo-de-paz-en-1991/
Correa Peraza, Hernando. 2009. Los Rostros de la Violencia: Colombia 1930-1958. Bogotá: Fondo de Publicaciones Universidad Sergio Arboleda.
Cramer, Christopher, and Elisabeth J. Wood. 2017. Introduction: Land rights, restitution, politics, and war in Colombia. Journal of Agrarian Change 17: 733-738.
Cramer, Christopher. 2003. Does inequality cause conflict?. Journal of International Development 15: 397-412.
Cronin-Furman, Kate, and Milli Lake. 2018. Ethics Abroad: Fieldwork in Fragile and Violent Contexts. Political Science & Politics 51, 3: 607–614.
Currea-Lugo, Victor de. 2014. ¿Por qué negociar con el ELN?. Bogotá: Pontificia Universidad Javeriana. Dahl, Robert. 1971. Polyarchy: Participation and Opposition. New Haven: Yale University Press. Daly, Sarah Zukerman. 2014. The Dark Side of Power-Sharing: Middle Managers and Civil War
Recurrence. Comparative Politics 46, 3: 333-353. ———. 2016. Organized Violence After Civil War: The Geography of Recruitment in Latin America.
Cambridge: Cambridge University Press. De Niro, Robert (Director). 1993. A Bronx Tale [Movie]. New York: TriBeCa Productions. De Soysa, Indra. 2002. Paradise is a Bazaar? Greed, Creed, and Governance in Civil War, 1989-99. Journal of
Peace Research 39, 4: 395–416. Deas, Malcom. 1995. Canjes violentos: reflexiones sobre la violencia política en Colombia. In Dos
Ensayos especulativos sobre la violencia en Colombia, edited by Malcolm Deas and Fernando Gaitan Daza. Bogotá: Tercer Mundo Editores.
Defensoría del Pueblo. 2003. Proyecto Apoyo Defensorial en Las Zonas de Rehabiliatación y Consolidación: Primer Informe de Actividades. Bogotá.
338
Degregori, Carlos Iván. 1990. El surgimiento de Sendero Luminoso. Lima: IEP. Delgado, Jorge E. 2015. Colombian Military Thinking and the Fight against the FARC-EP Insurgency,
2002–2014. Journal of Strategic Studies 38, 6: 826-851. Departamento Administrativo Nacional de Estadística (DANE). 1993. Censo general. ———. 2005. Censo general. de Rouen, Karl, and David Sobek. 2004. The Dynamics of Civil War Duration and Outcome. Journal
of Peace Research 41, 3: 303–320. Dixon, Paul. 2009. ‘Hearts and Minds’? British Counter-Insurgency Strategy in Northern Ireland.
Journal of Strategic Studies 32: 445-474. Dube, Oeindrila, and Suresh Naidu. 2015. Bases, Bullets, and Ballots: The Effect of US Military Aid on
Political Conflict in Colombia. The Journal of Politics 77, 1: 249-267. Dufort, Philippe. 2017. A typology of military factions in the Colombian officer corps: origins and
evolution of Colombian counter-insurgency. Canadian Journal of Latin American and Caribbean Studies 42, 3: 320-349.
Duica Amaya, Liliana. 2013. Geografía de la violencia en el Carmen de Bolívar 1990-2010. Masters Thesis. Universidad de los Andes, Bogotá.
Duncan, Gustavo. 2006. Los Senores de La Guerra: de Paramilitares, Mafiosos y Autodefensas En Colombia. Madrid: Planeta.
———. 2014. Drug Trafficking and Political Power Oligopolies of Coercion in Colombia and Mexico. Latin American Perspectives 41, 2: 18–42.
Duque Daza, Javier. 2017. Arauca: el saqueo de las regalías entre el ELN, los paramilitares y los políticos. Razón Pública, March 6. www.razonpublica.com/index.php/conflicto-drogas-y-paz-temas-30/10078-especial-radiograf%C3%ADa-de-la-corrupci%C3%B3n-i.html
Durkheim, Émile. 1933. The division of labor in society. New York: Macmillan. ———. 1961. Moral Education: A Study in the Theory and Application of the Sociology of Education. New
York: The Free Press of Glencoe. Eaton, Kent. 2006. The Downside of Decentralization: Armed Clientelism in Colombia. Security Studies
15, 4: 533–62. Echandía, Camilo. 2006. Dos décadas de escalamiento del conflict armado en Colombia, 1986–2006. Bogotá:
Editorial Kimpres. Edwards, Michael (ed). 2011. The Oxford Handbook on Civil Society. Oxford: Oxford University Press. Eidt, Robert. 1967. Modern Colonization as a Facet of Land Development in Colombia, South
America. Yearbook of the Association of Pacific Coast Geographers 29: 21-42. Ejército de Liberación Nacional. 2007. “Crónica del surgimiento del Frente Domingo Laín”, January
14. www.cedema.org. El Espectador. 2008. Abatido el segundo comandante del frente 35 de las Farc. April 27.
———. 2012. Abatido cabecilla del frente 37 de las Farc. September 27. https://www.elespectador.com/noticias/judicial/abatido-cabecilla-del-frente-37-de-farc-articulo-377858
———. 2014. El Frente Domingo Laín, mitos y realidades de una máquina de guerra. July 7. www.elespectador.com/noticias/paz/el-frente-domingo-lain-mitos-y-realidades-de-una-maquin-articulo-502321
———. 2015. Condenan a 42 años de prisión a responsables del crimen de la jueza Gloria Constanza Gaona. September 4. https://www.elespectador.com/noticias/judicial/condenan-42-anos-de-prision-responsables-del-crimen-de-articulo-583933
339
———. 2017. Memorias de víctimas de detenciones arbitrarias en Montes de María. September 26. https://www.elespectador.com/colombia2020/justicia/verdad/memorias-de-victimas-de-detenciones-arbitrarias-en-montes-de-maria-articulo-855810
Eland, Ivan. 2013. The Failure of Counterinsurgency: Why Hearts and Minds Are Seldom Won. Westport: Praeger.
Escobar, Cristina. 1982. El movimiento campesino en Sucre. Bachelors Thesis. Universidad de los Andes. Bogotá.
———. 1998. Clientelism, Mobilization and citizenship: Peasants Politics in Sucre, Colombia. Doctoral thesis, University of California, San Diego.
Esquivia Ballestas, Ricardo, and Barbara Gerlach. 2009. The Local Community as a Creative Space for Transformation: The View from Montes de María. In Colombia: Building Peace in a Time of War, ed. by Virginia Marie Bouvier. Washington D.C.: USIP.
Evans, Ryan. 2014. ‘The Population Is the Enemy’: Control, Behaviour, and Counter-insurgency in Central Helmand Province, Afghanistan. In The New Counter-insurgency Era in Critical Perspective, edited by C.W. Gventer, D.M. Jones, M.L.R. Smith. London: Palgrave Macmillan.
Fajardo, Darío. 2015. Estudio sobre los orígenes del conflicto social armado, razones de su persistencia y sus efectos más profundos en la sociedad colombiana. In Contribución al entendimiento del conflicto armado en Colombia. Bogotá: Comisión Histórica del Conflicto y sus Víctimas.
Falchetti de Sáenz, Ana María. 1995. El oro del gran Zenu: Metalurgia prehispánica en las llanuras del Caribe colombiano. Bogotá: Banco de la República.
Falleti Tulia G., and James Mahoney. 2015. The comparative sequential method. In Advances in Comparative-Historical Analysis, edited by James Mahoney and Kathleen Thelen. Cambridge: Cambridge University Press.
Falleti, Tulia G. 2016. Process tracing of extensive and intensive processes. New Political Economy 21, 5: 455-462.
Fals Borda, Orlando. 1976. Capitalismo, hacienda, y poblamiento: su desarrollo en la costa atlántica. Bogotá: Punta de Lanza.
Farrington, C. (ed). 2008. Global Change, Civil Society and the Northern Ireland Peace Process: Implementing the Political Settlement. London: Palgrave MacMillan.
Fearon, James D., and David D. Laitin. 2003. Ethnicity, Insurgency, and Civil War. The American Political Science Review 97, 1: 75-90.
Fichtl, Eric. 2003. Colombia: Araucan nightmare - Life and death in Tame. August 4. https://reliefweb.int/report/colombia/colombia-araucan-nightmare-life-and-death-tame
———. 2004. Civilian “Collaborators” in Colombia’s Conflict. March. http://web.archive.org/web/20071212182804/www.colombiajournal.org/collaboration.htm
Fioretos, O., Tulia Falleti, and A. Sheingate. 2016. Historical Institutionalism in Political Science. In The Oxford Handbook of Historical Institutionalism, edited by O. Fioretos, Tulia Falleti, and A. Sheingate. Oxford: Oxford University Press.
