ABSTRACT 
 
 
 
Title of disertation: MICRODYNAMICS OF ILEGITIMACY AND COMPLEX 
URBAN VIOLENCE IN MEDEL?N, COLOMBIA 
 
 Robert Dale Lamb, Doctor of Philosophy, 2010 
 
Disertation directed by: Profesor John D. Steinbruner 
 School of Public Policy, University of Maryland 
 
 
For most of the past 25 years, Medel?n, Colombia, has been an extreme case of 
complex, urban violence, involving not just drug cartels and state security forces, but also 
stret gangs, urban guerilas, community militias, paramilitaries, and other nonstate 
armed actors who have controlled micro-teritories in the city?s densely populated slums 
in ever-shifting aliances. Before 2002, Medel?n?s homicide rate was among the highest 
in the world, but after the guerilas and militias were defeated in 2003, a major 
paramilitary aliance disarmed and a period of peace known as the ?Medel?n Miracle? 
began. Policy makers facing complex violence elsewhere were interested in finding out 
how that had happened so quickly. The research presented here is a case study of 
violence in Medel?n over five periods since 1984 and at two levels of analysis: the city as 
a whole, and a sector caled Caicedo La Siera. The objectives were to describe and 
explain the paterns of violence, and determine whether legitimacy played any role, as the 
literature on social stability suggested it might. Multilevel, multidimensional frameworks 
for violence and legitimacy were developed to organize data collection and analysis. The 
 
study found that most decreases in violence at al levels of analysis were explained by 
increases in teritorial control. Increases in collective (organized) violence resulted from a 
proces of ?ilegitimation,? in which an intolerably unpredictable living environment 
sparked internal opposition to local rulers and raised the costs of teritorial control, 
increasing their vulnerability to rivals. As this violence weakened social order and the 
rule of law, interpersonal-communal (unorganized) violence increased. Over time, the 
?true believers? in armed political and social movements became marginalized or 
corrupted; most organized violence today is motivated by money. These findings imply 
that state actors, facing resurgent violence, can keep their tenuous control over the 
hilside slums (and other ?ungoverned? areas ) if they can avoid ilegitimizing 
themselves. Their priority, therefore, should be to establish a tolerable, predictable daily 
living environment for local residents and busineses: other anti-violence programs wil 
fail without strong, permanent, and respectful governance structures. 
 
 
 
 
 
 
 
 
MICRODYNAMICS OF ILEGITIMACY AND 
COMPLEX URBAN VIOLENCE IN 
MEDEL?N, COLOMBIA 
 
 
By Robert Dale Lamb 
 
 
 
 
Disertation submited to the Faculty of the Graduate School of the 
University of Maryland, College Park, in partial fulfilment of 
the requirements for the degre of 
Doctor of Philosophy 
2010 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
Advisory Commite: 
 
 Profesor John D. Steinbruner, Chair 
 Doctor Nancy Galagher 
 Profesor Karol Soltan 
 Doctor David A. Crocker 
 Profesor Carol Le Graham 
 
 
? Copyright by 
Robert Dale Lamb 
2010 
 
 ii 
Dedication 
Dale Russel Lamb 
1943?1981 
 
Patricia Anne Russel 
1970?1994 
 
 iii 
Acknowledgments 
This research was supported in part by a World Politics and Statecraft Felowship 
from the Smith-Richardson Foundation, and in part by a graduate felowship in the 
Advanced Methods of Cooperative Security Program at the Center for International and 
Security Studies at Maryland (CISM), funded by the John D. and Catherine T. 
MacArthur Foundation and the Carnegie Corporation of New York. Their support is 
greatly appreciated. 
I want to thank the members of my disertation comite: my chair, John 
Steinbruner, who always gave me enough rope to hang myself with, but never let me get 
to the point of hanging: his sound and timely interventions have kept this project on 
track; Nancy Galagher, who has been my felowship supervisor and a constant mentor 
throughout the proces; Karol Soltan, who chalenged me to succinctly summarize the 
literature on legitimacy in a way that would distinguish it from the literature on rational 
choice (se the introduction to Appendix C for my response); David Crocker, whose 
clases on development and democracy sparked my interest in the relationship betwen 
legitimacy and governance; and Carol Graham, whose work on the economics of 
happines has inspired me to look more deeply at the development literature. Al have 
offered useful advice, encouragement, and flexibility as my research topic evolved. 
I presented an early draft of my research topic at a seminar in Moscow sponsored 
by CISM and the Institute for U.S. and Canada Studies of the Russian Academy of 
Sciences (ISKRAN). The critical fedback I got there set me on a diferent, and what I 
 iv 
believe to have been a more productive, path; I thank especialy Thomas Scheling for his 
incisive comments, and the students and faculty of ISKRAN for their fedback. Kevin 
Jones, Tim Gulden, Tom Cosentino, Nate Freier, and many other CISM felows have 
provided helpful fedback and insights along the way as wel. I thank Bruce Giley for 
leading me early on to some useful references on measuring legitimacy. 
Carlos Salinas deserves special mention, due not only to his many years of 
friendship and support, but also to the encouragement and opportunities he has given me 
over the years to study human rights in general and Colombia in particular. In that vein, I 
am grateful as wel to Desmond Arias for suggesting that, if I was looking to study a 
place where the violence was once brutal and complex but that had recently become safe 
enough to live in, I should consider Medel?n. I appreciate John Hamre and Craig Cohen 
at the Center for Strategic and International Studies (CSIS) for offering me a visiting 
felowship, and plenty of flexibility, during the final stages of writing so that I could earn 
income while finishing my thesis ? and for offering me my dream job when it was 
complete. Thanks go also to Sanho Tre, Amanda Martin, Jana Silverman, Jim 
Zackrison, Cristina Espinel, Charlie Roberts, Jes?s Balb?n, Pablo Angarita, Juan 
Guilermo Buenaventura, Juan Pablo Barios, and Ricardo San?n for putting me in touch 
with needed contacts and for helping orient me to the physical or intelectual landscapes 
of Medel?n and Bogot?. Clara Atehort?a, Olga Jaramilo, Mary Luz Restrepo, and Diego 
and ?ngela Trujilo deserve special thanks for their friendship and their honest insiders? 
perspectives on the cultural, political, and linguistic idiosyncrasies of their city. 
My biggest thanks in Colombia, however, have to go to my main research 
asistant, Janeth Restrepo, without whose asistance this study would be impoverished. I 
 v 
found Janeth by dumb luck, and she ended up being the most competent and dedicated 
asistant I could have hoped for: she understood and anticipated my research needs, set 
up and acompanied me on sensitive field interviews, acted as my interpreter for slang-
speaking subjects, and occasionaly took over in the middle of interviews when she saw I 
was mising some nuance, the value of which I did not recognize until the transcripts 
came back. I very much appreciate her inteligence and patience, her sense of humor, and, 
above al, her friendship. 
Not al of the organizations or research asistants I approached with requests for 
asistance or data were helpful, so I want to expres my gratitude to those staf members, 
too numerous to mention individualy, who did give me the help I needed, including 
those at the office of the Government Secretariat (Secretar?a de Gobierno de Medel?n), 
the Medel?n office of the national public prosecutor (Fiscal?a General de la Naci?n), the 
Department of National Statistics Administration (DANE: Departamento Administrativo 
Nacional de Estad?stica), the national police department (Polic?a Nacional), the National 
Institute of Forensic Medicine and Sciences (Instituto Nacional de Medicina Legal y 
Ciencias Forenses), the National Labor School (Escuela Nacional Sindical), the People?s 
Training Institute (IPC: Instituto Popular de Capacitaci?n), and the Medel?n C?mo 
Vamos project at the ofice of the mayor. I thank especialy Juan Carlos Palou and Maria 
Victoria Llorente at the Fundaci?n Ideas para la Paz for permision to use a dificult-to-
aces dataset. I am enormously grateful as wel to al the subject-mater experts who 
took time to speak with me, to the many residents of Caicedo La Siera (mostly unnamed 
for security reasons) who shared their life stories with me, and to the countles residents 
 vi 
of Medel?n who chated with me informaly and educated me on the nuances of their 
city?s culture and history. 
I would not be who I am today without the unconditional love and thoughtful 
support that my mother, Barbara Lamb, has always given me. She has always been my 
biggest fan and probably does not realize how much I have appreciated it. I am grateful to 
her and to the rest of my family for their love and support over the years. 
Before I entered graduate school, I was craving a higher level of intelectual 
engagement and real-world influence than my carer as a journalist and research asistant 
was offering me. Sometime after the terorist atacks of 11 September 2001, my wife, 
Amy Muhlbach, saw me writing something that later inspired her to leave a sticky note 
on my computer: ?Bob, go to grad school.? In the years that followed she tolerated a lot, 
and when I proposed that we pack up and move to South America so I could do field 
research for a year, she did not hesitate; while there, she kept me sane and on track. I 
cannot imagine a more supportive friend or loving wife. Thank you. 
 
 vii 
Table of Contents 
Dedication......................................................................................................................ii 
Acknowledgments........................................................................................................iii 
Table of Contents........................................................................................................vii 
List of Tables..............................................................................................................xiv 
List of Figures..............................................................................................................xv 
Abreviations.............................................................................................................xvi 
Chapter 1. Introduction................................................................................................1 
1.1. Research Design...................................................................................................3 
1.2. Findings..............................................................................................................11 
1.3. Implications and Recommendations....................................................................15 
1.4. Roadmap............................................................................................................19 
Part I. Ilegitimacy and Violence in Medel?n, Colombia...........................................22 
Chapter 2. Background: Medel?n and Caicedo La Siera........................................26 
2.1. Medel?n: From the Conquest of Nutibara to the Rise of Cocaine........................29 
2.1.1. A Brief Tour of the Valey of Aburr?...........................................................29 
2.1.2. Bad-Ases, Punks, and Low-Lifes Take Over the City.................................36 
2.1.3. Pablo Escobar Builds the Medel?n Cartel....................................................47 
2.2. Caicedo La Siera: A Tour through East Central Medel?n..................................72 
Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992...................79 
3.1. Medel?n: The Insurgency Comes to the City......................................................86 
 vii 
3.1.1. Peace Camps Train Future Wariors.............................................................86 
3.1.2. Militias Emerge to Defend against Crime.....................................................92 
3.2. Caicedo La Siera: From Invasions to Statelets...................................................99 
3.2.1. The Invasions Bring Crime, Violence, and Gangs........................................99 
3.2.2. The Rise of M-6&7....................................................................................104 
3.2.3. The Rise of La Ca?ada...............................................................................108 
Chapter 4. Gangs and the Corruption of Guerila Militias: 1992-1998.................110 
4.1. Medel?n: A War of Al against Al...................................................................113 
4.1.1. The Militias Start to Lose Control..............................................................113 
4.1.2. Coosercom and the Peace Talks that Brought War.....................................119 
4.1.3. Gangs Learn a Leson from the Militia Experience....................................123 
4.1.4. The City Responds to the Violence............................................................126 
4.2. Caicedo La Siera: Unlocking the Doors to More Violence...............................133 
4.2.1. Two Children Fight, and the Result Is War.................................................133 
4.2.2. Life in an Urban War Zone........................................................................135 
4.2.3. ?Mama Luz? and the Community Peace Talks............................................149 
4.2.4. The Ilegitimation of the Militias................................................................160 
Chapter 5. Mafias and the Emergence of Paramilitaries: 1998-2002......................164 
5.1. Medel?n: ?Don Berna,? ?Double Zero,? and the Complex War..........................181 
5.1.1. ?Double Zero? Enters Medel?n...................................................................181 
5.1.2. ?Don Berna? Joins the Fight.......................................................................185 
5.1.3. Cacique Nutibara Plays ?Monopoly?...........................................................186 
5.1.4. The Batle for the West..............................................................................188 
 ix 
5.2. Caicedo La Siera: ?Job,? ?Scab,? ?Doll,? and the Final Batle............................192 
5.2.1. Metro Takes Over M-6&7..........................................................................192 
5.2.2. ?Job? Takes Over Cacique Nutibara............................................................196 
5.2.3. The Batle for the East................................................................................197 
5.2.4. Cacique Nutibara Targets Metro.................................................................201 
Chapter 6. Traffickers and the Corruption of Paramilitaries: 2002-2007..............207 
6.1. Medel?n During the ?Miracle? Period...............................................................223 
6.1.1. The Fal of the ?Political? Paras..................................................................223 
6.1.2. The Fal of the ?Narco? Paras.....................................................................232 
6.1.3. The Rise of the ?City of Eternal Spring?.....................................................241 
6.2. Caicedo La Siera: ?Job,? ?Mem?n? and the Quiet War.......................................256 
6.2.1. ?Mem?n? and the Death of ?Alberto Ca?ada?..............................................256 
6.2.2. An Uneasy Peace, with Hope for the Future...............................................265 
Chapter 7. A Window of Oportunity, Closing Quickly: 2007-2009.......................271 
7.1. Medel?n: Dark Forces and Black Eagles...........................................................276 
7.2. Caicedo La Siera: Rising Fear amid Hope for the Future.................................284 
7.2.1. The Fear of Strangers Returns....................................................................284 
7.2.2. Selective Violence Returns.........................................................................288 
7.2.3. The Hopes of the Community....................................................................291 
Part I. Analysis and Conclusions.............................................................................301 
Chapter 8. Complex Violence in Medel?n................................................................304 
8.1. Paterns of Violence..........................................................................................308 
8.2. Homicide Data..................................................................................................311 
 x 
8.3. Self-Inflicted Violence......................................................................................319 
8.4. Interpersonal Violence......................................................................................323 
8.4.1. Intrafamilial Violence................................................................................323 
8.4.2. Communal Violence..................................................................................328 
8.5. Collective Violence..........................................................................................340 
8.5.1. Social Violence..........................................................................................345 
8.5.2. Political Violence.......................................................................................348 
8.5.3. Economic Violence....................................................................................352 
Chapter 9. Legitimacy and Ilegitimacy in Medel?n................................................356 
9.1. Proxy Indicators................................................................................................363 
9.2. Causal Indicators..............................................................................................365 
9.2.1. Transparent................................................................................................368 
9.2.2. Credible.....................................................................................................368 
9.2.3. Justifiable...................................................................................................369 
9.2.4. Equitable....................................................................................................371 
9.2.5. Acesible..................................................................................................372 
9.2.6. Respectful..................................................................................................372 
9.3. Analysis............................................................................................................373 
9.3.1. Legitimacy of Nonstate Armed Actors.......................................................374 
9.3.2. Legitimacy of Community Actors..............................................................390 
9.3.3. Legitimacy of State Actors.........................................................................407 
Chapter 10. Implications...........................................................................................412 
10.1. Summary of Findings......................................................................................412 
 xi 
10.2. Implications for Policy....................................................................................414 
10.2.1. Protect and Respect Residents of Peripheral Barios.................................414 
10.2.2. Protect Busineses from Protection Rackets (?Vacunas?)..........................418 
10.2.3. Prepare for ?Invasions? by Internaly Displaced Persons...........................418 
10.2.4. Kep Data-Collection Programs Funded and Transparent.........................419 
10.3. Implications for Strategy.................................................................................420 
10.4. Implications for Theory..................................................................................423 
Part II. Apendices..................................................................................................428 
Apendix A. Policy, Globalization, and Governance...............................................429 
A.1. Definitions.......................................................................................................432 
A.2. Policy and Globalization..................................................................................445 
A.3. Policy, Governance, and ?Ungoverned? Areas..................................................463 
Apendix B. Complex Violence................................................................................481 
B.1. Manifestations of Violence...............................................................................490 
B.2. Context of Violence.........................................................................................495 
Apendix C. Legitimacy and Ilegitimacy................................................................501 
C.1. What Is Legitimacy, and What Does It Do?......................................................507 
C.1.1. Legitimacy, Loyalty, Support, Right, and Duty..........................................523 
C.1.2. Ilegitimacy and Opposition.......................................................................526 
C.2. What Else Does What Legitimacy Can Do?.....................................................529 
C.2.1. Agrement and Habit.................................................................................530 
C.2.2. Seduction, Persuasion, and Compromise...................................................531 
C.2.3. Force, Coercion, and Barter.......................................................................534 
 xii 
C.2.4. Deception..................................................................................................536 
C.3. Legitimacy of What?........................................................................................536 
C.4. Legitimacy Acording to Whom?.....................................................................541 
C.5. Legitimacy by What Proceses?.......................................................................548 
C.6. Legitimacy by What Criteria?..........................................................................549 
C.7. Measurement Framework.................................................................................559 
Apendix D. Research Design and Hypotheses........................................................564 
D.1. Motivation.......................................................................................................564 
D.2. Method: Single-Case Study with In-Case Variation..........................................568 
D.3. Data Sources....................................................................................................571 
D.4. Symbols...........................................................................................................576 
D.5. Model Hypotheses...........................................................................................578 
Model Hypothesis 1: Not Legitimacy...................................................................584 
Model Hypothesis 2: Legitimacy as Causal Phenomenon.....................................594 
Model Hypothesis 3: Legitimacy as Caused Phenomenon....................................597 
Model Hypothesis 4: Legitimacy as Intervening Phenomenon.............................600 
Model Hypothesis 5: Legitimacy as Antecedent Phenomenon..............................603 
Model Hypothesis 6: Legitimacy as Fedback.....................................................604 
D.6. Rejected Hypotheses........................................................................................605 
Hypothesis i.........................................................................................................606 
Hypothesis i........................................................................................................607 
Hypothesis ii.......................................................................................................608 
Hypothesis iv.......................................................................................................609 
 xii 
Hypothesis v........................................................................................................610 
Hypothesis vi.......................................................................................................611 
Hypothesis vii......................................................................................................612 
Hypothesis vii.....................................................................................................613 
Hypothesis ix.......................................................................................................614 
D.7. Synthesized Models.........................................................................................615 
Bibliography..............................................................................................................619 
 xiv 
List of Tables 
Table 2-1. Brief Glossary of Key Spanish, Lunfardo, and Parlache Terms.....................40 
Table 8-1. Correlation Coeficients among Five Measures of Homicides.....................312 
Table 8-2. Crime Report Correlations with Diferent Homicide Measures...................337 
Table B-1. Contexts and Manifestations of Violence, with Examples..........................497 
Table C-1. A Framework for Measuring Legitimacy...................................................559 
Table C-2. Criterial Measures of Legitimacy...............................................................562 
Table D-1. Six Model Hypotheses...............................................................................584 
Table D-2. Nine Rejected Hypotheses and Two Synthesized Models...........................606 
 
 
 xv 
List of Figures 
Figure 8-1. Five Measures of Homicides in Medel?n, 1984-2008................................313 
Figure 8-2. Homicides in Medel?n, by Month, 1987-2009...........................................314 
Figure 8-3. Comparison of Homicides across Levels of Analysis.................................316 
Figure 8-4. Homicide Rate in Medel?n, 1967-2008.....................................................319 
Figure 8-5. Suicides in Medel?n, 1990-2006...............................................................320 
Figure 8-6. Trends in Suicides versus Homicides.........................................................321 
Figure 8-7. Trends in Common-Crime Reports versus Homicides................................334 
Figure 8-8. Social Cleansing in Antioquia versus Homicides in Medel?n....................347 
Figure 8-9. Trends in Political Violence versus Homicides..........................................350 
 
 
 xvi 
Abreviations 
ACCU: Autodefensas Campesinas de C?rdoba y Urab?, Peasant Self-Defense Forces of 
C?rdoba and Urab? (paramilitary) 
ACM: Autodefensas Campesinas de Magdalena Medio, Peasant Self-Defense Forces of 
Magdalena Medio (paramilitary) 
AGC: Autodefensas Gaitanistias de Colombia, Gaitanist Self-Defense Forces of 
Colombia (paramilitary) 
AP: Asociated Pres 
AUC: Autodefensas Unidas de Colombia, United Self-Defense Forces of Colombia 
(paramilitary network) 
BCE: before common era; equivalent to BC (before Christ) 
Coosercom: Cooperativa de Seguridad y Servicios a la Comunidad, Community Services 
and Security Cooperative (demobilized militias) 
DANE: Departamento Administrativo Nacional de Estad?stica, Department of National 
Statistics Administration (state census bureau) 
DR: demobilization, disarmament, and reinsertion 
DoD: United States Department of Defense 
DoS: United States Department of State 
ELN: Ej?rcito de Liberaci?n Nacional, National Liberation Army (guerila) 
EPL: Ej?rcito Popular de Liberaci?n, People?s Liberation Army, sometimes translated as 
Popular Liberation Army (guerila) 
 xvii 
FARC: Fuerzas Armadas Revolucionarios de Colombia, Revolutionary Armed Forces of 
Colombia (guerila) 
FTO: Foreign Terorist Organization (DoS designation) 
GDP: gross domestic product 
GWOT: Global War on Terorism 
IDP: internaly displaced person 
JAC: Junta de Aci?n Comunal, Community Action Board (state body) 
JAL: Junta Administradora Local, Local Administration Board (state body) 
M-19: Movimiento Abril de 19, April 19
th
 Movement (guerila) 
M-6&7: Milicias 6 y 7 de Noviembre, November 6-7 Militia (urban militia) 
MAS: Muerte a Secuestadores, Death to Kidnappers (death squad) 
MoD: Colombian Ministry of Defense 
MPP: Milicias Populares del Pueblo y para el Pueblo Popular, Militias of the People and 
for the People (urban militia) 
MPVA: Milicias Populares del Vale de Aburr?, People?s Militias of the Abur? Valey 
(urban militia) 
Pepes: Perseguidos por Pablo Escobar, People Persecuted by Pablo Escobar (death squad) 
PLO: Palestine Liberation Organization 
PUI: Plan Urban?stico Integral, Comprehensive Urban Plan 
UGA/SH: ?Ungoverned Areas and Threats from Safe Havens? (Lamb, 2008) 
UNDP: United Nations Development Programe 
UP: Uni?n Patri?tica, Patriotic Union (political party) 
USOUTHCOM: United States Southern Command 
Lamb, Robert D. | Chapter 1. Introduction 1 
Chapter 1. Introduction 
If you have been paying atention, but not too closely, you might recognize the 
following story. Medel?n, Colombia, known worldwide as the hometown and 
headquarters of Pablo Escobar and his Medel?n Cartel, was at one point during the early 
1990s the second most violent city in the world, behind only Beirut. Escobar?s death in a 
rooftop shootout with police in 1993 did litle to curb the violence, as the urban wars 
there only become more complex during the 1990s: wars betwen rival stret crews and 
rival drug gangs; betwen drug gangs and community vigilante groups; betwen mafia-
backed drug gangs and guerila-backed militias acting as community self-defense 
vigilantes; betwen guerila-backed militias and mafia-backed anti-guerila paramilitary 
self-defense vigilantes; and betwen the state security forces and guerilas, gangs, 
militias, and ? wel, who else you got? By 2002, the homicide rate in Medel?n was 
among the highest in the world. It was Falujah before Falujah was Falujah. Even Beirut 
had become somewhat of a tourist destination by the turn of the milennium. 
But something funny happened the following year: after Colombia?s security 
forces succesfully ousted the last of the guerila-backed militias from the city, the 
biggest, baddest, anti-guerila?narco-terorist?paramilitary?vigilante?pimp?hit-squad?
racketering mafia ? the description is not an exaggeration ? laid down its weapons, 
declared peace, and rentered society as responsible, and forgiven, citizens. In 2004, 
homicides plummeted. By 2006, the murder rate was actualy higher in Washington, 
Baltimore, Detroit, and Miami than it was in Medel?n. Huge community libraries and 
Lamb, Robert D. | Chapter 1. Introduction 2 
public-transportation cable cars were being built in the former war zones of the city?s 
hilside periphery. People were staying out late, dancing downtown again. Hippie 
backpackers from Europe and North America were showing up unannounced, hanging 
out for a few eks, then moving on to their next cheap tourist destination with hardly an 
unpleasant experience to report in their travel blogs. There was a housing boom, a 
busines boom, a self-confidence boom. Articles were appearing in the international pres 
with titles like ?Sustaining the Medel?n Miracle? and ?Medel?n?s Makeover.?
1
 
The city was an international succes story, and policy makers facing complex 
urban violence in their own countries began showing up looking for advice. U.S. policy 
makers started talking about replicating its aid package to Colombia in other countries, 
such as Mexico, where drug violence was starting to look like Medel?n?s in the 1980s, or 
Afghanistan, where the drug trade had long funded civil wars and insurgencies but 
increasingly benefited pro-Taliban forces. These contexts may or may not have been 
comparable to the situation in Medel?n. Nevertheles, to many observers, Medel?n had 
become an example of how to overcome a very complex and very violent policy problem 
in a very short period of time, and many were interested in the answer to a question: How 
did this happen? 
It turns out, however, that their question was premature. The decline in the 
homicide rate since 2003, the most widely cited indicator of the ?Medel?n Miracle,? 
started to reverse itself sometime during 2007 or 2008 and skyrocketed during 2009: 
during just the first seven months of 2009, Medel?n witnesed more murders than it had 
                                                
1
 Anthony Faiola, ?Sustaining the Medelin Miracle: Colombia Strugles to Hold on to Gains from 
Globalization,? The Washington Post, 11 July 208; Forest Hylton, ?Medel?n?s Makeover,? New Left 
Review 44 (207): 70. 
Lamb, Robert D. | Chapter 1. Introduction 3 
in al of 2008. In short, judging from the data, the miracle was over: the murder rate had 
returned to the city?s 2003 levels. 
Before policy makers around the world start taking lesons from Medel?n?s 
experience ? and before critics start pointing told-you-so fingers at Medel?n?s elite ? 
the question to ask is not how the supposed miracle happened, but rather: What actually 
happened? 
1.1. Research Design 
To answer that, I moved to Medel?n in August 2008, when it was stil unclear 
whether the ?miracle? was ending, and spent the next ten months geting to know the 
people and the culture, consulting experts, interviewing residents, collecting documents 
and data, reading books and newspapers, and engaging in informal conversations with 
dozens of common people, cab drivers, teachers, social workers, and friends of friends. I 
developed a framework, based on one developed in the public-health field, to 
systematicaly analyze violence: not only homicides, not only violence by or against 
certain populations, and not only violence during a single period of time ? the standard 
approaches of most work on violence in Medel?n ? but the full range of physical and 
psychological forms of violence involving many diferent populations over five more or 
les distinct periods over the past 25 years. From previous research, I knew that most 
problems of violence and social instability in history have been strongly asociated with 
some kind of legitimacy deficit, and so, finding few orks that had dealt with legitimacy 
in Medel?n both systematicaly and thoroughly, I modified and expanded a framework I 
Lamb, Robert D. | Chapter 1. Introduction 4 
had previously developed for analyzing legitimacy, so that it could be used to analyze 
legitimacy and ilegitimacy across multiple levels of analysis.
2
 
Because life takes place at multiple levels but is experienced most imediately at 
the micro level, I studied the case at two levels of analysis: the city of Medel?n as a 
whole, and a smal sector within the city. To capture within-case variation, I studied both 
levels over five periods in recent history, divided acording to whether homicides (as a 
proxy for violence) were declining or rising: the period 1984 to 1992 experienced a 
dramatic increase in homicides at both levels of analysis; the period 1992 to 1998 
witnesed fluctuations in the magnitude of homicides but with an overal downward 
trend; this was followed by another spike in violence betwen 1998 and 2003; the period 
2003 to 2007 experienced a dramatic decline in homicides at both levels; and, finaly, 
beginning in 2007 or 2008 homicides began an ominous rise that continued into 2009. 
That design made it possible to compare the thre periods in which homicides were rising 
with each other and with the two periods in which homicides were faling; and to 
compare the two declining periods with each other and with the thre rising periods. 
Other forms of violence turned out either to track the paterns in homicides reasonably 
wel or to go against them in ways that did not invalidate the division of time. 
Using these frameworks as a systematic way to analyze and organize the 
quantitative and qualitative data I collected, I tried to answer thre basic questions that 
sem never to have been asked before: 
                                                
2
 The previous research was Robert D. Lamb, Ungoverned Areas and Threats From Safe Havens, final 
report of the Ungoverned Areas Project (Washington: Under Secretary of Defense for Policy, 2 January 
2008), http:/ww.cism.umd.edu /papers/display.php?id=306. The earlier framework was discussed in 
Robert D. Lamb, ?Measuring Legitimacy in Weak States? (paper presented at the Graduate Student 
Conference on Security, Georgetown University, Washington, 18 March 205). 
Lamb, Robert D. | Chapter 1. Introduction 5 
1. What have the overal paterns of violence been during the past 25 years? 
2. What has caused those paterns to change? 
3. What role did legitimacy or ilegitimacy play in those changes? 
The first question ? what were the paterns of violence in Medel?n? ? was 
meant to go beyond those works that study only one type of violence (such as 
homicides),
3
 only one category of perpetrator (such as gangs or paramilitaries),
4
 only one 
type of victim (such as displaced populations or trade unionists),
5
 only one brief period of 
time (such as annual human rights reports),
6
 or only one case within the city (such as an 
                                                
3
 e.g. Marleny Cardona, et al., ?Homicidios en Medel?n, Colombia, Entre 190 y 202: Actores, M?viles y 
Circunstancias,? [?Homicides in Medel?n, Colombia, from 1990 to 2002: Victims, Motives and 
Circumstances,?] Cadernos de Sa?de P?blica 21, no. 3 (205): 840. 
4
 e.g. Ralph Rozema, ?Urban DR-Proceses: Paramilitaries and Criminal Networks in Medel?n, 
Colombia,? Journal of Latin American Studies 40, no. 3 (208): 423; Edwin Cruz Rodr?guez, ?Los 
Estudios Sobre el Paramilitarismo en Colombia,? [?Studies of Paramilitarism in Colombia,?] An?lisis 
Pol?tico No. 60 (207): 117; Gilberto Medina Franco, Una Historia de las Milicias de Medel?n (Historia 
Sin Fin) [A History of the Militias of edel?n (A Never-Ending Story)] (edel?n, Colombia: Instituto 
Popular de Capacitaci?n?IPC, 206), htp:/bibliotecavirtual.clacso.org.ar/ar/libros/colombia 
/ipc/historiamilicias.pdf (acesed 10 April 209); Ramiro Ceballos Melguizo and Francine Cronshaw, 
?The Evolution of Armed Conflict in Medelin: An Analysis of the ajor Actors,? Latin American 
Perspectives 28, no. 1 (201): 110; Alonso Salazar J., No Nacimos Pa? Semila: La Cultura de las Bandas 
Juveniles de Medell?n [Born to Die: The Culture of Youth Gangs in Medell?n], 5? ed. (Bogot?: Centro de 
Investigaci?n y Educaci?n Popular?CINEP, 191). 
5
 e.g. Clara In?s Atehort?a Aredondo, ?Caracterizaci?n del Desplazamiento Forzado Intraurbano: Medel?n 
2000-2004? [?Characterizing Intra-Urban Forced Displacement: Medel?n 200-2004?] (Master?s thesis, 
Universidad de Antioquia, 207). 
6
 e.g. C?sar Augusto Mu?oz Restrepo, ed., Que los Arboles Dejen Ver el Bosque: Derechos Humanos en 
Antioquia?2005 [That the Tres Might Let the Forest See: Human Rights in Antioquia?2005] (Medel?n, 
Colombia: Instituto Popular de Capacitaci?n (IPC) de la Corporaci?n de Promoci?n Popular, 206); 
Instituto Popular de Capacitaci?n (IPC), ?Situaci?n de Violencia, Conflicto Urbano y Derechos Humanos 
en Medell?n,? [?The Status of Violence, Urban Conflict, and Human Rights in Medel?n,?] Observatorio de 
Derechos Humanos y Conflicto Armado (204); Roberto Armando Moreno Bedoya, Conflicto Urbano y 
Derechos Humanos en Medel?n: Balance desde Diferentes Sectores Sociales 202 [Urban Conflict and 
Human Rights in Medel?n: An Asesment of Diferent Social Sectors, 202], 1a ed., Re Lecturas, No. 26 
(Medel?n, Colombia: Instituto Popular de Capacitaci?n (IPC) de la Corporaci?n de Promoci?n Popular, 
2003); Rafael Rinc?n P., ed., Antioquia, Fin de Milenio: ?Terminar? la Crisis del Derecho Humanitario? 
[Antioquia at the End of the Milenium: Wil the Humanitarian Crisis End?] (Medel?n, Colombia: 
Instituto Popular de Capacitaci?n (IPC) de la Corporaci?n de Promoci?n Popular, 1999). 
Lamb, Robert D. | Chapter 1. Introduction 6 
event or a neighborhood).
7
 To get an acurate picture of the overal paterns of violence 
in the city, I took two approaches. I started by modifying the World Health 
Organization?s violence framework (taking some inspiration from the frameworks that 
Caroline O.N. Moser and Cathy McIlwaine had developed through their work on 
violence in Colombia) to be sure that I was capturing both physical and psychological 
forms of violence in a way that took the relationship betwen the diferent types of 
perpetrators and victims into acount.
8
 The result was a framework that studies violence 
that individuals can commit against themselves, against intimate partners and family 
members, or against members of the public, and that groups can commit against others 
for social, political, or economic reasons. The framework, in other words, specifies 
twelve categories of violence: two manifestations (physical and psychological) across six 
?contexts? (individual, interpersonal-intrafamilial, interpersonal-communal, collective-
social, collective-economic, and collective-political). 
                                                