Fisiy, Cyprian. 1992. Power and Privilege in the Administration of Law. Leiden: African Studies Centre. Fitzgerald, David. 2013. Learning to Forget: US Army Counterinsurgency Doctrine and Practice from Vietnam to
Iraq. Palo Alto: Stanford University Press. Fjelde, Hanne, and Indra De Soysa. 2009. Coercion, Co-optation, or Cooperation?: State Capacity and
the Risk of Civil War, 1961—2004. Conflict Management and Peace Science 26, 1: 5–25. Fluharty, Vernon Lee. 1966. Dance Of The Millions: Military Rule And The Social Revolution In Colombia
1930-1956. Pittsburgh: University of Pittsburgh Press. Flynn, Vincent. 1984. Model Villages in the Ixil Region. Cultural Survival Quarterly 8, 4. Fortna, Virginia Page. 2004. Does Peacekeeping Keep Peace? International Intervention and the
Duration of Peace after Civil War. International Studies Quarterly 48, 2: 269-292.
340
Fujii, Lee Ann. 2010. Shades of truth and lies: Interpreting testimonies of war and violence. Journal of Peace Research 47, 2: 231–241.
———. Interviewing in Social Science Research A Relational Approach. London: Routledge. Fundación Ideas para la Paz (FIP). 2014. Dinámicas del conflicto armado en Arauca y su impacto
humanitario. Bogotá: FIP.
———. 2015a. El ABC del ELN: Evolucion del frente de guerra oriental. Bogotá: FIP. ———. 2015b. El ELN y la industria petrolera: ataques a la infraestructura en Arauca. Bogotá: FIP. ———. 2016. Votando por la paz: Entendiendo la ventaja del “No”. Bogotá: FIP. Galula, David. 1964. Counterinsurgency Warfare: Theory and Practice. Westport: Praeger. Garay Salamanca, Luis Jorge, Eduardo Salcedo-Albarán, Isaac de León-Beltrán, and Bernardo
Guerrero. 2008. La captura y reconfiguración cooptada del estado en Colombia. Bogotá: Taller Imprenet. Garay Salamanca, Luis Jorge, Eduardo Salcedo-Albarán, and Natalia Duarte. 2017. Elenopolítica:
Reconfiguración Cooptada del Estado en Arauca, Colombia. Bogotá: Vortex. George, Alexander L., and Andrew Bennett. 2005. Case Studies and Theory Development in the Social Sciences.
Boston: MIT Press. Giddens, Anthony, and David Held (eds). 1982. Classes, Power and Conflict Classical and Contemporary
Debates. Berkeley: University of California Press. Gill, Leslie. 2016. A Century of Violence in a Red City Popular Struggle, Counterinsurgency, and Human Rights
in Colombia. Durham: Duke University Press. Giraudy, Agustina, Eduardo Moncada, and Richard Snyder. 2019. Subnational Research in
Comparative Politics. In Inside Countries: Subnational Research in Comparative Politics, ed. Giraudy, Moncada, and Snyder. New York: Cambridge University Press. 3–54.
Gómez Rivas, Daniela. 2016. Informe Regional Arauca. In El Contexto en Contexto: Un Análisis en Cinco Regiones Colombianas, 1998-2014, edited by Andrés Dávila Ladrón de Guevara, Gustavo Salazar Arbeláez, and Alexander González-Chavarría. Bogoá: Javeriana.
González, Fernán, Ingrid Bolívar, and Teófilo Vásquez. 2003. Violencia política en Colombia: de la nación fragmentada a la construcción del estado. Bogotá: ODECOFI-CINEP.
González González, Fernán. 2003. ¿Colapso parcial o presencia diferenciada del Estado en Colombia?: una mirada desde la historia. Colombia Internacional 58: 124-158.
———. 2014. Poder y Violencia en Colombia. Bogotá: ODECOFI – CINEP. Goodhand, Jonathan. 2005. Frontiers and Wars: the Opium Economy in Afghanistan. Journal of
Agrarian Change 5: 191-216. Goodwin, Jeff. 2001. No Other Way Out: States and Revolutionary Movements, 1945-1991. Cambridge:
Cambridge University Press. Gootenberg, Paul. 2003. Between Coca and Cocaine: A Century or More of U.S.-Peruvian Drug
Paradoxes, 1860–1980. Hispanic American Historical Review 83, 1: 119–150. Gould, Roger. 1995. Insurgent Identities: Class, Community, and Protest in Paris from 1848 to the Commune.
Chicago: University of Chicago Press. Granovetter, Mark. 1985. Economic Action and Social Structure: The Problem of
Embeddedness. American Journal of Sociology, 91: 481-510. Greer, Donald. 1935. Incidence of the Terror during the French Revolution A Statistical Interpretation.
Cambridge: Harvard University Press. Griffin, Keith., Azizur Khan, and Amy Ickowitz. 2002. Poverty and the Distribution of Land. Journal
of Agrarian Change 2: 279-330. Gruber, Barbara, and Jan Pospisil. 2015. 'Ser Eleno': Insurgent identity formation in the ELN. Small
Wars and Insurgencies 26, 2: 226-247.
341
Grzymala-Busse, Anna. 2010. The best laid plans: The impact of informal rules on formal institutions in transitional regimes. Studies in Comparative International Development 45, 3: 311-333.
Guelke, Adrian. 2003. Civil Society and the Northern Ireland Peace Process. Voluntas: International Journal of Voluntary and Nonprofit Organizations 14, 1: pp. 61-78.
Guevara, Ernesto. 2002. Guerrilla Warfare. Lanham: Rowman & Littlefield Publishers. Gutiérrez de Pineda, Virginia. 1968. Familia y Cultura en Colombia. Medellín: Editorial Universidad
de Antioquia. Gutiérrez Sanín, Francisco, and Mauricio Barón. 2005. Re-Stating the State: Paramilitary Territorial
Control and Political Order in Colombia (1978–2004). Crisis States Crisis States Research Center working papers series 1, no. 66. London: Crisis States Research Center, London School of Economics and Political Science. 1–31. http://eprints.lse.ac.uk/id/eprint/28178
Gutiérrez-Sanín, Francisco, and Jenniffer Vargas. 2016. El despojo paramilitar y su variación: quiénes, cómo, por qué. Bogotá: Editorial Universidad del Rosario.
———. 2017. Agrarian elite participation in Colombia's civil war. Journal of Agrarian Change 17: 739-748.
Gutiérrez Sanín, Francisco, and Elisabeth J. Wood. 2017. What Should We Mean by “Pattern of Political Violence”? Repertoire, Targeting, Frequency, and Technique. Perspectives on Politics 15, 1: 20–41.
Gutiérrez Sanín, Francisco. 2008. Telling the Difference: Guerrillas and Paramilitaries in the Colombian War. Politics and Society 36, 1: 3–34.
———. 2014. El orangután con sacoleva. Cien años de democracia y represión en Colombia (1910-2010). Bogotá: Debate.
———. 2019. Clientelistic Warfare: Paramilitaries and the State in Colombia (1982–2007). Oxford: Peter Lang.
Gutiérrez Lemus, Omar Jaime, and José Jairo González Arias. 2008. Situación actual de conflicto y exploración de escenarios posibles de paz y desarrollo en Arauca. Proyecto Políticas Públicas de Paz, Primer Informe Consultoría.
Gutiérrez Lemus, Omar Jaime. 2010. Arauca: espacio, conflicto e institucionalidad. Análisis Político 69: 3–34.
Hack, Karl. 1999. "Iron Claws on Malaya": The Historiography of the Malayan Emergency. Journal of Southeast Asian Studies 30, 1: 99-125.
Hall, Peter, and Rosemary Taylor. 1996. Political Science and the Three New Institutionalisms. Political Studies 44: 936-957.
Hall, Peter. 2003. Aligning Ontology and Methodology in Comparative Politics. In Comparative Historical Analysis in the Social Sciences, edited by James Mahoney and Dietrich Rueschemeyer. Cambridge: Cambridge University Press.
Haydu, Jeffrey. 1998. Making Use of the Past: Time Periods as Cases to Compare and as Sequences of Problem Solving. American Journal of Sociology 104, 2: 339-371.
Hazleton, Jacqueline. 2017. The “Hearts and Minds” Fallacy: Violence, Coercion, and Success in Counterinsurgency Warfare. International Security 42, 1: 80-113.
Hegre, Håvard. 2004. The Duration and Termination of Civil War. Journal of Peace Research 41, 3: 243–252.
Helmke, Gretchen, and Steven Levistky. 2004. Informal Institutions and Comparative Politics: A Research Agenda. Perspectives on Politics 2, 4: 725-740.
Henderson, James. 1985. Why Colombia Bled: A History of the Violence in Tolima. Tuscaloosa: University of Alabama Press.
Herbst, Jeffery. 2000. States and Power in Africa: Comparative Lessons in Authority and Control. Princeton: Princeton University Press.
342
Hirschi, Travis. 1969. Causes of Delinquency. Berkeley: University of California Press. Hobsbawm, Eric. 1975. The Age of Capital: 1848–1875. London: Weidenfeld & Nicolson. Horowitz, Shale. 2002. Explaining Peasant-Farmer Hegemony in Redistributive Politics: Class-,
Trade-, and Asset-Based Approaches. Comparative Studies in Society and History 44, 4: 827-851. Human Rights Watch. 1996. Colombia’s Killer Networks The Military-Paramilitary Partnership and
the United States. New York: Human Rights Watch. https://www.hrw.org/report/1996/11/01/colombias-killer-networks/military-paramilitary-partnership-and-united-states
———. 1998. War Without Quarter: Colombia and International Humanitarian Law. New York: Human Rights Watch. www.hrw.org/legacy/reports98/colombia
———. 2020. The Guerrillas Are the Police. January 22. https://www.hrw.org/report/2020/01/22/guerrillas-are-police/social-control-and-abuses-armed-groups-colombias-arauca
Humphreys, Macartan, and Jeremy Weinstein. 2006. Handling and Manhandling Civilians in Civil War. American Political Science Review 100, 3: 429–447.
Humphreys, Macartan. 2005. Natural Resources, Conflict, and Conflict Resolution: Uncovering the Mechanisms. Journal of Conflict Resolution 49, 4: 508–537.
Huntington, Samuel. 1968. Political Order in Changing Societies. New Haven: Yale University Press. Ibañez, Ana María. 2008. El desplazamiento forzoso en Colombia: un camino sin retorno hacia la pobreza. Bogotá:
Universidad de Los Andes. Isacson, Adam, and Abigail Poe. 2009. After Plan Colombia: Evaluating "Integrated Action", The
Next Phase of U.S. Assistance. Center for International Policy. Jentzsch, Corinna, Stathis N. Kalyvas, and Livia Isabella Schubiger. 2015. Militias in Civil Wars. The
Journal of Conflict Resolution 59, 5: 755-69. Josua, Maria, and Mirjam Edel. 2015. To Repress or Not to Repress—Regime Survival Strategies in
the Arab Spring. Terrorism and Political Violence 27, 2: 289-309. Kaldor, Mary. 1999. New and Old Wars: Organized Violence in a Global Era. Hoboken: Blackwell
Publishing. Kalyvas, Stathis. 2003. The Ontology of "Political Violence": Action and Identity in Civil Wars.
Perspectives on Politics 1, 3: 475-494. ———. 2006. The Logic of Violence in Civil War. Cambridge: Cambridge University Press. Kaplan, Oliver. 2013. Protecting Civilians in Civil War: The Institution of the ATCC in Colombia.
Journal of Peace Research 50, 3: 351–67. ———. 2017. Resisting War: How Communities Protect Themselves. Cambridge: Cambridge University
Press. Karl, Terri Lynn. 1990. Dilemmas of Democratization in Latin America. Comparative Politics 23, 1: 1-
21. ———. 1997. The Paradox of Plenty: Oil Booms and Petro-States. Berkeley: University of California Press. Karl, Robert. 2017. Forgotten Peace: Reform, Violence, and the Making of Contemporary Colombia. Berkeley:
University of California Press. Keen, David. 1998. The Economic Functions of Violence in Civil Wars. London: Routledge. Kilcullen, David. 2009. The Accidental Guerrilla. Oxford: Oxford University Press. ———. 2010. Counterinsurgency. Oxford: Oxford University Press. Kittel, Bernhard, and David Kuehn. Introduction: Reassessing the Methodology of Process Tracing.
European Political Science 12: 1–9. Kline, Harvey. 2009. Showing Teeth to the Dragons: State-Building by Colombian President Álvaro Uribe Vélez,
2002-2006. Tuscaloosa: The University of Alabama Press.
343
Kline, Harvey. 2015. Fighting Monsters in the Abyss: The Second Administration of Colombian President Álvaro Uribe Vélez, 2006-2010. Tuscaloosa: The University of Alabama Press.
Kramer, Mark. 2005. Guerrilla Warfare, Counterinsurgency and Terrorism in the North Caucasus: The Military Dimension of the Russian – Chechen Conflict. Europe-Asia Studies 57, 2: 209-290.
Lacina, Bethany, Nils Gleditsch, and Bruce Russett. 2006. The Declining Risk of Death in Battle. International Studies Quarterly 50, 3: 673-680.
Lackman, Michael. 1985. British Boer War And The French Algerian Conflict: Counterinsurgency For Today. B.S. Thesis. University of Montana.
Larratt-Smith, Charles. 2020. Navigating Formal and Informal Processes: Civic Organizations, Armed Nonstate Actors, and Nested Governance in Colombia. Latin American Politics and Society 62, 2: 75–98.
Las 2 Orillas. 2014. La infiltrada que terminó estafada después de haber posibilitado el rescate de Fernando Araújo. August 9. https://www.las2orillas.co/la-infiltrada-que-termino-estafada-despues-de-haber-posibilitado-el-rescate-de-fernando-araujo/
Lawrence, Thomas Edward (T.E). 2005. T.E. Lawrence in War and Peace: an Anthology of the Military Writings of Lawrence of Arabia. London: Greenhill.
Leal Buitrago, Francisco. 2003. La Doctrina de Seguridad Nacional: materialización de la Guerra Fría en América del Sur. Revista de Estudios Sociales 15: 74-87.
Leal Buitrago, Francisco. 2006. La politica de seguridad democrática 2002-2005. Análisis Político 57: 3-30.
Leech, Garry. 2003. Informers for a Day. April 7. https://garryleech.com/2003/04/ LeGrand, Catherine. 1986. Frontier Expansion and Peasant Protest in Colombia, 1850–1936. Albuquerque:
University of New Mexico Press. ———. 1989. Colonization and Violence in Colombia: Perspectives and Debates. Canadian Journal of
Latin American and Caribbean Studies 14, 28: 5-29. Leites, Nathan Constantin, and Charles Wolf, Jr. 1970. Rebellion and Authority: An Analytic Essay on
Insurgent Conflicts. Santa Monica: RAND Corporation. Lekha Sriram, Chandra, Julie A. Mertus, John C. King, Julian Hemming, Elizabeth Levy Paluck,
Chandra Lekha Sriram, Julie Norman, et al. 2009. Surviving Field Research: Working In Violent and Difficult Situations. London: Routledge.
Lenin, Vladimir. 1975. "The Dual Power". The Lenin Anthology. London: Norton. Levitsky, Steven, and Lucan Way. 2012. Beyond Patronage: Violent Struggle, Ruling Party Cohesion,
and Authoritarian Durability. Perspectives on Politics 10, 4: 869-889. ———. 2013. The Durability of Revolutionary Regimes. Journal of Democracy 24, 3: 5-17. Lichbach, Mark. 1994. What Makes Rational Peasants Revolutionary? Dilemma, Paradox, and Irony
in Peasant Collective Action. World Politics 46, 3: 383-418. Licklider, Roy. 1995. The Consequences of Negotiated Settlements in Civil Wars, 1945-1993. The
American Political Science Review 89, 3: 681-690. Ling, Peter. 2006. Social Capital, Resource Mobilization and Origins of the Civil Rights Movement.
Journal of Historical Sociology, 19, 2: 202-214. Long, Austin. 2006. On “Other War:” Lessons from Five Decades of RAND Counterinsurgency Research. Santa
Monica: RAND. López Hernández, Claudia. 2010. Y refundaron la patria. De cómo mafiosos y politicos reconfiguraron el estado
colombiano. Bogotá: Debate. Loy, Jane. 1978. Rebellion in the Colombian Llanos: The Arauca Affair of 1917. The Americas 34, 4:
502–531. ———. 1981. Forgotten Comuneros: The 1781 Revolt in the Llanos of Casanare. Hispanic American
Historical Review 61, 2: 235–257.
344
Lubkemann, Stephen. 2005. Migratory Coping in Wartime Mozambique: An Anthropology of Violence and Displacement in ‘Fragmented Wars.’ Journal of Peace Research 42, 4: 493–508.
Lyall, Jason. 2009. Does Indiscriminate Violence Incite Insurgent Attacks?: Evidence from Chechnya. Journal of Conflict Resolution 53, 3: 331–362.
———. 2010. Are Coethnics More Effective Counterinsurgents? Evidence from the Second Chechen War. American Political Science Review 104, 1: 1–20.
Machuca Pérez, Diana Xiomena. 2016. El Impacto de La Insurgencia y el Conflicto Armado en la ANUC: El Caso de Sucre. Masters Thesis. Universidad Nacional de Colombia, Bogotá.