7
 e.g. Pablo Emilio Angarita Ca?as, H?ctor Galo, and Blanca In?s Jim?nez Zuluaga, eds., Din?micas de 
Guera y Construci?n de Paz: Estudio Interdisciplinario del Conflicto Armado en la Comuna 13 de 
Medel?n [Dynamics of War and Construction of Peace: An Interdisciplinary Study of the Armed Conflict in 
Medel?n?s Comuna 13], 1a ed. (Medel?n, Colombia: Universidad de Antioquia?INER, Universidad de 
Medel?n, Corporaci?n Regi?n, y Instituto Popular de Capacitaci?n?IPC, 208); John Jaime Correa 
Ram?rez, ?Memorias de Pilos y Violencias: Bario Castila, Medel?n 1950-2000,? [?A Memoir of Crooks 
and Violence: Barrio Castila, Medell?n 1950-2000,?] in Identidades, Localidades y Regiones: Hacia Una 
Mirada Micro e Interdiciplinaria, ed. Renzo Ram?rez Baca and ?lvaro Acevedo Tarazona (Medel?n, 
Colombia: La Careta Editores, 207), 125; Pilar Ria?o-Alcal?, Dwelers of Memory: Youth and Violence 
in Medell?n, Colombia (New Brunswick, N.J: Transaction Publishers, 206); Juan Diego Alzate Giraldo, 
?Alg?n D?a Recuperaremos la Noche: La Construci?n de la Amenaza y el Miedo en Bario Caicedo?Las 
Estancias? [?Someday We?l Get Back the Night: The Construction of Threat and Fear in Bario Caicedo?
Las Estancias?] (Undergraduate thesis, Universidad de Antioquia, 205). 
8
 The definitions and categories used in this chapter were partly informed by: Etiene G. Krug, et al., eds., 
World Report on Violence and Health (Geneva: World Health Organization (WHO), 202); Caroline O.N. 
Moser and Cathy McIlwaine, ?Latin American Urban Violence as a Development Concern: Towards a 
Framework for Violence Reduction,? World Development 34, no. 1 (205): 89; Caroline O.N. Moser, 
?Urban Violence and Insecurity: An Introductory Roadmap,? Environment and Urbanization 16, no. 2 
(204): 3; Cathy McIlwaine and Caroline O.N. Moser, ?Violence and Social Capital in Urban Por 
Comunities: Perspectives From Colombia and Guatemala,? Journal of International Development 13, no. 
7 (2001): 965; Panos D. Bardis, ?Violence: Theory and Quantification,? Journal of Political & Military 
Sociology 1, no. 1 (1973): 121. 
Lamb, Robert D. | Chapter 1. Introduction 7 
Then, I tried to obtain data (both quantitative and qualitative) for al twelve 
categories of violence not only for the city of Medel?n itself but for Colombia as a whole 
and for a smal sector within Medel?n caled Caicedo La Siera. This was an efort to 
understand both the national context and the local details of the violence taking place in 
the city. Quantitative data were not available for al twelve categories of violence at al 
thre levels of analysis across al 24 and a half years of the study period, as would have 
been ideal. However, this study was the first to gather into one place almost al of the 
quantitative data that were actualy available (even if only via proxies), that were 
reasonably reliable, and that were made available by their owners or could be found in 
the public domain. Moreover, qualitative data could be found or infered for many of the 
mising data points, making possible general observations about yearly trends within 
each category of violence (whether it rose, fel, peaked, or hit a minimum in any given 
year), but not, unfortunately, about relative magnitudes across categories. This made 
possible a richer description of the paterns of violence than had previously been 
available from any other source. This framework is described in detail in Appendix B and 
the paterns of violence that were found are discussed in Chapter 8. 
The second question ? what caused the paterns of violence to change? ? was 
meant to go beyond a descriptive acount of violence in Medel?n and toward an 
explanation (and its asociated set of causal mechanisms) that could inform the work of 
policy makers facing dificult decisions about how best to lower the level and intensity of 
diferent kinds of violence. The vast literature on violence in Medel?n does not want for 
candidate explanations. The most thorough review of studies about violence in and 
around Medel?n published to date found a wide variety of explanations for violence, 
Lamb, Robert D. | Chapter 1. Introduction 8 
atributing changes in violence to psychological, cultural, economic, social, political, 
historical, and other phenomena.
9
 To determine which were correct I did two things. First 
I reviewed as many of these studies as possible and, relying on the findings of those with 
transparent methods and credible designs, extracted a set of candidate explanations, or 
hypotheses, for the paterns of violence. Second, I looked in greater detail at the cases 
using my own frameworks and my own interviews to se which made sense, meaning I 
first made a credible case in favor of each hypothesis by creating a narative of the causal 
mechanisms implied by it, then identified the evidence that each of these causal 
naratives had to ignore in order to make the case credible, and finaly revised or 
combined those that ignored the least evidence to develop a new causal narative that 
acounted for the most evidence. These hypotheses are discussed in Appendix D. 
The third and final question that I addresed in this study ? what role did 
legitimacy or ilegitimacy play in the changes in the paterns of violence? ? was meant 
to go beyond those works that study conflict as something that takes place mainly 
betwen and among armed actors, whether agents of the state or agents of nonstate 
entities, and that therefore ignore the role played by members of the communities where 
the violence takes place. Community members are not only victims or collaborators in 
violence, but also perpetrators, enablers, denouncers, informants, pacifiers, supporters, 
financiers, and recruits. Understanding the roles played by nonstate actors and state 
actors, but not by community actors, therefore mises a significant part of the story. 
                                                
9
 Pablo Emilio Angarita Ca?as, ed., Balance de los Estudios Sobre Violencia en Antioquia [An Evaluation 
of Studies about Violence in Atioquia] (Medel?n, Colombia: Municipio de Medel?n y Editorial 
Universidad de Antioquia, 201); Carlos . Ortiz S., ?Los Estudios Sobre la Violencia en las Tres ?ltimas 
D?cadas,? [?Studies of Violence in the Last Thre Decades,?] Bolet?n Socioecon?mico CIDSE, No. 2425 
(192). 
Lamb, Robert D. | Chapter 1. Introduction 9 
To get an acurate picture of the role of legitimacy, I took two approaches. First, I 
modified and expanded a framework I had previously developed for analyzing 
legitimacy. The intention was to overcome the greatest weaknes of legitimacy research: 
measurement validity. Most authors who atempt to measure legitimacy acknowledge that 
they cannot be certain that it is legitimacy and not something else that they are 
measuring. Those who asert that legitimacy resides primarily in individual beliefs 
acknowledge the dificulties of recal and other biases inherent in measuring opinions. 
Those who asert that legitimacy resides primarily in group behaviors acknowledge the 
dificulty of determining whether certain behaviors derive from belief rather than 
coercion. And those who asert that legitimacy resides in the objective characteristics of 
the structure under study acknowledge that their outsider judgment of the system?s 
legitimacy may wel difer from that of insiders. Yet these authors draw the data or 
observations that underlie their studies of legitimacy from only one or, at best, two of 
these levels of analysis (micro, meso, or macro). My framework is intended to be used to 
look for evidence at al thre levels and, by doing so, to provide a higher degre of 
certainty about what is being measured: if individuals say they believe some structure to 
be legitimate, and groups act as if they believe that structure to be legitimate, and that 
structure has characteristics that suggest it operates legitimately, then it is very dificult 
(albeit not impossible) to argue that legitimacy is not at work in the structure; but if one 
of those levels does not agre with the others, that suggests that something other than 
legitimacy is at play (coercion, for example). Furthermore, this framework does not 
measure only proxies for legitimacy, nor does it measure only causal indicators: rather, it 
Lamb, Robert D. | Chapter 1. Introduction 10 
measures both a proxy variable and six causal indicators (transparent, credible, 
justifiable, acesible, equitable, and respectful). 
Second, I tried to obtain or derive data for as many of these variables as possible 
across al thre levels of analysis (individual, group, and system) by exploring published 
works and opinion polls and interviewing experts and residents about violence, 
legitimacy, governance, and teritorial control in Medel?n and the sector Caicedo La 
Siera. This was an efort to understand both the microdynamics and the multi-level 
dynamics of legitimacy. Using this framework to organize the analysis, I tried to find 
evidence related to the legitimacy or the ilegitimacy of community actors, nonstate 
armed actors, and state actors. I was not able to extract qualitative data for al seven 
indicators for al thre types of actor at al thre levels of analysis for al 24 and a half 
years of the study period. However, this study was the first even to atempt such a 
systematic analysis of the dynamics of legitimacy amid the complex violence of 
Medel?n. As such, its findings provide a richer explanation of the causal mechanisms 
among teritorial control, governance, legitimacy, and violence in Medel?n than had 
previously been found (or sought), and therefore provides a sound foundation upon which 
to base future research on the dynamics of complex urban violence more generaly. The 
framework and the analysis of the dynamics of legitimacy are discussed in detail in, 
respectively, Appendix C and Chapter 9. 
The answers that emerged from these thre questions ? what were the paterns of 
violence, what caused them to change, and what role did legitimacy play? ? provide a 
glimpse of what it might take for policy makers to succed in sustainably reducing 
Lamb, Robert D. | Chapter 1. Introduction 11 
violence in places such as Medel?n. These findings and their implications are discussed 
in the next two sections. 
1.2. Findings 
The paterns of violence ? answering the first question ? are described in detail 
in Chapter 8. Violence is the dependent variable, but it is not described here as a single 
number, such as the number of homicides or an index of violence. Rather, the dependent 
variable used for this study was a qualitative description of the relative magnitudes of 
diferent forms of violence as they have changed over time. The emphasis of that 
description by necesity had to be on the thre forms of collective (i.e. organized) 
violence (social, economic, and political) and one form of interpersonal (i.e. unorganized) 
violence (communal); reliable data, whether quantitative, qualitative, or impresionistic, 
were simply unavailable for the other categories of violence (se Appendix B for an 
extended discussion of these diferent types of violence). Given that limitation, two main 
observations stood out, one regarding a long-term trend, one regarding a short-term 
dynamic. 
The most significant long-term trend observed over the 25-year study period was 
a gradual ?corruption? of collective-social and colective-political violence by collective-
economic violence; that is, the implicit ratio of economic violence to political or social 
violence has increased over time, although due to the paucity of reliable data, this ratio 
has not been able to be quantified. This has come about as violent organizations 
dedicated to ilicit profits (primarily narcotraficking mafias and hit squads) infiltrated 
and corrupted violent organizations dedicated to social order (e.g. community self-
Lamb, Robert D. | Chapter 1. Introduction 12 
defense groups), political change (e.g. anti-guerila paramilitaries), or both (e.g. urban 
militias). In other words, over time, the ?true believers? in the city?s armed social and 
political movements either lost out to organized criminals or were corupted by the lure 
of easy money, to the point that there are now very few true believers left: almost 
everyone involved in organized violence today is in it for the money. 
The most significant short-term observation was that most of the violence 
throughout the study period was interpersonal-communal (i.e. common, rather than 
organized, crime). This is so even during the peaks when the highest-profile forms of 
violence were collective-political (e.g. police masacres of young people in peripheral 
barios, asasinations of Patriotic Union party members, wars betwen militias and 
paras), collective-social (e.g. stret-gang turf wars, ?social cleansing? death squads), and 
collective-economic (e.g. organized crime, drug wars). This is explained by observing 
that these spikes in collective violence created a context in which the rule of law as too 
weak to control interpersonal-communal violence, and those who engaged in the later 
often used that broader context to their own advantage, either by recruiting or deceiving 
collective-violence organizations to cary out personal vendetas, or simply by taking 
advantage of the environment of impunity. Or to put it more simply: the gang wars made 
social control impossible, and violent crime increased as a result. 
The study?s second and third questions ? what explains the paterns of violence, 
and what role did legitimacy or ilegitimacy play in the relationship betwen teritorial 
control and violence? ? can be answered together (the roles of legitimacy and 
ilegitimacy are analyzed in detail in Chapter 9). The answer has thre parts: 
Lamb, Robert D. | Chapter 1. Introduction 13 
1. Teritorial control has reduced violence. Each statelet or micro-teritory 
within the city (and, from 2003 to 2007, in the city as a whole) has 
experienced a decline in violence once a unitary actor has emerged capable of 
keeping other actors out. 
2. The cost of teritorial control has been positively correlated with ilegitimacy. 
Legitimacy has reduced the cost, and ilegitimacy has raised the cost, of 
gaining and holding teritory. 
3. Ilegitimacy has been positively correlated with unpredictability in the manner 
and outcomes of governance. Communities for whom life has become 
dangerously unpredictable have been likely to withdraw support from, oppose, 
or begin to support rivals of, those actors who had been controlling their 
teritory. 
A synthesis of these findings yields the following explanation of the relationship 
betwen teritorial control and increases in violence, including the role of ilegitimacy. 
Within Medel?n, control over micro-teritories has required resources (e.g. money, 
weapons, communication, transportation) and people (e.g. recruits, financiers, facilitators, 
denouncers, informants, etc.) to counter opposition from external rivals (i.e. external to 
the micro-teritory), from internal rivals, and from rivals to the internal alies of the 
controlling group. For resource-rich groups (for example, a mafia-backed gang with 
aces to narcotraficking income), legitimacy has not been necesary to hold teritory 
because whatever support was needed for these fights could be purchased or coerced 
using resources they already had. To maintain control, therefore, the controlling group 
needed only to engage in a strategy of ilegitimacy-avoidance (maintaining a reasonably 
Lamb, Robert D. | Chapter 1. Introduction 14 
predictable living environment for residents). However, when that strategy has failed ? 
when their abuses, moral transgresions, or losses of credibility have caused life in the 
micro-teritory to become unpredictable and intolerable ? then the level of moraly 
driven non-compliance and outright opposition has risen, increasing both the number of 
rivals and the level of resources required to coerce or purchase needed support. This has 
provided strategic opportunities for existing external and internal rivals to recruit more 
people and resources into their own fight against the controlling group, making a real 
contest for control tenable. As a consequence, the controlling group and its rivals then 
engaged directly in collective violence against each other and each other?s supporters, or 
alowed themselves to be drawn into private grudges through false denunciations. 
Moreover, as the controling group focused on defense against rivals, it necesarily 
shifted resources and atention away from maintaining general order in their micro-
teritories. Inatention to order entailed a breakdown in social order and the rule of law 
that created an atmosphere of impunity for common crime asociated with interpersonal-
communal violence, which both fed back into problems of governance and further 
damaged the credibility and therefore the legitimacy of the controlling group. Thus have 
both collective violence and interpersonal-communal violence risen as a consequence of 
contests over control of micro-teritories. 
When one side or another in such contests has managed to prevail and take (or 
retake) full control over the micro-teritory, it no longer had a need for violence against 
rivals or rivals? supporters. As a consequence, collective violence asociated with the 
conflict has falen, and the controlling group could again dedicate atention and resources 
to protecting internal alies and maintaining order, so interpersonal-communal violence 
Lamb, Robert D. | Chapter 1. Introduction 15 
asociated with the breakdown in social order or the rule of law has falen as wel, which 
has fed back into improvements in governance in a virtuous cycle. Thus have both 
collective violence and communal violence falen as a consequence of control over 
micro-teritories. 
These explanations, more than any others, acount for the dynamics observed in 
dozens of micro-teritories within Medel?n, including in Caicedo La Siera, across al of 
the time periods studied. They obviously leave out some details, such as the role of other 
categories of violence, the specific types of behavior that generate ilegitimacy, leading 
indicators for an iminent ilegitimation, longer-term trends, and auxiliary proceses 
such as the legitimation of violence. Some of these factors are relevant and are discussed 
in detail in later chapters. But others have been left out because no evidence could be 
found to confirm or dispute their role, or because, while they might contribute to the 
explanation, they were not necesary for it. 
The most important asumption underlying these findings is that they apply only 
to the short-term: No actor in control of any micro-teritory in Medel?n?s violent 
periphery has managed to avoid delegitimizing or ilegitimizing itself long enough to find 
out what would have happened in the long term. 
1.3. Implications and Recommendations 
These findings are suggestive ? only suggestive ? of propositions that merit 
further study in terms of their implications for general theory and strategies to implement 
anti-violence policies. They also point to a set of recommendation for those who make 
policy for Medel?n. 
Lamb, Robert D. | Chapter 1. Introduction 16 
Each of the thre main findings enumerated in the previous section is potentialy 
generalizable to theory. The first finding (that, within certain bounds, violence is 
negatively correlated with teritorial control) tends to confirm the work of Kalyvas, albeit 
with a modified set of asumptions.
10
 The second (that teritorial control is positively 
correlated with ilegitimacy) is consistent with the canon of legitimacy theory, but 
restates the general correlation betwen legitimacy and stability as a theory of 
ilegitimacy and does so in terms of the mechanism through which ilegitimacy leads to 
instability. And the third (that ilegitimacy is positively correlated with unpredictability) 
represents a new proposition that merits further study. 
Each of these thre findings also potentialy has a central implication for 
strategies to reduce violence as wel: The first implies that any strategy of violence 
reduction should begin with a capability to prevent rival actors from governing. The 
second implies that, in any efort to atain or sustain teritorial control, resource-poor 
actors should engage in a strategy of legitimation while resource-rich actors should 
engage in, at minimum, a strategy of ilegitimacy-avoidance. And the third suggests that 
the esential element of an ilegitimacy-avoidance strategy is predictability, whose main 
components might be transparency (acurate, correct, and comprehensive publicity of the 
rules, rights, duties, and identities of those who are in control) and credibility (capable 
and non-arbitrary enforcement or fulfilment of rules, rights, and duties), although these 
two components might prove to be neither necesary nor sufficient to predictability. 
These propositions should be tested in other contexts before these strategies are 
implemented outside of Medel?n. 
                                                
10
 Stathis N. Kalyvas, The Logic of Violence in Civil War. 
Lamb, Robert D. | Chapter 1. Introduction 17 
Specific recommendations to those who make policy for Medel?n are provided in 
the final chapter of this disertation. The most important of these recommendations 
require the most imediate atention: Today, state actors who currently enjoy a modest 
and tenuous degre of public support in the micro-teritories in which they currently have 
a (weak) presence (but formerly did not) should consider themselves to be resource-poor. 
This normaly would suggest a strategy of legitimation. But because the ilicit actors who 
are the main perpetrators of the collective-economic violence we are seing today are 
engaged in a contest for control over markets rather than over teritory, state actors 
should be able to hold that teritory in the short term with a only strategy of ilegitimacy-
avoidance. As the violence in the city has begun to return to its formerly high levels, city 
and national security forces have been sent in to the peripheral barios to keep order. The 
imediate priorities of the officials who manage those security forces should be: 
1. to ensure that residents of those barios know al relevant laws, their rights and 
obligations under the law, and the rights and obligations of the security forces 
patrolling their neighborhoods; 
2. to ensure that rules against corruption within the security forces are strictly 
enforced; these first two recommendations are intended to ensure 
transparency, which contributes to predictability; 
3. to give the security forces the troops/officers, resources, and training they 
need to protect residents? rights and enforce the law hile engaging in 
operations against nonstate armed actors; this recommendation is intended to 
ensure credibility, which also contributes to predictability; and 
Lamb, Robert D. | Chapter 1. Introduction 18 
4. to fully empower, support, and expand the local Imediate Atention 
Command units (CAI: Comandos de Atenci?n Inmediata) to act as 
intermediaries betwen residents and security forces to ensure that residents 
do not turn to nonstate armed actors to resolve their grievances as they have in 
the past; this recommendation is intended to improve the state?s credibility 
relative to that of the nonstate armed actors. 
These recommendations, if implemented, should help to maintain a predictable 
and tolerable quality of life and avoid any ilegitimizing actions that would raise the costs 
of control and begin a new cycle of violence. It should be added, however, that, if the 
nonstate armed actors who are the target of security forces? operations decide to contest 
control of the peripheral barios, then the state should recognize that its resources wil be 
inadequate for the task and that it wil need to enlist the active and voluntary support of 
the community. 
To reiterate: If the curent trends in violence in Medel?n were to reach a point 
where state actors begin to lose what litle credibility they had earned as protectors of 
order over the past five years, and if state security forces were to engage in needles 
abuse or disrespect of citizens, then the costs of maintaining the tenuous control that city 
and national officials do have in the peripheral barios today wil rise so rapidly that only 
a dramatic flood of resources ? at a level that is currently unplanned for and perhaps 
unavailable ? would be able to prevent a return to city?s violent past. If that were to 
happen, one would have to characterize the ?Medel?n Miracle? as a fluke at worst or a 
mised opportunity at best. As of this writing (October 2009), that moment has not yet 
arived, but the window of opportunity is rapidly closing. 
Lamb, Robert D. | Chapter 1. Introduction 19 
1.4. Roadmap 
The remainder of this disertation is divided into thre parts. Part I (Chapters 2-7) 
tels the story of what happened in Medel?n. Chapter 2 gives brief tours of the city and 
one sector within it caled Caicedo La Siera, which is the embedded case that was 
studied for the details of the dynamics taking place in the city as a whole; that chapter 
also provides historical context for the full study period. Chapters 3-7 describe what 
happened in each of the five case periods, focusing on the diferent kinds of armed actor, 
the forms of violence they employed, and their relationships with each other and with 
members of the communities whose micro-teritories they controlled. Chapters 3 and 4 
describe the rise and fal of guerila-backed militias in the city, with their rise coming in 
response to an explosion of violent crime in the peripheral barios and their fal 
coinciding with their corruption by criminal elements. Chapters 5 and 6 describe the rise 
and fal of the paramilitaries in Medel?n, with their rise facilitated by the corruption of 
the militias and backed by rural mafias funded in large part by narcotraficking proceds, 
and their fal taking the form of a demobilization, ultimately corrupted by the infiltration 
of drug money. Chapter 7 describes the chalenge that the recent increase in violence in 
Medel?n is posing to the very tenuous control and very tenuous legitimacy that the state 
had managed to establish during the preceding period in those barios that had been under 
the control of the militias and the paras. 
Part I (Chapters 8-10) presents an analysis of what happened in Medel?n. 
Chapter 8 answers the study?s first question: what were the paterns of violence in 
Medel?n? Chapter 9 answers the study?s second two questions: what explains the 
paterns of violence in Medel?n, and what role did legitimacy or ilegitimacy play? 
Lamb, Robert D. | Chapter 1. Introduction 20 
Chapter 10 lays out the study?s findings, discusses their implications for theory and 
strategy, and offers a set of recommendations to policy makers. 
Part II includes four appendices that provide context and further detail for the 
study. Appendix A elaborates many of the key concepts employed throughout the study 
and defines terms that I use in ways that might be considered idiosyncratic (e.g., statelet, 
control, etc.) or controversial (e.g., state, subsidiarity, etc.). It also provides a broader 
discussion of how these concepts and the findings of this study might be relevant to 
today?s policy-making environment, a context in which globalization presents deep and 
unresolved chalenges, in terms of both the types of security threats that emerge today 
and the way those threats can and cannot be addresed through policy. (One such 
chalenge was the topic of this study: to determine what role legitimacy played as 
violence waxed and waned and shifted its shape over the past 25 years, a period during 
which, worldwide, the acesibility and afordability of global communication and 
transportation networks grew exponentialy, the world went from being majority rural to 
half urban, and identities and loyalties simultaneously expanded beyond the boundaries 
of states and contracted toward local communities and smal groups, making obsolete 
existing academic asumptions about governance and governability, states and statelets.) 
Appendix B introduces the concept of complex violence, looking specificaly at how that 
concept can be used to make a study of complex security environments such as Medel?n 
tractable. It then walks through the framework used in Chapter 8 to analyze the paterns 
of complex violence as they evolved in Medel?n. Appendix C introduces a scalable 
conceptualization of legitimacy, that is, a framework that may be used (as it was for 
Chapter 9) to analyze, measure, or develop a strategy to influence legitimacy at micro, 
Lamb, Robert D. | Chapter 1. Introduction 21 
meso, macro, or multiple levels of analysis. Appendix D discusses the motivation behind 
this study, the method and data sources used to cary it out, and its findings in terms of 
the hypotheses that were tested, rejected, and synthesized. 
 
Lamb, Robert D. | Part I. Ilegitimacy and Violence in Medel?n, Colombia 22 
Part I. Ilegitimacy and Violence in Medel?n, Colombia 
If there is one country where it is dificult to write about violence, that is 
Colombia. And yet, year after year, more and more research reports, 
articles, and books are accumulated. As if by surounding violence with 
words we believe we can exorcise it or at least corner it. But in vain. And 
so the paradox takes on the characteristic of a symptom: imersed as we 
are in its daily frenzy, it is nearly imposible for us to kep our distance, 
yet, with tenacious efort, the majority of what is writen tries to kep to 
the cold language of statistics or to the dispassioned discourse of 
typologies and cause/efect explanations. It is a subtle way of looking at 
violence without leting oneself get trapped in its whirlwinds, but also of 
impeding its ?comprehensibility.? 
 ?Jes?s Mart?n-Barbero (1997)
1
 
 
There has been a lot of interest among policy makers worldwide in explaining the 
?Medel?n Miracle? ? the city?s transformation, beginning in 2002, from one of the most 
complex and violent cities in the world to one that is eminently stable and livable. If the 
policies put in place in Medel?n by the local and national governments worked so wel 
there, it is thought, perhaps they could be replicated in other places experiencing complex 
urban violence as wel. Policies that are transplanted from one place to another, however, 
rarely work so wel in their new home without appropriate modifications to acount for 
                                                
1
 ?Si hay un pa?s donde sea dif?cil escribir sobre violencia ?se es Colombia. Y, sin embargo, los reportes de 
investigaci?n, los art?culos y libros se acumulan en modo creciente a?o tras a?o. Como si al cercarla con 
palabras crey?ramos poder conjurarla o al menos acoralarla. Pero en vano. Con lo que la paradoja adquiere 
rasgos de s?ntoma: sumergidos como estamos en su v?rtigo cotidiano nos es casi imposible tomar distancia, 
pero, en tenaz esfuerzo, la mayor?a de lo que se escribe trata de mantenerse en el fr?o lenguaje de la 
estad?stica o en el distanciado discurso de las tipolog?as y las explicaciones causa/efecto. Es una sutil 
manera de mirar la violencia sin dejarse atrapar en su remolino, per tambi?n de imposibilitar su 
?comprensi?n?.? Jes?s Mart?n-Barbero, prologue to Maquinaciones Sutiles de la Violencia, [The Subtle 
Machinations of Violence,] by Gisela Daza and M?nica Zuleta (Bogot?: Siglo de Hombres Editoriales y 
DIUC?Universidad Central, 197). 
Lamb, Robert D. | Part I. Ilegitimacy and Violence in Medel?n, Colombia 23 
local conditions. So what were the local conditions in Medel?n that might explain the 
dramatic changes there? Why did the positive trends that began in 2002 start to reverse 
themselves in 2008? And what broader policy lesons should be drawn from this initial 
succes and its incipient reversal? To answer these questions, I compare the conditions 
present during the ?Medel?n Miracle? period, betwen 2002 and 2007, to what was going 
on in the city and the country the last time homicides fel, from 1991 to 1998; and I 
compare the conditions present today, as homicides are rising again, to the last two times 
homicides rose in the city, from 1984 to 1991 and again from 1998 to 2002. In these 
comparisons, I pay close atention not only to homicide levels and the public policies 
implemented to lower them, but to the overal paterns of violence present in the city, the 
most significant events that took place at the local and national levels, and the 
microdynamics of legitimacy during each period. 
The explanation that emerges is neither as straightforward as that ofered by 
official government sources (briefly, that strong-state and good-governance policies 
worked) nor as conspiratorial as that offered by the country?s political left (namely that 
the government made secret pacts with paramilitaries to ignore their crimes in exchange 
for their keeping the homicide rate under control). While both of these explanations have 
a basis in reality ? in fact, combining parts of both does get closer to the truth ? the 
first does not identify the causal mechanisms through which specific policies supposedly 
led to a decline in violence while the second does not adequately explain the actual 
behavior of key state actors with respect to the paramilitaries with whom they supposedly 
had made the pacts. What both explanations are mising is an acount of the role that 
Lamb, Robert D. | Part I. Ilegitimacy and Violence in Medel?n, Colombia 24 
legitimacy ? or more acurately, ilegitimacy ? has played and continues to play in the 
changes in the paterns of violence in the city. 
Part I provides evidence for the proposition that, in Medel?n, ilegitimacy was an 
intervening phenomenon in a positive-fedback cycle of collective and interpersonal-
communal violence asociated with contests over teritorial control. This proposition is a 
variant of the long-established asociation betwen legitimacy and stability in social 
structures,
2
 but it refines and explains that relationship: legitimacy is corelated with 
stability at least partly because its opposite enables and magnifies the forces that lead to 
instability; teritorial control is possible, if costly, in the absence of legitimacy, but it is 
significantly les likely to be sustained in the presence of ilegitimacy. The history of 
violence in Medel?n over the past 25 years is evidence of this. At both the city level and 
the level of smal sectors within the city, conflicts over control of micro-teritories have 
been nearly constant, but within those micro-teritories have alternated periods of 
violence with periods of relative stability. The periods of low violence and stability have 
tended to be asociated with succesful ilegitimacy-avoidance (sometimes also 
acompanied by legitimation) by the armed actor in control of each micro-teritory, 
whereas the periods in which violence has risen ? primarily interpersonal-communal 
and collective physical violence
3
 (especialy homicides) ? have tended to be preceded 
by failures in ilegitimacy-avoidance, usualy characterized by abuses against community 
members and transgresions of community values by the armed actor in control. More 
                                                
2
 Se, for example, Moris Zelditch, Jr., ?Theories of Legitimacy,? in The Psychology of Legitimacy: 
Emerging Perspectives on Ideology, Justice, and Intergroup Relations, ed. John T. Jost and Brenda Major 
(New York: Cambridge University Pres, 201), 3. 
3
 i.e. physical violence comited either by one or two individuals acting on their own (comunal) or by 
members of groups organized for social, economic, or political purposes (colective); see Apendix B. 
Lamb, Robert D. | Part I. Ilegitimacy and Violence in Medel?n, Colombia 25 
broadly, the story of Medel?n has been the story of a society incrementaly legitimizing 
itself to itself ? and incrementaly delegitimizing diferent forms of violence and the 
armed actors perpetrating violence. As the epigraph points out, these stories has been 
buried beneath the ?daily frenzy? of the violence itself. They emerge clearly only once 
one looks closely at that violence and the human stories underlying it. 
Part I of the study tels some of those stories. Chapter 2 provides extensive but 
important background information on the city of Medel?n and the sector within it that 
was studied in detail, Caicedo La Siera, with a particular emphasis on the underworld 
figures who had set the stage for the violence to come. After that, in Chapters 3-7, the 
relevant history of Medel?n and Caicedo La Siera is given for each of the five study 
periods: 1984-1992, 1992-1998, 1998-2002, 2002-2007, and 2007 to the first half of 
2009. The first two periods were dominated by the rise and fal of the guerila militias; 
the second two periods, by the rise and demobilization of the anti-guerila paramilitaries 
and their criminal alies; the final period, by questions about the relationship betwen the 
state and the underworld. These histories are then analyzed in Part I. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 26 
Chapter 2. Background: Medel?n and Caicedo La Siera 
When I was in high school in southern New Jersey, thre friends and I spent a 
wekend afternoon playing Monopoly, a notoriously long board game in which 
individual players compete to purchase enough of the board?s properties and utility 
companies that nobody else can earn enough Monopoly Money from rent and service 
payments to stay in the game. After a few hours of play, none of us had yet emerged as a 
clear winner. Then my friend John, rulebook in hand, pulled Bob M. out into the kitchen. 
Mat and I heard nothing but whispers and giggles for about five minutes, after which 
John and Bob returned to announce that they were merging al of their asets and that we 
should henceforth consider them a single player. We weren?t sure this was within the 
rules, but John, ever the businesman, refered us to the rulebook, which apparently did 
not explicitly forbid mergers and acquisitions. Within a few rounds, both Mat and I 
realized we couldn?t compete (Johnbob now held almost half the properties on the board), 
so we held our own kitchen-merger meting. The competition betwen Johnbob and 
Matbob was cut-throat and it breathed new life into the game ? for maybe an hour. 
Dinnertime was approaching, and we were al ready for the game to be over. But 
Monopoly can be addictive: you can?t just end the game before a winner emerges. As our 
energy level waned, one of us sheepishly proposed something that also was not explicitly 
forbidden by the rulebook: a grand merger of Johnbob and Matbob. The idea was 
imediately acepted and, the four of us, now a very large (as far as group games go) 
corporation, celebrated our having achieved a true monopoly over the board?s entire 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 27 
teritory. We put away the pieces, folded up the board, and went our separate ways, home 
to dinner. 
Add a few more players and some brutal motorcycle asasinations to this 
scenario, and you would have more or les the history of ilegal armed activity in 
Medel?n from the late 1980s to the demobilization of the Cacique Nutibara Bloc, which 
had co-opted, absorbed, or conquered every militia, gang, mafia, and paramilitary group 
of any importance in Medel?n before it laid down its weapons in December 2003. A 
cacique (ka-SEE-kay) is a chief, and Cacique Nutibara was a powerful indigenous chief 
of the 16
th
 century. His armies had conquered teritories and trade routes in the Aburr? 
Valey (where modern-day Medel?n is) and the surounding region of what today is the 
department (province) of Antioquia. Nutibara was notorious for the public display of the 
heads and limbs of his enemies as trophies.
1
 His notoriety and his valiant but ultimately 
unsuccesful eforts to defend his chiefdom against the Spanish conquerors are stil 
celebrated today in local lore. The city of Medel?n named an important hil after 
Nutibara. A hotel and other busineses have named themselves after him as wel. And so 
did a paramilitary-narco-terorist mafia, which conquered the teritories and trade routes 
controlled by the underworld of the same region that the cacique himself had conquered 
nearly 500 years earlier ? and did so about as brutaly. The diference is that in 2003 the 
Bloque Cacique Nutibara (the Cacique Nutibara Bloc) was not dismantled as a result of 
defeat: it handed in its weapons and dismantled itself in the wake of its mergers and its 
victories over al of its competitors ? in other words, it achieved a monopoly, turned in 
its pieces, and went home. 
                                                
1
 Frank Salomon and Stuart B. Schwartz, South America, vol. 3, The Cambridge History of the Native 
Peoples of the Americas (Cambridge: Cambridge University Pres, 199), 598. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 28 
Or did it? Most people think the Bloque Cacique Nutibara (BCN) and several 
related paramilitary blocs (or ?paras?) demobilized in name only. BCN had been 
afiliated with both a counterinsurgent paramilitary confederation and a narcotraficking 
mafia. Since its anti-guerila objectives had been achieved by early 2003 ? the last of 
the leftist militias was ejected from East Central Medel?n in February of that year (se 
Chapter 3) ? BCN no longer had a need to continue as a paramilitary force. But since 
nobody asked it to demobilize as a criminal force, its members ?demobilized? only as 
paras and many of them erely continued their criminal operations under diferent 
names. The peace that descended upon Medel?n after BCN?s demobilization, most 
people believe, is mostly explained by the fact that the underworld economy and the 
violence that went with it was now almost completely controlled by a single strongman 
without any real competitors within the city: without competitors, there is no war. 
If this were the whole story it would be easily explained by existing theories of 
violence and conflict, such as Kalyvas?s theory that violence in civil war is a function of 
teritorial control: violence is lowest when one side or the other fully controls the teritory 
in question, highest when control is incomplete and contested.
2
 In broad outline, the 
findings of this study certainly support that theory, as far as it goes. But the Kalyvas 
theory and the version of events offered in the previous paragraph do not acount for why 
and how such contests over teritorial control begin in the first place. To understand that, 
one has to study the local social and political dynamics of the micro-teritory in the 
period leading up to the conflict, as wel as the broader context in which it takes place. In 
                                                
2
 Stathis N. Kalyvas, The Logic of Violence in Civil War. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 29 
Medel?n, there were many such conflicts, many micro-teritories and statelets
3
 that 
alternated betwen violence and relative calm. BCN?s final victory over Medel?n?s 
underworld, its members? imediate post-monopoly demobilization, and their evolving 
relationship with important state actors since then ? indeed, the sheer significance of the 
fact that a monopoly over ilicit violence was even achieved in Medel?n ? can be 
understood fully only after a detailed review of the history of those complex micro-
conflicts. And that history begins with how the violence in Medel?n became so complex 
in the first place. This chapter tels that story by reviewing the historical background and 
the social, political, and economic contexts in which it al began. 
2.1. Medel?n: From the Conquest of Nutibara to the Rise of Cocaine 
2.1.1. A Brief Tour of the Valley of Abur? 
When the Roman consul Quintus C?cilius Metelus Pius was sent to western 
Hispania during the first century BCE to subjugate a rebelious proconsul, he began by 
establishing a series of military bases. One ended up being named after him: Metelinum. 
Over the centuries, the base expanded into a vilage, and ?Metelinum? contracted into 
?Medel?n.? It was near this vilage of Medel?n, in the province of Badajoz, Spain, about 
midway betwen Madrid and the Atlantic Ocean, that Hern?n Cort?s was born in 1485 
and Gaspar de Rodas was born in 1518. Cort?s, of course, was the Spanish conquistador 
who later subjugated the Aztec Empire in Mexico. Rodas, les known in world history, 
                                                
3
 A statelet is a teritory that is controled in such a way that it would qualify as a state acording to most 
academic and legal definitions but that is not internationally recognized as such. See Apendix A. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 30 
was a conquistador who became governor of Antioquia, a region of the central range of 
the northern Andean mountains of South America. Rodas was appointed governor after 
the death of his former protector and colleague, the marshal Jorge Robledo. Robledo, 
while exploring the region in 1541, had glimpsed a valey in south-central Antioquia and 
sent an aid to check it out. The aid, Jer?nimo Luis T?jelo, found it when he arived during 
the night of 23 August 1541, and the Spaniards dubbed it the Vale de San Bartolom? (St. 
Bartholomew Valey). The natives who lived there, led by the tribal chief known as 
Cacique Nutibara, were fierce but poor; had they ben fierce but rich, Robledo would 
likely have tried to conquer them. Instead, war with the natives was postponed for some 
decades. In the end, however, Nutibara and his armies were defeated by the Spaniards, 
the native populations in the area were almost completely annihilated, and the survivors 
were enslaved. The natives had caled the valey ?Aburr?? and, despite their defeat at the 
hands of the Spaniards, it was this name, rather than San Bartolom?, that was the one that 
stuck. Governor Rodas established Spain?s first presence in the Aburr? Valey in the late 
16
th
 century. Permanent setlements followed over the next century, and in 1674 the 
Spanish royalty approved a request to name the setlements after the Old World 
birthplace of the region?s Spanish colonizers: La Vila de Nuestra Se?ora de la 
Candelaria de Medel?n. At the time, Medel?n was home to about 3,000 setlers.
4
 
Antioquia (an-TYOH-kya) was a fairly independent region among the Spanish 
colonial possesions; its main interactions with the outside world were economic, with 
                                                
4
 Information on the distant history of Medel?n in this and the next thre paragraphs is drawn from Gerard 
Martin, et al., Medel?n: Transformaci?n de Una Ciudad [Medel?n: Transformation of a City] (Medel?n, 
Colombia: Alcald?a de Medel?n and Banco Interamericano de Desarolo, 209); Forest Hylton, ?edel?n: 
The Peace of the Pacifiers,? NACLA Report on the Americas, published electronicaly by North American 
Congres on Latin America (NACLA), htp:/nacla.org/node/459 (acesed 13 April 209); Forest 
Hylton, ?Medel?n?s Makeover?; and comon knowledge in the city. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 31 
outside political and cultural influences playing a much smaler part there than in other 
Spanish setlements. Its capital at the time was a smal vilage caled Santa Fe de 
Antioquia, some fifty kilometers northwest of the Aburr? Valey. Early on, gold mining, 
with the hardest labor done by slaves, was its most important commercial activity; 
banking followed, and, later, as world demand for coffe grew, descendents of the 
colonists ? or paisas (PIE-suhs), as the people who setled Antioquia came to be known 
? helped finance the development of the ?coffe axis? south of Antioquia. During the 
Wars of Independence that began in 1810, Ecuador, Venezuela, Colombia, and Panama 
(Panama was then stil part of Colombia) seceded from Spain and formed a single 
independent country (named the Republic of Colombia), but despite a decade of fighting 
in other parts of the country and region, isolated Antioquia was never realy afected by 
the war. (Ecuador and Venezuela seceded from Colombia in 1830; Panama separated 
from Colombia in 1903.) Gold and cofe exports thus fueled the region?s economic 
growth, bringing wealth and prestige to many landowners and traders. Medel?n, situated 
betwen the crops and mines to its south and west and the capital city of Santa Fe de 
Antioquia to its northeast, became an increasingly important trading post, atracting 
wealthy traders, suppliers, and most importantly merchant bankers. Its local political and 
economic importance continued to grow until it finaly was named capital of the 
Antioquia in 1826. 
During the Industrial Revolution in the late 19
th
 century, merchants diversified 
beyond coffe to light industry, the city began a broad range of urban-planning eforts, 
and by the early 20
th
 century, ?Antioquia ? having emerged unscathed from the thre-
year civil conflict of 1899-1902, known as the War of a Thousand Days ? had moved to 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 32 
the centre of national economic life, and Medel?n became an important nexus for 
investment, speculation and the acumulation of value.?
5
 Among such investments were 
textile factories, the growth of which in the 20
th
 century made the Aburr? Valey one of 
the most important textile centers in the country (and today it is one of the most important 
centers of fashion in Latin America). 
In the beginning of the 20
th
 century Medel?n had fewer than 60,000 inhabitants. 
By 1928 that figure had doubled, and at mid-century there were more than 350,000 
people living in the city. Growth was rapid thereafter: The population exceded one 
million by 1973, was nearly 1.5 milion in 1985, and reached 1.6 milion by 1993, a 50% 
increase in just 20 years. The 2005 census found more than 2.2 milion people living in 
Medel?n ? 37 times the size of the population one hundred years earlier ? and 
2.3 milion in 2008.
6
 Antioquia is today Colombia?s sixth largest department by area and, 
with nearly six milion people, the second largest by population (after Bogot?). Medel?n 
is the department?s largest city. 
The Aburr? Valey is esentialy the Medel?n River water basin: al the creks 
and streams from the surrounding mountains drain into the river. Betwen Antioquia?s 
clay-and-mud geology and its industries? les than environmentaly friendly waste-
disposal practices, the Medel?n River today is litle more than a concrete channel for 
muddy, polluted water, although eforts have been made recently to clean it up. If the city 
of Medel?n is the heart of the valey, the Medel?n River is its backbone, and the 
geography of the area is best understood by reference to it, as locals do. Imagine yourself 
                                                
5
 Ibid. 
6
 Departamento Administrativo Nacional de Estad?sticas (DANE), ?Censo 205,? http://ww.dane.gov.co 
(acesed 2 August 209); Gerard Martin, et al., Medel?n: Transformaci?n de Una Ciudad, 43. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 33 
floating down the river. The muddy-water tour would begin in the south and take you 
northward through the valey?s ten municipalities, starting in Caldas, pasing through La 
Estrela, then past Sabaneta, Itag??, and Envigado before reaching the city of Medel?n. 
The municipality of Itag?? (e-ta-GWE), just south of Medel?n and on the west bank of 
the river, is the first stop of the Metro, an above-ground public-rail system that runs along 
the river from Itag?? in the south to Belo in the north, with a branch running from the 
city center to the west side of Medel?n, and two cable-car systems taking trafic up the 
sides of the mountains, one to the West Central slope, one to the Northeastern slope. As 
you float northward on the river, the areas that were setled first ? such as El Poblado on 
the south side of the city and Candelaria in the city center ? are situated to your right, on 
the east bank; as the city?s population grew, the west bank became setled as wel, and 
today the entire valey is inhabited on both sides. Newer arivals have generaly setled 
the areas farther and farther from the river, until today even the mountainsides are packed 
with everything from luxury high-rises to tin-roof shacks. 
In the valey, when people say ?up,? they mean away from the river (eastward on 
the east side, westward on the west side); ?down? means toward the river. Strets and 
roads are aranged more or les as a grid where the geography alows it. Those roadways 
that take you away from the river (east-west) and toward or up the mountainsides are 
caled calles (KAH-jheyss) and are numbered sequentialy, beginning with Cale 1 in the 
Poblado neighborhood in the south, to Cale 50 running through the middle of the city 
center, and to Cale 126 at the city?s northern border. Roads running paralel to the river 
(north-south) are caled carreras, also numbered sequentialy, with Carera 1 at the top of 
the mountains on the east side, Carera 43 running through the city center, Carera 80 at 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 34 
the foot of the western mountainside, and higher numbers as you climb up the mountain. 
On the mountainsides themselves, most of which were not setled acording to the city?s 
plans but rather were setled ilegaly, the grid system breaks down severely, with houses 
and strets emerging wherever there was space, acording to nobody?s plans. Many of the 
step, winding roads are litle more than mud paths or concrete stairways; some residents 
have to climb more than 300 stairs to get home. 
The municipality of Medel?n (meh-dheh-JHEN) is the heart of the Abur? 
Valey. The whole valey has a population of about 3.5 milion people, and al but about 
one milion of them live in the city of Medel?n itself. The city is divided administratively 
into six zones, 16 comunas (wards), and 271 barrios (neighborhoods), with betwen 11 
and 26 barios in each comuna; the city has a few rural zones, caled corregimientos, as 
wel. The zones are the Northeastern (Zone 1: Comunas 1-4), Northwestern (Zone 2: 
Comunas 5-7), East Central (Zone 3: Comunas 8-10), West Central (Zone 4: Comunas 
11-13), Southeastern (Zone 5: Comuna 14), and Southwestern (Zone 6: Comunas 15-16). 
As a general rule, and with a few key exceptions, the farther from the river and 
the higher the elevation, the poorer the bario and the lower the quality of life (and during 
some periods, the greater the violence). Aside from overcrowding and poverty, many 
residents of these peripheral barios suffer a major geological risk: mudslides. Heavy 
rains wreak havoc on the clay mountains, blocking roads nearly every time it rains, and 
occasionaly wiping out as many as a dozen families living in mountainside shacks. (The 
dark joke in Medel?n is that, on the hilside communities, the only law that functions is 
the law of gravity.
7
) The bright side of this region?s climate, however, is the temperature. 
                                                
7
 Alonso Salazar J., No Nacimos Pa? Semila: La Cultura de las Bandas Juveniles de Medel?n, 190. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 35 
With an elevation of about 1,500 meters (4,900 fet) above sea level and a humid 
subtropical climate, the valey enjoys year-round temperatures hovering around 22?C 
(about 72?F) and rarely venturing beyond a range of about 15?C?30?C (60?F?80?F). For 
this reason Medel?n is known as the City of Eternal Spring. (During the city?s most 
violent times, another dark joke played on the Spanish version of this nickname: ?La 
Ciudad de la Eterna Primavera? became ?La Ciudad de la Eterna Balacera,? or the City of 
the Eternal Shootout. A similar joke played on the city?s other nickname: ?Medalo,? or 
Medalion ? a reference to the region?s pride and its historicaly important role in the 
country?s gold trade ? became ?Metralo,? which would translate, roughly, as Machine-
Gun City.) 
As the rains raise the river and cary you northward on your floating tour of the 
valey, you?l pas housing of al economic levels, from wood-and-tin shacks near (and 
some in) the city dump to middle-clas brick houses to high-clas high-rises. You?l float 
by the city center, under a few bridges, and past a few industrial centers. And soon the 
river wil cary you out of the city limits, past the Metro?s last stop, and through the 
northern municipalities of Belo, Copacabana, Girardota, and Barbosa. After that, the 
river?s name changes to Porc? and flows into the Nech? River, which spils into 
Antioquia?s largest river, the Cauca, which continues north through the department of 
Bolivar and merges with largest river in the country, the Magdalena. The Magdalena 
River is the boundary betwen the departments of Magdalena and Bolivar, then, at the 
coast, betwen Magdalena and Atl?ntico; its mouth is the port city of Baranquila, and its 
waters cary you into the Caribbean Sea ? where the ships of conquistadors, and then 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 36 
pirates, were the first to link the international trade in Colombian goods with violence 
and social and teritorial control, a link that has only strengthened since then. 
2.1.2. Bad-Ases, Punks, and Low-Lifes Take Over the City 
The textile factories that opened in metropolitan Medel?n in the early part of the 
20
th
 century produced a working clas that enjoyed decent wages and benefits. This 
industrial precedent, plus the urban-planning system and the busines-friendly regulatory 
climate, atracted light industry to what many industrialists considered one of the most 
progresive cities in the country. Together with the Catholic Church, the families who 
monopolized the diferent industrial sectors helped to build a paternalistic social order 
based on conservative values: ?A model of social control based in early industrial sites 
and factories was developed, and its discipline-oriented content, inspired by the work 
ethic, spread over the rest of the city.?
8
 Self-discipline, and loyalty and respect for 
authority, became the key both to geting and keeping a job that provided a reasonably 
comfortable living, and to geting into heaven. The strength of these social structures was 
such that La Violencia, Colombia?s civil war ? like the Wars of Independence and the 
War of a Thousand Days before it ? hardly touched Medel?n proper during the late 
1940s and 1950s: there was litle violence within the city limits and the city?s economy 
continued growing strongly. While the elite industrialists were themselves mostly 
Conservative supporters, they ?prohibited partisan propaganda in their factories? and 
continued to hire supporters of the Liberal party without serious discrimination: ?the 
Medel?n elite consciously promoted an image of the city as an ?oasis? of peaceful 
                                                
8
 Ramiro Cebalos Melguizo, ?The Evolution of Armed Conflict in Medel?n,? Latin American Perspectives 
116, no. 28 (2001): 110. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 37 
capitalist productivity, beneficial to the nation, thanks to the social responsibility of its 
major industrialists.?
9
 (La Violencia was fought nationwide betwen members of the 
Liberal and Conservative parties over the course of about a decade, ending in 1958 with a 
power-sharing agrement caled the National Front, through which Liberals and 
Conservatives would alternate power. The National Front excluded other political parties, 
however, and the most radical of them took to the mountains to begin another civil war 
that continues today.) 
Beginning in the 1960s, inexpensive exports from Asian manufacturers began to 
depres demand for Medel?n?s products on the global market, while the same cheap 
Asian imports into Colombia began to undercut its manufacturing sector. At the same 
time, coffe prices on the world market were beginning to decline, and the resulting 
economic presure on smalholder farmers in Colombia?s coffe axis led to a rapid 
increase in migration to the city. Supply of labor was rapidly outpacing the availability of 
jobs, and despite Medel?n?s record of succes in urban planning, the city simply was not 
prepared to absorb the influx of rural migrants: 
Squater neighbourhoods sprouted up the gren hilsides on either side of 
the Medel?n River, especialy in the northern Aburr? Valey: warens of 
hand-built dwelings constructed from cheap brick, wood, cinder blocks or 
bareque [cane and mud], interconnected by step flights of steps. ? 
Within a few decades these fast-growing slums would house half the city?s 
2.2 milion population. Meagre state resources were funneled through 
neighbourhood commites, the Juntas de Aci?n Comunal. But the fact 
that police and army units were sent in to demolish hilside setlements 
was a symptom of the crisis of authority on the city?s new frontiers.
10
 
                                                
9
 Forest Hylton, ?Medel?n?s Makeover.? 
10
 Ibid. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 38 
These ?invasions? (the word used in Colombia for mas migrations into ilegal 
setlements) drove the growth of Medel?n?s population, which nearly tripled betwen 
1951 and 1973, and a created a ?paralel city? of ?pirate barios? representing the 
emerging divide betwen center and periphery.
11
 For the first time, workers, both new 
migrants and longtime residents, the uneducated as wel as those with university degres, 
faced a job market that simply could not absorb them. As in many other countries during 
the 1960s, there was campus and labor unrest and growing radicalization among a smal 
but vocal minority. In short, the traditional authority structures that had helped maintain 
social order in Medel?n were breaking down and the quality of life was deteriorating for 
much of the population. 
Despite its conservative exterior, Medel?n had always had a vibrant but represed 
subculture that suggested a degre of quiet tolerance for vice. Prostitution and alcoholism 
were common, with several red-light districts in the city, and especialy in the tougher 
neighborhoods marijuana consumption was common among the boys and young men 
who hung out on stret corners with their buddies, a formation commonly refered to as a 
gallada [ga-JHAH-da]. The galadas, a much looser collection of people than a crew or a 
gang, were a social problem ostly for the communities they hung out in, since stret 
fights betwen galadas were not uncommon, and they were known as wel to commit 
robberies and sometimes to haras and rape women. Medel?n had a high rate of 
imprisonment as wel, mostly from pety crimes comited by galadas or by pilos [PE-
jhohs], the generic slang term for a young punk or a smal-time crook who was not 
                                                
11
 Pablo Emilio Angarita Ca?as, ?Comentario a la Investigaci?n ?Estado del Arte Sobre Pol?tica Criminal y 
Violencia Juvenil?,? [?Coments on the Study ?The State of the Art in Crime Policy and Juvenile 
Violence?,?] in Balance de los Estudios Sobre Violencia en Antioquia, ed. Pablo Emilio Angarita Ca?as 
(Medel?n, Colombia: Municipio de Medel?n y Editorial Universidad de Antioquia, 201), 76. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 39 
necesarily part of a gang but who was certainly involved in crime.
12
 During the 1960s, 
as stronger drugs became more readily available and overcrowding began causing serious 
social problems, encapuchados or capuchos (hooded vigilantes) and ?civil defense? 
groups (death squads) began appearing in some neighborhoods ? often with the approval 
or direct support of the city?s conservative elites ? to undertake ?social cleansing,? that 
is, to ?clean up? these human sources of public disorder: drug addicts, prostitutes, 
homosexuals, vagrants, criminals, homeles children, and other low-status groups and 
individuals who were sometimes refered to as desechables (disposable people).
13
 
Crime, however, was not just the domain of pilos, galadas, and desechables. 
Smuggling was common as wel, often as a side job by otherwise respected businesmen, 
gentlemen who went to the best parties and were members of the best social clubs. They 
smuggled everything from stolen cars and appliances to emeralds and cigaretes, using 
specialized routes to get some goods (e.g. cars and appliances) that had been stolen from 
North America into Colombia, other goods (e.g. emeralds and marijuana) out of 
Colombia to North America, and yet other goods (e.g. cigaretes) to diferent places 
within Colombia or betwen Colombia and the tax-fre zones of places such as Panama. 
Some of these smugglers were family-based, but others were more powerful but discret 
criminal consortia involving corrupt oficials and otherwise legitimate businesmen, 
sometimes caled pesados (heavyweights) whose way of doing busines was later to be 
typified by Medel?n?s mafia oficinas (se Chapter 5). 
                                                
12
 John Jaime Corea Ram?rez, ?Memorias de Pilos y Violencias.? 
13
 Luis Fernando Quijano Moreno, Interview No. 5. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 40 
Table 2-1. Brief Glosary of Key Spanish, Lunfardo, and Parlache Terms 
autodefensa: self-defense group; sometimes synonymous with (right-wing) paramilitary, but there 
have been both leftist and nonaligned comunity self-defense or civil-defense groups as well 
banda: gang, usualy a crime gang or an organized-crime ring, with a relatively closed membership 
and somewhat formal organizational structure 
barrio: neighborhod; the smalest administrative unit on oficial maps (whose borders may difer 
from popular understandings of the borders); there are 271 barios in Medel?n as of 209 
bazuco: cocaine paste, which is cheap and highly adictive 
camaj?n: an earlier term for malevo 
capucho: a slang term for encapuchado 
chichipato: a ?low-life? individual or gang member who comits crime in his or her own 
neighborhood, often but not exclusively as a result of a drug addiction 
combo: crew or street gang; les organized and with loser membership requirements than a banda 
comuna: ward; an administrative unit containing a group of barios; there are 16 comunas in Medel?n 
coregimiento: rural subdivision; an administrative unit into which departamentos are divided 
departamento: department, or province (like a U.S. State); an administrative unit headed by a 
governor; Colombia has 32 departamentos; Medel?n is the capital of the department of Antioquia 
desachables: ?disposable? people; members of low-status social groups, such as drug dealers, adicts, 
prostitutes, homosexuals, and stret children; often the target of social cleansing 
encapuchado: hoded vigilante, usualy asociated with social cleansing but sometimes as part of a 
death squad, hit squad, or other armed group 
escuadr?n de la muerte: death squad; kils for political or social reasons, rather than for hire 
gallada: a lose colection of young men, sometimes involved in crime; more closed and organized 
than a parche, but more open and les organized than a pandilla or a banda 
gato: cat, slang term for guerila 
guerila: a member of an insurgent group who engages in guerila warfare, or the group itself 
Junta de Acci?n Comunal (JAC): Comunity Action Board; an elected body that represents 
comunities at the barrio level or smaller 
Junta Administradora Local (JAL): Local Administration Board; an elected body that represents 
comunities at the comuna level, usually in cooperation with the JACs within its jurisdiction 
limpieza social: social cleansing, the kiling or displacing of desechables 
malevo: a ?bad-ass? street tough who wears flashy atire, carries a knife, and manages relatively 
smal-time criminal activities; mostly predates the 1980s 
oficina: office; a term used to refer to a criminal organization that connects customers with 
contractors (often for assassinations) 
paisa: a person born and raised in or around Antioquia; think of them as the ?Texans? of Colombia; 
the cofe-advertising character Juan Valdez is always portrayed in typical paisa atire 
pandila: crew or crime gang, similar to a combo, usualy more open and les organized than a banda 
para: short for paramilitar, or paramilitary 
paraco: another term for a para 
paramilitar: a member of a paramilitary group, or the group itself; usually aligned with the state 
p?jaro: ?bird,? slang for para 
parche: ?patch,? slang for a group of friends or the place they hang out 
parce: ?bro,? ?mate,? ?budy,? ?dude?; a name you call a friend or a member of your parche 
pesado: ?heavyweight,? slang for a man with a lot of power, sometimes involved in organized crime 
pilo: smal-time crok, or young punk involved in crime 
pistolero: gunman 
sicariato: asasination, or the social phenomenon of young people working as hired asasins 
sicario: hired asasin, usualy working as part of a hit squad or gang of asasins 
zona: zone; an administrative unit containing one to four comunas; there are 6 zonas in Medel?n 
NOTE: Common usage of some terms (e.g. banda, combo, pandilla) may differ from these 
more formalized definitions, which are intended as a rough guide to how the author uses 
them in this work. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 41 
But in the 1950s through the early 1970s, the most visible sign that an underworld 
existed ? standing somewhere betwen the high-society gentleman smugglers and the 
?low-life? disposable people ? was a particular, iconic figure of Medel?n?s rougher 
cityscapes. Picture a Texan. Take of the cowboy hat and six-shooter. Make him a guy 
who caries a knife and knows how to use it. He walks like a 1980s gangster, dreses like 
a 1970s pimp, and smokes wed like a 1960s hippie. He talks a combination of stret jive 
and Pig Latin. And he listens to tango. With some exaggeration, you now have in your 
mind an image of the stereotypical Colombian malevo, a word that translates, roughly, as 
?bad-ass.? In the 1950s and into the 1960s, this figure was sometimes caled a camaj?n 
(kah-mah-HAHN), which is sometimes translated (inacurately) as ?pimp,? but malevo 
was the more common term by the 1970s.
14
 You?re picturing him as a Texan because 
paisas (people from Antioquia and some nearby regions) are a lot like Texans: proud, 
provincial, and teritorial about where they live, and, in their social relations, friendly and 
hospitable but pragmatic and instrumental. The paisa malevos displayed these traits with 
an aggresive self-confidence that, acording to their chroniclers, was hard to mis. 
Hanging out in their galadas, many of the camajanes and malevos spoke to each 
other in a stret jive that was a modified form of Lunfardo, the gheto slang that was 
spoken in the slums and outskirts of Buenos Aires and Montevideo during the late 1910s 
and 1920s when stret poetry was first set to tango music. Soon thereafter, Lunfardo-
laced tango records from Argentina and Uruguay introduced the rest of the Spanish-
speaking world to a whole new set of slang terms (most having to do with sex, money, 
and life on the stret) and to a Pig Latin-like way of playing with words whereby 
                                                
14
 John Jaime Corea Ram?rez, ?Memorias de Pilos y Violencias.? 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 42 
syllables were reversed (amigo becomes gom?a, bario becomes rioba, etc.).
15
 Tango 
records were readily available in Medel?n beginning in the 1930s, and paisas have been 
crazy about the music and the culture surrounding it ever since: Medel?n has a tango 
museum, tango clubs, tango schools, tango competitions, tango festivals, tango-themed 
restaurants, and even tango-playing dive bars in grity neighborhoods. The genre?s 
greatest star, Carlos Gardel, died in a plane crash in Medel?n while on tour in 1935, and 
the city has adopted him as their own. Although the music was refined or rewriten for 
upper-clas audiences in the 1930s and 1940s, the most beloved tango songs among 
common people remained those that spoke to themes that the camajanes and malevos of 
Medel?n could wel relate to: stret brawls, unfaithful women, broken hearts, dreams of a 
beter life. (The first tango song ever recorded with lyrics was Carlos Gardel?s 1917 
rendition of ?Mi Noche Triste,? or ?My Sad Night,? which tels the heartbreaking tale of 
a pimp who has been dumped by his favorite whore ? anticipating Thre 6 Mafia?s hip-
hop hit, ?It?s Hard Out Here for a Pimp,? by almost 90 years.) Medel?n?s malevos 
listened to tango in the 1950s through the 1970s the way American gangsters listen to 
?gangsta?-style hip-hop today; it was an integral part of the underworld culture.
16
 
These Lunfardo-speaking bad-ases and their imitators were the public face of 
Medel?n?s underworld up through the early 1970s, when the smal-time trade in 
contraband cigaretes and marijuana started to be transformed by the global expansion of 
the marijuana trade. Before then, however, the malevo was not part of any roving gang of 
                                                
15
 H?ctor Romay, Dicionario de Lunfardo y Terminolog?a Popular [Dictionary of Lunfardo and Folk 
Terminology] (Buenos Aires: Bureau Editor, 204). 
16
 Fernel Ocampo M?nera, Hablemos de Grandes Valores de Tango y Algo M?s [Let?s Talk about Tango?s 
Greatest Figures and More] (Manizales, Colombia: Editorial Manigraf, 205), 256; Carolina Santamar?a 
Delgado, ?Bambuco, Tango and Bolero: Music, Identity, and Clas Strugles in Medel?n, Colombia, 1930-
1953? (PhD disertation, University of Pitsburgh, 2006), ch. 4. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 43 
young asasins (as he would be in the 1980s) but rather was a smal-timer working alone 
or with the rest of his galada, carying out ad hoc criminal operations with other smal-
timers, or taking contracts from the heavyweights. The malevos made their money 
ilegaly and sometimes violently, and they showed off whatever wealth they had in style, 
but their control of vice and their participation in ilegal smuggling ?never took on the 
nature of the crime busines or created social repercussions that would turn them into a 
problem of public order, and thus the city absorbed this colorful personage as part of the 
urban landscape.?
17
 
As Medel?n approached the 1980s, however, the malevos were becoming an 
anachronism, mere predecesors to the real bad-ases who emerged during that violent 
decade: the asasin gangs, the drug dealers, the mafia enforcers. An influx of refugees 
from rural war zones brought to the hilside squater setlements a typicaly rural and 
paisa way of speaking, which young people, forging new identities through their shared 
marginalization, mixed together with the malevos? tango-inspired Lunfardo. The result 
was a distinctly paisa urban slang that became known as Parlache, the language of 
Medel?n?s youth and gang culture in the 1980s. Parlache slang is what was spoken in 
popular books and films that portrayed Medel?n?s youth and the notorious violence of 
the era, facilitating the spread of Parlache, first to Medel?n?s universities and then out to 
the rest of the country. The film Rodrigo D: No Future, for example, acurately portrayed 
Medel?n?s ?Generaci?n No Futuro? (the No-Future Generation, people who were 
adolescents during the 1980s). It tels the story of a punkero (a punk-rocker) whose only 
ambition is to form a punk band he could play drums in. Filming on location, the director, 
                                                
17
 Ramiro Cebalos Melguizo, ?The Evolution of Armed Conflict in Medell?n.? 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 44 
a Medel?n native, used young people from some of the city?s most violent barios in the 
north of the city, and the young actors more or les improvised the story in front of the 
camera, talking the way they normaly talked, acting the way they normaly acted, and 
dresing the way they normaly dresed. The story was fictional but, to their generation, 
familiar: The guys in the neighborhood deal drugs and steal cars, are frequently on the 
run from somebody, sometimes get kiled, and sometimes kil out of boredom. While he 
himself does not engage in such behavior, Rodrigo does hang out with the punk gangsters 
from time to time, but he has no real friends, nobody to realy trust. He also has no job 
and can barely aford drumsticks, but he holds out hope of finding a used drum kit he 
might one day buy cheap. He can?t slep at night, and he wanders the strets during the 
day. He?s bored and depresed, and in the end he takes his own future away. In real life, 
half of the movie?s main actors didn?t live past age 20 themselves.
18
 But their foul-
mouthed slang caught on in the country?s popular culture, living on as Colombia?s 
adopted slang. 
In one of the peripheral barios I visited, a young woman told me in an informal 
conversation that, some years ago, she was invited to somebody?s 21
st
 birthday party, 
which was being planned as a grand community afair. She said that, when she asked 
why the birthday was such a big deal, the person who had invited her looked at her funny 
and said, ?Because we never get to celebrate a 21
st
 birthday!?: too few of the young men 
in the neighborhood had ever lived that long, so a 21
st
 birthday was something to 
celebrate. This was surely an exaggeration, but not, perhaps, by much. In the 1980s, tens 
of thousands of people in Medel?n died violent deaths, the majority of them young men 
                                                
18
 Rodrigo D: No Futuro, DVD, directed by Victor Manuel Gaviria (Medel?n, Colombia: Liberty 
Multimedia, 1989). 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 45 
from the overcrowded hilside barios, especialy in the north of the city, where social 
conditions were deteriorating rapidly, traditional authority was being replaced by a 
materialist culture, the formal economy was not providing adequate employment, and a 
lot of young people, bored, unemployed, and poor, turned to the jobs that were most 
readily available to those wiling to take on the risks: seling drugs, stealing cars and 
motorcycles, robbing people at gunpoint, and even asasinating people for pay. Those 
who resisted the lure of the underground economy had few options: stay in the bario, try 
to find a job or stay in school, keep your head low, avoid stray bullets, and hope you 
don?t acidentaly get on anyone?s bad side (the wrong glance at the wrong person at the 
wrong time could get you kiled); or save yourself by leaving the bario and waiting 
things out (many families throughout the city sent their tenage boys and girls to live 
with relatives outside of Medel?n). Many who stayed survived, but a lot of others died in 
the crossfire. 
Drug use skyrocketed. Most commonly, people smoked marijuana. The few ho 
could aford it snorted cocaine. The very poorest, however, smoked bazuco, cocaine paste 
that was an intermediate product during the production of cocaine; it was readily 
available, relatively cheap, and extremely addictive. It had a similar efect on Medel?n?s 
poorest neighborhoods in the 1980s as crack cocaine did in the United States. Many 
addicts would do almost anything (robbery, pety theft, prostitution) to get the money 
they needed for their next hit, and crime commited by bazuco addicts skyrocketed. 
(Bazuco addicts were sometimes caled chichipatos, a term that was used to mean, more 
or les, ?low-lifes? but that usualy refered to people or gangs who commited crimes in 
their own neighborhoods, often as a result of drug addiction.) Another popular drug was 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 46 
glue, snifed (then and today) mostly by stret children to numb the body and mind. 
Another popular film in Colombia, Vendadora de Roses (The Rose Seler), by the same 
man who directed Rodrigo D ? who used the same method of hiring young people from 
the barios where the film was set ? portrayed the life of glue-snifing stret children 
(and, as with Rodrigo D, most of the film?s child actors had falen to the same fate as 
much of the rest of their generation: they were dead or in prison within a few years of 
filming). Chichipatos and glue-snifing stret children became common targets of 
vigilante violence: they were at the top of the list of people who were considered 
?disposable.? 
The blame for al of this social disruption cannot be lain at the fet of Medel?n?s 
most famous criminal, Pablo Escobar. ?El Patr?n,? ?El Doctor,? or ?Don Pablo,? as he 
was variously known throughout his carer, did not create the social and economic 
conditions that made life so dificult for so many in Medel?n?s poorest barios; in fact, 
for some he provided the means ? money, identity, respect ? to escape those 
conditions. His main contribution to his hometown and to his country, however, was to 
vastly acelerate the social degeneration brought about by those conditions. His story is 
worth teling in some detail, not only because his story was the story of the 1980s, but 
also because it set the stage for al that was to follow in Medel?n and because the 
convergence of threats sen worldwide today ? the use of insurgent and terorist tactics 
for economic rather than purely political purposes, the shifting of aliances betwen and 
among diferent types of ilicit actor (today a guerila, tomorow a traficker, the next 
day a paramilitary), etc. ? was anticipated, and in many cases created, by the 
innovations of Escobar?s Medel?n Cartel. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 47 
2.1.3. Pablo Escobar Builds the Medel?n Cartel 
On 2 December 1993, Pablo Escobar Gaviria placed a telephone cal to his son, 
Juan Pablo, who was in hiding with the rest of his family at a luxury apartment building 
in Bogot?. Sixten months after his dramatic escape from his custom-built luxury prison, 
Escobar?s family remained among the few loyalists the fugitive drug lord could trust 
unconditionaly. Over the course of more than a year, in the most masive manhunt in 
Colombian history, a special unit composed of the National Police and the Colombian 
Army ? along with rival cartel leaders and a shadowy death squad formed explicitly for 
the purpose ? had succeded in dismantling much of Escobar?s narcoterorist empire. At 
least a hundred of his closest asociates had been kiled, dozens of others captured or 
surrendered, his fortunes dwindling, much of his property destroyed by fire and bombs. 
Juan Pablo warned his father not to stay on the phone long enough for police to trace the 
cal. But Escobar must have felt isolated, perhaps needing the comfort of a familial voice. 
Maybe he had gotten cocky, or lost his razor-sharp instinct for survival. Or maybe he was 
just tired of being alive. Whatever the reason, this much is clear: He stayed on the phone 
too long, and the cal was traced to the house in an upscale neighborhood of Medel?n 
where Escobar was hiding. At 3:15 p.m., 23 special police troops surrounded and entered 
the building. ?I?m hanging up,? were Escobar?s last known words, ?because something 
funny?s going on here.?
19
 
Pablo Escobar Gaviria was born in 1949 in Rionegro, a mixed-income smal town 
les than an hour?s drive from downtown Medel?n, to a farmer and a school teacher. 
                                                
19
 The acount of Escobar?s last day is based primarily on Gabriel Garc?a M?rquez, News of a Kidnaping, 
trans. Edith Grosman (New York: Penguin Boks, 198), 289; and Kevin Fedarko, ?Pablo Escobar?s Dead 
End,? Time Magazine, 13 December 193. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 48 
Known as an ambitious child in his youth, he was driven to overcome the poverty his 
mother had always complained about, and some say he?d always dreamed of becoming 
president of Colombia. Local lore suggests Escobar got his start in crime as a schoolboy 
stealing gravestones, sand-blasting the names off of them, and seling the slabs to 
Panamanian smugglers, although this claim has never been verified. Whatever the case, 
he apparently made a name for himself in the early 1970s during the ?Marlboro Wars? 
among rival gangs traficking in smuggled cigaretes. From cigaretes to marijuana, from 
marijuana to cocaine: at each step in his ascent through the ranks of smal-time criminal 
gangs, he proved himself a worthy and ruthles adversary. In this sense, he was not al 
that diferent from the other cocaine trafickers who were emerging in 1970s Colombia: 
Most leading cocaine capos, including Pablo Escobar, got their start in the 
late 1960s as underlings in networks of contraband imports of U.S. 
manufactured goods run by older contrabandistas. These networks linked 
Miami and Col?n, Panama, to Turbo, Antioquia?s Caribbean port in 
Urab?, as wel as the string of towns in the Antioquian lowlands leading 
out to it. In keeping with regional tradition, Escobar and his generation 
were ambitious contraband entrepreneurs: Each had his own labor 
networks based on kinship and friendship, and collectively, they quickly 
displaced or kiled the old men who had trained them.
20
 
But Escobar was particularly ruthles in this regard. He made his name by specializing 
more in protection and enforcement than in the development of production or distribution 
systems. He threatened to kil anyone who got in his way or refused his demands. In 
1976, he was arested on drug charges, but the case documents were lost after a few years 
and he was never prosecuted; the judges and prosecutors involved in the case had 
reported receiving death threats; the aresting officer was kiled years later. His threats 
                                                