Mahoney, James, and Kathleen Thelen. 2010. Explaining institutional change: agency, ambiguity and power. Cambridge: Cambridge University Press.
Mahoney, James. 2000. Path Dependence in Historical Sociology. Theory and Society 29, 4: 507-548. Malamud, Andrés. 2004. The increase in bilateral conflicts in Latin America: its consequences in and
outside the region. Working Paper. Madrid: Real Instituto Elcano. Mamdani, Mahmood. 1996. Citizen and Subject: Contemporary Africa and the Legacy of Late Colonialism.
Princeton: Princeton University Press. Mampilly, Zacariah. 2007. Stationary Bandits: Understanding Rebel Governance. PhD Dissertation. Los
Angeles: University of California. ———. 2011. Rebel Rulers: Insurgent Governance and Civilian Life During Civil War. Ithaca: Cornell
University Press. Margulis, Matias E., Nora McKeon, and Saturnino M. Borras. 2013. Land Grabbing and Global
Governance: Critical Perspectives. Globalizations 10, 1: 1-23. Marks, Thomas. 2002. Colombian Army Adaptation to FARC Insurgency. Strategic Studies Institute. US
Army War College. Mason, T. David, and Patrick J. Fett. 1996. How Civil Wars End: A Rational Choice Approach. Journal
of Conflict Resolution 40, 4: 546–568. Mason, T. David. 1998. "Take Two Acres and Call Me in the Morning": Is Land Reform a Prescription
for Peasant Unrest?. The Journal of Politics 60, 1: 199-230. Matanock, Aila M., and Paul Staniland. 2018. How and Why Armed Groups Participate in Elections.
Perspectives on Politics 16, 3: 710–27. Mauceri, Philip. 2004. States, Elites, and the Response to Insurgency. In Politics in the Andes - Identity,
Conflict, Reform, edited by Jo-Marie Burt and Philip Mauceri. Pittsburgh: University of Pittsburgh Press.
Medina Gallego, Carlos. 1996. ELN: Una historia contada a dos voces. Entrevista con 'el cura' Manuel Pérez y Nicolás Rodríguez Bautista, 'Gabino'. Bogotá: Rodriguez Quito Editores.
Metelits, Claire. 2009. Inside Insurgency: Violence, Civilians, and Revolutionary Group Behavior. New York: New York University Press.
Migdal, Joel. 1974. Peasants, Politics, and Revolution. Pressures toward Political and Social Change in the Third World. Princeton: Princeton University Press.
Mills, Greg, David Kilcullen, Dickie Davis, and David Spencer. 2016. A Great Perhaps? Colombia: Conflict and Convergence. London: Hurst.
Mitchell, Neil, Sabine C. Carey, and Christopher K. Butler. 2014. The Impact of Pro-Government Militias on Human Rights Violations. International Interactions 40, 5: 812-836.
Mkandawire, Thandika. 2002. The Terrible Toll of Post-Colonial 'Rebel Movements' in Africa: Towards an Explanation of the Violence against the Peasantry. The Journal of Modern African Studies 40, 2: 181–215.
Molano, Alfredo. 2004. Loyal Soldiers in the Cocaine Kingdom: Tales of Drugs, Mules, and Gunmen. New York: Columbia University Press.
345
Monografia Politico Electoral (MPE). 2007a. Departamento de Sucre 1997 a 2007. Bogotá D.C.: Corporación Nuevo Arco Iris.
Monografia Politico Electoral (MPE). 2007b. Departamento de Bolívar 1997 a 2007. Bogotá D.C.: Corporación Nuevo Arco Iris.
Moody, James, and Douglas R. White. 2003. Structural Cohesion and Embeddedness: A Hierarchical Concept of Social Groups. American Sociological Review 68, 1: 103-127.
Moore, Barrington. 1966. Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making of the Modern World. Boston: Beacon Press.
Mosinger, Eric. 2017. Brothers or others in arms? Civilian constituencies and rebel fragmentation in civil war. Journal of Peace Research 55, 1: 62-77.
Mukherjee, Shivaji. 2018a. Colonial Origins of Maoist Insurgency in India: Historical Institutions and Civil War. Journal of Conflict Resolution 62, 10: 2232-2274.
———. 2018b. Historical legacies of colonial indirect rule: Princely states and Maoist insurgency in central India. World Development 111: 113-129.
Mustafa, Daanish, and and Katherine E. Brown. 2010. The Taliban, Public Space, and Terror in Pakistan. Eurasian Geography and Economics 51, 4: 496–512.
Nagl, John. 2002. Learning to Eat Soup with a Knife: Counterinsurgency Lessons from Malaya and Vietnam. Chicago: University of Chicago Press.
Naseemullah, Adnan, and Paul Staniland. 2016. Indirect Rule and Varieties of Governance. Governance 29: 13-30.
Oberschall, Anthony. 1973. Social Conflict and Social Movements. Englewood: Prentice-Hall. Observatorio del Programa Presidencial de Derechos Humanos y Derecho Internacional Humanitario
de la Vicepresidencia de la República. Olson, Mancur. 1993. Dictatorship, Democracy, and Development. The American Political Science Review
87, 3: 567–576. Oquist, Paul. 1980. Violence, Conflict, and Politics in Colombia. New York: Academic Press. Orjuela, Camilla. 2003. Building Peace in Sri Lanka: A Role for Civil Society?. Journal of Peace Research
40, 2: 195-212. Ortiz Sarmiento, Carlos Miguel. 1985. Estado y subversión en Colombia : la violencia en el Quindío. Bogotá :
Fondo Editorial CEREC. Ostrom, Elinor. 1990. Governing the Commons: The Evolution of Institutions for Collective Action.
Cambridge: Cambridge University Press. Otto J.S., and N.E. Anderson. Cattle Ranching in the Venezuelan Llanos and the Florida Flatwoods:
A Problem in Comparative History. Comparative Studies in Society and History 28, 4: 672–683. Paffenholz, Thania, and Christoph Spurk. 2006. Civil Society, Civic Engagement, and Peacebuilding.
Social Development papers 36. Washington: The World Bank. Paige, Jeffery. 1978. Agrarian Revolution. New York: Free Press. Pakenham, Thomas. 1979. The Boer War. London: Weidenfeld & Nicolson. Palacios, Marco. 2006. Between Legitimacy and Violence: A History of Colombia, 1875–2002. Durham: Duke
University Press. Parkinson, Sarah. 2013. Organizing Rebellion: Rethinking High-Risk Mobilization and Social
Networks in War. American Political Science Review 107, 3: 418–432. Pearlman, Wendy. 2020. Mobilizing From Scratch: Large-Scale Collective Action Without Preexisting
Organization in the Syrian Uprising. Comparative Political Studies. Pécaut, Daniel. 1987. Orden y violencia: Colombia 1930-1953, Volumen 1. Bogotá: Siglo Veintiuno
Editores. Peñate, Enrique. 1991. Arauca: Politics and Oil in a Colombian Province. Master’s thesis, University of
346
Oxford. ———. 1998. El sendero estratégico del ELN: del idealismo guevarista al clientelismo armado.
Working paper 15. Pérez, Jesús María. 2010. Luchas Campesinas y Reforma Agraria: Memorias de un Dirigente de la ANUC en la
Costa Caribe. Bogotá: Puntoapartes. Pérez Bareño, Leonel. 2015. Tame, Sarmientos y burros: Ex alcaldes de Tame, Arauca, recuerdan su gestión.
Bogotá: Taller de Edición Rocca. Petersen, Roger. 2001. Resistance and Rebellion: Lessons from Eastern Europe. Cambridge: Cambridge
University Press. Petraeus, David. 2006. Counterinsurgency: Field Manual 3-24. Boulder: Paladin Press. Pierson, Paul, and Theda Skocpol. 2002. Historical Institutionalism in Contemporary Political Science.
In Political Science: State of the Discipline, edited by I. Katznelson and H.V. Milner. New York: W.W. Norton.
Pierson, Paul. 2004. Politics in Time: History, Institutions, and Social Analysis. Princeton: Princeton University Press.
Piombo, Jessica. 2007. Terrorism and U.S. Counter-Terrorism Programs in Africa: An Overview. Strategic Insights 6, 1.
Plakoudas, Spyridon. 2016. Population transfers in counter-insurgency: a recipe for success?. Small Wars & Insurgencies 27, 4: 681-701.
Plataforma de Organizaciones de Desarrollo Europeas en Colombia (PODEC). 2011. Análisis del Plan de Consolidación de Montes de María. Bogotá D.C.
Plazas Díaz, Leidy Carolina. 2017. Los inicios del Frente Domingo Laín del ELN en Arauca, 1970–1978. Procesos Históricos 31, 16: 4–16.