20
 Forest Hylton, ?Medel?n: The Peace of the Pacifiers.? 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 49 
were not idle; few dared to defy him, because they knew the consequences.
21
 Mark 
Bowden, who wrote a best-seling book about Escobar, has emphasized that ?Pablo 
Escobar didn?t create the cocaine busines.? 
He had no entrepreneurial or management skils to speak of. It was just 
that everybody was afraid of him. If anybody discovered a trade route or a 
new way of doing things, Pablo would come knocking and say, ?OK, you 
work for me now.? You couldn?t say no to him.
22
 
Others have made the same point: ?By the mid-1970s, [Escobar] had established a 
monopoly on protection. ? Others were beter at exporting cocaine ? purchasing coca 
paste in Bolivia and Peru, flying it to Colombia, and refining it in clandestine laboratories 
before shipping it to market ? but they had to pay Escobar for each kilo they moved.?
23
 
By 1978, Escobar was wealthy enough to buy a country house and ranch on the 
outskirts of Medel?n and convert the property into an amusement park, a zoo for exotic 
animals, and the headquarters of what would become known as the Medel?n Cartel. A 
cartel is a ?consortium of independent organizations formed to limit competition by 
controlling the production and distribution of a product or service,?
24
 and the Medel?n 
Cartel, unlike later traficking organizations that would be caled by the same term, was 
exactly that. 
                                                
21
 Margaret DiCanio, Encyclopedia of Violence: Frequent, Comonplace, Unexpected (New York: 
iUniverse, 204); Douglas Cruickshank, ?Death of a Drug Lord: In ?Killing Pablo,? Mark Bowden Details 
the 16-Month Game of Cat and Mouse That Finaly Tok Down Medel?n Cartel Founder Pablo Escobar?
With the Help of the U.S. Government,? Salon, htp:/archive.salon.com/news/feature/2001 
/05/24/bowden/print.html (acesed 13 April 209). 
22
 Ibid. 
23
 Forest Hylton, ?Medel?n: The Peace of the Pacifiers.? 
24
 ?cartel,? Dictionary.com, WordNet 3.0, Princeton University, htp:/dictionary.reference.com/ 
browse/cartel (acesed April 12, 209). 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 50 
Fabio Ochoa Restrepo was a horse breder and succesful businesman who ran a 
side busines smuggling stolen home appliances and contraband whiskey.
25
 In the mid-
1970s, Escobar persuaded Ochoa?s sons, Jorge Luis, Juan David, and Fabio Jr., to use 
their family?s smuggling routes and contacts to start transporting cocaine into the United 
States. Jorge and his brothers took over the smuggling busines from their father in 1976 
and grew it into an international distribution system for cocaine. By 1981, however, the 
Ochoa system was at its limit; the Ochoas and Escobar needed a way to move their 
growing backlog of cocaine to market. That?s where Carlos Enrique Lehder Rivas came 
in. Lehder got his start in crime supplying his family?s used-car busines in Medel?n with 
stolen vehicles. After moving into the North American market in the mid-1970s, he found 
himself in a U.S. prison on a charge of car theft. There, he met George Jung, a marijuana 
traficker whose key innovation had been to smuggle large amounts of pot into the United 
States from Mexico using smal aircraft flying under-radar. After they were paroled, 
Lehder and Jung formed a smal-time cocaine-smuggling partnership whose sole purpose 
was to raise enough money to buy a smal airplane and a smal island in the Bahamas so 
they could develop a big-time narcotics transport network. They succeded. From 
Norman?s Cay island, they began transporting to Miami ever-larger shipments of 
Colombian cocaine on behalf of Ochoa, Escobar, and others, including Jos? Gonzalo 
Rodr?guez Gacha, a former emerald smuggler who got hooked in to the cocaine busines 
in Bogot? and met Escobar in 1976. In April 1981, Ochoa and Escobar caled a meting 
                                                
25
 Information for this and the next five paragraphs comes from Guy Gugliota and Jef Len, Kings of 
Cocaine: Inside the Medel?n Cartel, an Astonishing True Story of Murder, Money, and International 
Coruption (New York: Simon and Schuster, 1989); Patrick Clawson and Rensselaer W. Le, The Andean 
Cocaine Industry, 1st ed. (New York: St. Martin?s Pres, 196); Margaret DiCanio, Encyclopedia of 
Violence: Frequent, Comonplace, Unexpected; Forest Hylton, ?edell?n: The Peace of the Pacifiers?; 
and other sources as cited. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 51 
to clarify the busines and financial relationships among the numerous organizations, and 
the cartel arangement was formalized. Drugs flowed into Miami, cocaine prices 
plummeted there, and demand for the cartel?s product soared. They were earning milions 
of dollars with each shipment. The cartel set up a system of ?bundling? smal-time 
suppliers? products into larger shipments for a fe, giving even mom-and-pop cocaine 
producers an easy way to sel as litle as a kilo at a time wholesale. Escobar was the main 
enforcer of these deals, and he took upon himself as wel the responsibility of 
maintaining a favorable political and legal climate in Medel?n and Colombia as a whole. 
To maintain that climate, Escobar used two things: plata and plomo. In Medel?n, 
plata, or silver, is slang for ?cash?; plomo, or lead, is, obviously, what the busines end of 
a bullet is made of. To stay out of prison and keep his operations safe from law 
enforcement, his men offered ?plata o plomo?: a generous bribe if the judge, police 
officer, journalist, prosecutor, or politician were cooperative, a rain of bullets if they were 
not. The plata came from the drug trade ? hundreds of milions of dollars per month by 
the mid-1980s ? and the plomo came from the poor barios of his home town, where 
smal stret gangs were contracted on an ad hoc basis to cary out select asasinations 
and other acts of violence. By the end of the decade, it was estimated that more than 
3,000 young people in Medel?n were working as hired asasins, members of one of the 
dozens of asasination gangs that had emerged. 
Escobar could recruit from these barios because he was loved and admired by 
many who lived there (a historical fact denied by many today). In addition to being a 
poor-kid-made-good to whom they could relate, Escobar was known as a wealthy 
businesman who was very generous with his plata: ?Don Pablo? had funded a social-
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 52 
works program caled Medel?n sin Tugurios (Medel?n without Slums), and built soccer 
fields and stadiums, churches, schools, and even entire barios where poor people living 
in shacks could be relocated into brick houses with electricity and running water. He 
owned a newspaper, Medel?n C?vico, and a television program, Antioquia al D?a, to 
publicize the good works that he and his felow ?businesmen? were undertaking (and to 
contrast those good works with that of ?corrupt? and ?lazy? politicians). To launch his 
political carer, Escobar also founded a populist political movement, Civismo en Marcha 
(Good Citizenship on the March). His goal at the time was to build his image as a paisa 
Robin Hood. And he succeded. 
But that public image hid a much darker side, which was to be exposed only after 
his disastrous entry into politics. That darker side involved recruiting tenagers from the 
poor barios where he was most popular, primarily in the Northeast zone, to become paid 
kilers. Their tactical innovation (later adopted worldwide) was the drive-by shooting, 
which required four elements: a weapon, a motorcycle, a driver, and the parrilero, the 
shooter, who would sit or stand on the parrilo, or gril, on the back of the motorcycle and 
fire at the intended victim as they drove past without stopping. But that method was not 
used exclusively: any method that could get somebody kiled and get the kiler some 
money would end up being employed. In response to the enormous money-making 
potential of joining a hired-assassin gang, many young people formed themselves into 
groupings variously caled bandas, combos, or pandilas, terms that al translate roughly 
as ?gang,? or into smal ?busines offices,? caled simply oficinas ? a natural evolution 
from the galadas, malevos, punks, pesados, and vigilantes of the earlier era.
26
 The drive-
                                                
26
 Se discusion of these terms in the next section. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 53 
by shootings and other crude methods used by these young asasins for their contracts 
from Escobar (and from others who wanted somebody dead) often resulted in the deaths 
of innocent bystanders, which frequently began a cycle of revenge kilings. As a result, 
violence became commonplace in Medel?n?s peripheral barios during the 1980s, and it 
began to eat away at the social fabric, tear families apart, and shift power relations away 
from those with authority (parents, priests) toward those with weapons, daring, and no 
scruples. Gabriel Garc?a M?rquez summarized the efect of easy money on Colombian 
society. ?The idea prospered,? he wrote: ?The law is the greatest obstacle to happines; it 
is a waste of time learning to read and write; you can live a beter, more secure life as a 
criminal than as a law-abiding citizen ? in short, this was the social breakdown typical 
of al undeclared wars.?
27
 The social breakdown and pervasive insecurity led many 
people in the city?s peripheral barios either to join or support whatever gang could 
protect them, or to form vigilante and self-defense groups, which opened the doors to the 
entry of guerila-backed militias later in the decade (se next section). 
Meanwhile, and despite the populist image Escobar and his colleagues had so 
carefully cultivated, the cartel members? own ?addiction to the use of violence for 
political ends?
28
 ensured that their peaceful image would eventualy be shatered by the 
reality of what their busineses entailed. Two sets of events prompted Escobar to declare 
war, first against the leftist rebels fighting an insurgency against the Colombian state, and 
later against the Colombian state itself. The consequences would publicly expose the dark 
                                                
27
 Gabriel Garc?a M?rquez, News of a Kidnaping. 
28
 Patrick Clawson and Renselaer W. Le, The Andean Cocaine Industry. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 54 
side of the Medel?n Cartel and force Escobar out of public philanthropy and politics and 
into hiding for the rest of his life. 
The first event was a kidnapping. By the 1980s, Colombia?s leftist guerilas had 
turned to kidnapping and extortion to finance their insurgency. In late 1981, a smal urban 
insurgent group, the M-19 (Movimiento 19 de April, the April 19
th
 Movement) ? whose 
taking of the Palace of Justice several years later would, ironicaly, be partly funded by 
Escobar ? made the mistake of geting on Escobar?s bad side by kidnapping Martha 
Nieves Ochoa, the sister of the Ochoa brothers, and demanding a very high ransom for 
her release. The cartel refused to pay M-19 even a single peso. Instead, on 3 December 
1981, they sent an airplane over a sports stadium during a high-profile soccer match to 
rain leaflets onto the crowd below. The leaflet announced the formation of a new 
organization caled Muerte a Secuestadores (MAS), or Death to Kidnappers. It was 
signed by Escobar and more than two hundred of the country?s other top ?businesman? 
? drug trafickers, every one, although this was not yet publicly known ? declaring 
their intention to put an end to kidnapping by Colombia?s insurgents. 
Its stated objective was the public and imediate execution of al those 
involved in kidnappings, beginning from the date of the mesage. The 
statement offered 20 milion pesos ([then about] $300,000) for information 
leading to the capture of a kidnapper and promised that the kidnappers 
would be hanged from a tre, shot and marked with the MAS?s sign. 
Kidnappers in jail could expect to be murdered; if that was impossible, 
retribution would fal on friends in jail and on close family members.
29
 
The Ochoa sister was released within thre months, unharmed, and without the M-19 
having received any ransom money. (Ironicaly, she would become a go-betwen in 
                                                
29
 Margaret DiCanio, Encyclopedia of Violence: Frequent, Comonplace, Unexpected, 308. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 55 
negotiations to release hostages kidnapped by Pablo Escobar later in the decade.
30
) But 
MAS continued its kiling spre nevertheles, targeting insurgents, suspected insurgents, 
leftist sympathizers, and suspected sympathizers al over the country. Copycat groups 
emerged, and soon any number of homicides having nothing at al to do with kidnappers 
were being atributed to this new death squad, a convenient cover for personal grudges 
and profesional vendetas. MAS enjoyed some degre of public support ? few ould 
shed tears over the death of a kidnapper ? and Lehder, who used far too much of his 
own product for Escobar?s comfort, publicly bragged about his links to the death squad 
and his contributions to the Liberal Party. MAS thus became the first death squad to mix 
narcotraficking with paramilitary counterinsurgency and coercive propaganda against 
establishment politics, an early prototype for the paramilitary and mafia groups that 
would join forces some fiften years later. 
The second set of events leading to Escobar?s fal from social grace was an 
election campaign. In June 1982 Escobar ran for and was elected to the Chamber of 
Deputies as an alternate for Liberal Party member Jairo Ortega Ram?rez, boss of the 
political machine in a town just south of Medel?n. As a public official who could stand in 
for Ortega when Ortega was unavailable, Escobar enjoyed both a legitimate public ofice 
and official imunity from prosecution for certain crimes. It was the highest ? indeed 
the only ? public office he would ever reach, and it was, at best, a consolation prize for 
Escobar. Powerful members of the Liberal Party considered the young drug trafickers 
(both Escobar and Lehder had entered politics) to be a danger not only to Colombia but 
likely to their own future carers in politics as wel, given the great wealth and populist 
                                                
30
 Gabriel Garc?a M?rquez, News of a Kidnaping, 87. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 56 
backing these ?businesmen? enjoyed.
31
 Among Escobar?s strongest political critics was 
the charismatic and popular politician Luis Carlos Gal?n, a former minister of education 
and ambasador to Italy who had founded a disident faction within the Liberal Party 
caled Nuevo Liberalismo, or New Liberalism, in 1979. Escobar himself had tried to join 
the New Liberalism party in 1982, but Gal?n had learned about his secret involvement in 
the drug trade, exposed it to a crowd of 5,000 people in Medel?n, and ejected Escobar 
from the party that same year. 
For Escobar, whose identity had been wrapped up at least as much in politics as it 
was in trade, this was a devastating blow, a rejection of al the good works he had built 
his reputation upon, and he took it as a deep, personal insult. It was only after being 
kicked out of the New Liberalism party that Escobar ran for the Chamber of Deputies 
office as a Liberal Party candidate, ?but he had not forgotten the insult and unleashed an 
all-out war against the state, and in particular against the New Liberalism.?
32
 
The source of the relentles increase in the homicide rates of Medel?n and 
Colombia as a whole ? which, within a decade, would place Colombia among the most 
violent countries in the world and Medel?n in the company of cities such as Beirut ? can 
be traced to the events of 1982 and 1983. After then-President Belisario Betancur 
appointed Senator Rodrigo Lara Bonila as justice minister, as his government?s New 
Liberalism representative, in August 1983, Ortega, in the Chamber of Deputies, tried to 
tarnish Lara Bonila?s reputation by claiming the justice minister had received large 
                                                
31
 Douglas Cruickshank, ?Death of a Drug Lord: In ?Kiling Pablo,? Mark Bowden Details the 16-Month 
Game of Cat and Mouse That Finaly Tok Down Medel?n Cartel Founder Pablo Escobar ? With the 
Help of the U.S. Government.? 
32
 Gabriel Garc?a M?rquez, News of a Kidnaping, 21. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 57 
campaign contributions from drug dealers. ?Lara Bonila flung the acusation back at his 
acusers, and went on to raise questions about Pablo Escobar [Ortega?s alternate] and the 
Death to Kidnappers movement.?
33
 The justice minister went on to isue indictments 
against Escobar, Lehder, and other members of the Medel?n Cartel; the media started 
picking up stories about the drug lords and their influence in al aspects of polite society 
(politics, soccer, busines, etc.); and the United States? requests for the extradition of 
Lehder and other Colombian trafickers under a treaty the two countries had recently 
signed was beginning to weigh heavily on the cartel leaders. Escobar withdrew from 
politics and went into hiding in January 1984. In March, a vast cocaine complex that the 
Ochoas had opened the previous year was busted in a dramatic operation run by a unit 
that reported to Lara Bonila. On 30 April 1984, hit men from Los Quesitos, a Medel?n 
hit squad, kiled Lara Bonila in Bogot? in a drive-by shooting, and the Betancur 
government began a masive crackdown on Escobar and other narcotrafickers in 
response; the president signed the outstanding extradition order for Lehder a wek later. 
The response by the government was so strong that the leaders of the Medel?n Cartel 
temporarily relocated to Panama, from where they sent a mesage to Betancur offering to 
close shop, forfeit al their asets to the state, surrender, and cooperate with crop-
substitution and other government projects, in exchange for amnesty and a guarante that 
they wouldn?t be extradited to the United States. But political leaders balked at 
negotiating with criminals, the United States made it clear it opposed any such 
setlement, and the proposal was rejected. 
                                                
33
 Margaret DiCanio, Encyclopedia of Violence: Frequent, Comonplace, Unexpected, 310. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 58 
And so it went for the rest of the decade, a cat-and-mouse game in which the 
government was hit hard by the cartel, then hit the cartel hard in return, each side with 
proposals and counterproposals, kidnappings by the cartel, crackdowns by the 
government, back and forth. The cartel leaders took to refering to themselves publicly as 
The Extraditables, part of an efort to persuade the electorate that their political leaders 
were harming Colombian sovereignty by refusing to overturn the extradition treaty with 
the United States; in case persuasion didn?t work, they applied ?plata o plomo? to the 
same end. Judges and police were cowed, corrupted, or kiled, especialy after the Palace 
of Justice disaster in 1985, when the M-19 stormed the building that housed the Supreme 
Court, and the ensuing firefight with the Army left more than 100 hostages, including 
half of the 21 Supreme Court Justices, dead). The rule of law as soon almost completely 
broken, nowhere greater than in the city of Medel?n. In September 1988, then-president 
Virgilio Barco made a speech that semed to invite the trafickers to the bargaining table. 
Although they claimed the invitation was actualy aimed at the emerging paramilitary 
groups, Barco administration officials nevertheles met with Medel?n Cartel leaders 
several times over the next ten months to discuss a negotiated setlement ? while at the 
same time preparing for Operation Primavera, a wide-ranging, and relatively succesful, 
atack against the cartel?s production facilities. The talks broke of on 18 August 1989, 
when Pablo Escobar collected on a political debt from the man who had ruined him 
politicaly seven years earlier. ?Luis Carlos Gal?n, who was protected by eighten wel-
armed bodyguards, was machine-gunned on the main square in the municipality of 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 59 
Soacha, some ten kilometers from the presidential palace.?
34
 Gal?n had been running for 
president in the 1990 election, and al indications semed to suggest he would win. 
After Gal?n?s asasination, the bloodiest phase yet of Colombia?s drug war 
ensued: President Barco declared a state of emergency aimed at the narcotrafickers, and 
the trafickers declared ?absolute and total war against the government.?
35
 The rest of 
1989 saw the Extraditables set off on a violent campaign of kiling, bombing, and 
kidnapping, while Barco created the Elite Corps, a squadron of 300 troops charged with 
batling drug trafickers, which they did with enthusiasm and brutality. But during this 
time of violence, the Extraditables also were in touch with a group of respected public 
figures, known as the Notables, who pased an offer for setlement to the government. In 
January 1990, the Extraditables, as a gesture, declared a truce, released some hostages, 
handed laboratories and explosives over to the government, and offered, again, to end the 
drug trade. During a two-month period of relative calm, official negotiations ensued ? 
but the administration was divided, the U.S. took a hard line, and, again, the negotiations 
failed. The violence resumed. 
In mid-1990, Gal?n?s campaign manager, C?sar Gaviria, won the presidential 
election, took ofice on August 7, and beginning in September made serious eforts to 
negotiate peace with the trafickers in terms aceptable to both sides. The main 
achievement: surrender of the thre Ochoa brothers (at the urging of their sisters), 
                                                
34
 Gabriel Garc?a M?rquez, News of a Kidnaping, 22. 
35
 The acount of Escobar?s surender, imprisonment, and escape is drawn partly from an unpublished 
background paper the author wrote at the request of Amnesty Inernational in 1997, for which the author 
retained al rights. Sources for that paper incude: Alma Guilermoprieto, ?Exit el Patr?n,? The New Yorker 
69, no. 35 (1993): 72; Patrick Clawson and Rensselaer W. Le, The Andean Cocaine Industry; Gabriel 
Garc?a M?rquez, News of a Kidnaping; Guy Gugliotta and Jef Len, Kings of Cocaine; and Semana, ?Yo 
Fui el Creador de los Pepes,? [?I Was the Creator of the Pepes,?] 31 May 194, 38. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 60 
betwen December 1990 and February 1991. They received imunity from extradition 
and relatively short jail terms. 
The incarceration of the Ochoa brothers and other drug bosses did not result in an 
end to Colombia?s violence, however ? far from it. The decades-long civil war was stil 
in full force. In the beginning of 1991, for example, guerila groups launched a 
coordinated offensive that Gabriel Garc?a M?rquez caled ?the bloodiest escalation of 
guerila violence in the history of the country.?
36
 The army and the police responded in 
kind ? with arbitrary arests, torture, forced disappearances, executions, and masacres 
? targeting suspected guerila sympathizers, members of the left-wing Patriotic Union 
(UP: Uni?n Patri?tica) political party and their families, and unarmed civilians living in 
areas where guerilas were thought to be active. Paramilitary groups with ties to the 
security forces threatened and murdered journalists, teachers, students, judges, lawyers, 
union leaders, and the desechables (?disposable? people). Betwen the war of the 
narcotrafickers and the war of the guerilas, combined with the everyday violence 
unleashed in cities such as Medel?n by gangs of hired asasins and by regular citizens 
with guns and grudges, 1991 was the year when violence peaked in Colombia as a whole 
and Medel?n in particular (that year, Medel?n?s violence was exceded in the world only 
by that of Beirut, Lebanon). 
Meanwhile, Pablo Escobar continued to pay young people from the slums of 
Medel?n to murder police officers (during Escobar?s carer, at least 500 oficers were 
asasinated in Medel?n alone), and police officers and death squads continued to 
masacre young people from those same slums. Even after his cohorts from the cartel had 
                                                
36
 Gabriel Garc?a M?rquez, News of a Kidnaping. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 61 
turned themselves in, Escobar continued to take and hold hostages, and he kiled at least 
one of them to improve his bargaining position. His car and truck bombs kiled innocent 
pasersby in addition to his political targets. In March 1991, he threatened to dynamite 
into rubble the historic district of Cartagena de Indias on Colombia?s Caribbean coast if 
police operations in Medel?n did not cease by April. His violence was legendary ? and 
so was his personality. Garc?a M?rquez wrote: ?No Colombian in history ever possesed 
or exercised a talent like his for shaping public opinion. And none had a greater power to 
corrupt. The most unsetling and dangerous aspect of his personality was his total 
inability to distinguish betwen good and evil.?
37
 Then-president Gaviria wanted Escobar 
in prison. The security forces wanted him dead. 
By this time, public opinion was firmly behind the movement to change the 
constitution to forbid extradition ? Escobar?s pet project. He was wiling to surrender to 
the authorities, but only under certain circumstances, and throughout the months of 
negotiations through intermediaries, pasage of a constitutional ban on extradition 
remained his one, absolutely inflexible demand. In June 1991, when Colombia pased its 
first new Constitution in a hundred years, he got his wish: non-extradition was 
incorporated as a fundamental right of citizens. 
Gaviria, reasoning that it was not in the country?s interest for Escobar to remain at 
large and acknowledging that conventional law enforcement had otherwise failed, 
granted some of Escobar?s other wishes as wel. The location of the prison was of 
Escobar?s own choosing: a former drug rehabilitation center in a remote area of 
Envigado, the Medel?n suburb where Escobar grew up and where the population was 
                                                
37
 Ibid. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 62 
almost completely loyal to him. It was said that Escobar had actualy owned the center 
but sold it to Envigado?s mayor specificaly for use as his personal prison. Escobar also 
got to supervise the selection of the prison?s security guards ? he hired half, the mayor, a 
loyalist, hired the rest ? and Colombia?s security forces were not permited on its 
grounds. Finaly persuaded that his surrender would proced as agred, Escobar fred his 
remaining hostages and entered La Catedral (The Cathedral), as the prison came to be 
known, on 19 June 1991. 
Over the next thirten months, safe from his mortal enemies from the Cali cartel 
(as wel as from the Colombian security forces and the U.S. Drug Enforcement 
Administration [DEA]), and hidden from the eyes of government and military officials, 
Escobar transformed his prison into a luxury residence, rivaled, some said, only by 
Colombia?s best five-star hotels. Army engineers had already built him a profesional, 
iluminated soccer field as part of the surrender deal, but Escobar went far beyond that. 
His people smuggled in the best construction materials available, and they built a high-
tech communications system, a library, a gym, and a discotheque; he bought expensive 
art and planted a garden (of marijuana); the ?cels? where he and a few other 
?incarcerated? cartel chiefs slept had amenities such as a large-scren television, a hot 
tub, and a water bed; Escobar hosted elaborate parties and was frequently visited by 
family, friends, colleagues, and prostitutes; and he possesed machine guns and other 
weapons. And when government inteligence sources discovered that the Cali bosses had 
bought bombs and were in the market for an atack plane, an air-raid bunker was added to 
the mix. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 63 
More significantly, however, Escobar continued to run his empire from prison, 
which is exactly why he wanted to be imprisoned in Colombia and not in the United 
States, where he would have been cut off from his organization. He continued to collect 
?taxes? from the Medel?n trafickers who had remained fre ? a total of about 
$100 milion per month. Those trafickers, and Escobar himself, viewed Escobar as the 
big boss ? esentialy a mafia ?godfather? ? even though he was not necesarily the 
biggest traficker within the Medel?n syndicate. But he was the leading propagandist and 
negotiator in the batle over extradition, and for that the other trafickers paid him what 
amounted to a war tax. When he surrendered, he is said to have told them, ?I am the price 
of peace,? and demanded that the Medel?n traficking groups continue to pay him for his 
sacrifice.
38
 Escobar?s luxury acommodations became a minor scandal within Colombia 
and around the world when word got out about them. But the real scandal was ?the 
unusualy large number of tortured and mutilated corpses that had been turning up on the 
outskirts of town,? acording to Alma Guilermoprieto, writing in The New Yorker in 
October 1993. ?The rumor was that the dead men were among Escobar?s own lieutenants 
and most trusted busines partners, that they had ben kidnapped and taken to La 
Catedral, and that there, under the boss?s supervision, at least a dozen of them had been 
acused of betrayal and then tortured and kiled.?
39
 
In fact, it was the torture and murder of four such partners ? Fernando ?El 
Negro? Galeano, Mario Galeano, Gerardo ?Kiko? Moncada, and Wiliam Julio Moncada, 
who had complained to Escobar about a stash of cash his men had stolen from them ? 
                                                
38
 Ibid; Patrick Clawson and Renselaer W. Le, The Andean Cocaine Industry. 
39
 Alma Guilermoprieto, ?Exit el Patr?n.? 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 64 
and the murder of a score of their asociates in early July 1992, that sealed Escobar?s fate. 
Word got to Colombia?s chief prosecutor, Gustavo de Grief, about the prison 
interogations, and after an investigation de Grief met with President Gaviria, who 
decided he could no longer justify keeping Escobar at La Catedral. They hastily made 
plans to transfer him to a government-controlled high-security prison in Bogot?. 
Escobar learned of the plans on the evening news and took two hostages: 
Colombia?s vice minister of justice and the national director of prisons, who for unknown 
reasons had gone to talk with Escobar that evening. On the night of 21 July 1992, Gaviria 
sent around 400 or 500 police and army troops to Envigado to surround the prison. 
Diferent acounts conflict, but early the next morning the troops apparently started 
shooting, and Escobar, wearing blue jeans and sneakers (not women?s clothes, as some 
had reported), along with a handful of his felow inmates, bribed some guards (some say 
with a plate of food, others say with money), shot their way out of the prison (acounts 
difer), and escaped into the mountains without their hostages. 
Another humiliation for Colombia ? and for Gaviria. Escobar had made a fool 
out of the president and Gaviria was no longer in the mood to negotiate. After the drug 
lord?s dramatic escape, which made headlines around the world, the government of 
Colombia offered a $1.4 milion reward (later increased to $5 milion) for information 
leading to his arest. The prosecutor general filed charges that could be used against 
Escobar ? and stick ? in the event he was captured. Gaviria sent uniformed police 
house-to-house in Envigado and formed a Search Bloc (Bloque de B?squeda), made up 
of elite police and Army troops, to find the fugitive. Gaviria even ignored Escobar?s 
initial offer to return to La Catedral as wel as his later offers ? made when it became 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 65 
apparent that Gaviria wasn?t impresed with the terms of his earlier incarceration ? to 
surrender unconditionaly and be confined in the ?most humble and modest jail.? 
True to form, within a month of his escape, Escobar began to wage a new 
campaign of violence that would turn out to be even more deadly than the one he had 
undertaken in the months before his retirement to La Catedral. By April 1993, Escobar 
was blamed for eleven car-bomb explosions in Bogot?, Medel?n, and other cities, which 
kiled more than sixty people and injured hundreds. (The homicide rate was lower in 
Colombia and in Medel?n during the year 1992 than in 1991 only because the unrelated 
guerila campaign, which had raised homicide to such high levels in 1991, had already 
ended; stil, on a month-by-month basis, the homicide rate in Medel?n in December 1992 
was higher than that of any month in 1991, primarily due to Escobar?s campaign against 
the state, which he waged from his various hideouts in the Aburr? Valey and possible 
elsewhere.) 
In August 1992, Colombia had asked the United States to increase its role in the 
search for Escobar. The State Department imediately posted a $2 milion reward for 
information leading to the drug lord?s capture. The U.S. Justice Department indicted him 
on charges of bombing a 1989 Avianca flight over Colombia (among the victims were 
two Americans). The CIA and DEA sent additional agents to Bogot?. And U.S. Southern 
Command (SOUTHCOM) provided electronics experts and equipment, including 
airplanes for surveilance, to Colombia?s government. It has been estimated that about 
400 Americans participated in the search for Escobar. 
But Escobar was nowhere to be found (though his lawyers continued to offer his 
surrender). One acount suggests Escobar went to Brazil to hide, but most people 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 66 
believed he was simply taking advantage of the extensive network of safe houses he had 
built among his supporters in and around Medel?n over the years. 
In November 1992, Gaviria declared a state of emergency that permited his 
government to restrict civil liberties when deemed necesary. In January 1993, Escobar 
declared an al-out war against the government, and a car bomb in Bogot? that kiled 
twenty-one people showed just how serious he was. Gaviria increased the reward for his 
capture to $6.7 milion. The violence escalated. But the worst was yet to come. 
On 31 January 1993, a powerful bomb exploded at a farm owned by Escobar?s 
mother. Two days later, a group nobody had ever heard of took responsibility. They 
caled themselves Perseguidos por Pablo Escobar (People Persecuted by Pablo Escobar), 
or the Pepes for short. At first, people believed the Pepes were a popular uprising against 
the man who had terorized the country for over a decade. But the firepower this death 
squad unleashed over the next months made it clear that far more powerful forces were 
behind it. The DEA, in cables released in response to a Fredom of Information Act 
request by Amnesty International USA, documented some of the atacks caried out by 
the Pepes: 
2 Feb [1993]. Pepes claimed responsibility for thre Medel?n area car 
bomb atacks on 31 Jan. against Escobar family residences. They vowed 
?to work towards his annihilation.? The body of Luis Alberto Isaza 
Estrada was found shot numerous times in Las Palmas section of 
Medel?n. A sign found on him read: ?For working for the narcoterorist 
and baby kiler Pablo Escobar.? ? 
10 Feb. An unidentified body, later found to be Gustavo Adolfo Posada 
Ortiz, was found in El Poblado section of Medel?n. Posada was 
considered by the Pepes to be an acomplice to Escobar, and had been 
kidnapped. This kiling was the second to be atributed to the Pepes. 
11 Feb. In El Retiro bario, seven young motorcyclists were asasinated 
by approximately twenty individuals in four vehicles. The Pepes were said 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 67 
to be responsible because of the asociations the victims had with 
members of the Medel?n Cartel.
40
 
Such atacks continued for the next ten months. In March 1993, the Pepes atempted to 
kil the mayor of Envigado and kidnapped and kiled several of Escobar?s lawyers before 
declaring a cease fire to give Escobar a chance to surrender. The cease fire didn?t last 
long, and Escobar and the Pepes continued to wage their war. By summer, many of the 
Medel?n Cartel?s top bosses had been kiled or had surrendered to the authorities. 
Escobar?s children?s school teacher was kiled, as was his wife?s brother. In al, more 
than 125 people asociated with Escobar (some estimates are much higher) were dead by 
the end of the year. 
If the Pepes were not a popular uprising against Escobar, as some claimed, who 
were they? In 1994, in an interview ith the Colombian wekly Semana, Fidel Casta?o 
Gil, a national paramilitary leader and former narcotraficking asociate of Escobar, 
revealed that he had organized the death squad. The friendship betwen Escobar and the 
virulent anti-guerila Casta?o (whose alias was ?Rambo?) had grown sour, he said, when 
Escobar helped the ELN guerilas with a masive weapons shipment. Acording to 
Casta?o, the two had had a serious argument over the shipment and never spoke to each 
other again. In 1992, Casta?o declined Escobar?s invitation to visit him at La Catedral 
with the Galeanos and Moncadas ? and he thereby escaped their fate. Afterwards, 
Casta?o said, ?several people sought me out to ask me to lead a self-defense group 
                                                
40
 Unpublished briefing prepared by the author for Amnesty International USA, 1997 (author?s private 
colection). 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 68 
against Pablo Escobar. I came up with the name Perseguidos por Pablo Escobar.? The 
group, he said, was funded by ?industrialists, politicians, ranchers.?
41
 
Although Casta?o wouldn?t reveal who else participated in the Pepes? actions, it 
is clear he had help from a wide cross section of the population. The most obvious 
supporters were the leaders and foot soldiers of the Cali cartel, who in 1993 were already 
supplanting their Medel?n rivals as the top trafickers in Colombia ? but who were 
much more discret in their activities and much les hierarchical in their organization. 
Les obvious, perhaps, to those outside Colombia were members of the police and the 
military, who moonlighted for the Pepes because they wanted Escobar dead and didn?t 
want to go through the bureaucracy to achieve that end. Moreover, Clawson and Le, 
citing a Bogot? television report, wrote: ?Pepes was apparently linked to another anti-
Escobar group, Colombia Libre (Fre Colombia), that also emerged in early 1993. 
Colombia Libre, which was composed of former Escobar asociates and asorted 
busines men and industrialists, offered a reward of $5 milion for Escobar?s head.?
42
 The 
legal and criminal elements worked side by side in the hunt for Escobar: 
Escobar?s enemies collaborated enthusiasticaly with the Colombian 
government. Colombia Libre claims to have provided inteligence to the 
authorities to support forty operations against Escobar?s organization. 
Similarly, leading Cali trafickers maintained a network of informants in 
Medel?n and ? acording to several acounts ? also deployed high-tech 
tracking devices to intercept Escobar?s communications. The information 
obtained from these channels was pased on to the police and to the DAS 
[the Department of Administration Security, Colombia?s equivalent to the 
FBI]. ? In sum, the government?s succesful liquidation of Escobar ? 
owed much to the eforts of Colombia?s cocaine establishment.
43
 
                                                
41
 Semana, ?Yo Fui el Creador de los Pepes.? 
42
 Patrick Clawson and Renselaer W. Le, The Andean Cocaine Industry. 
43
 Ibid. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 69 
That cocaine establishment did not go away with Escobar?s death. On the contrary, it 
evolved into something more insidious, more subtle, and far more dificult to take down. 
But Escobar is the one who put al the elements in place. Before him, crime and violence 
in Medel?n were smal-time. Even the country?s civil war did not bring anywhere near 
the level of violence that Escobar?s drug war and asasin gangs brought to Medel?n. 
?I?m hanging up, because something funny?s going on here.? On the day after his 
44
th
 birthday, Escobar heard people entering the safehouse on the first floor: the Search 
Bloc. He escaped through a second-storey window to the roof, wearing nothing more 
than blue jeans rolled up at the ankles and a dark blue t-shirt with a maroon stripe at the 
sleve. He didn?t even have time to put his shoes on. He had been running from these 
men for 499 days, never staying in the same place for more than six hours at a time, never 
staying on the phone for more than two minutes at a time, until just moments ago while 
talking to his son. He had put a very high price on the head of each of the men who now 
were running up his stairs: $27,000 per Search Bloc member kiled. Hundreds of their 
colleagues had already lost their lives making his young asasins rich, and on the roof, 
his one remaining bodyguard could no longer protect him from them. After it was over, 
someone took a picture: eight men with high-powered rifles, some in uniform, some in 
plain clothes, thumbs up and al smiles, surrounding a pale, fat corpse draped over the 
Spanish roof tiles, his arms unnaturaly splayed about his head, as if to protect himself, 
too late, from the seven bullets that had pierced his face and neck. 
During his rise to become one of the wealthiest people in the world, Pablo 
Escobar built entire barios for marginalized people who had been living in shacks, 
ignored by the state and disdained by mainstream society. He provided income and social 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 70 
recognition of sorts for thousands of young people who otherwise had faced poverty and 
obscurity. In some of those barios, Don Pablo is stil admired, stil prayed for at Sunday 
Mass.
44
 Many poor families entered the middle clas, and some, the very upper clases, as 
a result of the opportunities he provided them. ?Among the thousands of supporters who 
gathered last Friday afternoon hoping to glimpse Escobar?s body before it was lowered 
into his grave, few remembered that more than 20 years ago, he had launched his 
ascension to head the world?s most powerful drug organization by seling tombstones he 
had stolen,? wrote Time magazine in December 2003, repeating the dubious legend. 
?Pablo Escobar?s carer was ending exactly where it began ? in a Medel?n 
graveyard.?
45
 
Much of the ilicit wealth that entered Medel?n during Escobar?s reign was 
laundered and ultimately invested in legitimate busineses ? restaurants, shopping mals, 
high-rise apartment buildings ? that contributed to the city?s development and continue 
to produce legitimate income and economic opportunities today, the way the Kennedy 
family?s Prohibition-era bootleg profits were reinvested so that subsequent generations of 
Kennedys could enjoy wealth without stigma. In Medel?n, few today wil talk about 
where their families? wealth came from; few il acknowledge that some of the luxury 
office buildings and high-end shopping centers of the city?s most prestigious 
neighborhoods were built with drug profits several times removed. Outside of the poor 
barios where the state has only recently begun to pay them any atention ? and even 
                                                
44
 ElTiempo.com, ?As? se est? viviendo la guera entre bandas en las comunas del nororiente de Medell?n,? 
[?This is how the gang war is going in Medel?n?s northeastern wards,?] 9 April 209, 
http:/ww.eltiempo.com/colombia/justicia/-asi-se-esta-viviendo-la-guera-entre-bandas-en-las-comunas-
del-nororiente-de-medelin_4956840-1 (acesed 12 April 209). 
45
 Kevin Fedarko, ?Pablo Escobar?s Dead End,? Time Magazine, 13 December 193. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 71 
among many who stil live in those barios ? just about everybody, cab drivers, store 
owners, people of al stripes, consider Pablo Escobar to be a source of shame, not of 
wealth or development, because they know that the rest of the world asociates the name 
of their city with Escobar?s most notorious innovations: the mas-production and large-
scale, global distribution of cocaine; the drive-by shootings of motorcycle asasins; and 
the urban teror and warfare against the state and society in the service, above al, of 
private economic gains rather than any broader political program. They want to erase his 
memory, break the word asociation betwen Medel?n and cartel, and live in peace and 
with pride. 
But what Pablo Escobar left behind in 1993 has not made it easy for them to do 
this. He left behind thousands of corpses, many of which have never been found, and tens 
of thousands of broken families. As such, he also left behind debts of blood, young boys, 
even some girls, vowing to avenge the deaths of their fathers and brothers and friends and 
colleagues. He left behind the lure of easy money, the opportunity to sel one?s soul and 
take another?s in exchange for just enough cash to buy a loud stereo system and a nice 
pair of shoes. He left behind a broken system of authority relations, in which neither 
one?s family, nor one?s community, nor one?s church, nor one?s government had the 
credibility needed to enforce norms of behavior that could enable peaceful coexistence 
and maintain social, familial, or personal stability. And perhaps most significantly, he left 
behind the fragments of a vast money-making system, a network of asasins, growers, 
distributors, producers, racketers, launderers, trade routes, and a system for recruiting 
from a nearly endles supply of marginalized and often desperately poor people without 
the protection of the state or the respect of the society on whose periphery they were 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 72 
scraping out a living. The blood had barely been washed away by the rain on the rooftop 
when his underlings and enemies began fighting a war for control over the remnants of 
this empire. 
2.2. Caicedo La Siera: A Tour through East Central Medel?n 
No two people in Medel?n, given a map, would agre on exactly where ?Caicedo 
La Siera? is. That is because Caicedo and La Siera are simultaneously two separate 
places, many diferent places, and exactly the same place, and their history and 
geography explain why is this is so. Caicedo La Siera is in Zone 3, the East Central zone 
of the city that includes Comuna 8 (Vila Hermosa), Comuna 9 (Buenos Aires), and 
Comuna 10 (La Candelaria). Comuna 10 includes the city center, made up mostly of 
downtown busineses, and has long had among the highest murder and crime rates in the 
city (making al Zone 3 violence data useles as a proxy for violence in Comuna 8, since 
Comuna 10 dominates any data aggregated at the zone level); Comunas 8 and 9 are 
mostly hilside residential communities. Geographicaly, Zone 3 is the Quebrada Santa 
Elena (Santa Elena Crek) water basin: the Santa Elena collects rainwaters from the 
smaler creks in the central mountains on the eastern side of the city and caries them 
down through the city center and into the Medel?n River. Comunas 8 and 9 are the upper 
Santa Elena water basin, with the crek serving as the boundary betwen the two: 
Comuna 8 is the north bank, Comuna 9, the south bank. In Comuna 10, the downtown 
area, the crek runs beneath Cale 52, caled Avenida La Playa (or Beach Avenue, so 
named ? or so locals claim ? because of its proximity to this particular body of water), 
which is one of the main axes of the city center. In fact, the Santa Elena crek has been 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 73 
one of the more important landmarks in the city?s history. It was originaly named An?, 
and the area where it drained into the Medel?n river was one of the first areas setled, in 
the 17
th
 century: the city?s first church, La Candelaria, was built there, and the main plaza 
(now caled Parque Ber?o) was established nearby; the area then grew into what is today 
downtown Medel?n.
46
 
At the top of the mountains of the Santa Elena water basin lies the border betwen 
the city of Medel?n and the corregimiento (rural subdivision) of Santa Elena, historicaly 
one of the main points of entry into the city. Santa Elena and the towns of Rionegro and 
Guarne are both east of the city, and some of the first roads into Medel?n from these 
towns ran along the Santa Elena Crek. In the late 19
th
 century, the road along the 
northern bank of the crek ? today, the upper part of Cale 52?La Playa ? had its first 
stop in Medel?n at an area known as Las Estancias, long the site of the only church 
betwen the top of the mountain and La Candelaria downtown. At first unpaved, this 
step and winding road was traversed mostly by mule. As the city started to become more 
populated, however, there was a need to get more and more products from the towns east 
of the city into the markets at the city center, and to get more and more people from the 
city center up to church on Sunday. The solution to that transportation problem gave this 
area, once known exclusively as Las Estancias, the name (perhaps, more acurately, the 
nickname) of Caicedo. 
During the 1920s, a trolley system was built in downtown Medel?n, but it reached 
only as far east as La Toma, the bridge over the Santa Elena Crek that connects the 
                                                
46
 Information from this and the next thre paragraphs comes largely from Diego R?os, personal 
comunication (Interview No. 4), 13 March 209; Juan Diego Alzate Giraldo, ?Alg?n D?a Recuperaremos 
la Noche?; Gloria Naranjo Giraldo, Medel?n en Zonas (Medel?n, Colombia: Corporaci?n Regi?n, 192), 
86; and common knowledge in Medel?n. 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 74 
winding road coming down the mountain (upper Cale 52?La Playa) to the road that goes 
directly downtown (lower Cale 52?La Playa). At La Toma is a bario that city planners 
caled ?Caicedo.? Being on the south bank of the Santa Elena crek, bario Caicedo is in 
Comuna 9, not Comuna 8. Yet almost nobody in Medel?n today cals that bario 
?Caicedo?; to most people, it?s just part of Buenos Aires: everyone ?knows? that Caicedo 
is in Comuna 8, north of the crek. During the 1930s and 1940s, the city?s population was 
growing quickly as the worldwide depresion forced farmers off their lands. But the 
downtown trolley system had no service beyond La Toma, and there was a growing 
demand for mas transit to take trafic from downtown up to the hilside communities 
where the les-expensive housing was. Microbus companies emerged to met that 
demand. At first the buses caried goods and pasengers from the troley stop at Caicedo, 
across La Toma bridge, and up just a few blocks on Cale 52?La Playa to a bus stop 
caled La Planta (near Carera 23), but later the routes were extended to another bus stop, 
Estrechura/Canelones (near Cra. 19), and so on through the middle of the century. A new 
bus stop was added every decade or so as the roads were paved and demand justified the 
cost of expansion: to Aguinaga (Cra. 16), then to San Antonio (Cra. 13), and finaly to 
Tres Esquinas (Cra. 9), where the bus route split of into thre directions to serve the 
growing peripheral barios: one route went through bario Vilatina (and later into to what 
is now bario San Antonio), one through bario Vila Liliam to bario Vila Turbay, and 
one through bario Vila Liliam to bario La Siera. Because the brightly colored buses 
originated at the bario by the bridge, they al had ?CAICEDO? painted on the front in 
big white leters. Over time people started refering to the areas the bus routes served by 
the same name. Today, Caicedo, in the popular imagination, is no longer the area where 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 75 
the bus routes start, but the area where the bus routes end: the destination has become the 
origin. (Later, as the bus routes became established, some bus owners started 
diferentiating the thre routes at the top by painting ?LA SIERA? or ?VILATINA? 
on their buses, although many stil cary the name Caicedo.) 
Caicedo, therefore, is not one place but, in a way, four: (1) the sector 
encompasing al or part of the six barios in the upper reaches of Comuna 8 that serve as 
the destinations of the Caicedo bus routes: San Antonio, Las Estancias, La Siera, Vila 
Liliam, Vila Turbay, and Vilatina; (2) the corridor encompasing al or part of those six 
barios plus al or part of the thre barios along the northern bank of the Santa Elena 
crek that Cale 52 runs through: El Pinal, La Libertad, and Sucre; (3) the Caicedo 
(sometimes speled Caycedo) that is stil an oficialy designated bario in Comuna 9 in 
the city?s maps and development plans; and (4) the Caicedo that designates the JAC 
(Junta de Aci?n Comunal, or Community Action Board) in Comuna 8 that is carved out 
of the southern end of bario Sucre at La Toma bridge, just across the crek from the 
Comuna 9 Caycedo. For the purposes of this study, ?Caicedo? is the first of the four: the 
six barios of upper Comuna 8. Some in Medel?n would agre with this usage; many 
others would object strenuously. But Caicedo is only half of the story: the case under 
study is Caicedo La Siera. Where is La Siera? 
Like Caicedo, the name of the areas caled ?La Siera? has a storied history as 
wel, and it, too, refers to four overlapping places. A sierra is a mountain ridge, usualy 
jagged like the teth of a saw (the Spanish word sierra literaly means ?saw?), and the 
uppermost part of Comuna 8 is nothing if not step and jagged. Oficialy, La Siera is 
(1) the uppermost bario of Comuna 8, acording to the city maps, just next to Vila 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 76 
Turbay; and it also is (2) the name of a JAC that actualy is in bario La Siera but that 
does not encompas al of that bario. In the popular imagination, things are a litle 
diferent, depending on whether you live in the area or outside of it, and perhaps 
depending on how long you have lived in the area. Because it is so far from the city 
center, and so high in elevation, and because much of that section is hidden by Cero Pan 
de Azucar (Sugarloaf Hil) at the top of the mountain, few people who do not live there 
have ever sen or visited it. To them, La Siera is the place where the buses marked ?LA 
SIERA? go (under the same logic that drove the evolution of Caicedo?s identity); and 
since most have never actualy taken those buses, this means La Siera is either (3) the 
same six barios as Caicedo (as in the previous paragraph), or (4) al or part of the four 
barios above Quebrada La Castro (Castro Crek, which runs down Pan de Azucar): Las 
Estancias, Vila Liliam, Vila Turbay, and, of course, bario La Siera. For the purposes 
of this study, ?La Siera? or ?bario La Siera? wil be the first of these four options, the 
uppermost bario, while ?Caicedo La Siera? ? the subject of this case study ? wil 
refer to the fourth option: the four barios above La Castro. Unfortunately, while many in 
Medel?n would have no strenuous objection to this usage, others certainly would, since 
for some people ?Caicedo?La Siera? refers specificaly to bario La Siera in the sector 
they think of as Caicedo, just as ?Caicedo?Las Estancias? would be bario Las Estancias 
in the same area. Nevertheles, the area under study needs to be specified and named, and 
Caicedo La Siera ? the barios Las Estancias, Vila Liliam, Vila Turbay, and La Siera 
? is no worse a name than other available options. 
Caicedo La Siera was chosen as the sector through which to study the details of 
the dynamics of legitimacy in Medel?n partly because it was one of the peripheral areas 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 77 
whose history has been fairly typical in the city: ignored by the state, plagued by violence 
and gangs, taken over by militias, subjected to war by paramilitaries, taken over by 
organized crime, recognized and partly improved by the city, and finaly faced with the 
fear of a return to worse times. It was chosen as wel, however, because my research 
asistant had once worked there and so had good contacts within the bario: it was a place 
I could operate in, and where I could be confident that residents and former residents 
would talk to me. (Unfortunately, midway through the list of interviews that had been 
planned, violence and fear spiked, and several people warned me that it had become too 
dangerous for my interview subjects to be sen with an outsider and too dangerous for 
outsiders to enter the bario, so I was forced to terminate my field research there early 
and rely instead on the interviews that had been taken up to that point.) Each of the next 
five chapters has a subsection that looks specificaly at Caicedo La Siera, and these are 
based primarily on these interviews and on other primary sources that I was lucky enough 
to encounter, including a documentary film ade about the sector, a book of memoirs 
writen by local residents, and several interview transcripts by previous researchers. 
The paterns of violence are substantialy more dificult to identify in Caicedo La 
Siera than in Medel?n and Colombia, since most data are not disaggregated at the bario 
level; in other words, Caicedo La Siera has the same data problems as the two higher 
levels of analysis ? data are not collected for some forms of violence, the quality of 
much of the data that are collected is questionable, much of the data have coverage only 
for recent years, and much of the data that have been collected are not made publicly 
available in a useable form by those who manage it ? but with the additional chalenge 
that the data are over-aggregated. Nevertheles, a combination of the quantitative data 
Lamb, Robert D. | Chapter 2. Background: Medel?n and Caicedo La Siera 78 
that are available and the qualitative impresions of residents and experts provides 
enough of a glimpse into the key paterns to proced with the analyses in the chapters that 
follow. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 79 
Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 
Flying from Medel?n to most of Colombia?s major cities takes les than an hour, 
but by bus you have the opportunity to stare out the window at gorgeous natural scenery 
for six, 10, or 15 hours straight. Outside of Medel?n, rural Colombia ofers stunning 
views of lush mountains, gren valeys, dramatic cloud formations, and vast, undulating 
pastures. Today you can enjoy such trips with litle concern for your safety: the Army has 
retaken control over most major highways, which until just a few years ago had been 
subject to frequent guerila or paramilitary checkpoints where driver and pasenger alike 
were shaken down, their documents checked, their money and valuable goods stolen; 
sometimes the name on somebody?s document would match a name on some list and the 
pasenger would be taken off the bus to disappear forever. 
The situation was worse away from the highways, and even worse the farther you 
got from the highways. The longest-running civil war in the Western Hemisphere, 
beginning in the 1960s, was being fought in Colombia, and most of the fighting was 
taking place in rural areas, far from regular atention and protection from the central 
government. The war began just a few years after the previous civil war, known today 
only as La Violencia ? The Violence ? ended in a setlement betwen the two main 
waring factions: the Conservative Party and the Liberal Party. The setlement created the 
National Front, basicaly a power-sharing agrement betwen the two parties that 
excluded al other political parties from the political system. Disident liberals and 
members of radical parties objected to the arangement, and some of them took to 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 80 
Colombia?s remote mountains and thick jungles, formed themselves into military units, 
and, inspired by the Communist revolutions in the Soviet Union, China, or Cuba, 
launched an insurgency whose main approach was guerila warfare. Or rather, they 
launched several insurgencies, as the leftist groups did not al join together in solidarity 
? and sometimes hated one another with a pasion exceded only by their hatred of the 
National Front. 
The National Army was sent out to fight them from time to time, not always very 
efectively, and to its forces were added the eforts of armed civilian militias. Over the 
course of several decades, none of the thre categories of fighter ? guerila, military, 
and paramilitary ? could claim the moral high ground in terms of respect for 
international norms for human rights and the conduct of war: those were honored in word 
and occasionaly in deed, but there was never any doubt that this was a dirty war, as La 
Violencia was before it. As a Colombian physician who has taken it upon himself to 
study his country?s violence has observed: 
I think that noone in the country today has any doubt that Colombia is not 
just a violent country, it?s a barbarous country; barbarity is not respecting 
the most minimal humanitarian [standards], barbarity is not recognizing 
anyone [who should enjoy] exclusion from the conflict, barbarity is 
[geting] a practicaly sadistic satisfaction from violent actions, which 
could be ilustrated with multiple examples ?. We are in a phase pasing 
from a grand cycle of violence to a situation of barbarity ?.
1
 
                                                
1
 ?Pienso que nadie hoy en el pa?s tiene duda de que Colombia ya no es solamente un pa?s violento, es un 
pa?s b?rbaro; barbarie es el no respeto a ning?n m?nimo humanitario, barbarie es el no reconocimiento de 
cualquier exclusi?n en el conflicto, barbarie es la satisfacci?n pr?cticamente s?dica en la acci?n violenta, la 
cual se puede ilustrar con m?ltiples ejemplos ?. Estamos en una fase de paso de un gran ciclo de violencia 
a una situaci?n de barbarie ?.? Sa?l Franco Agudelo, ?Comentarios a la Investigaci?n ?Estado del Arte 
Sobre Violencia Urbana en Antioquia?,? [?Coments on the Study ?State of the Art in Urban Violence in 
Antioquia?,?] in Balance de los Estudios Sobre Violencia en Antioquia, ed. Pablo Emilio Angarita Ca?as 
(Medel?n, Colombia: Municipio de Medel?n y Editorial Universidad de Antioquia, 201), 185. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 81 
By the 1980s, vilagers, farmers, and ranchers were regularly being atacked by guerilas 
who wanted resources (goods, land, recruits, money) and threatened death and 
displacement to those who did not supply them. Then they were atacked by state security 
forces who acused them of supplying resources to guerilas; then atacked by 
paramilitaries who acused them of being guerilas; then atacked by the guerilas again, 
who acused them of snitching to the authorities and to the paras. The atack could be 
anything from a threatening note, to the kiling of their livestock, to a threat to leave or 
suffer the same fate as others have, to the selective kiling and disembowelment of 
community leaders, to masacres of entire vilages. 
For many, just one atack, one threat, was enough to drive them from their homes: 
Colombia, even today, has long had some of the highest numbers of internaly displaced 
populations (IDP) in the world. Amanda Uribe had heard a rumor that the leader of 
paramilitaries in her rural vilage ? ?they were the ones who governed the town, they 
were the authorities? ? didn?t like her son. ?One fateful and biter morning they knocked 
on the door of my house. It was a group of armed and hooded men.? 
The boss of these cruel, heartles asasins headed for my son?s room, 
where he was stil sleping and had no idea what was happening in our 
house. ? ?Get up, man, let?s go, we have to talk.?? For us, it was 
impossible to do anything to save him. ? [I heard] two shots about two 
blocks from the house. My reaction was imediate. I remember I let out a 
piercing scream loaded with al the pain of my heart. ? Barely two weks 
had pased since Alex?s death when we decided to move to Medel?n.
2
 
                                                
2
 ?? los paramilitaries eran los que gobernaban el pueblo, elos eran la autoridad. ? Una fat?dica y amarga 
ma?ana tocaron a la puerta de mi casa. Era un grupo de hombres armados y encapuchados. ? El jefe de 
esos imp?os, desalmados, asesinos, se dirigi? al cuarto de mi hijo que a?n dorm?a y que no se hab?a 
enterado de lo que eestaba sucediendo en nuestra casa. ? ?Lev?ntese hombre y salgamos que tenemos que 
conversar.? ? Para nosotros era imposible hacer algo para salvarlo. ? [Escuch?] dos disparos como a dos 
cuadras de la casa. Mi reacci?n fue inmediata. Recuerdo que yo pegu? un grito desgarrador cargado de 
dolor de mi coraz?n. ? Hab?an pasado apenas quince d?as de la muerte de Alex cuando decidimos 
desplazarnos para Medell?n.? Amanda Uribe, ?Tres Sucesos Amargos,? [?Thre Biter Events,?] in Jam?s 
Olvidar? Tu Nombre, ed. Patricia Nieto (Medel?n, Colombia: Alcald?a de Medel?n, 206), 31. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 82 
Atacks, threats, masacres, vandalism, and fear in the countryside drove many families 
into the cities, where they hoped to find safety and jobs. Most, of course, found neither. 
Instead, they setled in the periphery of places such as Medel?n, which had few 
jobs and no land where they could setle legaly. They were forced to become squaters 
on unoccupied lands on the mountainsides or near the city dump. They had to pirate 
electricity and water, since the city would not recognize the legality of their setlements 
or provide adequate development asistance, public or social services, or police 
protection. Many of these setlements were on hilsides that were at high risk of deadly 
mudslides (the family quoted above later lost their shack in Medel?n in a mudslide). 
They were on their own. Worse stil, many of these displaced families were considered 
outsiders, a threat, by those who were already living there on the city?s periphery. Partly 
this was a reflection of the in-group versus out-group mentality that is very strong in 
Colombian culture, partly it was a reflection of economic realities: the old-timers were 
not necesarily poor, but they usualy earned litle more than a marginal living and did 
not appreciate having to compete with the newcomers for resources that already were too 
scarce. They certainly did not appreciate their ways, their customs, their diferences.
3
 
Conditions were poor, tensions were high, and life was hard. 
In the communities where the Medel?n Cartel recruited its young asasins and 
dealers, not everyone was impresed by the lure of the mafia?gangster?asasin lifestyle. 
In fact, it is safe to say that most people in these marginalized communities strongly 
disapproved of what many of their young men and boys were being drawn into ? and 
were drawing them into. Part of this was virtue, a belief that, however poor you are, you 
                                                
3
 Such sentiments were expresed to me by a resident of East Central Medel?n: ?Cristiana,? personal 
comunication (Interview No. 9), 20 March 209. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 83 
do not kil, you do not threaten: we are good people, and that is not what good people do. 
Part of it was fear: once fred from the behavioral constraints of social norms against 
certain uses of violence, too many young people began acting in ways that were deeply 
harmful to their own communities, and most residents, afraid, wanted to maintain order. 
The older generation especialy disapproved of the metal-heads, pot-heads, and punks 
who were hanging around on stret corners playing loud music, seting a bad example for 
children.
4
 The malevos, the bad-ases of the earlier era, who had caried knives but rarely 
used them, had become pistoleros, carying around handguns and using them far too 
readily. The glue-snifing stret children never went away: however many were 
?cleansed,? there were always more to replace them: more orphans, more boys fleing 
abusive parents, more girls fleing pedophile stepfathers, and too much misery and 
hunger to handle without a constant, mind-numbing glue high. The ?low-life? chichipatos 
were stil around as wel, smoking their bazuco (cheap cocaine paste), stealing and 
whoring to score their next hit, desperate and violent as ever (se ? 2.1.2). Crime was 
rising. 
?They atacked you at six in the morning, at noon, at night, leaving the house. 
Catching the bus! They?d try to kil you for a hit of bazuco, they?d kil you out of spite,? 
a construction worker told a reporter about life in the ?20
th
 of July? neighborhood in 
Comuna 13, in the west of the city, during the ?invasions? of the early 1980s. He 
described life in a place that the city would not recognize as a legal setlement, where 
public services had to be stolen (more than a few people died trying to run pirated copper 
wires from the electric grid to their shacks), and where the nearest police station had too 
                                                
4
 ?Germ?n,? personal comunication (Interview No. 16), 6 April 209. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 84 
few officers for too many people with too many problems (and where responding to a cal 
would sometimes require a climb of a hundred stairs or more): 
There were a lot of low-lifes with cheap pistols and a lot of thieves around 
here. ? The soap disappeared from public restrooms and the clothes from 
the clotheslines. And they?d even come into people?s houses and take the 
pots from the stove ? stil boiling! ? and leave the poor people without 
lunch. ? [The police] were worthles ? four oficers who never even 
went out on patrol. They holed themselves up in there, taking care of each 
other. ? But most serious was the disunity. There were so many problems 
and people didn?t work together. There was mistrust and resentment over 
problems that had happened: that this one guy stole this other guy?s lot, 
that those people were throwing trash and shit into their neighbor?s yard, 
that what?s-his-name was corrupting some guy?s daughter, ? that that one 
woman?s taking that other woman?s husband, that a step-father?s running 
off with his step-daughter, that some woman was caried off by the police 
for sticking a table knife into her man?s leg. Those were the most common 
disputes.
5
 
But such disputes often turned violent, if they hadn?t been already, and anyone who tried 
to get involved or get the police involved to resolve a dispute could end up geting 
themselves kiled. 
In a way it?s no wonder that so many young people joined gangs; that more did 
not speaks to the eforts of the many community leaders and common people who risked 
their lives to try to maintain some semblance of stability and community solidarity, of 
which there was very litle at the time. 
                                                
5
 ?A usted lo atracaban a las seis de la ma?ana, al mediod?a, en la noche, saliendo de la casa. ?Cogiendo el 
bus! Lo atracaban por un bazuco, y por una inquina lo mataban. ? Hab?a mucho chichipato con chang?n y 
mucho ladr?n por ah?. ? Los jabones desaparec?an de los ba?os comunales y la ropa de los alambres. Y 
hasta se entraban a las casas a sacar las olas de los fogones, ?hirviendo todav?a!, y dejaban a la pobre gente 
sin almuerzo. ? [La polic?a] no serv?a para nada ? cuatro agentes que ni siquiera sal?an a patrular. 
Permanec?an ah? encerados, cuid?ndose los unos a los otros. ? Pero lo m?s grave era la desuni?n. Tantos 
problemas que hab?a, y la gente no trabajaba unida. Viv?an con desconfianza y resentimientos por 
problemas que habian pasada; que el uno le rob? el lote al otro, que aquel le est? tirando la basura y la 
mierda al vecino, que fulano le perjudic? la hija a perano, ? que aquela mujer le est? quitando el marido a 
la otra, que un padrastro se larg? con una hijastra, que a tal se?ora se la llev? la polic?a por clavarle a su 
marido un trinchete en una pierna. Esas eran las peleas m?s comunes.? Ricardo Aricapa Ardila, Comuna 
13: Cr?nica de Una Guera Urbana [Comuna 13: Chronicles of an Urban War], 2a ed. (Medel?n: 
Editorial Universidad de Antioquia, 205), 18. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 85 
The variety of smal stret crews and gangs and the variety of reasons for joining 
them were broad. Some gangs were microenterprises, something young people would 
join because they could make or steal money. The communities they lived in were poor, 
but they could make money by stealing from the wealthiest people in their own barios or 
by stealing in the richer parts of town; by seling pot to tenagers, glue to stret children, 
or bazuco to addicts in their own barios, or pot and cocaine to wealthier consumers in 
wealthier barios; or by operating a plaza de vicio where they sold drugs retail or 
prepared drugs wholesale (for example, by cutting pot leaves and packaging them into 
joints for retail sale, or cutting a kilogram of pure cocaine with nonreactive substances 
and packaging it for resale). Many entrepreneurial gangs made money by ?seling? 
protection to the scores of corner stores, smal busineses, and transportation services that 
operate in even the poorest barios (a payment for such ?services? is caled a vacuna, or a 
?vacine? against atack), which could earn them a few dollars per store or per bus every 
wek; depending on how much teritory they controlled, this could amount to a 
substantial sum of money in a place where the most common alternative job, stret 
vendor, would earn them just a few dollars a wek in total. 
Some gangs were manifestations of long-standing family feuds and blood debts; 
one gang war in East Central Medel?n began with an argument betwen children from 
diferent blocks: their older brothers got involved, someone died, and the revenge cycle 
began, the later political wars being litle more than cover for existing hostilities.
6
 Some 
gangs were groupings of people with similar interests, often perverse; one smal gang in 
the Northeast was dedicated to raping a very specific demographic: men in their 40s and 
                                                
6
 ?Mama Luz? Edna Garc?a Copete, personal comunication (Interview No. 1), 1 April 209. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 86 
50s ariving home from work late at night.
7
 Early on there was not much organization ? 
a thousand flowers bloomed. As time progresed, however, smaler crews would evolve 
into criminal gangs, and specialized gangs ? such as the notorious sicarios (asasins) 
formed by the Medel?n Cartel ? would begin operating as independent contractors, 
hired by narcotrafickers, politicians, businesmen, and common people alike to cary out 
specific crimes, ?resolve? conflicts, or take care of grudges. A smal-time hitman could 
earn a hundred dolars for a job, with more important targets bringing in hundreds and 
sometimes thousands of dollars ? and there was no shortage of customers and victims in 
the city. 
3.1. Medel?n: The Insurgency Comes to the City 
3.1.1. Peace Camps Train Future Warriors 
The real growth in nonstate capacity for collective violence in Medel?n can be 
traced to a specific date of national significance: 24 August 1984. That was the date the 
government of President Belisario Betancur signed a cease-fire with the Movimiento 19 
de Abril (M-19) guerilas to begin negotiations for peace. During the peace talks, M-19 
set up ?peace camps? (campamentos de paz) in Medel?n and other major cities as a way 
to build popular support for when they would become a political organization once the 
peace agrements concluded. In Medel?n such camps were set up in five barios: Popular 
and Zamora in Comuna 1 in the Northeast, Moravia and Castila in Comunas 4 and 5 just 
north of the city center, and Vilatina in Comuna 8 in East Central Medel?n. Young 
                                                
7
 Gilberto Medina Franco, ?Una Historia de las Milicias de Medel?n (Historia Sin Fin),? 23. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 87 
people would show up every day, more than 50 at a time at the camp in Popular, for 
example.
8
 Many had otherwise been spending their days on stret corners doing drugs; at 
the peace camps, by contrast, they were atending lectures and conferences on 
revolutionary politics, doing community theater, organizing sancocho block parties 
(sancocho is a type of stew often made communaly during outdoor parties), and helping 
to distribute groceries to poor families. The camps were helping the communities 
maintain order. 
In addition to propagandizing and providing social services, however, M-19?s 
peace camp organizers were also hedging their bets in case the peace talks broke down. 
As one member explained, ?we were secretly giving people military schooling: 
asembling and disasembling firearms, doing inteligence and recovery work.?
9
 Of the 
scores of young people atending the camps, they would select the ones who showed the 
most interest in politics and give them extra training on the side. But the camps were in 
place for les than a year. At the national level, the peace talks, which the army and most 
police departments never supported in the first place, broke down in January 1985, and 
the camps were shut down later in the year. ?In like the middle of ?85, the government 
put out a decre in which they outlawed the camps, because we were making more 
wariors and not thinking about peace. Then they let loose on us ? they raided our 
headquarters, arested our people, lowered our flag. Things got so rough we decided to 
                                                
8
 Alonso Salazar J., No Nacimos Pa? Semila. 
9
 ?? por debajo de cuerda le d?bamos escuela militar a la gente: armar y desarmar fieros, hacer 
inteligencia y recuperaciones.? ??ngel,? interview ith Ibid., 7. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 88 
close the camps and head back to the mountains.?
10
 Some of the newly trained young 
people went with them. Others, fleing persecution, just disappeared from the barios. 
Others, however, stayed put. A lot of them continued doing social work and 
community service, but some, realizing that they now had a very useful set of skils, 
turned their community-organizing and military training to other uses. One was a young 
man named Ignacio. Nacho, as everyone caled him, was an active participant in the 
peace camps, and after they were shut down he continued working with some of his camp 
mates, and they recruited others. Back when the peace camps had ?sponsored community 
sancochos, retreats, or other activities to win the comunity?s afection, they organized 
commisions to solicit cash donations from the busines community ? [and these] 
donations were made without major reservation, since the M-19?s camps had offered 
security and recreation to these forgotten sectors.? But after the camps were dismantled, 
what was once a community-organizing efort eventualy became a protection racket: 
?the donations that at first were given wilingly became obligatory payments or vacunas,? 
the word locals use (it means ?vacine?) to refer to money extorted from local busineses 
for ?security? services.
11
 M-19 found it necesary to warn Nacho to stop using their name 
for such activities, and so his community group, now basicaly a gang, came to be known 
as Los Nachos. After he was kiled in a gun batle, Los Nachos lost what litle political 
                                                
10
 ?Como en la mitad del a?o 85 el gobierno sac? un decreto en el que prohibi? los campamentos, porque 
est?bamos preparando m?s guerrileros y no pensando en la paz. Entonces se desat? una persecuci?n, nos 
allanaban la sede, deten?an la gente, nos ariaban la bandera. La cosa se puso tan complicada que decidimos 
cerrar los campamentos y echarnos otra vez pal monte.? ??ngel,? interview with Ibid. 
11
 ?.. cuando los campamentos realizaban un sancocho comunitario, una retreta y otra actividad para 
ganarse el afecto de la comunidad, organizaban comisiones para solicitar la colaboraci?n en especie a los 
negocios del sector. .. El aporte del comercio se daba entonces sin mayores reparos pues los campamentos 
del M-19 hab?an ofrecido seguridad y recreaci?n a estos sectores olvidados. Cuando los jefes del M-19 
salieron de El Popular, los aportes que en principio eran de buena voluntad se convirtieron en cuota 
obligatoria o vacuna ..? Gilberto Medina Franco, ?Una Historia de las Milicias de Medel?n (Historia Sin 
Fin),? 24. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 89 
ideology their founder had once imposed upon them.
12
 In 1986 they linked up with one of 
Pablo Escobar?s men to help recruit j?venes kamikases (kamikaze kids), young asasins 
who were paid to kil important public oficials and who were not expected to survive the 
atempt. The Nachos?s ?community service? got increasingly ugly. One day, some 
members of Los Nachos set a smal bus on fire, in full public view, because its owner had 
not paid his vacuna. It turned out, however, that thre tenagers who worked for the 
owner were stil inside. They could not escape the flames, and their ?neighbors watched 
in horror this Dantesque scene? of thre screaming boys burning to death. Community 
outrage forced the state to respond in 1987, but of the 25 members who were captured 
only five were ultimately charged; the rest continued their outrages against the 
community.
13
 
Los Nachos were only one of many such groups. ?Los Nachos, Los Calvos, Los 
Monta?eros, Los Pelusos, and other litle gangs emerged,? one resident of the northern 
barios told Alonso Salazar Jaramilo (elected Medel?n?s mayor in 2008) for a best-
seling book of interviews he published in 1990 about youth gangs in the city. The 
resident continued: 
Those gangs were made up of two or thre grown-ups and a bunch of litle 
shits who grew into thugs, 13-, 14-, 15-year-old kids doing the work of the 
devil. They charged taxes, 2,000 pesos a wek for shops and 5,000 for 
buses, demanding it right in the strets as if they were the law, robbing gas 
trucks. ? The heyday was in 1986 and 1987, [when] the gangs controlled 
the whole bario. Life changed completely, everyone holed themselves up 
                                                