Podder, Sukanya. 2013. Non-State Armed Groups and Stability: Reconsidering Legitimacy and Inclusion. Contemporary Security Policy 34, 1: 16-39.
Pontecorvo, Gillo. 1966. The Battle of Algiers [Movie]. Milan: Rizzoli. Popkin, Samuel. 1979. The Rational Peasant: The Political Economy of Rural Society in Vietnam. Berkeley:
University of California Press. Porch, Douglas, and Jorge Delgado. 2010. “Masters of today”: Military Intelligence and
Counterinsurgency in Colombia, 1990–2009. Small Wars and Insurgencies 21, 2: 277–302. Porch, Douglas. 2012. The Hunt for Martín Caballero. Journal of Strategic Studies 35, 2: 243-270. Porch, Douglas. 2014. Counterinsurgency Exposing the Myths of the New Way of War. Cambridge: Cambridge
University Press. Porras, Eduardo. n.d. Mesa Campesina Regional de los Montes de María. Documento Estratégico en Tierras
y Desarrollo Rural. Porras, Eduardo. 2014. Conflictos, violencias y resistencias en los Montes de María: un análisis de
temporalidad extendida. In Territorio y Conflicto en la Costa Caribe, edited by Fernán González et al. Bogotá: Odecofi-CINEP.
Portes, Alejandro. 1998. Social Capital: Its Origins and Applications in Modern Sociology. Annual Review of Sociology 24: 1-24.
Price, Jessica. 2019. Keystone Organizations Versus Clientelism: Understanding Protest Frequency in Indigenous Southern Mexico. Comparative Politics 51, 3: 407-435.
Programa Naciones Unidas para el Desarrollo (PNUD). 2010. Los Montes de María: Análisis de la Conflictividad. PNUD Informe.
Raleigh, Clionadh, and Caitriona Dowd. 2013. Governance and Conflict in the Sahel’s ‘Ungoverned Space’. Stability: International Journal of Security and Development 2, 2.
Ramírez, María Clemencia. 2011. Between the Guerrillas and the State: The Cocalero Movement, Citizenship, and Identity in the Colombian Amazon. Trans. Andy Klatt. Durham: Duke University Press.
347
Rappaport, Joanne. 2007. Civil Society and the Indigenous Movement in Colombia. Social Analysis 51, 2: 107-123.
Rausch, Jane. 1984. The Tropical Plains Frontier: The Llanos of Colombia, 1531–1831. Albuquerque: University of New Mexico Press.
———. 1993. The Llanos Frontier in Colombian History 1830–1930. Albuquerque: University of New Mexico Press.
———. 1999. Colombia: Territorial Rule and the Llanos Frontier. Gainesville: University Press of Florida. ———. 2013. Territorial Rule in Colombia and the Transformation of the Llanos Orientales. Gainesville:
University of Florida Press. Reno, William, and Jahara Matisek. 2018. A New Era of Insurgent Recruitment: Have ‘New’ Civil
Wars changed the Dynamic?. Civil Wars 20, 3: 358-378. Reno, William. 2003. Political Networks in a Failing State: The Roots and Future of Violent Conflicts
in Sierra Leone. International Politics and Society 2: 29-43. República de Colombia. 2003. Política de Defensa y Seguridad Democrática. Ministerio de Defensa Nacional.
https://www.oas.org/csh/spanish/documentos/Colombia.pdf Restrepo, A.R., and M.C. Rodríguez. 2000. Flor de Abril : La Corriente de Renovación Socialista, de las Armas
a la Lucha Política Legal. Bogotá: Corporación Nuevo Arco Iris. Reyes Posada, Alejandro, Liliana Duica, and Wilber A. Pedraza. 2010. El despojo de tierras por paramilitares
en Colombia. Bogotá: Mimeo. Reyes Posada, Alejandro. 1978. Latifundio y Poder Político: La Hacienda Ganadera en Sucre. Bogotá: CINEP. ———. 2009. Guerreros y Campesinos: El Despojo de la Tierra en Colombia. Bogotá: Ariel. Richani, Nazih. 2002. Systems of Violence: The Political Economy of War and Peace in Colombia. Albany: SUNY
Press. Rivera Cusicanqui, Silvia. 1987. The Politics and Ideology of the Colombian Peasant Movement: the Case of
ANUC. Bogotá: CINEP. Robinson, James. 2013. Colombia: Another 100 Years of Solitude? Current History 112, 75: 43–48. Rodríguez, J.J. 1993. El Ejército del Siglo XX de Reyes a López. In Historia de las Fuerzas Militares de
Colombia, edited by A. Valencia Tovar. Bogotá: Planeta. Rojas, Cristina. 2009. Securing the State and Developing Social Insecurities: The Securitisation of
Citizenship in Contemporary Colombia. Third World Quarterly 30, 1: 227-245. Roldán, Mary. 2002. Blood and Fire: La Violencia in Antioquia, Colombia, 1943-1953. Durham: Duke
University Press. Romero, Mauricio. 2000. Changing Identities and Contested Settings: Regional Elites and the
Paramilitaries in Colombia. International Journal of Politics, Culture, and Society 14, 1: 51-69. ———. 2003. Paramilitares y autodefensas, 1982–2003. Bogotá: Planeta Colombiana. Ron, James. 2000. Territoriality and Plausible Deniability: Serbian Paramilitaries in the Bosnian War.
In Death Squads in Global Perspective, edited by B.B. Campbell and A.D. Brenner. New York: Palgrave Macmillan.
Ronderos, María Teresa. 2014. Guerras Recicladas. Bogotá: Penguin Colombia. Ross, Michael. 2004. What Do We Know about Natural Resources and Civil War? Journal of Peace
Research 41, 3: 337–356. Rotberg, Robert. 2002. The new nature of nation-state failure. The Washington Quarterly 25, 3: 83-96. Rubio, Mauricio. 2005. Illegal Armed Groups and Local Politics in Colombia. Journal of Drug
Issues 35, 1: 107–130. Rubin, Michael. 2020. Rebel Territorial Control and Civilian Collective Action in Civil War:
Evidence from the Communist Insurgency in the Philippines. Journal of Conflict Resolution 64, 2–3: 459–489.
348
Rucinque, Hector F. 1972. Colonization of the Sarare Region of Eastern Colombia. Madison: University of Wisconsin Press.
Rueschemeyer, Dietrich, Evelyne Huber Stephens, and John D. Stephens. 1992. Capitalist Development and Democracy. Chicago: University of Chicago Press.
Saénz Rovner, Eduardo. 2002. Colombia años 50: industriales, política y diplomacia. Bogota: Universidad Nacional de Colombia.
Safford, Frank, and Marco Palacios. 2002. Colombia: Fragmented Land, Divided Society. Oxford: Oxford University Press.
Salehyan, Idean. 2009. Rebels without Borders: Transnational Insurgencies in World Politics. Ithaca: Cornell University Press.
Sambanis, Nicholas. 2001. Do Ethnic and Nonethnic Civil Wars Have the Same Causes?: A Theoretical and Empirical Inquiry. Journal of Conflict Resolution 45, 3: 259–282.
Sánchez, Gonzalo, and Donny Meertens. 2001. Bandits, Peasants, and Politics: The Case of La Violencia in Colombia. Austin: University of Texas Press.
Sarbahi, Anoop. 2014. Insurgent Population Ties and the Variation in the Trajectory of Peripheral Civil Wars. Comparative Political Studies 47, 10: 1470-1500.
Sarmiento, Libardo. 2015. Arauca 1983-2015: Fin de un ciclo histórico y transición incierta. Informe de la Fundación Paz y Reconciliación.
Sauders Robert. 2011. Partners for Peace: Cooperative Popular Resistance and Peacebuilding in the Israeli-Palestinian Conflict. In Nonviolent Resistance in the Second Intifada. Middle East Today, edited by M.C. Hallward and J.M. Norman. New York: Palgrave Macmillan.
Scott, James. 1976. The Moral Economy of the Peasant. New Haven: Yale University Press. ———. 2000. Seeing Like a State: How Certain Schemes to Improve the Human Condition Have Failed. New
Haven: Yale University Press. Seawright, Jason, and John Gerring. 2008. Case Selection Techniques in Case Study Research: A Menu
of Qualitative and Quantitative Options. Political Research Quarterly 61, 2: 294–308. Semana. 1985. La Petro-Guerrilla. June 17. https://www.semana.com/economia/articulo/la-petro-
guerrilla/6573-3 ———. 1989. La ejecucion de un monseñor. November 6.