12
 ??ngel,? interview ith Alonso Salazar J., No Nacimos Pa? Semila, 76. 
13
 ?? los vecinos observaban hororizados la dantesca escena.? Gilberto Medina Franco, ?Una Historia de 
las Milicias de Medell?n (Historia Sin Fin),? 24. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 90 
their houses at six in the afternoon. They started kiling each other, they 
fought over busines afairs, retribution, turf disputes.
14
 
While pervasive interpersonal-communal, collective-economic, and collective-
social violence by criminals and smal-time gangs in the early part of the decade was 
what opened the doors to the M-19 peace camps ? since the guerila camps offered a 
means of social order to communities desperate for it ? those peace camps had the ironic 
longer-term efect, beginning in 1984, of building the capacity of gangs to commit larger-
scale acts of collective violence.
15
 The following year, homicides in Medel?n began to 
rise in earnest: the rate in 1985 was 109 murders per 100,000 inhabitants of the city, more 
than double the previous year?s rate of 46 per 100,000; it hit 131 in 1986, 157 in 1987, 
204 in 1988, and 258 in 1989 ? a quintupling of the murder rate in just five years.
16
 This 
increase was not due solely to the crime gangs that emerged from the peace camps. This 
was also the period in which Pablo Escobar was recruiting young people from the 
peripheral barios, especialy in the Northeastern zone, to form gangs specializing in 
asasinations. Once formed, these gangs took contracts (as wel as rewards for kiling 
police) not only from Medel?n Cartel members but from anyone wiling to pay their 
                                                
14
 ?Surgieron Los Nachos, Los Calvos, Los Monta?eros, Los Pelusos y otras banditas ? Esas bandas eran 
formadas por dos o tres mayores y una manada de culicagados crecidos a matones, peladitos de 13, 14, 15 
a?os haciendo las del diablo. Cobraban impuestos, de dos mil pesos semanales a las tiendas y cinco mil a 
los colectivos, requisaban en la cale como si fueran la ley, atracaban los caros surtidores. ? En 1986 y 
1987 fue el auge total, las bandas controlaban todo el bario. La vida cambi? completamente, todo el 
mundo se enceraba en las casas a las 6 de la tarde. Entre elos empezaron matarse, se peliaban por enredos 
de negocios, de venganzas o disput?ndose el teritorio.? ??ngel,? interview ith Alonso Salazar J., No 
Nacimos Pa? Semila, 86. 
15
 In this study, comunal violence is unorganized violence that takes place in public places by one or two 
individuals acting on their own, and collective violence is more organized violence perpetrated by members 
of a group for social, political, or economic reasons. Se Apendix B. 
16
 Data from Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden Civil, Unidad de 
Convivencia Ciudadana; and Fiscal?a General de la Naci?n, Unidad de Reaci?n Inmediata. Good data are 
scarce for other forms of violence in Medel?n before 190. Se Chapter 8. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 91 
price. Asasination became both a commodity and a carer path for young people who 
had no other job prospects; an Army study in the late 1980s found 120 asasin gangs 
with maybe 3,000 members whose average age was 16 years.
17
 This was also the period 
in which a new leftist political party, the Patriotic Union (UP: Uni?n Patri?tica), emerged 
on the national stage, and was nearly wiped out by hit men and death squads made up of, 
or hired by, state actors and paramilitaries. Many UP-afiliated students and profesors at 
the University of Antioquia in Medel?n, for example, were asasinated, and human-
rights defenders and labor-union leaders suffered many losses at both the national and the 
local levels as wel; it was common knowledge in the city that the security forces used the 
asasin gangs for such political purposes. In Medel?n in 1990, there were 52 forced 
disappearances, 85 extrajudicial executions, and 17 masacres by unknown perpetrators; 
in 51 combat events, 180 people died, including 147 civilians, and 110 people were 
injured, 88 of them bystanders.
18
 Stil, these specific instances of political violence were a 
smal portion of the 5,424 homicides registered in the city that same year.
19
 Most, it is 
widely believed, were not instances or organized violence at al but rather a manifestation 
of pervasive violent crime amid a widespread breakdown in the rule of law and in the 
institutions of social order that had been chalenged by an uncomfortable mixing of rural 
and urban cultures in the context of a population explosion, a mixing of political and 
criminal motives in the context of a national conflict, and a mixing of local and global 
markets in the context of a narcotraficking explosion. Clearly the capacity for violence 
                                                
17
 Cited in Alonso Salazar J., No Nacimos Pa? Semila, 187. 
18
 Data from Centro de Recursos para el An?lisis de Conflictos (CERAC) Base de Datos de Conflicto 
(CCDB-CERAC V.8.). 
19
 Secretar?a de Gobierno Municipal de Medell?n. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 92 
? inter-personal communal, and collective-political, -social, and -economic ? had al 
been greatly augmented, and the social bariers to its use, greatly diminished. 
3.1.2. Militias Emerge to Defend against Crime 
While the peace camps were one of the factors in the growth in this capacity for 
violence in the mid-1980s, they also were one of the two main sources of the 
communities? capacity to defend themselves against the power of violent gangs 
beginning in the late 1980s; the other source was the long tradition in Colombia of 
vigilante justice. Some of the political leftists who had emerged from the peace camps ? 
in contrast to those who simply took advantage of the training without caring much about 
its political content ? drew both on their peace-camp training and on this vigilante 
tradition to form community self-defense militias. Meanwhile, some of the vigilantes 
who ?had formerly been anonymous ?entrepreneurs? of social cleansing in the barios? in 
the previous era now ?became the leaders of new militia groups? that were being formed 
to protect those communities. ?The activities of the death squads and social-cleansing 
groups, frequent in Medel?n during the 1970s, had also created an atmosphere favorable 
to the militia?s [sic] approach of taking justice into one?s own hands.? 
The militia groups became widely known throughout the city during 1990 
and 1991. Many people were surprised at the spectacle of armed and 
hooded young people who proclaimed themselves the armed power of the 
barios, but they had been known for some time in a les public way for 
their readines to exterminate criminals and drug addicts, especialy 
bazuco dealers.
20
 
The first communities where the gangs started to be confronted head-on by the 
groups that came to be known as militias were in the Northeastern zone, especialy in 
                                                
20
 Ramiro Cebalos Melguizo, ?The Evolution of Armed Conflict in Medel?n? (references omited). 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 93 
Bario Popular in Comuna 1. Acording to legend, it was ?Frederico,? a National 
Liberation Army (ELN: Ej?rcito de Liberaci?n Nacional) guerila from the Bajo Cauca 
region in northern Antioquia, who first showed up in 1987 to confront a gang that had 
threatened his family. (ELN is a Colombian insurgent group that was founded in the 
1960s by former members of a leftist political party excluded from the National Front 
power-sharing agrement that ended the Colombian civil war, La Violencia, in 1958; like 
other radicals, they took to the mountains and launched a guerila war.) But Frederico 
didn?t have the skils to organize the community beyond that initial confrontation. 
Another ELN member did. ?Julio? showed up in the same area not long after Frederico. A 
felow militant later described Julio?s goals and the early chalenges the militias faced in 
Medel?n: 
Julio came from distant lands and he didn?t just bring some automatic 
weapons, he also came armed with al kinds of ideas about how to 
confront the gangs, and he developed a proposal to cary out in Medel?n a 
project identical to [one caried in the city of Barancabermeja involving] 
workers militias. Nevertheles, Julio ran into a big obstacle: while the 
violence of Baranca [sic] involved a clearly political confrontation 
betwen guerilas and social organizations on the one hand, and the 
armed forces and paramilitaries on the other, what there was in Medel?n 
was this esoteric mix of political elements along with common violence.
21
 
But Julio presed on. In 1987, he made a dramatic, public gesture against four gangsters 
who had been extorting some residents of bario Popular. The gangsters had been 
demanding wekly payments to ?maintain the virginity? of the bario?s young women 
                                                
21
 ?Julio se desplaz? desde lejanas tieras y trajo no s?lo algunas armas autom?ticas, sino que tambi?n vino 
armado con un mont?n de ideas de c?mo enfrentar las bandas, elabor? la propuesta de construir en 
Medel?n un proyecto a imagen y semejanza de lo que eran las milicias obreras ? de Barancabermeja. Sin 
embargo, Julio se choc? con un gran obst?culo: mientras en la violencia vivida en Barranca hab?a una clara 
confrontaci?n pol?tica entre guerrila y organizaciones sociales de un lado, y fuerzas armadas y 
paramilitares en el otro, en Medel?n lo que hab?a era una mezcla enrarecida de elementos pol?ticos, pero 
tambi?n de violencia lumpesca.? Quoted in Gilberto Medina Franco, ?Una Historia de las Milicias de 
Medel?n (Historia Sin Fin),? 14. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 94 
and girls. Julio, ?in the middle of the stret and in front of the entire community,? 
confronted the four extortionists: he executed two of them on the spot and gave the other 
two an opportunity to change their behavior (they fled the bario instead). With this, ?the 
militia?s rules of the game were established: an execution as a leter of introduction [to 
the community,] ? persuasion and dialog as a way to pacify the delinquents, and al of 
this, of course, to win the community?s approval ?.?
22
 For some of the smaler gangs, 
this demonstration and others like them were enough to encourage them to disband 
voluntarily; the rest, lacking the discipline and weapons of the militias, fel in batle. To 
build on this foundation, Julio pulled together ELN members, local social and community 
groups, young victims of gang violence, and former participants in the M-19?s peace 
camps. 
Jos? Ricardo Barero Tapias describes the militias as having evolved in thre 
phases, each characterized by a representative militia. The first was represented by the 
Popular Militias of the People and for the People (MPP: Las Milicias Populares del 
Pueblo y para el Pueblo), which emerged in 1988 in barios Popular 1, Popular 2, La Isla, 
and Santo Domingo in the Northeast, as wel as in several barios of the Northwest. 
Founded by a former guerila named Pablo Garc?a, and joined both by other ex-guerilas 
such as Julio and by members of the community, MPP, like similar militias during this 
period, was dedicated to countering the gangs and common criminals that were 
threatening the security and tranquility of their communities, before turning to social 
work and community service, but they later overeached in their eforts to expand into a 
                                                
22
 ?? se establecieron las reglas del juego de las milicias: el ajusticiamiento como carta de presentaci?n ?, 
la persuasi?n y el di?logo con la delincuencia como v?a de pacificaci?n, todo esto por supuesto buscando la 
aprobaci?n de la comunidad ?.? Quoted in Ibid., 15. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 95 
broader political movement. The second phase was represented by he People?s Militias of 
the Aburr? Valey (MPVA: Milicias Populares del Vale de Aburr?), which was founded 
in 1990 and operated in the Northeastern and East Central zones more as negotiators of 
peace with and betwen gangs than as enforcers like the MPP. The MPVA were 
political organizations created specificaly ?with the intention of being a ?legitimate 
authority? in their area of influence, favoring broad and democratic forms of participation 
and social organization and trying to neutralize the criminal groups through disuasive 
[rather than coercive] mechanisms,? although they, too, later overeached, becoming first 
litle more than a private security firm and over time more of an extortive criminal 
organization. The third phase was represented by the Milicias Bolivarianas (MB), which 
were imported into several barios from outside of the city by people who had intended to 
protect communities not only from gangs but from right-wing death squads and state 
security forces as wel, but they failed to gain any traction within the communities 
because their managers did not understand local conditions wel enough to be efective.
23
 
The early militias were succesful because they acted, in the words of one of their 
founders, as ?a state inside the state?
24
 ? what I cal a statelet (se ? A.1). Francisco 
Guti?rez San?n and Ana Mar?a Jaramilo atributed this initial succes to the facts that the 
militias provided security to communities suffering from a severe breakdown in social 
order, that they achieved a great deal of social support as a result of this security and the 
                                                
23
 ?? se crearon con la intenci?n de ser una ?autoridad leg?tima? en el ?rea de su influencia, propiciando 
formas amplias y democr?ticas de participaci?n y organizaci?n social y tratando de neutralizar a los grupos 
delincuenciales mediante mecanismos disuasivos.? Jos? Ricardo Barero Tapias, ?Las Milicias Populares y 
el ELN Como Casos de Estudio: El Anclaje Cultural del Conflicto Pol?tico,? [?People?s ilitias and the 
ELN as Case Studies: The Cultural Anchor of the Political Conflict,?] Revista Javeriana (201): 213. 
24
 Quoted in Francisco Guti?rez San?n and Ana Mar?a Jaramilo, ?Crime, (Counter-)Insurgency and the 
Privatization of Security: The Case of Medel?n, Colombia.? 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 96 
social work that followed, and that they were explicitly dedicated to encouraging and 
promoting the communities? own moral values ? within the poor barios they controled. 
Outside those barios, they did not even pretend to be a state within a state. One former 
militant explained the dichotomy: ?[Y]ou have to steal, but only from the rich; you can be 
an asasin or an asailant, but outside of the bario; the state is the enemy and you have 
to hit it hard. So a crook can be a militia member inside the bario but a thief in another 
part of the city ..?
25
 
Despite their behavior outside of their barios, and despite the fact that the militias 
did act as statelets, they were not fuly antagonistic to state actors, but rather were in a 
both ?competitive and mutualistic? relationship with them: 
Indeed, the extreme levels of corruption of the authorities alowed the 
militias to link their revolutionary discourse to the concrete anti-criminal 
motives they operated on. At the same time, in their everyday activities, 
the militias relied on the state ? and on the very celebrated bureaucratic 
eficacy of Medel?n officials. As a militia member used to say, their 
struggle consisted of ?military undertakings in the night? and ?social work 
in the day time.? And, to be eficacious, social work required a state 
presence.
26
 
Many of the militias worked particularly wel with the city?s Community Action Boards 
(JACs: Juntas de Aci?n Comunal), which had been formed several decades earlier with 
the explicit purpose of acting as the city?s most local link to the people and as the 
people?s most direct means of aces to the city. As such the JACs have tended to be the 
component of the state with the greatest legitimacy acording to the people of the barios 
                                                
25
 ?? hay que robar pero a los ricos, usted puede ser un sicario, un asaltante pero fuera del bario, el Estado 
es el enemigo y hay que darle duro. As? un pilo puede ser miliciano en el barrio y en otra parte de la ciudad 
un delincuente ?.? Gilberto Medina Franco, ?Una Historia de las Milicias de Medel?n (Historia Sin Fin),? 
12. 
26
 Francisco Guti?rez San?n and Ana Mar?a Jaramilo, ?Crime, (Counter-)Insurgency and the Privatization 
of Security.? 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 97 
of the JACs? jurisdiction. The JACs also have tended to have the greatest flexibility of al 
the city?s agencies with respect to the organizations they can work with, because they act 
exclusively within certain barios with very litle city supervision. Their support for the 
militias? social workers during the late 1980s, therefore, was litle diferent from their 
support for non-militia social workers and community organizations. Pablo Garc?a, the 
leader of MPP in bario Popular in the city?s Northeast, won the support not only of the 
community but of Popular?s JAC as wel, after ?cleansing? the bario of its most 
?disposable? elements.
27
 Despite the presence of the JACs, it was the militias, and not the 
state, that were clearly in control of those barios; returning to the definition of control 
introduced in Chapter 1 (and elaborated in Appendix A), the militias had the ability to 
prevent the state from governing in their barios, but they nonetheles selectively alowed 
the state to govern certain functions, and indeed facilitated the delivery of certain state 
services when it helped them consolidate their support within those communities. 
Stil, in 1989, the state security forces began confronting the militias. What they 
found, however, was not what they expected: They had expected the militias to be as 
poorly armed and undisciplined as the gangs they had displaced. What they found instead 
were wel armed, wel trained, and disciplined military units that knew the strets and 
aleys of the barios and had strong backing from the community. The police were no 
match. Within a few years, the state would be forced both to operate through the criminal 
gangs that the militias opposed and, in paralel, to negotiate peace with the militias (se 
? 4.1.2). 
                                                
27
 Ibid. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 98 
This study?s top-level findings ? that, in Medel?n, teritorial control has reduced 
violence, and ilegitimacy is positively correlated with the costs of teritorial control ? 
are partly supported by the dynamic betwen the gangs and the militias during this first of 
the five study periods. As this section demonstrates, the second finding (ilegitimacy) 
certainly is supported: The gangs who had controlled micro-teritories had ilegitimized 
themselves by their abuses against community members and their transgresions against 
the communities? conservative values, opening themselves to chalenge by the militias, 
who legitimized themselves to those same communities by protecting community 
members and enforcing the communities? values. But the fact that the militias had 
undertaken an explicit strategy of legitimation to gain teritorial control suggested that 
some of the alternative hypotheses ? that legitimacy is either a cause or an antecedent 
condition for teritorial control and subsequent reductions in violence ? might be true. 
(Se Chapter 9 for a complete analysis of the dynamics of legitimacy.) 
A closer look at the dynamics at a finer level of analysis, however, would show 
that other groups were able to win and keep teritorial control without an explicit strategy 
of legitimation, suggesting that legitimacy, while helpful, is neither necesary nor 
sufficient to teritorial control. This might be explained in part by reference to the relative 
amount of resources available: the later gangs who succesfully controlled statelets (as 
opposed to the earlier smal-time gangs who had merely controlled micro-teritories) 
were flush with cash from narcotraficking and sicariato (hired asasination) ? they had 
partnered or contracted with outside traficking networks that were connected to the 
global cocaine trade ? while the militias who succesfully controlled statelets, even if 
they enjoyed modest support from rural guerilas, had to depend mostly on local sources 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 99 
of funding, such as protection rackets and smal-time drug sales. This observation 
strongly supports the second finding, because if ilegitimacy is positively correlated with 
the costs of teritorial control, then one would expect a resource-poor organization to 
succed only if it engaged in a strategy of legitimation, whereas a resource-rich 
organization could succed with merely a strategy of ilegitimacy-avoidance. This 
dynamic can be sen by looking a litle more closely at what happened with some of the 
gangs and militias in Caicedo La Siera. 
3.2. Caicedo La Siera: From Invasions to Statelets 
3.2.1. The Invasions Bring Crime, Violence, and Gangs 
Danilo?s family moved to Vila Turbay, in the uppermost part of Comuna 8, 
around 1984, when he was about 14 years old.
28
 The area was stil mostly unoccupied. 
?The roads weren?t paved, there was no running water, there was no sewage. ? It was 
like farmland. There were very few houses ?.?
29
 Of the houses that were there, some had 
been built ilegaly, in a proces Colombians cal an ?invasion,? in which people would 
arive from another part of the city or from outside of the city, find unoccupied land, and 
build shacks or houses; sometimes the police would arive (perhaps caled in by annoyed 
residents of nearby barios) to chase away the squaters and raze their homes, and 
                                                
28
 ?Danilo? is not his real name. He and several other residents I interviewed consented to my using their 
real names for this study, since, they said, their backgrounds or their current views are already known in the 
comunity. Nonetheless, as a precaution, I have decided to uses aliases for those whose past participation 
in illicit activity or curent position in comunity leadership might, in the future, put them in a vulnerable 
position with the armed actors who have recently returned to the area. 
29
 ?Las v?as estaban sin pavimentar, no hab?a acueducto, no hab?a alcantarilado, ? eso era como una finca. 
Eran muy pocas casas ?.? ?Danilo,? personal comunication (Interview No. 10), 30 March 209. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 100 
sometimes they wouldn?t. If enough people invaded and managed to stay put, the ?pirate 
bario? might eventualy be recognized by the city as a legal bario and get its own 
Community Action Board (JAC, Junta de Aci?n Comunal) and city-budget money. But 
before that point (and sometimes even after), the quality of life was usualy poor, since in 
most instances the city could not legaly provide services to setlements it did not 
officialy recognize (although some leaders of the squater communities have managed, 
with hard work, to persuade someone in the government to help). Like many others in 
Vila Turbay, Danilo?s family, originaly from Envigado just south of Medel?n, had lived 
in the coffe zone south of Antioquia before moving to Vila Turbay and buying a smal 
house that an earlier squater had built. 
Even though there were few public services in the area, at least there were no 
violent gangs. ?At the time, Medel?n ? was stil very healthy. There was peace.?
30
 But 
the population of this sector was growing rapidly, and 1984 was the year Pablo Escobar 
went into hiding and declared war against the Colombian state ? and the year before 
Medel?n?s homicide rate would begin its relentles climb. The peace in Danilo?s 
neighborhood would not last long. 
As more and more people invaded the uppermost sectors of Comuna 8, and the 
resulting overcrowding began to put more and more social and economic presures on 
residents, smal stret gangs or crews started to form. Most started out as nothing more 
than a group of young people hanging out with their neighbors, in harmles, loose 
gatherings caled parches in the local slang. But disputes with neighboring parches would 
sometimes crystalize one group?s identity and create incentives for more disciplined self-
                                                
30
 ?En ese momento Medel?n ? era muy sano. Hab?a paz.? ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 101 
defense, and if a slightly more closed, more formalized grouping resulted, it would 
become a crew or stret gang (described as a gallada, a pandila, or a combo, depending 
on the speaker). Some of these grew, some of them erged together, and some of them 
(caled either combos or bandas, again depending on the speaker) turned to crime. In 
Danilo?s neighborhood, ?there was a combo up here, there was another one down there. 
Sometimes it was block to block, [but] they were like organizing themselves. It was a 
phenomenon that grew very fast.?
31
 Danilo himself would soon become part of that 
phenomenon. 
Danilo?s arival in Vila Turbay at the beginning of the study period coincided not 
only with the rapid population growth in Caicedo La Siera, which encouraged the 
formation of crews and stret gangs, and not only with the opportunities aforded by the 
rise of narcotraficking citywide, which encouraged the formation of criminal gangs, but 
also with a new phase in peace negotiations betwen guerilas and the national 
government, which within a few years would lead to the formation of guerila militias. 
As discussed earlier, it was during the peace talks in 1984 that the M-19 guerilas set up 
their peace camps in several barios throughout the city. One was near Caicedo La Siera, 
in bario Vilatina, just on the other side of La Castro crek from Las Estancias and Vila 
Liliam. It was at the Vilatina peace camp, which was shut down in 1985, that Danilo and 
others who would later become members of stret gangs, crime rings, militias, and 
paramilitaries got the weapons and military training that, over the decade and a half that 
followed, not only helped them defend their families and neighbors against criminals and 
their turf against neighboring gangs, but also enabled them to turn minor disputes into 
                                                
31
 ?[Aqu?] hab?a un combo, abajo hab?a otro. A veces era entre cuadras, se fueron como organizando. Eso 
fue un fen?meno que creci? muy r?pido.? ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 102 
violent conflicts, and major conflicts into al-out war; in other words, the peace camps ? 
in combination with the growth of narcotraficking and the urban population and in the 
context of a formerly rural civil war beginning to afect the cities ? were the primary 
drivers of the change in Medel?n from mostly interpersonal-communal violence to the 
more high-profile collective violence (or, to be more precise, they enabled people to add 
collective violence to the communal violence that was already present and that also 
increased during the same period). By the late 1980s, people were geting kiled over 
disputes about teritory, money, women, pride, drugs, slights, revenge, expensive shoes, 
and control over drug houses and brothels. Sometimes it wasn?t enough to kil a single 
individual; sometimes the whole family would be kiled. Frequent shootings led to many 
deaths by stray bullets. Many families were forced to move away. 
Like many other residents interviewed for this study, Danilo atributed the rise in 
violence in Caicedo La Siera to the formal economy?s weak job market, the easy money 
and high-prestige lifestyle aforded by the ilicit economy, and, at the root of both, the 
absence of the state: 
I?d say that maybe [it was] a mater of employment or, since a lot of 
people are from other places ?, a lack of opportunities. Wel, and also 
that, back then, it was the time of narcotraficking ? and [it was] 
becoming like a busines, like a lifestyle. ? This was happening because 
there was no state presence there ?.
32
 
In fact, the state did have a presence in Caicedo La Siera, but as several longtime 
residents pointed out, it was limited to schools, health clinics, the JACs, and the 
                                                
32
 ?Digamos que la cuesti?n de pronto [es] de empleo, o como mucha gente son de otras partes ? la falta 
de oportunidades. Y tambi?n pues que en ese tiempo estaba lo del narcotr?fico ? y eso se va 
convirtiendose como en un negocio, como una forma de vida.? ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 103 
occasional police post; it was otherwise absent.
33
 Nevertheles, it would be more acurate 
to say that the state?s absence was les a cause of the rise in violence than a failure to 
prevent that rise. 
Around 1987, a retired police officer known as ?Desiderio? took advantage of this 
absence ? he knew the police were focused far more on the ilicit activity taking place in 
the northern part of the city ? and got some people together to form a crime ring that 
operated in upper Caicedo La Siera near the border with coregimiento Santa Elena. The 
Desiderio gang robbed buses ? sometimes going as far as seting up checkpoints ? and 
eventualy became enough of a nuisance to people pasing through the community to 
draw the atention and intervention of the police. Other smal gangs began emerging 
around the same time: One in the Santa Luc?a sector, along Santa Elena crek in Las 
Estancias. Another in an area just north of Santa Luc?a caled Las Mirlas. A man people 
caled ?Gamuza? formed a gang in a part of Lower Vila Liliam caled La Arenera. Jairo 
Alberto Ospina Olaya, who became known as ?Alberto Ca?ada,? created a gang that came 
to be known as La Ca?ada and ended up controlling most of Lower Vila Liliam, 
contesting parts of Las Estancias, and operating in other parts of the city. And a man 
named Hugo formed the November 6 and 7 Militia (M-6&7: Milicias 6 y 7 de 
Noviembre), which ended up controlling La Siera, Vila Turbay, Upper Vila Liliam, and 
nearby areas in Comuna 8 at the border of corregimiento Santa Elena. Both the M-19 and 
the National Liberation Army (ELN: Ej?rcito de Liberaci?n Nacional) formed guerila 
militias in areas just outside of Caicedo La Siera: M-19 in and around Vilatina, ELN in 
Ocho de Marzo, the uppermost bario of Comuna 9, across the crek from Las Estancias 
                                                
33
 C?sar Mendoza Gonz?lez, personal comunication (Interview No. 6), 17 March 2009; ?Mama Luz? 
Interview No. 1. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 104 
and near the bario that Pablo Escobar had built during his ?Medel?n without Slums? 
political campaign (se ? 2.1.3). The M-6&7 militia clearly had leftist sympathies: its 
name celebrated the date of the M-19?s atack on the Palace of Justice in 1985, an event 
any Vilatina peace camper surely would have known a lot about. But while it did have 
friendly relations with the guerila militias in the area, M-6&7 was not directly 
controlled by any of them. (It was M-6&7 that Danilo joined, but he did not go into much 
detail about his life as a militant.)
34
 
These are only the groups that eventualy rose to prominence in and around 
Caicedo La Siera. Interspersed among them were dozens of other smaler crews, crime 
rings, asasins-for-hire, drug dealers, poseurs, and wanna-be?s, many of whom were 
involved in turf batles and stret fights, and many of whom were eventualy defeated or 
absorbed by stronger adversaries who had either the backing of elements of the Medel?n 
Cartel or the backing of the people of their community. The gangs that ended up 
dominating Caicedo La Siera were La Ca?ada and M-6&7, and the story of their rise to 
dominance is instructive with respect to questions about violence, control, governance, 
and legitimacy. 
3.2.2. The Rise of M-6&7 
Of the two, the November 6 and 7 Militia (M-6&7) ? refered to informaly in 
the area as los de arriba, which translates as those from above or those people up there, 
                                                
34
 ?Danilo,? Interview No. 10; ?Mama Luz,? Interview No. 1; ?Comandante Fernando,? interview ith 
Arleison Arcos Rivas, ?Ciudadan?a Armada: Aportes a la Interpretaci?n de Procesos de Defensa y 
Aseguramiento Comunitario en Medell?n: El Caso de las Milicias Populares? [?Armed Citizenry: 
Contributions to the Interpretation of Comunity Protection and Defense Proceses in Medel?n: The Case 
of the People?s Militias?] (Master?s thesis, Universidad de Antioquia, 2005). 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 105 
refering to its location in the upper part of Caicedo La Siera, and more recently refered 
to, retrospectively, as the La Siera gang (as a consequence of a documentary about them; 
se ? 5.2) ? is generaly considered to have been les of a gang and more of a milicia 
popular, which is alternately translated as popular militia and people?s militia. 
Ideologicaly, the M-6&7 militia was aligned with the guerilas, and many of its 
members had participated in the peace camps (se ? 3.1.1), where they were indoctrinated 
in revolutionary ideology and antiestablishment propaganda. However, they also learned 
how to do social work and community service, and helped set up neighborhood-watch 
programs to protect against crime. As a result, the young peace campers had already won 
the afection of many in their own communities, who were not generaly bothered by the 
peace campers? anti-state rhetoric (since the state was not doing much for them in any 
event) but who were very much bothered by the drug use, the prostitution, the robberies, 
the rapes, and the murders that were taking place near their homes. When some of the 
peace campers later took up arms to take a more direct role in combating crime and gang 
violence, the community was ready to welcome and support them. ?They came in doing 
social work. They also went in looking at how people could stop geting robbed,? Danilo 
said, adding that ?the first ones who went in were people from the bario itself.?
35
 
In a 1999 interview, Hugo, the founder of M-6&7, explained the guerilas? 
strategy. After the peace camps shut down, he said, ?we started to look inside the 
structure we had in our organization.? 
In the organization there were groups of people who were sympathizers to 
the organization, who weren?t militants but who could be channeled into 
doing it, into developing this kind of semiclandestine structure, into taking 
                                                
35
 ?Ellos entraron haciendo trabajo social. Entraron tambi?n mirando como que la gente no se dejara robar 
? Incluso ? los primeros que entraron ah?, era gente misma del bario.? ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 106 
advantage of that kind of semiclandestine structures [sic] to launch the 
November 6 and 7 Militia project. So the militia project is [sic] born with 
people who were operating in the whole sector, from Caicedo on up ? but 
it was a group that was very limited, let?s say, with ten or twelve people 
from each one of those barios. And then the ned arose to strengthen the 
relationship with those communities and expand the militias? base so that 
it wouldn?t be such a select group, but that we would cal for the people to 
participate as wel. ? We started out working basicaly with adults, 
centering on activities to solve the need for security that was very much 
felt by the community. Then [targeting] some drug markets, some criminal 
gangs. That wasn?t what fundamentaly brought us together. We also tried 
to link it to some actions, even if what was being done was a day of 
protest ? the political side of the busines was geting introduced to 
them.
36
 
The militia went far beyond propaganda, social work, and neighborhood 
protection. It engaged in an active and extensive ?social cleansing? campaign as wel, 
targeting gangsters, robbers, rapists, child molesters, prostitutes, stret-corner drug users, 
and others who were harming the community or ofending its conservative sensibilities. 
At first the militias would warn the offenders to stop the offending activity; then they 
would threaten them, offering them the choice of stopping the activity or leaving the 
bario; and then they would kil the ofenders in a public enough way that the execution 
would serve as both punishment and warning. Even though they would later overeach in 
this regard ? they would end up targeting unfaithful spouses, homosexuals, curfew 
                                                
36
 ?? nosotros comenzamos a explorar dentro de la estructura que ten?amos. En la organizaci?n hab?a unos 
grupos de gente simpatizante de la organizaci?n que no eran militantes pero que se pod?an canalizar para 
hacer, desarollar ese tipo de estructuras semiclandestinas [sic] y se aprovechan ese tipo de estructuras 
semiclandestinas para lanzar el proyecto de milicias 6 y 7 de noviembre. Entonces el proyecto de milicias 
nace [sic] con gente que operaba en todo el sector, desde Caicedo hasta ariba ? pero era un grupo 
digamos muy reducido, eran diez o doce personas de cada uno de esos barrios. Y luego se plante? la 
necesidad de afianzar la relaci?n con esas comunidades y ampliar la base de las milicias, que esto no sea un 
grupo como tan selecto sino que convoquemos la gente tambi?n a que participe. ? Iniciamos, arancamos 
trabajando con gente adulta fundamentalmente, y en torno a actividades a resolver necesidades de 
seguridad, muy sentidas de la comunidad. Entonces, unas plazas de vicio, unas bandas de delincuentes. No 
fundamentalmente era eso lo que nos convocaba. Y adicionalmente intent?bamos articular algunas aciones 
siquiera que se hac?an a las jornadas de protesta ? se le iba introduciendo pues ese matiz pol?tico al 
asunto.? Punctuation edited for clarity. ?Comandante Hugo,? interview with Arleison Arcos Rivas, 
?Ciudadan?a Armada,? 205. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 107 
breakers, stret children, bad fathers, and other people who, acording to the conservative 
standards of the community, were undesirable or ?disposable? ? the militia?s social 
cleansing campaign was, for the first several years, broadly supported by those in the 
community who had been longing for a return to order. By 1990, the M-6&7 had 
managed to oust most of the groups and individuals who had been creating disorder, to 
establish teritorial control, and to return some degre of social control to La Siera 
(about 50 city blocks in size) and Vila Turbay (about 30 city blocks); it occasionaly 
exercised control over Upper Vila Liliam (about 50 city blocks) as wel. It had betwen 
200 and 300 troops under its command. 
Not everybody was happy with M-6&7?s control. Some had political objections to 
its leftist ideology, but I found only one reference to this being a factor in anyone?s 
decision to leave the bario (and in that case the interviewe semed to suggest the 
ideological diference centered primarily on the use of drugs: several young men left M-
6&7?s teritory and joined the La Ca?ada gang because of M-6&7?s drug-use 
prohibitions).
37
 Instead, what I found was that the militia?s opponents were mainly the 
people who had been threatened by the militias and the families and friends of the 
victims. Many gang members (and their families) who were forced to leave the area 
controlled by los de arriba ended up setling not too far away, down in Las Estancias and 
Lower Vila Liliam, where some tried simply to blend in and stay out of the conflict and 
others (like the drug users just cited) actualy joined or sought protection from los de 
abajo, the gang that controlled the areas below: La Ca?ada. 
                                                
37
 ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 108 
3.2.3. The Rise of La Ca?ada 
The Spanish word ca?ada [cahn-YAH-da] means ?stream? and in Caicedo La 
Siera locals use that word to refer to Castro Crek (Quebrada La Castro), which flows 
down Sugarloaf Hil (Cero Pan de Azucar) from corregimiento Santa Elena and into 
Santa Elena Crek. Castro Crek serves as the lower boundary of Caicedo La Siera (i.e. 
it separates Vila Liliam and Las Estancias above from Vilatina and San Antonio below). 
La Ca?ada is also the nickname residents gave to the area near the stream (primarily in 
Lower Vila Liliam). Since that is the area where the gang founded by Jairo Alberto 
Ospina Olaya emerged, the gang picked it up as its own name and its founder picked its 
up as his alias, ?Alberto Ca?ada.? 
Unlike M-6&7, Alberto and La Ca?ada did not have political aspirations. While 
they certainly engaged in a degre of community protection, they were primarily a 
criminal organization that controlled the local drug markets (caled plazas de vicio in 
Spanish) and smuggling routes, did some contract work (such as asasinations and 
smuggling) for Medel?n?s narcotrafickers and other customers, and made enough money 
in the ilicit economy to pay of the police and the army to just ?let them work.?
38
 La 
Ca?ada later became a silent partner of the state in its fight against the militias. It ended 
up controlling, either directly or through proxy stret crews, a number of neighborhoods 
in lower Caicedo La Siera (plus others in surrounding areas), each about 20 city blocks 
in size: Lower Vila Liliam, including a section caled La Arenera; Santa Luc?a, in upper 
Las Estancias along the crek; and Las Mirlas in upper Las Estancias just north of Santa 
Luc?a. 
                                                