———. 1992. El gran negocio de la guerrilla. August 3. https://www.semana.com/nacion/articulo/el-gran-negocio-de-la-guerrilla/17890-3
———. 2002. El nuevo narcotráfico. September 23. www.semana.com/nacion/articulo/el-nuevo-narcotrafico/54190-3
———. 2003. No se movía una aguja sin permiso de la guerrilla. November 24. www.semana.com/nacion/articulo/no-movia-aguja-permiso-guerrilla/62088-3
———. 2004. El Miedo Arrinconado. October 30. https://www.semana.com/nacion/articulo/el-miedo-arrinconado/68919-3
———. 2006. Enemigos íntimos. April 15. https://www.semana.com/nacion/articulo/enemigos-intimos/78357-3
———. 2007a. Con la baja de ‘Martín Caballero’, el Bloque Caribe de las Farc queda herido de muerte. October 25. https://www.semana.com/on-line/articulo/con-baja-martin-caballero-bloque-caribe-farc-queda-herido-muerte/89051-3
———. 2007b. ¡Libre! January 6. https://www.semana.com/nacion/articulo/libre/82860-3 ———. 2008. Por muerte de sindicalistas en Arauca, destituido 20 años un coronel del Ejército.
September 2. https://www.semana.com/on-line/articulo/por-muerte-sindicalistas-arauca-destituido-20-anos-coronel-del-ejercito/94848-3
349
———. 2009a. "¡Ya cayó, ya cayó!" October 3. https://www.semana.com/nacion/articulo/ya-cayo-cayo/108253-3
———. 2009b. Once infantes de marina acusados de desaparición y homicidio. December 10. https://www.semana.com/nacion/justicia/articulo/once-infantes-marina-acusados-desaparicion-homicidio/110853-3
———. 2009c. Capturado el “hombre de las finanzas” de las Farc en Arauca. April 13. https://www.semana.com/nacion/conflicto-armado/articulo/capturado-hombre-finanzas-farc-arauca/102033-3
———. 2009d. Se voló importante jefe del ELN. October 7. https://www.semana.com/nacion/conflicto-armado/articulo/se-volo-importante-jefe-del-eln/108423-3
———. 2009e. Cómo el Ejército se alió con el ELN en Arauca. January 19. https://www.semana.com/nacion/articulo/como-ejercito-alio-eln-arauca/99226-3
———. 2020. Fiesta de sangre: así fue la masacre de El Salado. February 16. https://www.semana.com/nacion/articulo/masacre-de-el-salado-como-la-planearon-y-ejecutaron-los-paramilitares/557580
Skocpol, Theda. 1979. States and Social Revolutions. Cambridge: Cambridge University Press. Sinha, Aseema. 2005. The Regional Roots of Developmental Politics in India: A Divided Leviathan.
Bloomington: Indiana University Press. Sistema de Información de Desarrollo Rural (SIDER). Slater, Dan, and Erica Simmonds. 2010. Informative Regress: Critical Antecedents in Comparative
Politics. Comparative Political Studies 43, 7: 886–917. Sluka, Jeffrey. 1989. Hearts and Minds, Water and Fish: Support for the IRA and INLA in a Northern Irish
Ghetto. Bingley: JAI Press. Souleimanov, Emil Aslan, and Huseyn Aliyev. 2017. How Socio-Cultural Codes Shaped Violent Mobilization
and Pro-Insurgent Support in the Chechen Wars. London: Palgrave MacMillan. Staniland, P. 2012. States, Insurgents, and Wartime Political Orders. Perspectives on Politics 10, 2: 243-
264. ———. 2014a. Networks of Rebellion: Explaining Insurgent Cohesion and Collapse. Ithaca: Cornell University
Press. ———. 2014b, Counter-insurgency and Violence Management. In The New Counter-insurgency Era in
Critical Perspective, edited by C.W. Gventer, D.M. Jones, and M.L.R. Smith. London: Palgrave Macmillan.
Staniland, Paul. 2015. Militias, Ideology, and the State. Journal of Conflict Resolution 59, 5: 770–793. Stanton, Jessica A. 2015. Regulating Militias: Governments, Militias, and Civilian Targeting in Civil
War. Journal of Conflict Resolution 59, 5: 899–923. Steele, Abbey. 2017. Democracy and Displacement in Colombia’s Civil War. Ithaca: Cornell University Press. Steenkamp, Willem. 1989. South Africa’s Border War: 1966-1989. Warwick: Helion and Company. Steinmo, Sven, Kathleen Thelen, and Frank Longstreth (eds). 1992. Structuring Politics: Historical
Institutionalism in Comparative Analysis. Cambridge: Cambridge University Press. Stewart, Francis (ed). 2008. Horizontal Inequalities and Conflict Understanding Group Violence in Multiethnic
Societies. London: Palgrave MacMillan. Stoddart, D.R., and J.D. Trubshaw. 1962. Colonisation in Action in Eastern Colombia. Geography 47:
47–53. Suykens, Bert. 2015. Comparing Rebel Rule Through Revolution and Naturalization: Ideologies of
Governance in Naxalite and Naga India. In Arjona et al. 2015. 138–57. Sweig, Julia E. 2002. What Kind of War for Colombia? Foreign Affairs.
Taber, Robert. 1965. War of the Flea: The Classic Study of Guerrilla Warfare. New York: L. Stuart. Támara, Edgardo. 1997. Historia de Sincelejo. De los Zenúes al Packing House. Bogotá: Impreandes
Presencia. Tate, Winifred. 2007. Counting the Dead: The Culture and Politics of Human Rights Activism in Colombia.
Berkeley: University of California Press. Taydas, Zeynap, and Dursun Peksen. 2012. Can states buy peace? Social welfare spending and civil
conflicts. Journal of Peace Research 49, 2: 273–287. Thelen, Kathleen. 1999. Historical Institutionalism in Comparative Politics. Annual Review of Political
Science 2, 1: 369-404. Thomson, Andrew. 2018. Outsourced Empire How Militias, Mercenaries, and Contractors Support US Statecraft.
London: Pluto Press. Thoumi, Francisco. 2002. Illegal Drugs in Colombia: From Illegal Economic Boom to Social Crisis.
The Annals of the American Academy of Political and Social Science 582, 1: 102–116. Thyne, Clayton, and Ryan Schroeder. 2012. Social Constraints and Civil War: Bridging the Gap with
Criminological Theory. Journal of Politics 74, 4: 1066-1078. El Tiempo. 1991. En Arauca, la justicia soy yo. November 13.
www.eltiempo.com/archivo/documento/MAM-189787 ———. 1992. Arauca: El gobierno de la guerrilla. November 1.
https://www.eltiempo.com/archivo/documento/MAM-233646 ———. 1995a. El ELN maneja en Arauca 10 mil millones de pesos. April 2.
www.eltiempo.com/archivo/documento/MAM-308471 ———. 1995b. Guerrilla continua exterminio de mujeres en Saravena. May 12.
https://www.eltiempo.com/archivo/documento/MAM-325166 ———. 1996a. Radiografía del Domingo Laín. May 5.
www.eltiempo.com/archivo/documento/MAM-289644 ———. 1996b. FARC asesinan a 11 policías en Chalán II. March 14.
https://www.eltiempo.com/archivo/documento/MAM-337533 ———. 1996c. Guerrilleros colombianos famosos atacaron el puesto venezolano de Carabobo.
February 10. https://www.eltiempo.com/archivo/documento/MAM-363266 ———. 1997a. Arremetida guerrillera en Arauca, Tolima, y Antioquia. April 15.
https://www.eltiempo.com/archivo/documento/MAM-742853 ———. 1997b. ELN tumba helicóptero: 24 muertos. July 7.
https://www.eltiempo.com/archivo/documento/MAM-622218 ———. 1997c. Asesinado presidente de la ANUC. June 11.
https://www.eltiempo.com/archivo/documento/MAM-588373 ———. 1997d. Las bengalas salvaron a San Jacinto. February 8.
https://www.eltiempo.com/archivo/documento/MAM-583920 ———. 1997e. Asesinado subgerente de telecom en Sincelejo. September 2.
https://www.eltiempo.com/archivo/documento/MAM-620934 ———. 1998a. La Costa, en la mira de la subversión. Bolívar, noche llena de crueldad en Córdoba.
August 5. https://www.eltiempo.com/archivo/documento/MAM-834787 ———. 1998b. Historia de dos guerrillas en Arauca. July 2.
https://www.eltiempo.com/archivo/documento/MAM-820634 ———. 1999. Siguen combates en Arauca. June 25.
https://www.eltiempo.com/archivo/documento/MAM-921149 ———. 2001. Parálisis en Arauca por FARC. September 29.