38
 Ibid. 
Lamb, Robert D. | Chapter 3. Crime and the Emergence of Guerila Militias: 1984-1992 109 
Up to around 1991 or 1992, M-6&7 and La Ca?ada consolidated a great deal of 
teritorial control over Caicedo La Siera, and each within its own borders tried to 
maintain a degre of community support: M-6&7 primarily through a strategy of 
legitimation, La Ca?ada primarily by virtue of the security and benefits it provided to 
friendly locals and the wealth and power it was able to acumulate through the ilicit 
economy (and, later, the backing of the state). As long as both generaly protected and 
respected members of the community ? as long as they did not revert to the pervasively 
abusive and dangerous behavior of the previous generation of gangs and criminals ? the 
community did not oppose them in any serious way (and those who tried, died, or were 
forced to leave). Many people even actively supported, rooted for, or verbaly defended 
the side that controlled their own bario, a reflection of the local culture?s strong in-group 
mentality. Both would remain in control of their respective teritories ? with occasional 
disputes betwen them and against other, leser gangs in the area ? throughout the 
1990s. By the end of that decade, however, La Ca?ada would enjoy the backing of state 
actors, narcotrafickers, and paramilitaries in opposition to M-6&7, enabling it to stay in 
control of its teritory despite its failure to ever realy legitimize itself to the community. 
M-6&7 would be another story entirely: it would delegitimize and then 
ilegitimize itself to the people in its own teritory and lose their backing, while 
simultaneously fighting opposition from La Ca?ada, the state, the paramilitaries, the 
guerilas ? and its own members. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 110 
Chapter 4. Gangs and the Corruption of Guerila Militias: 1992-1998 
In the last chapter, the poor hilside barios on Medel?n?s periphery had started 
out as nearly ?ungoverned? but relatively stable areas. There was violence and crime, but 
they were at a level that most residents could tolerate and addres within existing social 
structures. Evidence suggests that their tolerance was due to the strength of community?s 
social institutions ? there were usualy a church, some schools, reasonably strong 
families, and nosey neighbors ? and those social institutions had the ability to maintain a 
respectable degre of social control, despite the scant presence of state institutions (such 
as police) or other armed actors wiling and able to maintain order. In short, control was 
maintained by the community. But an influx of outsiders into those communities put 
presure on their social institutions, and by the end of the 1980s, one might say that social 
control lost ground to teritorial control under the law of the jungle: the criminals, crews, 
and gangs who emerged amid the overcrowding fought over micro-teritories and local 
aces to global ilicit markets, and they clearly had the capability to prevent at least 
some forms of outside governance and self-governance. But as rulers of the bario they 
were completely ilegitimate in the eyes of the bario communities: the insecurity that 
resulted from their methods of micro-teritorial control created a demand for a return to 
social control. 
In upper Caicedo La Siera, that demand was met by the November 6 and 7 
Militia (M-6&7), which, at least initialy, enjoyed some support from most of the 
community based on a relationship of legitimacy. In the terminology of the framework 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 111 
used in this study (se Appendix C), M-6&7 made their rules known (transparent), 
demonstrated their capability to enforce them (credible), became advocates for the 
communities? own moral values (justifiable, or even admirable), could be and often were 
approached by the community to have conflicts and disagrements resolved (acesible), 
and by al acounts treated most community members with dignity (respectful). (The 
bario?s low-status members ? the so-caled disposable people ? certainly could said to 
have been treated inequitably, but I could find no information to support any claims about 
how equitably other members of the community felt the militants had treated them.) 
In lower Caicedo La Siera, the dynamic was diferent: no militia emerged to take 
control, either because they never tried to take control of some blocks or because where 
they tried they failed because the gangs that were there were too wel armed. The most 
significant of these wel armed gangs to emerge was La Ca?ada, whose strength derived 
from the economic resources it was able to bring to the fight. La Ca?ada did not emerge 
with any political or social program; rather, it grew based on its ability to win economic 
resources from the ilicit economy, primarily through contracts from, and drug sales for, 
the Medel?n Cartel. La Ca?ada did enjoy some degre of support from the community, 
which derived from the fact that they did protect favored residents against rival gangs and 
the militants from above, and took some other steps to keep the community happy. ?If the 
people from the community are happy, you can manage things beter,? one resident said, 
since it gives people a reason to act as informants; plus, in a community where a lot of 
people have already suffered, he added, further harming them risked turning them into 
sapos (frogs), or informants, for La Ca?ada?s rivals.
1
 La Ca?ada?s control, therefore, 
                                                
1
 ?Si la gente de la comunidad est? contenta usted maneja mejor las cosas.? ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 112 
derived primarily from its military strength (it could win acquiescence through coercion) 
and its economic strength (it could win support through barter), and only secondarily 
from any relationships of legitimacy: Its rules were generaly known (transparent) and 
usualy enforced (credible), and several residents mentioned that the gangsters were 
respectful and acesible (e.g. they could be asked to resolve conflicts in the community). 
And La Ca?ada did not explicitly work to embody or promote community values (for 
example, they permited drug use, which most in the conservative communities could not 
approve of), nor did it enforce its own rules equitably (people they didn?t like did not 
generaly enjoy their protection). In short, rather than maintaining teritorial control 
primarily through a strategy of legitimation (actively making oneself worthy of the 
community?s support), they semed implicitly to be engaging in a strategy of 
ilegitimacy-avoidance (working merely to avoid triggering a moral obligation of 
opposition). 
The fact that both groups could maintain control shows that, at least in the short 
term, having legitimacy acording to the community is a useful, but neither a necesary 
nor a sufficient, condition to winning or maintaining teritorial control, as long as the 
group in control stays within certain bounds of behavior with respect to that community. 
Legitimacy lowers the costs of teritorial control, which is why the relatively resource-
poor M-6&7 had to engage in a legitimation strategy, while the relatively resource-rich 
La Ca?ada could get away with weak legitimation eforts and focus mainly on 
ilegitimacy-avoidance. 
As the present chapter wil demonstrate, the resource-poor militants in both 
Caicedo La Siera and Medel?n as a whole would end up not only failing in their 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 113 
legitimation strategies but also actualy ilegitimizing themselves by failing to maintain 
stability in their teritories and as a consequence would ultimately lose those teritories. 
4.1. Medel?n: A War of Al against Al 
4.1.1. The Militias Start to Lose Control 
A man was outside in the stret near his house in northern Medel?n, so drunk that 
he was chatering, acording to one witnes, like a parot. His relatives, giving up on 
trying to get him home by themselves, went to the local militia commander to ask for his 
help in geting him back inside the house. The commander tels the story: 
This man was a father in the family, an honorable person who got out of 
control only when he drank too much. As the comander of that zone, I 
remember having sent two young guys to help the family get the drunk 
man inside. After a few minutes, some blasts were heard: the guys kiled 
the man in the coldest of blood right in the middle of the crying family. 
The explanation the guys gave was that the drunk had insulted them. 
Despite al the metings we had with the neighborhood, the expulsion of 
the two young guys, we were never able to overcome the il-wil within 
the community on that stret.
2
 
The leftist militias that had taken control and brought order to some of Medel?n?s 
most violent barios in the late 1980s and early 1990s ? primarily in the Northeastern 
zone, but also in the Northwestern, downtown, East Central, and West Central areas ? 
began experiencing the pain of governing once they had established a monopoly over the 
                                                
2
 ?Este se?or era un padre de familia, una persona honorable, que s?lo cuando se tomaba unos tragos se 
descontrolaba. Como jefe de zona recuerdo haber enviado dos muchachos para ayudar a la familia a entrar 
al borachito. A los pocos minutos se escucharon varias detonaciones: los muchachos mataron al boracho 
con la mayor de la sangre fr?a en medio del lanto de su familia; la explicaci?n que dieron los muchachos 
fue que el borracho los insult?. A pesar de las reuniones que se hicieron con el vecindario, la expulsi?n de 
los muchachos de la organizaci?n, no se logr? subsanar el malestar dentro de la comunidad de la cuadra.? 
Gilberto Medina Franco, ?Una Historia de las Milicias de Medel?n (Historia Sin Fin),? 52. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 114 
use of force in their tiny statelets. Incidents such as that involving the drunk became 
common enough to begin driving wedges betwen the militias and the communities that 
had once welcomed them. Thre things corrupted the militias? ability to realize their 
original vision for the city: internal divisions, the lure of easy money, and overeaching. 
First, as time went by and the militias grew in both size and local power, internal 
disagrements grew into internal divisions, and soon the militias were generating splinter 
groups that sometimes chalenged their parent organizations violently. This was the case 
with the Popular Militias of the People and for the People (MPP: Las Milicias Populares 
del Pueblo y para el Pueblo), which controlled and governed much of the city?s Northeast 
and was expanding westward and southward by 1991. The organization had been wel 
governed internaly, with a central command coordinating and guiding decentralized 
units, some governing areas as smal as a city block, and the MPP even opened a militia 
school in January 1991 that taught urban warfare and political doctrine. Acording to one 
first-person acount, however, as MPP?s numbers grew, midlevel commanders began 
launching their own batles to win community recognition for themselves; 
they worked hard to score the greatest number of conquests over the 
chicks in the bario. Personal friction with other leaders became frequent, 
each making a verbal display of their exploits, al of which was bringing 
about a tribal style of resolving internal diferences. That?s how MPP got 
broken up into cliques and afinity groups that had les and les in 
common.
3
 
In the face of these internal divisions and pety disputes, some of the most radical of the 
MPP?s commanders took a hard line, both politicaly and teritorialy: they criticized as 
traitors moderates and others on the left who would consider negotiating peace, violently 
                                                
3
 ?Cada uno de esos jefes lideraba su propia batala por ganarse el reconocimiento de las comunidades; se 
esforzaba en conquistar el mayor n?mero de sardinas del bario. Los roces personales con otros l?deres se 
volvieron frecuentes, cada uno hac?a un despliegue verbal de sus haza?as, todo lo cual fue imponiendo un 
estilo tribal para resolver las diferencias internas. As? quedaron fracturadas estas MPP en grupos de 
afectos y simpat?as que cada vez ten?an menos en com?n.? Ibid., 49. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 115 
atacked any mainstream politician who tried to campaign in or near their teritory, and in 
the end had to turn to former gangsters who knew how to use weapons to help them keep 
control of the teritory they stil held. This hard-line response involved a great deal of 
violence, and as more and more residents of the teritories they controlled were 
victimized ? they or their families were afected directly by the disputes and indirecty 
by the disorder those disputes brought about ? the MPP increasingly alienated the 
communities they operated in. The alienation opened the door to splinter groups and 
chalengers endeavoring to win the community support the MPP had lost. The People?s 
Militias of the Aburr? Valey (MPVA: Milicias Populares del Vale de Aburr?) emerged 
in the Northeast during this period and operated by working, as nonviolently as possible, 
with and through community groups that had been alienated by the MPP. In the 
Northwestern zone, some midlevel MPP commanders joined up with a group of Popular 
Liberation Army (EPL: Ej?rcito Popular de Liberaci?n) guerilas to form a splinter 
group caled the Workers Command (COB: Comandos Obreros). Other splinter groups 
emerged as wel, and the batles for teritory returned, bario by bario, just as during the 
militias? wars against crews and gangs.
4
 
Second, the broader political vision that most of the militias originaly had would 
have required more resources than were imediately available from voluntary 
community sources, and some militants responsible for raising resources began playing 
acounting tricks, seking ?voluntary? donations more coercively, or hiring their units 
                                                
4
 Jos? Ricardo Barero Tapias, ?Las Milicias Populares y el ELN Como Casos de Estudio?; Gilberto 
Medina Franco, ?Una Historia de las ilicias de Medel?n (Historia Sin Fin),? 43. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 116 
out for private security services.
5
 Moreover, acording to Tom?s Ernesto Concha Sanz, 
who got to know the militants during the later peace negotiations, ?Although they had 
altruistic atitudes, such as defending the interests of the community and protecting them 
against atacks from crooks, it was obvious that the low level of culture [and] the almost 
nonexistent political education [of al but a few of their members] ? was going to 
determine that they would progresively fal back into atitudes similar to those practiced 
by the criminals they said they were fighting.?
6
 In many cases, those atitudes included 
rationalizing their involvement in the drug trade and other ilicit busineses as a way to 
finance their operations; with time, their visions of the greater good were lost amid the 
high-prestige lifestyle aforded them by such easy money. Acording to a former militia 
leader, it was this ambition to expand into a broader political movement that led to their 
downfal: 
The militias, at first, began to defend their bario and gave their lives for 
their family and the bario; in a second stage, some militants and directors 
began to manage high revenues and other kinds of busineses. Then, 
economic interests were set up around the militias. A lot of outsiders came 
in with busines proposals for narcotraficking, crime, ? a barn, a 
supermarket, atracted by the security the militias offered; then the 
disintegration problem began: many of the decisions the militias were 
making were for economic interests. I remember cases of directors 
responsible for economic projects who pilfered piles of money. The 
corruption was huge.
7
 
                                                
5
 Jos? Ricardo Barero Tapias, ?Las Milicias Populares y el ELN Como Casos de Estudio?; Tom?s Ernesto 
Concha Sanz, ?El Caso de la Milicias o Acuerdo de Santa Elena,? [?The Case of the Militias, or the Santa 
Elena Acords,?] in Experiencias de Intervenci?n en el Conflicto Urbano, ed. Ana Daza A. (Medel?n, 
Colombia: Alcald?a de Medel?n, 201), 213. 
6
 ?Aunque ten?an actitudes altruistas como la de defender los intereses de la comunidady protegerla de los 
ataques de los ?pilos?, era obvio que su bajo nivel cultural, su casi nula formaci?n pol?tica ? iba a 
determinar que, progresivamente, cayeran en actitudes similares a loas que ejerc?an los delincuentes que 
dec?an combatir.? Ibid. 
7
 ?Las milicias, en un primer momento, empezaron a defender su bario y daban la vida por su familia y el 
bario; en una segunda etapa algunos milicianos y dirigentes empezaron a manejar ingresos altos y otro tipo 
de negocios. Entonces, alrededor de las milicias se constituyeron intereses econ?micos. Mucha gente de 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 117 
Moreover, during Pablo Escobar?s war against the state, he began to use revolutionary 
and nationalistic rhetoric to win popular support for his campaign to outlaw the 
extradition of Colombian citizens. ?This [rhetoric] blurred the militia/criminal 
dichotomy, as [Escobar] started to tolerate and even encourage the activity of the militias 
in some parts of the city, and to funnel arms and other resources to some of them.?
8
 
Finaly, the militias? ?social cleansing? campaigns started to grate on the 
communities. At first welcomed as the only available mechanism for maintaining social 
order, the kiling of people for dangerous (and, later, merely disapproved-of) behaviors 
meant that, over time, more and more people in the communities were feling the loss of 
a family member or friend to social cleansing. A corner drug dealer might not be 
somebody any community would necesarily welcome in the neighborhood, but the drug 
dealer is often a family member of someone in that very community and, if the 
community is poor, he might actualy be that family?s only source of income ? so his 
death would sting twice. Over time, more and more families felt that double sting and 
wanted it to end. Combined with the lure of easy money from narcotrafickers, and the 
radicalization and fragmentation brought about by the militias? growing pains, this 
overeaching during what Arcos Rivas cals the ?late period? of the militias resulted in a 
?greying of the lines betwen gangs and militias?: ?This period, betwen 1991 and 1994, 
was characterized by the resort to summary executions, arbitrary rule, the community?s 
                                                                                                                                            
afuera ven?a a proponer negocios, el narcotr?fico, la delincuencia, instalaban que un granero o un 
supermercado atra?dos por la seguridad que brindaba la milicia, entonces, empez? el problema de 
descomposici?n, muchas de las decisiones que tomaban las milicias eran por intereses econ?micos; me toc? 
casos de dirigentes responsables de proyectos econ?micos que despilfarraban un mont?n de dinero. La 
corupci?n era grande.? Quoted in Jos? Ricardo Barero Tapias, ?Las Milicias Populares y el ELN Como 
Casos de Estudio,? 217. 
8
 Francisco Guti?rez San?n and Ana Mar?a Jaramilo, ?Crime, (Counter-)Insurgency and the Privatization 
of Security.? 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 118 
ignorance about civil and military decisions, [and] the resort to murder as a tool for 
resolving conflicts within the group.?
9
 Those behaviors caled into question their 
legitimacy acording to the very communities whose legitimacy they had explicitly 
solicited when they had first emerged. 
Moreover, it was the most violent period in Medel?n?s history. It was in 1992 that 
Pablo Escobar escaped from prison and waged his final war against the Colombian state, 
enlisting many of his asasin gangs to kil police officers and other officials and 
funneling resources to the militias who were also in a war against the state. It was also the 
year his cartel asociates started turning on him in response to the torture, murder, and 
mutilation of four of his most powerful busines partners in July 1992 (the surviving 
asociates would form the anti-Escobar death squad, the Pepes, in February 1993). With 
multiple wars being fought simultaneously, plus the general lawlesnes that usualy 
acompanies urban war, December 1992 was the deadliest month in Medel?n?s history: 
the city?s Secretar?a de Gobierno reported 647 homicides in the city that month ? an 
average of nearly 21 per day. The year as a whole also experienced peaks in: suicides 
(105 self-inflicted deaths, which would not be exceded again until 1998), combat 
betwen and among al kinds of armed actors (57 separate events, which would not be 
exceded again until 2002), forced disappearances (64 people, which would only be 
                                                
9
 ?Me refiero concretamente al periodo tard?o de las Milicias Populares, que coincide con el engrisamiento 
de las bandas y milicias. Este periodo, entre 191-1994, se caracteriza por el recurso a las ejecuciones 
sumarias, la arbitrariedad, el desconocimiento de la comunidad para la toma de decisiones militares y 
c?vicas, el recurso al homicidio como instrumento de resoluci?n de conflictos al interior del grupo.? 
Arleison Arcos Rivas, ?Ciudadan?a Armada,? 108. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 119 
surpased in 1996), and masacres by paramilitary groups (5 separate events, which 
would not be surpased until 2001).
10
 
4.1.2. Coosercom and the Peace Talks that Brought War 
It was during this extremely violent period that some of the militants, including 
MPP founder Pablo Garc?a and other relatively moderate leaders, decided it was time to 
negotiate a disarmament. Even though they were in control of entire barios, even entire 
zones of the city, these leaders made a calculation that, given the resources available to 
the state, and given the constant batles with the police, gangs, and rival militias, it was 
ultimately in their own and their communities? best interests to negotiate a setlement. 
?Iregular war is not sustainable,? even les so in an urban environment where the state 
has al the resources, explained C?sar Mendoza Gonz?lez, a human rights activist who 
has followed the demobilization. ?Day after day, they were geting worn out. That?s why 
they laid down their arms.?
11
 State actors likewise had recognized that at least part of the 
problem could be resolved by reincorporating at least the most moderate members of the 
militias into civilian life. In 1991, national- and department-level state actors made secret 
contacts with certain militia leaders to test the possibility of a negotiated agrement, then 
started creating the legal framework that would be needed to begin talks in public. If the 
militias were not, technicaly, armed political actors but were instead crews formed to 
protect themselves against criminals, or gangs formed to engage in crime (and diferent 
                                                
10
 Suicide data from Departamento Administrativo Nacional de Estad?stica (DANE), ?Estad?sticas Vitales,? 
http:/ww.dane.gov.co (acesed 2 August 2009); combat events and masacres by paramilitaries from 
Centro de Recursos para el An?lisis de Conflictos (CERAC), Bogot?, Colombia; and forced disapearances 
from Instituto Popular de Capacitaci?n (IPC), Medel?n, Colombia. 
11
 C?sar Mendoza Gonz?lez, Interview No. 6. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 120 
militias had characteristics of both), then there was no legal framework for negotiating 
with them: when the country?s Constitution was revised in 1991, it included clauses that 
were intended to help negotiate peace with political armed groups that were chalenging 
the state. That problem was solved by administrative fiat: the state declared the militias to 
be political actors, and negotiations began in February 1994 in Santa Elena, a town on the 
rural outskirts of the city.
12
 Homicides dropped 23 percent that month from a month 
before.
13
 
Present at the negotiations were representatives of MPP and MPVA on one side 
and representatives of the national, departmental, and local governments on the other, 
asisted by the good ofices of the Catholic Church and a number of non-governmental 
organizations (NGOs) such as Corporaci?n Region and Pastoral Social de Medel?n. But 
the negotiations themselves were a disaster: the state representatives ended up spending 
most of their energy protecting the militia representatives from each other. Soon after the 
negotiations began, acusations of treason led to internecine kilings among the militias; 
those at the negotiating table at Santa Elena were targeted both by other militia 
representatives also at Santa Elena and by the non-negotiating purists who considered al 
the negotiators to be traitors to the cause. ?Even worse, the negotiation proces weakened 
the broad social support for the militias. As soon as they abandoned their military 
activities, they forfeited the power that alowed them both to control the gangs and to 
promote their cherished moral order.?
14
 After the drop in February, when the negotiations 
                                                
12
 Tom?s Ernesto Concha Sanz, ?El Caso de la Milicias o Acuerdo de Santa Elena.? 
13
 Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden Civil, Unidad de Convivencia 
Ciudadana. 
14
 Francisco Guti?rez San?n and Ana Mar?a Jaramilo, ?Crime, (Counter-)Insurgency and the Privatization 
of Security.? 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 121 
began, homicides jumped 47 percent in March.
15
 Nonetheles, the negotiations were 
succesfully concluded after four months and the terms of what became known as the 
Santa Elena Acords were announced on 26 May 1994. But the exclusion of community 
representatives from the talks themselves, and the violence that acompanied the 
negotiations, made it inevitable that those terms would not be considered aceptable by 
the common people who lived in the barios who were the supposed beneficiaries; at the 
very least, the agrement would not enjoy the benefit of the doubt. 
The agrement did provide for a number of smal improvements in the quality of 
life in the barios the militias had controlled: infrastructure and other projects that 
generated a litle bit of employment, health and educational investments, recreational 
programs, help in improving and geting legal title to housing, and so on. But these 
eforts were underfunded and poorly managed. 
The most visible product of the negotiations, therefore, was the Community 
Services and Security Cooperative, known as Coosercom (La Cooperativa de Seguridad y 
Servicios a la Comunidad), an organization made up of several hundred demobilized 
militants who were to be paid by the state to maintain order in the barios under their 
jurisdiction and provide inteligence to the police regarding crime in those barios. 
Coosercom, in other words, was an organization of demobilized militants that was funded 
and legitimized by the state to continue doing what the militants had already been doing 
for several years (maintaining order), only now as a de facto appendage to the state 
security forces. In return for their demobilization, the cooperative members were given 
                                                
15
 Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden Civil, Unidad de Convivencia 
Ciudadana. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 122 
amnesty for al crimes, political or not, and received a monthly salary and five buildings 
to use as headquarters for their work in 32 barios.
16
 
With the il-wil generated (during the negotiations) betwen and among the 
militants who were negotiating, betwen them and the non-negotiating militants, and 
betwen al of the militants and many of the community members that had once 
supported them, there was no way this agrement would succed. It hardly lasted two 
months: the internecine kilings that acompanied the acusations of treason continued as 
they had during the negotiations, and almost as soon as the agrement was announced the 
government began receiving credible acusations of involvement in crimes and abuses by 
Coosercom members, who now had greater military capacity and state support, but very 
litle real oversight. 
And then, in July 1994, Pablo Garc?a, the founder of MPP and one of the lead 
negotiators of the Santa Elena Acords, was asasinated. The leader of the MPVA was 
arested for the crime. A new militia-on-militia war broke out, and by the time 
Coosercom was formaly disolved a few years later, al of its leaders and hundreds of its 
members had been kiled, and the organization was nothing more than a failed 
experiment: the amnesties agred to during the negotiations had been declared ilegal 
within the judicial system, and the legitimate political movement that the militants had 
hoped they would be able to develop and lead never got of the ground. Their political 
candidates did very poorly at election time, even within the communities they had once 
protected: they simply ?did not recognize the enormous gap betwen their imagined 
constituency, the manual workers, with their complex political culture, and their own 
                                                
16
 Francisco Guti?rez San?n and Ana Mar?a Jaramilo, ?Crime, (Counter-)Insurgency and the Privatization 
of Security?; Tom?s Ernesto Concha Sanz, ?El Caso de la Milicias o Acuerdo de Santa Elena.? 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 123 
rather obscure, revolutionary insider frame of reference.?
17
 What remained of the 
militants by mid-decade were relatively isolated vestiges of the ?political? militias, 
dedicated to their original vision, who never demobilized, some of whom continued to 
operate more or les as originaly envisioned and managed to maintain control over 
certain barios and even to expand their teritory, others of whom tried to stick with their 
original vision but overeached and failed as a result. These were surrounded by the 
?criminal? (or ?narco?) militias who were militants in name only, corrupted by the lure of 
easy money from narcotraficking and other criminal activities and litle diferent from 
the crews and gangs whose abuses they had originaly been formed to oppose. The last 
vestiges of Coosercom were dismantled in November 1996.
18
 
4.1.3. Gangs Learn a Leson from the Militia Experience 
Meanwhile, there continued to be many barios that the militias never took control 
over, those barios remaining instead in the hands of crews, crews confederated with 
larger gangs, or those larger gangs themselves, usualy with funding from 
narcotraficking or asasination contracts. A lot of those gangs were inconsiderate or 
abusive to their neighbors, able to maintain control of their barios mainly by virtue of 
their economic and military power (i.e. they were able to buy of, displace, or kil any 
significant source of opposition). But many others took a leson from the golden era of 
                                                
17
 Francisco Guti?rez San?n and Ana Mar?a Jaramilo, ?Crime, (Counter-)Insurgency and the Privatization 
of Security.? 
18
 Pablo Emilio Angarita Ca?as, H?ctor Galo, and Blanca In?s Jim?nez Zuluaga, Din?micas de Guera y 
Construci?n de Paz, 46. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 124 
the militias (pre-1991) ? that legitimacy lowers the costs of teritorial control ? and 
?were vigorously promoting communitarian causes to win the support of the population?: 
They learnt self-discipline and started to impose some basic regulations on 
criminal activity (you shal not steal in your own bario, etc.). Other 
groups ? opposed to both the gangs and the militias ? appeared, but they 
too semed to have learned that to maintain teritorial control they had to 
offer security, some kind of self-discipline, and a constructive, 
communitarian set of activities.
19
 
Stil, violence was rampant, and whatever the behavior of the gangs toward 
members of their own barios, they were undeniably abusive both to other gangs and to 
people from other barios. Henry Holgu?n (ohl-KGEN), for example, came of age 
during the gang wars of the late 1980s. He joined his local crew, La Banda de Lebr?n, 
and by the early 1990s had moved up in the ranks to become its leader. Operating in the 
Northwestern zone, Holgu?n grew Lebr?n into a gang of 180 members who, among other 
things, took contracts for narcotrafickers. After Escobar?s death in December 1993, they 
began absorbing smaler gangs and crews, and during the gang-militia wars of the early 
1990s went on the ofensive against the Milicias Bolivarianas, whose social cleansing 
campaigns were being caried out clumsily and never resulted in the community backing 
that the other militias had enjoyed, thereby opening the door to chalengers such as the 
Lebr?n gang. After defeating that militia in the Northeastern zone, Lebr?n?s war against 
other, non-militia gangs continued. 
As with most gangs who controled teritory, the people who lived in Lebr?n?s 
teritory could not venture into strets controlled by other gangs, and people from outside 
of its teritory were not alowed to enter, upon penalty of death. These gangs fought to 
                                                
19
 Francisco Guti?rez San?n and Ana Mar?a Jaramilo, ?Crime, (Counter-)Insurgency and the Privatization 
of Security.? 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 125 
win, with draconian border-control policies meant to help them keep definitive control 
over their own litle statelets: ?Those teritories became forbidden zones for the rest of 
the community. Crossing their borders could cost you your life.? Holgu?n?s story, or 
variations on it, was one of many that played out in scores of barios throughout the city 
during the 1990s. By 1995, it was estimated that there were 156 separate gangs operating 
in the city ? a number that surely was approximate (it came from a gangster-sponsored 
informal census) and that certainly changed frequently.
20
 By the end of the decade there 
were an estimated 8,000 people, most betwen the ages of 15 and 25, who were members 
of somewhere betwen 180 and 220 diferent gangs, crews, militias, and other types of 
armed groupings involved in ?criminal activities such as asasination-for-hire, 
kidnapping, extortion, bank robbery, vehicle theft, etc., whether on their own initiative or 
under contract to criminal, guerila, or self-defense organizations.?
21
 
While those figures might have been smaler in the 1990s than in the 1980s, and 
while each year after 1991 had fewer homicides than the previous, most acounts of the 
1990s nevertheles suggest that, in peoples? subjective judgments, the problem of 
violence was not geting any beter from where it had been during the earlier, deadlier 
years: the fear remained constant, and possibly grew, the longer the gang-militia, gang-
                                                
20
 ?Esos territorios se convirtieron en zonas vedadas para el resto de la comunidad. Traspasar sus fronteras 
pod?a costar la vida.? Semana, ?Paz de verdad,? [?Real peace,?] 8 May 200. 
21
 ?? actividades criminales como el sicariato, el secuestro, la extorsi?n, el asalto a bancos, el robo de 
veh?culos, etc., bien por su iniciativa o bien porque son contratados por organizaciones criminales, guerila 
y autodefensas.? Luis Fernando Duque, Antecedentes y Evoluci?n del Programa de Convivencia 
Ciudadana de Medel?n [Antecedents and Evolution of the Citizens Coexistence Program of Medel?n] 
(Medel?n, Colombia: Alcald?a de Medel?n, 200), 20. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 126 
gang, and militia-militia wars continued.
22
 By the mid-1990s people were tired of 
violence and ready for peace ? even the gangs (some of them, at least) were ready for a 
change. 
4.1.4. The City Responds to the Violence 
Peace negotiations betwen pairs of waring factions began taking place from 
community to community, some led by local parishes, others by determined community 
leaders, stil others by jailed former gangsters ? such as Holgu?n, who after defeating the 
Bolivarianas was arested and sent to Medel?n?s toughest prison, Belavista, where he 
had a change of heart and, with a couple dozen other former gangsters, succeded in 
negotiating cease fires and non-aggresion micro-pacts betwen gangs stil on the 
outside.
23
 But because there were scores, perhaps hundreds, of diferent armed actors 
fighting each other in ever-shifting aliances in Medel?n, there were probably a hundred 
micro-wars taking place at any given moment. While pairwise negotiations could help 
calm things down for a few blocks or half a bario, stabilizing the city as a whole, or at 
least large sectors of the city at a time, would require a much more systematic efort. In 
addition to the Santa Elena Acords with the militias, many other eforts were atempted 
by the public, private, and voluntary sectors, sometimes working independently, 
sometimes working together, beginning in the late 1980s and picking up speed in the 
mid-1990s. These eforts included 
                                                
22
 Marta In?s Vila Mart?nez, Luz Amparo S?nchez Medina, and Ana Mar?a Jaramilo Arbel?ez, Rostros del 
Miedo: Una Investigaci?n Sobre los Miedos Sociales Urbanos [Faces of Fear: A Study of Urban Social 
Fears] (Medel?n, Colombia: Corporaci?n Regi?n, 203). 
23
 CIDEAL, La Violencia en Medel?n y Colombia: Iniciativas Para la Soluci?n del Conflicto (1980-2004) 
[Violence in Medel?n and Colombia: Initiatives for Solving the Conflict (1980-2004)] (Medel?n, 
Colombia: CIDEAL?Pastoral Social?AECI, 205); Semana, ?Paz de verdad.? 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 127 
reinsertion proceses for amnestied members of diferent political groups; 
peace pacts betwen young gang and militia members and their reinsertion 
into social life; proceses to educate communicators in the management of 
violence isues with the support of various universities; support for the 
justice [system], especialy modernizing administration and management 
for judges; rapprochement betwen communities and the police; 
strengthening of family dynamics; and education campaigns for peace and 
coexistence in schools and communities.
24
 
One of the first citywide eforts to addres violence took place due to a change 
that had taken place nationaly: mayors, for the first time in Colombia?s history, were 
elected by popular vote beginning in 1989, a change that forced candidates for ofice to 
be more responsive to ? or at least give lip service to ? the concerns of citizens. The 
winner of Medel?n?s first popular mayoral election, Juan G?mez Mart?nez, had visited 
many of the city?s barios during the campaign to hear what the concerns were, and 
overwhelmingly the concern was over violence. During his two-year term, therefore, 
G?mez tried to involve the Juntas de Aci?n Comunal (JAC, Community Action Boards) 
in conflict resolution in the most violent or at-risk barios, but law enforcement was the 
primary focus of his administration?s approach to countering violence, which mainly 
involved increasing the number of Comandos de Atenci?n Inmediata (CAI, which are 
police units stationed in the barios) and investing in more police baracks throughout the 
city. 
But policy makers at the local and national levels recognized that a stronger 
police presence would not be sufficient, either in the short or the long term, to seriously 
                                                
24
 ?? procesos de reinserci?n de amnistiados de diferentes grupos pol?ticos, los pactos de paz entre j?venes 
integrantes de bandas y milicias populares y su reinserci?n a la vida social, los procesos de educaci?n a 
comunicadores en el manejo de temas de violencia con apoyo de varias universidades, el apoyo a la 
justicia, en especial la modernizaci?n en la administraci?n y gesti?n de los juzgados, el acercamiento entre 
la comunidad y la Polic?a, el fortalecimiento de la din?mica familiar, y campa?as de educaci?n para la paz 
y la convivencia en las escuelas y en al ?mbito comunitario.? Luis Fernando Duque, Antecedentes y 
Evoluci?n del Programa de Convivencia Ciudadana de Medel?n, 21. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 128 
counter violence, and in many ways could actualy increase it in the short term. Luis 
Fernando Duque, who designed the city?s most succesful peace program in the late 
1990s, uses a public-health analogy to explain why. Using law enforcement as the main 
strategy for countering violence, he told me, is a reactive strategy, like beginning to treat 
a disease only after it has begun to spread. In public health, reactive strategies ? such as 
disease management and rehabilitation ? are a form of tertiary prevention: the disease 
has already struck, so eforts are put into mitigating its efects and preventing its further 
spread. Law enforcement and criminal rehabilitation are the analogues to public policies 
for countering violence. Secondary prevention treats the disease upon infection but before 
the onset of symptoms (disease screning and preventive treatment are examples), 
analogous to eforts to convert gangsters into citizens before they commit crimes, for 
example. Primary prevention focuses on general environmental conditions and human 
behaviors to avoid infection in the first place (such as making clean water and vacines 
available, or advocating hand-washing or condom use), analogous to eforts to prevent 
youths from joining gangs or commiting crimes in the first place.
25
 What Medel?n 
needed, Duque told city officials, were more secondary and primary interventions so the 
city could get ahead of the problem.
26
 And in fact, during the 1990s, the trend in policy 
                                                