———. 2002. Tres departamentos en zonas de rehabilitación. September 22. https://www.eltiempo.com/archivo/documento/MAM-1336334
———. 2003a. Cronología de la zona de rehabilitación de Arauca. March 9. https://www.eltiempo.com/archivo/documento/MAM-965808
———. 2003b. Se rajó la zona de Arauca. May 20. https://www.eltiempo.com/archivo/documento/MAM-990532
———. 2003c. Gobierno congela el giro de regalías para Arauca. January 29. https://www.eltiempo.com/archivo/documento/MAM-981951
———. 2003d. Indignación por muerte de menor. April 19. https://www.eltiempo.com/archivo/documento/MAM-970632
———. 2003e. Otro carro bomba en Arauca. January 27. https://www.eltiempo.com/archivo/documento/MAM-998666
———. 2003f. Arauca: Entre la guerra y la estigmatización. March 9. https://www.eltiempo.com/archivo/documento/MAM-966599
———. 2004a. Puerto Rondón: El Pueblo donde el tiempo se detuvo. May 30. https://www.eltiempo.com/archivo/documento/MAM-1554643
———. 2004b. Las trampas vietnamitas de las FARC en los Montes de María. November 2. https://www.eltiempo.com/archivo/documento/MAM-1505948
———. 2004c. Asesinan a padres de militar. September 12. https://www.eltiempo.com/archivo/documento/MAM-1503120
———. 2004d. El Conflicto al borde de la vía. July 21. ———. 2004e. Nadie le quiere llevar las camas al DAS. February 14.
https://www.eltiempo.com/archivo/documento/MAM-1499863 ———. 2005. Una nueva estrategia para Arauca. April 13.
https://www.eltiempo.com/archivo/documento/MAM-1679730 ———. 2007a. Así se tendió el cerco a Martín Caballero en Montes de María. October 26.
https://www.eltiempo.com/archivo/documento/MAM-2704992 ———. 2007b. Martín Caballero había armado el frente 37 con hijos, sobrinos, primos y vecinos.
October 28. https://www.eltiempo.com/archivo/documento/MAM-2707513 ———. 2007c. Martín Caballero' armó el frente 37 de las Farc con hijos, sobrinos, primos y vecinos.
October 27. https://www.eltiempo.com/archivo/documento/CMS-3786559 ———. 2007d. Gobierno Uribe da por muerto al ERP. May 2.
https://www.eltiempo.com/archivo/documento/MAM-2479563 ———. 2007e. 12 guerrilleros de los Montes de María dejaron sus armas y volvieron a la vida civil.
April 25. https://www.eltiempo.com/archivo/documento/CMS-3530715 ———. 2008a. Infiltrados de la Policía hasta tomaron cerveza con jefes de FARC. August 3.
https://www.eltiempo.com/archivo/documento/MAM-3039515 ———. 2008b. 'Paras' colgaron y degollaron a algunas de las víctimas de la masacre de 'El Salado'.
July 29. ———. 2012a. 'Mellizo' dice que él mismo mató a jefe de su grupo paramilitar. May 20.
https://www.eltiempo.com/archivo/documento/CMS-11842491 ———. 2012b. Álvaro, el papá de los tres niños que asesinó el subteniente Muñoz. September 25.
https://www.eltiempo.com/archivo/documento/CMS-12253746 ———. 2013a. Renace la Unión Patriótica. July 10.
https://www.eltiempo.com/archivo/documento/CMS-12924130 Tirado Mejía, Álvaro. 1981. López Pumarejo: la revolución en marcha. In Nueva Historia de Colombia,
edited by Álvaro Tirado Mejía. Bogotá: Planeta. Tilly, Charles. 1964. The Vendée. Cambridge: Harvard University Press.
352
———. 1969. Collective Violence in European Perspective. In Violence in America: Historical and Comparative Perspectives, edited by Hugh Davis Graham and Ted Robert Gurr. Washington D.C.: National Commission on the Causes and Prevention of Violence.
———. 1978. From Mobilization to Revolution. New York: McGraw-Hill. ———. 1992. Coercion, Capital, and European States, AD 990-1992. Hoboken: Wiley-Blackwell. ———. 1999. Durable Inequality. Berkeley: University of California Press. ———. 2004. Social Movements, 1768-2004. Boulder: Paradigm Publishers. Tollefsen, Andreas, and Halvard Buhaug. 2015. Insurgency and Inaccessibility. International Studies
Review 17, 1: 6–25. Trampusch, Christine, and Bruno Palier. 2016. Between X and Y: how process tracing contributes to
opening the black box of causality. New Political Economy 21, 5: 437-454. Tribunal Superior de Bogotá Sala de Justicia y Paz. 2017. Versión Libre (Eugenio José Reyes Regino). Trinquier, Roger. 1964. Modern Warfare: A French View of Counterinsurgency. London: Pall Mall
Press. Tsai, Kellee. 2016. Adaptive Informal Institutions. In The Oxford Handbook of Historical Institutionalism,
edited by edited by Orfeo S. Fioretos, Tulia G. Falleti, and Adam Sheingate. Oxford: Oxford University Press.
Tsebelis, George. 1990. Nested Games: Rational Choice in Comparative Politics. Berkeley: University of California Press.
Tucker, Spencer C. 2015. U.S. Conflicts in the 21st Century: Afghanistan War, Iraq War, and the War on Terror. Santa Barbara: ABC-CLIO.
Unión Camilista-Ejército de Liberación Nacional (UC-ELN). 1990. Poder popular y nuevo gobierno: conclusiones Segundo Congreso. Bogotá: Ediciones Colombia Viva.
United States Government Accountability Office. 2005. Security Assistance: Efforts to Secure Colombia’s Caño-Limón-Coveñas Oil Pipeline Have Reduced Attacks, But Challenges Remain. Report to Congressional Requesters.
Valentino, Benjamin, Paul Huth, and Dylan Balch-Lindsay. 2004. "Draining the Sea": Mass Killing and Guerrilla Warfare. International Organization 58, 2: 375-407.
Van Ausdal, Shawn. 2009. Pasture, profit, and power: An environmental history of cattle ranching in Colombia, 1850–1950. Geoforum 40, 5: 707-719.
Van Isschot, Luis. 2015. The Social Origins of Human Rights: Protesting Political Violence in Colombia’s Oil Capital, 1919–2010. Durham: Duke University Press.
Vargas, Gonzalo. 2009. Urban Irregular Warfare and Violence Against Civilians: Evidence From a Colombian City. Terrorism and Political Violence 21, 1: 110-132.
Vargas, Jenniffer, and Sonia Uribe. 2017. State, war, and land dispossession: The multiple paths to land concentration. Journal of Agrarian Change 17: 749-758.
Varshney, Ashutosh. 2001. Ethnic Conflict and Civil Society: India and Beyond. World Politics 53, 3: 362-398.
Velasco, Juan David. 2016. El voto uribista en los municipios colombianos: Patrones y significados (2002-2014). Análisis Político 30, 89: 3-37, 2017.
Verdad Abierta. 2008. ‘El Oso’, Marco Tulio Pérez. December 29. https://verdadabierta.com/perfil-marco-tulio-perez-alias-el-oso/
———. 2009a. ¿Conspiración de silencio?: Estados Unidos y la masacre de El Salado. September 23. https://verdadabierta.com/iconspiracion-de-silencio-estados-unidos-y-la-masacre-de-el-salado/
———. 2009b. Dos ex ‘paras’ confiesan tres masacres en Arauca. September 21. https://verdadabierta.com/dos-ex-paras-confiesan-tres-masacres-en-arauca/
353
———. 2010a. Desmovilizados contaron cómo llegaron los ‘paras’ a Arauca. March 15. https://verdadabierta.com/desmovilizados-contaron-como-llegaron-los-paras-a-arauca/
———. 2010b. ‘Mellizo’ Mejía aseguró que entraron a Arauca con la complicidad de miembros del Ejército. July 21. https://verdadabierta.com/mellizo-mejia-aseguro-que-entraron-a-arauca-en-complicidad-de-miembros-del-ejercito/
———. 2011. Un singular pacto de paz Eln-Farc. September 28. https://verdadabierta.com/un-singular-pacto-de-paz-eln-farc/
———. 2012a. Los Méndez, antecesores de los ‘paras’ en los Montes de María. February 29. https://verdadabierta.com/los-mendez-antecesores-de-los-paras-en-los-montes-de-maria/
———. 2012b. La violencia sexual de los ‘paras’ de Arauca. July 9. https://verdadabierta.com/los-delitos-sexuales-de-los-paras-de-arauca/
———. 2014. Las verdades de las Farc que exigen en Montes de María. Available at: https://verdadabierta.com/las-verdades-de-las-farc-que-exigen-en-montes-de-maria/
Villanueva Martínez, Orlando. 2012. Guadalupe Salcedo y la insurrección llanera, 1949-1957. Bogotá: Universidad Nacional de Colombia.
Viterna, Jocelyn. 2006. Pulled, Pushed, and Persuaded: Explaining Women’s Mobilization into the Salvadoran Guerrilla Army. American Journal of Sociology 112, 1: 1-45.
Von Clausewitz, Carl. 1918. On War. London: Trubner & Co. Waked Sánchez, Nathalia. 2013. ¿De cocaleros a cacaoteros?: Incorporación de una comunidad
campesina del piedemonte araucano a una economía legal. Congreso Nacional de Sociologia. Walter, Barbara F. 2002. Committing to Peace: The Successful Settlement of Civil Wars. Princeton: Princeton
University Press. Watts, Stephen, Jason H. Campbell, Patrick B. Johnston, Sameer Lalwani and Sarah H. Bana. 2014.
Countering Others' Insurgencies. Santa Monica: RAND Corporation. Weinstein, Jeremy. 2006. Inside Rebellion: the Politics of Insurgent Violence. Cambridge: Cambridge
University Press. Wickham-Crowley, Timothy. 1992. Guerrillas and Revolution in Latin America: A Comparative Study of
Insurgents and Regimes Since 1956. Princeton: Princeton University Press. ———. 2015. Del Gobierno de Abajo al Gobierno de Arriba …and Back: Transitions To and From
Rebel Governance in Latin America, 1956–1990. In Arjona et al. 2015. 47–73. Wiegand, Krista E. 2009. Reformation of a Terrorist Group: Hezbollah as a Lebanese Political Party.
Studies in Conflict & Terrorism 32, 8: 669-680. Wolf, Eric. 1969. Peasant Wars of the Twentieth Century. Tulsa: University of Oklahoma Press. Wood, Elisabeth Jean. 2003. Insurgent Collective Action and Civil War in El Salvador. Cambridge:
Cambridge University Press. ———. 2006. Variation in Sexual Violence during War. Politics & Society 34, 3: 307–342. Wood, Reed M. 2010. Rebel capability and strategic violence against civilians. Journal of Peace Research
47, 5: 601–614. Woods, Kevin M. 2020. Smaller-scale land grabs and accumulation from below: Violence, coercion
and consent in spatially uneven agrarian change in Shan State, Myanmar. World Development 127. Yates, Joe. 2004. Criminological Ethnography: Risks, Dilemmas and their Negotiation. In Policing in
Central and Eastern Europe: Dilemmas of Contemporary Criminal Justice, edited by Gorazd Mesko, Milan Pagon, and Bojan Dobovsek. University of Maribor.
Zamosc, León. 1986. The Agrarian Question and the Peasant Movement in Colombia: Struggles of the National Peasant Movement (1967–1981). Cambridge: Cambridge University Press.
Zedong, Mao. 2005. On Guerrilla War. Dover Publications. Zhukov, Yuri M. 2015. Population Resettlement in War: Theory and Evidence from Soviet Archives.
Journal of Conflict Resolution 59, 7: 1155-1185.
354
Field Interview Guide 1) Regional Expert/Researcher, Bogotá D.C., January 2016. 2) Employee at Defensoria del Pueblo, Arauca municipality, January 2016. 3) Former Public Defender, Arauca municipality, January 2016. 4) Journalist, Arauca municipality, January 2016. 5) Prison Warden, Arauca municipality, January 2016. 6) Social Worker, Arauca municipality, January 2016. 7) Police Chief, Arauca municipality, January 2016. 8) Oil union boss, Arauca municipality, January 2016. 9) Journalist, Arauca municipality, January 2016. 10) Journalist, Arauca municipality, January 2016. 11) Religious Leader, Arauquita, January 2016. 12) Cattle rancher, Arauquita, January 2016. 13) Ex-mayor, Arauquita, January 2016. 14) Civic leader, Arauquita, January 2016. 15) Ex-mayor, Arauquita, January 2016. 16) Ex-teacher/Union leader, Saravena, January 2016. 17) Journalist, Saravena, January 2016. 18) Indigenous Leader, Saravena, January 2016. 19) Human Rights Defender, Saravena, January 2016. 20) Teacher, Saravena, January 2016. 21) Human Rights Defender, Saravena, January 2016. 22) Internally Displaced Person, Fortul, January 2016. 23) Activist, Fortul, January 2016. 24) Civil servant, Saravena, January 2016. 25) Teacher, Tame, January 2016. 26) Regional Expert/Researcher, Cartagena, February 2016. 27) Regional Researcher, Cartagena, February 2016. 28) Internally Displaced Person, El Carmen de Bolívar, February 2016. 29) Victims Rights Leader, Ovejas, February 2016. 30) Victims Rights Leader, Ovejas, February 2016. 31) Victims Rights Leader, El Carmen de Bolívar, February 2016. 32) Internally Displaced Person, San Jacinto, February 2016. 33) Victims Rights Leader, Ovejas, February 2016. 34) Teacher, El Carmen de Bolívar, February 2016. 35) Victims Rights Leader, Ovejas, February 2016. 36) Victims Rights Leader, Ovejas, February 2016. 37) Civic leader, El Carmen de Bolívar, February 2016. 38) Regional Expert/Lawyer, Sincelejo, February 2016. 39) Internally Displaced Person, San Jacinto, February 2016. 40) Victims Rights Leader, El Carmen de Bolívar, February 2016. 41) Teacher, San Jacinto, February 2016. 42) Civic leader, San Onofre, February 2016. 43) Local habitant, San Onofre, February 2016. 44) Internally Displaced Person, San Onofre, February 2016. 45) Local habitant, San Onofre, February 2016. 46) Civic leader, San Onofre, February 2016.
355
47) Former Army General, Bogotá D.C., December 2016. 48) Civil servant, Bogotá D.C., July 2016. 49) Human Rights Defender, Arauca municipality, July 2016. 50) Peace activist, Arauca municipality, July 2016. 51) Human Rights Defender, Arauca municipality, July 2016. 52) Human Rights Defender, Arauca municipality, July 2016. 53) Civil servant, Arauca municipality, July 2016. 54) Internally Displaced Person, Arauca municipality, July 2016. 55) Internally Displaced Person, Arauca municipality, July 2016. 56) Internally Displaced Person, Arauca municipality, July 2016. 57) Local agricultural worker, Arauca municipality, July 2016. 58) Local agricultural worker, Arauca municipality, July 2016. 59) Religious Leader, Arauquita, July 2016. 60) Local agricultural worker, Arauquita, July 2016. 61) Religious Leader, Arauquita, July 2016. 62) Civic leader, Arauquita, July 2016. 63) Human Rights Defender, Tame, July 2016. 64) Civic leader, Arauquita, July 2016. 65) Youth leader, Arauca municipality, July 2016. 66) Civic leader, Saravena, July 2016. 67) Business leader, Saravena, July 2016. 68) Former teacher, Saravena, July 2016. 69) Victims Rights Defender, Arauca municipality, July 2016. 70) Human Rights Defender, Tame, July 2016. 71) Local historian, Tame, July 2016. 72) Journalist/Former mayor, Tame, July 2016. 73) Former mayor, Tame, July 2016. 74) Teacher, Tame, July 2016. 75) Civic leader, Saravena, July 2016. 76) Local businessman, Fortul, July 2016. 77) Civic leader, San Jacinto, August 2016. 78) Agricultural worker, San Jacinto, August 2016. 79) Internally Displaced Person, San Jacinto, August 2016. 80) Internally Displaced Person, San Jacinto, August 2016. 81) Civic leader, San Jacinto, August 2016. 82) Taxi Driver, El Carmen de Bolívar, August 2016. 83) Former journalist, El Carmen de Bolívar, August 2016. 84) Victims Rights Leader, El Carmen de Bolívar, August 2016. 85) Victims Rights Leader, El Carmen de Bolívar, August 2016. 86) JAC President, El Carmen de Bolívar, August 2016. 87) Demobilized FARC fighter, El Carmen de Bolívar, August 2016. 88) Demobilized EPR fighter, Sincelejo, August 2016. 89) Demobilized AUC fighter, El Carmen de Bolívar, August 2016. 90) Civic leader, El Carmen de Bolívar, August 2016. 91) Agricultural worker, Ovejas, August 2016. 92) Internally Displaced Person, Ovejas, August 2016. 93) Regional expert, Bogotá D.C., September 2016. 94) Regional Expert/Lawyer, Sincelejo, August 2016.
356
95) Regional Expert/Researcher, Bogotá D.C., December 2016. 96) Internally Displaced Person, Bogotá D.C., December 2016. 97) Former general of the Naval Armada, Bogotá D.C., December 2016. 98) Agricultural worker, Fortul, December 2016. 99) Internally Displaced Person, Tame, December 2016. 100) Internally Displaced Person, Tame, December 2016. 101) Regional Expert/Lawyer, Sincelejo, December 2016.