25
 Luis Fernando Duque, personal comunication (Interview No. 19), 13 May 209; citing Hugh Rodman 
Leavel and E. Gurney Clark, Preventive Medicine for the Doctor in His Comunity: An Epidemiologic 
Approach, 2nd ed. (New York: Blakiston Division, 1958). 
26
 The recomendations of his research team have ben published in Luis Fernando Duque, ed., La 
Violencia en el Vale de Abur?: Su Magnitud y Programa Para Reducirla [Violence in the Abur? Valey: 
Its Magnitude and a Program to Reduce It] (Medel?n, Colombia: Previva?Prevenci?n de la Violencia y 
Otras Conductas de Riesgo, Universidad de Antioquia, and Area Metropolitana del Vale de Abur?, 205); 
Luis Fernando Duque, ed., Pol?tica P?blica Para la Promoci?n de la Convivencia y la Prevenci?n de la 
Violencia en el Valle de Aburr?: 2007-2015 [Public Policy for the Promotion of Peaceful Coexistence and 
the Prevention of Violence in the Abur? Valley: 207-2015], 2a ed. (Medel?n, Colombia: Previva?
Prevenci?n de la Violencia y Otras Conductas de Riesgo, Universidad de Antioquia, and Area 
Metropolitana del Vale de Abur?, 207); Luis Fernando Duque, ed., La Violencia en el Vale de Abur?: 
Caminos Para la Superaci?n [Violence in the Abur? Valey: Ways to Overcome It], 2a ed. (Medel?n, 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 129 
responses was toward more secondary interventions, with serious primary interventions 
finaly being put into place during the later part of the decade. 
Perhaps the most important antecedent to that later progres took place beginning 
in 1991, the year a National Constituent Asembly developed a new Constitution to 
replace the country?s Constitution of 1886. The Asembly was headed by representatives 
from both major political parties (the Liberal and the Conservative) and from the M-19 
Democratic Aliance (Alianza Democr?tica M-19), a political party founded in November 
1989 by demobilized April 19 Movement (M-19: Movimiento 19 de Abril) insurgents. 
The constitution that was pased on 16 July 1991 more explicitly speled out the 
subsidiarity relationship (se Appendix A) betwen the central government and the 
regional governments, abolished presidential appointment of governors and gubernatorial 
appointment of mayors in favor of direct election (which some cities such as Medel?n 
had already put in place), reformed the judicial system, established a Constitutional Court 
separate from the Supreme Court, created a legal framework for the negotiation of 
demobilization agrements with political groups chalenging the state?s authority, and 
banned the extradition of Colombian citizens. This later clause had been advocated 
violently by Pablo Escobar and the Extraditables (se ? 2.1.3) and agred to reluctantly 
by the political establishment as a way to end the war of the narcotrafickers; the war did 
not end imediately upon pasage, but it did spel the beginning of the end of the 
Medel?n Cartel and anticipated the end of the Cali Cartel. In 1996, a year after the Cali 
Cartel?s leaders were arested, the non-extradition clause of the 1991 constitution was 
repealed, opening the way for the extradition of scores of narcotrafickers to the United 
                                                                                                                                            
Colombia: Previva?Prevenci?n de la Violencia y Otras Conductas de Riesgo, Universidad de Antioquia, 
and Area Metropolitana del Vale de Abur?, 209). 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 130 
States over the following twelve years, a series of acts that had mixed results in the short 
term but that have shown early indications of having potentialy some positive results in 
the long term. The remaining constitutional reforms of 1991, especialy the establishment 
of the Constitutional Court and a legal framework for peace negotiations, would have 
important implications for the state?s ability to addres isues related to violence and 
legitimacy later in the country?s history. 
Other antecedents to the progres against violence that gained real traction in the 
mid-2000s took place during the 1990s as wel. On 22 May 1991, the administration of 
Colombian President C?sar Gaviria Trujilo released its National Anti-Violence Strategy, 
whose objectives were ?to guarante that the monopoly on the use of force be in the 
hands of the state?s armed institutions, to recover the justice [system?s] capacity to punish 
crime and counter impunity, and finaly to extend the state?s institutional reach into al 
parts of the national teritory.?
27
 Medel?n was the first city to benefit from this national 
strategy, with the creation of a Metropolitan Security Council headed jointly by the 
governor of Antioquia and the mayor of Medel?n, who worked in collaboration with 
Colombia?s National Security Council, a department-level advisory commision, and 
civil-society organizations to develop and implement a policy caled the Promotion of 
Peaceful Coexistence in Medel?n and its Metropolitan Area (Promoci?n de la 
Convivencia Pac?fica en Medel?n y su ?rea Metropolitana). This policy created Citizen 
Participation and Reconciliation Boards (Juntas de Participaci?n y Conciliaci?n 
Ciudadana) to involve communities in antiviolence programs and proceses, sponsored 
                                                
27
 ?Garantizar que el monopolio del uso de la fuerza est? en manos de las instituciones armadas estatales, 
recuperar la capacidad de la justicia para sancionar el delito y combatir la impunidad y, finalmente, ampliar 
el cubrimiento institucional del Estado en todo el territorio nacional.? Quoted in Luis Fernando Duque, 
Antecedentes y Evoluci?n del Programa de Convivencia Ciudadana de Medel?n, 23. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 131 
youth programs, established a special complaints commision, expanded local and 
departmental police forces and criminal-investigation capacity, established smal-arms 
control measures, reformed the way the justice system dealt with minors, and 
strengthened the bario-level JACs and the comuna-level Local Administration Boards 
(JAL, Juntas Administradores Locales).
28
 Although these eforts were more reactive than 
proactive, these were the first policies targeted specificaly at violence that atempted to 
acount for local conditions; al later policies and programs in Medel?n evolved from or 
were modeled after them. 
The most significant of these later eforts were the establishment of the office of 
the Peace and Coexistence Adviser (Asesor?a de Paz y Convivencia) in 1993, and the first 
city-sponsored anti-violence strategy, the Strategic Security Plan for Medel?n and its 
Metropolitan Area (Plan Estrat?gico de Seguridad para Medel?n y su ?rea 
Metropolitana), during the administration of mayor Luis Alfredo Ramos Botero (1992-
1994). The office of the Peace and Coexistence Adviser was established as the city?s 
representative to negotiate the failed demobilization, disarmament, and reinsertion (DR) 
agrement with the militias in 1994 discussed at the beginning of this section. The 
Strategic Security Plan, published by the city under the name Medel?n at Peace, was 
intended to improve the quality of life in the poorest and most violent barios by 
expanding citizen aces to democratic proceses, countering corruption, and improving 
social services and basic security. The Peace and Coexistence Adviser and Medel?n at 
Peace together established both the precedent and the institutional means for security 
                                                
28
 Herman Eduardo Nore?a Betancur, ?Los Paramilitares en Medel?n: La Desmovilizaci?n del Bloque 
Cacique Nutibara: Un Estudio de Caso? [?Paramilitaries in edel?n: The Demobilization of the Cacique 
Nutibara Bloc: A Case Study?] (Master?s thesis, Universidad de Antioquia, 207), 67; Luis Fernando 
Duque, Antecedentes y Evoluci?n del Programa de Convivencia Ciudadana de Medel?n, 25. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 132 
planning in Medel?n, even if their direct efects were disastrous in the case of the DR 
program and limited with respect to the degre to which they reduced violence in the city 
over the short term. The administration of Sergio Gabriel Naranjo P?rez (1996-1997), for 
example, incorporated many of the elements of his predecesor?s policies in his own 
security and development plans, which were partialy funded by a $15 milion loan from 
the Inter-American Development Bank (IADB).
29
 The succesful policies of subsequent 
administrations drew on these programs and institutions as wel. 
Despite setbacks, these and many smaler eforts by the public, private, and 
voluntary sectors did pay off in many smal ways throughout the city. At worst, they 
demonstrated that there were a lot of people in Medel?n and throughout the country 
sincerely dedicated to peace and the nonviolent resolution of conflicts. At best, they 
succeded in converting some uncounted number of gangsters and militants into citizens, 
and preventing some uncountable number of young people from becoming gangsters, 
militants, and criminals in the first place. This surely influenced the levels of violence in 
the city during the mid-1990s, which by most measures were declining or fluctuating 
from 1991 through 1998; the only figures available that showed a clear, overal rising 
trend during this period were for suicides, extrajudicial executions, and one of the two 
available measures of forced disappearances.
30
 But lawlesnes was stil the rule in many 
                                                
29
 CIDEAL, La Violencia en Medel?n y Colombia: Iniciativas Para la Soluci?n del Conflicto (1980-2004); 
Medel?n en Paz: Plan Estrat?gico de Seguridad para Medel?n y su ?rea Metropolitana [Medel?n at 
Peace: Strategic Security Plan for Medel?n and its Metropolitan Area] (Medel?n, Colombia: Alcald?a de 
Medel?n, 1994), cited in Luis Fernando Duque, Antecedentes y Evoluci?n del Programa de Convivencia 
Ciudadana de Medel?n, 26; Plan de Desarolo para la Ciudad: 193-1995 [Development Plan for the 
City: 193-1995] (Medel?n, Colombia: Alcald?a de Medel?n, 1992), and Plan de Desarolo de Medel?n: 
1995-1997 [Development Plan for Medel?n: 1995-1997] (Medel?n, Colombia: Alcald?a de Medel?n, 
1995), both cited in Herman Eduardo Nore?a Betancur, ?Los Paramilitares en Medel?n,? 70. 
30
 Suicide data from Departamento Administrativo Nacional de Estad?stica (DANE); extrajudicial 
executions, from Centro de Investigaci?n y Educaci?n Popular (CINEP); and forced disappearances, from 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 133 
of the city?s peripheral barios and the city center, and in 1998 the arival of 
paramilitaries in the city would usher in a new era of violence. 
What was happening in Caicedo La Siera during the 1990s, when the militias in 
much of the rest of the city were being corrupted by the seduction of easy money and 
their own power, when militias were fighting militias and gangs, when gangs were 
fighting militias and other gangs, when gangs and crews were shifting alegiances with 
other gangs and crews in response to threats and opportunities as they arose, and when a 
wide range of state and nonstate actors were desperately trying to bring peace to the city? 
The following section takes a look. 
4.2. Caicedo La Siera: Unlocking the Doors to More Violence 
4.2.1. Two Children Fight, and the Result Is War 
The exact dates are hard to pin down; diferent people have diferent memories of 
the initial dispute. But sometime during the mid-1990s, probably around 1996, a boy 
from a section of Las Estancias caled Santa Luc?a got into a fight and punched a boy 
from Lower Vila Liliam. Or so says a resident of Lower Vila Liliam; it?s likely that 
those from Santa Luc?a would remember it the other way around, that the boy from 
Lower Vila Liliam started the fight, and that everything that followed was instigated 
from that bario as wel. In any event, ?the next day, the brothers of the Santa Luc?a boy 
                                                                                                                                            
Instituto Popular de Capacitaci?n (IPC). CINEP provides data on forced disapearances as wel (showing 
wide fluctuations year to year), but its figures and IPC?s sem to have nothing to do with each other: for 
example, CINEP reports that disappearances fell from 13 to 3 between 191 and 192, while IPC reports 
that they more than doubled, from 29 to 64 in the same period. Neither organization?s standards for 
determining disappearances are transparent. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 134 
came to take it out on the brother [of the boy] from here, and from there the fight was 
on.? It started as a fight betwen two boys, and ?now it was a fight betwen adults, 
because now, to defend their litle brothers, the older brothers got involved, then one 
from there kiled one from here.?
31
 
Whoever was to blame for the initial fight and the needles escalation, the 
consequences were bad for everyone in both barios: Each side pulled together a crew, 
the crews took revenge for each kiling, the bigger gangs (already with a history of 
hostilities against each other) took sides and brought bigger weapons, and the blood feud 
entered a vicious cycle that escalated into war. Residents of Lower Vila Liliam could not 
enter or pas through Santa Luc?a, nor vice versa, without risking death at the hands of 
the local gangsters.
32
 Thus began the third gang war that Caicedo La Siera had suffered 
since the 1980s. 
The first gang war had peaked in 1990, as the crews and gangs in the area fought 
over teritory, with five main gangs eventualy consolidating control over micro-
teritories, some as smal as 20 or 30 city blocks (se ? 3.2): M-6&7 (led by Hugo) in La 
Siera, Vila Turbay, and Upper Vila Liliam; La Ca?ada (led by Alberto) in Lower Vila 
Liliam; La Arenera (led by Gamuza) in Lower Vila Liliam, closer to Las Estancias; a 
gang in Santa Luc?a, in upper Las Estancias along the crek; and a gang in the Las Mirlas 
neighborhood of upper Las Estancias just north of Santa Luc?a. The second gang war had 
peaked in 1994 for reasons few residents in the area sem to remember but almost 
                                                
31
 ?Entonces al d?a siguiente, los hermanos del ni?o en Santa Lucia le vinieron a reclamar al hermano de 
ac?, y desde ah? se fue formando la pelea. ? Ya fue entre adultos, porque ya por defender a su hermanitos, 
ya entraron fue los grandes, entonces ya uno de al? mato uno de ac?.? ?Mama Luz,? Interview No. 1. 
32
 Ibid. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 135 
certainly had to do with militia wars that took place during the Santa Elena Acords (se 
? 4.1.2) as wel as ongoing disputes over teritory. Most of those interviewed for this 
study, however, date the ?beginning? of the gang wars to the third war, the fight betwen 
the children, which started in late 1995 or probably sometime in 1996, since that?s the 
year homicides began to increase to their peaks in 1999-2000. The fourth gang war 
started when the paramilitaries entered the city and intervened in existing conflicts as a 
strategy to displace the guerilas (and rival narcotrafickers) beginning in 1998 and 1999 
(se ? 5.2). After the third war, however, the gang wars al start blending together in 
locals? memories, a series of batles and grudges and deaths and shootouts betwen 
barios, the most significant (with respect to the story of Medel?n?s violence, if not 
necesarily localy) taking place betwen the people who lived lower on the hilside and 
the people who lived higher up, with those in betwen ? in Upper Vila Liliam ? 
suffering the worst of it. 
4.2.2. Life in an Urban War Zone 
Cristiana has lived in Upper Vila Liliam for 50 years.
33
 At the time the area was a 
rural property owned by a gentleman for his country home. When she first arived she 
lived in a cane-and-mud house, a traditional style of construction caled bareque (ba-
RAY-kay), but she was able to buy a smal plot of land for 100 pesos and, with money 
she earned by seling creamed soups (cremas), built the house she lives in today. You 
would never know it was once a country house: surounded on al sides today by brick 
houses with Spanish-tile roofs, cramed together row after row, abutting the jagged 
                                                
33
 ?Cristiana? is not her real name; she requested anonymity out of concern for her safety. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 136 
concrete sidewalks with their high and uneven curbs, lining the asphalt strets buzzing 
with life, mostly young people walking by, groups of schoolchildren in uniform shouting 
or giggling as they pas, buses crawling up the step hil spewing black smoke, thumping 
reggaeton music blasting from the speakers of old cars, trucks overloaded with whatever 
needed to be delivered, entire families riding on one motorcycle, and stret vendors 
ringing bels and shouting ??aguacate! ?aguacate!? (?avocados! avocados!?). At the time 
there was no war, no conflict, she told me. Or rather, she told my research asistant, 
Janeth, who helped me understand Cristiana?s strongly acented paisa argot (and helped 
Cristiana understand my bafling gringo Spanish). We were siting in her tiny living room 
just off to the side of her tiny convenience store, as people popped in and interupted, 
asked for some item she didn?t have, then left again just as quickly. 
We used to live in peace. There have always been thieves [in the 
community] ? [but things] went bad only about 15 or 16 years ago [i.e. 
early 1990s] because [now they] were coming from other places to rob us. 
? [The displaced people from up there] caused us more problems because 
they came from other places, with other customs, and they started 
gangs.?
34
 
Cristiana worked as a volunter for the neighborhood Civic Center, the 
predecesors to the Community Action Boards (JACs, Juntas de Aci?n Comunal), and 
then continued serving the community once the JACs were founded, though she never 
held an official title. Today she is 70 years old and works at a nearby school, where most 
of the students come from broken, displaced, or extremely poor families, and where the 
school has frequently found itself at the front lines of the gang wars. Living and working 
right at the border betwen M-6&7 teritory and La Ca?ada teritory, Cristiana was very 
                                                
34
 ?? viv?amos en paz. Ladrones siempre los ha habido ? [pero esto] se vino a da?ar hace solo 15 a 16 
a?os porque vienen a robar de otras partes. ? [Los desplazados de ariba] causaron m?s problemas porque 
llegaban de otras partes, con otras costumbres y se forman las bandas.? ?Cristiana,? Interview No. 9. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 137 
careful not to criticize any of the combatants. ?They said [the conflict] was about 
defending land, but nobody realy knew hat it was about.?
35
 Many others have 
expresed the same ignorance. 
In her teling, both sides in the war were careful to avoid civilian casualties. 
?They realy took care of people. The war was betwen them, not with us. They didn?t 
touch one?s family nor one?s children,? she said, using uno (one) ? the paisa equivalent 
of the ?royal we? ? to refer to herself. Moreover, the gangs alerted her when a 
confrontation was about to take place, and she in turn alerted her neighbors and the 
school. They did this ?thanks to the authority and the respect I had in the bario.? 
My son spoke with the ones from above [M-6&7] and the ones from 
below [La Ca?ada]. He talked to them about peace, he organized socer 
championships for them, [and] they always took care of him. ? Me, they 
respected me, because they never did anything to me, or to my kids, either. 
They never threatened me, I didn?t have to leave. I used to manage a 
community restaurant, and some of them had eaten there.
36
 
She realized that they were looking out for her, she said, when she was leaving her 
restaurant one day and realized a man with a weapon was following her. He told her to go 
directly home and made sure she arived safely. And even though she knew most of 
them, she said that if, during the day, she were to pas a gangster from, say, M-6&7, he 
would gret her only as a stranger, ?as if we didn?t know each other? so that, if someone 
from La Ca?ada were to se it, it wouldn?t prejudice them against her and put her in 
danger. (And yet she also said that nobody realy knew ho in the community was a 
                                                
35
 ?[El conflicto se dio] que por defender tereno, pero en realidad nadie sab?a por qu?.? ?Cristiana,? 
Interview No. 9. 
36
 ?? gracias a la autoridad y respeto que ten?a en el bario. ? Mi hijo hablaba con los de ariba y con los 
de abajo. Les hablaba de paz, les hac?a campeonatos de f?tbol, [y] siempre lo cuidaban. ? A mi me 
respetaban porque no me hac?an nada y tampoco a mis hijos. No me amenazaron, no me tuve que ir. Yo 
manejaba el restaurante comunitario y algunos de elos hab?an comido al?.? ?Cristiana,? Interview No. 9. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 138 
gangster and who was not a gangster until one did something in public that indicated a 
gang afiliation. To this comment, however, she imediately added: ?At that time, one 
had to stay neutral? ? suggesting that perhaps this profesed ignorance about who was in 
what gang was more strategic than anything else.
37
) 
During the wars of the 1990s, especialy late in the decade (se next chapter), a lot 
of the students at the neighborhood school had to move out of the bario because of 
threats against their families. When a lot of students stopped showing up for clas, the 
school?s administrator, Alicia, told me, that?s when ?one realizes that things are 
happening.?
38
 Alicia said the gangs generaly kept the school as a sort of neutral zone, 
however; if nothing else, at least nobody was ever kiled on school grounds. She 
atributes this to the school?s very careful diplomacy with al sides in the war. ?One has 
to be very impartial, very stealthy, cautious about ? what you say. Some things just can?t 
be said to everyone.? She was refering to how she and her staf interacted with the 
gangsters, demonstrating to them what they taught their own students: ?Respect generates 
respect ?. If you can serve as a good role model, you?l earn the authority.?
39
 
As a result, Cristiana said in her living room, the gangs ?never did anything to the 
teaching staf,? nor to the students, at least not on school grounds.
40
 But the school could 
                                                
37
 ?Uno no sabe que alguien est? en una banda hasta que no se unta en la pomada. Uno no sab?a qui?n es 
qui?n, hasta que sab?a que estaba untado. En ese tiempo hab?a que ser neutro.? ?Cristiana,? Interview No. 9. 
38
 ?? se da cuenta de que pasan cosas.? ?Alicia,? personal comunication (Interview No. 7), 19 March 
2009. ?Alicia? is not her real name; she, to, requested anonymity to protect the students and staf at the 
schol. 
39
 ?Uno tiene que ser muy imparcial, muy sigiloso, cauteloso ? en lo que dice. A todas las personas no 
puede dec?rseles todo.? ?El respeto genera respeto. ? Si usted presta una buena funci?n, gana autoridad.? 
?Alicia,? Interview No. 7. 
40
 ?A los docentes no les hac?an nada.? ?Cristiana,? Interview No. 9. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 139 
not control what happened when their students got home. Not only did the school lose 
many students when their families were forced to leave their homes, but the lure of the 
gangster lifestyle in the bario was something that was very seductive, especialy to the 
boys. When asked what they wanted to be when they grew up, their answers reflected 
what they saw around them, and if they had family who were members of gangs ? and 
many of them did ? the answer was often: ?I?m going to be like my cousin. I?m going to 
be like my brother,? Cristiana said.
41
 Schoolyard problems often led to taunts and threats 
by one student to sic his or her gangster brother after another. 
Midway through our conversation, a stocky, middle-aged man walked into her 
living room and politely but confidently interupted, introducing himself as a community 
leader in the bario. We had not arived to Vila Liliam ore than 30 minutes before, and 
word had already goten to him that a couple of strangers were in the bario, interviewing 
Cristiana in her living room. Speaking with a smile and the polite but suspicious air of a 
politician, he asked us who we were and what we were doing in the bario, and expresed 
surprise and disappointment that we hadn?t gone to se him first. This interview as 
taking place in early 2009, during a time when it had become clear that many of the 
demobilized paramilitaries had renewed their mafia ties and were in the proces of 
threatening, corrupting, and even kiling community leaders throughout the city, 
including in Comuna 8, as a way of winning aces to city contracts (se ? 7.2). I didn?t 
have any idea whether this man was one of them. Janeth, my research asistant, didn?t 
know either, but she told me afterwards that she suspected he was just a guy with the 
personality of a politician. At the time, however, I found it jaring to discover just how 
                                                
41
 ?Yo voy a ser como mi primo. Yo voy a ser como mi hermano.? Ibid. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 140 
closely the communities kept tabs on visitors and just how quickly this community leader 
had responded to what semed to be an informal community inteligence network. I was 
imediately reminded that there had been a time in this part of the city, not too long ago, 
when a pasing stranger risked being kiled just for showing up unannounced, and a 
resident could be kiled just for being sen with an outsider. (Within two months I would 
be advised that those times had returned and that I could best protect my research 
subjects ? and myself ? by not visiting the bario any more.) Janeth answered his 
questions thoughtfully and smoothed things over quickly. By the time he left we had an 
interview scheduled with him for two weks later. 
During our interview ith him, Ernesto turned out not to be a para, just as Janeth 
had predicted. Moreover, he had no patience for any of the armed actors, and at our 
interview he painted a much darker picture of their efect on the community than 
Cristiana or Alicia had. ?In fact, we had problems during the course of the conflict,? he 
said, losing his politician?s air as he spoke. He had lived with his wife in a house on a 
stret that was the site of some of the most intense fighting of the militia-gang wars and 
the later militia-para and para-para wars (se next chapter), right at the border betwen 
the gangs from above and the gangs from below. ?We had a fast-food busines right there 
in front of the house,? Ernesto told us in the storeroom of one the other smal busineses 
he manages. His wife used to run the store while he worked on other projects. ?We think 
that it was their target, from its being in a lot of people?s eyes, because it was like a 
meting place. Because when one side wasn?t here tying things up, the others were. The 
fact of the mater,? he paused, ?is that she died by the force of arms in the conflict. They 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 141 
kiled her while she was working.?
42
 He didn?t say whether his wife had been targeted for 
assasination or had simply died in the crossfire. 
It was 2001 when Ernesto lost his wife, many months after los de arriba (the 
gangsters from the upper barios) had ceased being militias and had switched sides, and a 
time when the fourth gang war was already under way.
43
 It was an extremely dificult 
period, the one that began in the mid-1990s, with the worst of it starting around turn of 
the milennium. ?At that time there were a minimum of thre confrontations a wek,? 
Ernesto said. A lot of the batles took place right on his stret, and the only time the 
police showed up to do anything about it, he said, was the next morning when they would 
come up to recover the bodies. ?No, the law never intervened, because there were strong 
forces here, and al over the city. The problem was too big.?
44
 Everyone knew ho the 
gangsters were, he said, but everybody was afraid to report them to the police, because 
the gangsters would find out who snitched and have them kiled. (And as he himself had 
demonstrated to me just a couple of weks earlier, news spreads quickly in this area.) 
Just below Upper Vila Liliam, in Lower Vila Liliam, and above it, in Vila 
Turbay and La Siera, the communities there had what was practicaly a luxury in 
                                                
42
 ?Pero de hecho tuvimos problemas en el proceso del conflicto. ? Nosotros ten?amos un negocio de 
comidas r?pidas ah?, al frente de la casa. Nosotros pensamos que esto fue objeto, de estar en los ojos de 
muchos, porque como era un punto de encuentro. Porque cuando no estaban los unos ah? mecateando, 
estaban los otros. Lo cierto del caso es que ela falleci? bajo el efecto de arma del conflicto. A ela la 
asesinaron, trabajando.? ?Ernesto,? personal comunication (Interview No. 14), 1 April 2009. Even though 
?Ernesto? consented to my using his real name for this study, I am using an alias to protect his identity from 
the armed actors who have returned to the area. 
43
 The next chapter details the politics behind the third and fourth gang wars. Here I am going slightly 
beyond the present chapter?s nominal timeframe of 1992-1998 to discuss the communities? experiences of 
those wars, since the battles most in the comunities remember most powerfully began during the later 
part of this period and continued, in their minds practicaly unabated, until the truce of 2003. 
44
 ?En ese entonces a la semana eran, m?nimo, tres enfrentamientos. ? No, la ley no interven?a, porque 
hab?a una fuerza grande aqu? y en toda la ciudad, la problem?tica era grande.? ?Ernesto,? Interview No. 14. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 142 
comparison to the situation in Upper Vila Liliam: a gang that almost fully controlled 
their bario. With reasonably secure teritorial control, there were muchachos or mandos 
? guys or commanders (i.e. gangsters or gang leaders) ? whom they could go to for 
protection, or to resolve disputes, or to report neighbors who were spies for the other side 
or had been sen with enemy gangsters. The muchachos from above would take care of 
the people who lived in the upper barios (though not always equitably) and the ones 
from below would take care of those in the lower barios (also not equitably). In betwen, 
in Upper Vila Liliam, some people ? like Cristiana or Alicia ? did have enough 
authority or prestige within the community that they could usualy count on the 
muchachos from either side to look out for them and make sure they knew hen to stay 
out of the way, and family members and friends of the gangsters could usualy count on 
someone to protect them or avenge them as needed. But everyone else in Upper Vila 
Liliam was on their own; since control over their bario was contested fairly evenly, they 
were never secure enough to be able to depend on either side for protection. 
A lot of people were kiled having nothing to do with the conflict, innocent 
bystanders who didn?t hide or protect themselves when the bullets came flying. Plus, 
?there was another kind of victim in the conflict, and they were the ones who were 
atacked out of revenge, or because they didn?t share the ideology. They?d blacklist one 
family member, but then they?d go and finish of the entire family.? Busineses, such as 
his, were also targeted: 
It was common for 10, 30, 40 youths to show up, even women, hooded 
and everything. The men would turn out the lights, everyone outside ? 
and they?d line up 150 suspects and search them, and they?d pick out one 
or two and walk them 10 meters, or right there at the corner, and kil them. 
So that happened at a bunch of the busineses in this zone. In other words, 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 143 
it was a system of, you know, intimidation, to demonstrate power, and 
with weapons of al sizes.
45
 
Snitching to the police was useles and dangerous, and snitching to either gang was just 
dangerous. So most in the community, stuck in the middle, tried to stay quiet and tried to 
stay out of the way. 
Leady (pronounced ?lady?) was born in La Siera around the time M-6&7 was 
?cleaning up? the bario of its ?thieves, rapists, and drug addicts.? ?When I was born, the 
situation had already changed a litle with respect to [the earlier violence]. But another 
type of violence began,? she wrote, the new violence asociated les with common crime 
(i.e. interpersonal-communal) and more with teritorial control (i.e. collective). She told 
her diary a story she?d been told when she was seven years old, probably around 1997. 
Her father and her uncle, both living in La Siera (M-6&7 teritory), needed to get 
downtown for their jobs, and one day they walked down to the bus stop at Tres Esquinas 
(La Ca?ada teritory, in Lower Vila Liliam): 
there was a bus with pasengers ready to leave, when a man went up to the 
driver and said that he thought he had told him not to work, and the driver 
responded to this guy that, yes, he was going to work anyway. This guy 
pulled out a gun, put it to [the driver?s] head, and fired. The bus like took 
off and crashed into an electric pole.
46
 
                                                
45
 ?Hab?a otro tipo de victimas del conflicto y eran los que atacaban por venganzas o por que no compart?an 
la ideolog?a. Se le vetaban a una familia e iban acabar con toda la familia. ? [Era] com?n que legaran diez 
o legaran treinta o legaban cuarenta j?venes, hasta mujeres, tapados y todo. Se?ores apaguen la luz, todos 
afuera filados. ? y filaban ciento cincuenta sospechosos y los requisaban, y seleccionaban uno o dos, [y] 
los caminaban diez metros o ah? a la vueltesita [y] los mataban. Entonces eso pasaba en distintos negocios 
de aqu? de la zona. O sea era un sistema pues de aci?n para intimidar, para demostrar el poder, y con 
armas de todos los alcances.? Ibid. 
46
 ?? hab?a un bus con pasajeros listo para salir. Cuando un hombre se le acerc? al conductor del bus y le 
dijo que si no le hab?an dicho que no trabajara y ?l le respondi? al tipo que s? iba a trabajar. ?ste sac? un 
arma, se la puso en la cabeza y le dispar?. El bus, como iba a arrancar, se fue contra un poste de energ?a.? 
Leady Jhoana Reyes, ?Mi Diario,? [?My Diary,?] in Jam?s Olvidar? Tu Nombre, ed. Patricia Nieto 
(Medel?n, Colombia: Alcald?a de edel?n, 206), 52. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 144 
They had to walk the rest of the way to their jobs, and in the afternoon they tried to return 
by taxi. But it wasn?t easy: 
when they got to the place where the driver was kiled before, [the 
gangsters] made them get out and told the taxi driver to get lost. My father 
and his brother, very scared, got out, and they were asked where they were 
going. Realy afraid, they said they were going up above, without saying 
that they were going [al the way] to La Siera, because if they had said 
that, they would have gotten kiled.
47
 
He made it home, but her father decided that day to leave the bario, because it was too 
dangerous for him to keep trying to commute from the bario ?above,? La Siera, to the 
city center, since that required pasing through the bario ?below,? and the gangs were 
now enforcing an extremely strict border-control policy. ?I also remember that during 
that realy hard crisis the bario had then, we went without food for two weks, because 
the muchachos weren?t leting any cars go in or out, since they thought they might get 
infiltrated.?
48
 Another resident of La Siera remembered those two weks as wel: even 
the church got shot-up. ?You can stil se for yourself the bullet holes in al the houses.?
49
 
A resident of Vila Turbay, who had been a combatant in that conflict, gave me a tour of 
the war zone: and indeed, I could se for myself the bullet holes in the houses.
50
 That 
whole period was a time when communities were deeply divided and isolated. Residents 
of one bario were asumed to be enemies of the gangsters in control of the other. 
                                                
47
 ?? Regresaron en taxi y cuando iban legando donde antes hab?an matado al conductor, los hicieron 
bajar y le dijeron al taxista que se perdiera. Mi pap? y su hermano muy asustados se bajaron y les 
preguntaron que para d?nde iban, elos asustad?simos dijeron que para ariba, sin decir que para La Siera, 
porque si lo dec?an los mataban.? Ibid. 
48
 ?Tambi?n me acuerdo que en esa crisis tan dura que tuvo el bario nos quedamos sin comida durante 
quince d?as porque los muchachos no dejaban entrar ni salir ning?n caro, pues pensaban que pod?an ser 
infiltrados.? Ibid. 
49
 ?Todav?a puede verse los huecos de balas a las casas.? ?Boris,? personal comunication (Interview No. 
8), 20 March 2009. 
50
 ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 145 
One day at elementary school, Leady was taking her afternoon clases when some 
hooded gangsters showed up and said that nobody was alowed to leave the school until 
they said so. ?I remember we had a very Catholic teacher and she had us al praying over 
the next two hours.? 
Then they took al of the students out onto the patio and we stayed there 
until like six thirty in the evening. When we were to leave, some 
muchachos with hoods and guns divided us into two groups: those who 
were going above [to La Siera] and those who were going below. Those 
of us who were going up, like six of these guys acompanied us, and they 
took some [students] al the way home.
51
 
The contrast betwen the concern the gangsters and militants of the era showed to 
the schools and what semed to be an uter disregard for casualties elsewhere, especialy 
in Upper Vila Liliam, is sometimes jaring in these stories. What is clear is that M-6&7 
and La Ca?ada were not monolithic organizations; La Ca?ada, for example, had evolved 
from a single neighborhood gang into more of a rough aliance of crews and gangs in 
Lower Vila Liliam and the surrounding areas; it was not realy a hierarchical criminal 
organization, although discipline was stil enforced by its founding members, but more of 
a network manager. Some of its members were asasins, others robbed banks, others 
sold drugs, and others were just unemployed neighborhood kids looking for something 
interesting to do and an easier way to make money.
52
 With such diversity, there would 
clearly be some members with a big heart and others with no soul. The same could be 
said for M-6&7; it had been founded by idealists who legitimized themselves to the local 
                                                
51
 ?Me acuerdo que ten?amos una profesora muy cat?lica y nos puso a rezar durante dos horas seguidas. 
Despu?s nos bajaron a todos los alumnos al patio y nos quedamos al? como hasta las seis y media de la 
tarde. Cuando ?bamos a salir unos muchachos encapuchados y armados nos dividieron en dos grupos: los 
que iban para ariba y los que iban para abajo. A los que sub?amos no acompa?aron como seis de esos 
tipos, a algunos los llevaban a sus casas.? Leady Jhoana Reyes, ?Mi Diario.? 
52
 ?Germ?n,? Interview No. 16. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 146 
communities, but over time it was increasingly joined by people who were more like 
gangsters than leftist militants. And as the decade wore on, both loosened their standards 
of recruitment. ?How did the groups grow?? Ernesto asked aloud: 
Both groups: so as they suffered more casualties and deaths, the group 
would [recruit] relatives of the dead, for revenge. So they kept growing the 
group: ?Wel, they went and kiled my brother? Then bam, I?ve gotta start 
being part of that group.? [They grew also] with the vacuna system 
[protection rackets], because at the root of the conflict, that was what they 
used, like their scam ? ?man, we ned to buy weapons, we need to boost 
our troop strength?  so they went around geting cash from the 
communities. So anyone who doesn?t give it to them, then poof.
53
 
Over time, living in a place where shootings could break out at any moment, 
coping mechanisms became part of daily life. Ernesto described how people ended up 
being afected by the sound of smal bombs exploding: 
They were atrocious, both in their direct efects and in their intimidation 
of the community. With the slightest sound, people got realy scared and 
worried, in such a way that, when some confrontation was about to 
happen, at the slightest sound, at the slightest hint that they were geting 
closer, everyone suddenly closed their doors, their windows, and looked 
for a protected spot.
54
 
Sounds had to be interpreted for what they foretold. ?Almost always there are, like, 
preludes,? a young man told an anthropology student in 2002, talking about the wars that 
began in the 1990s. 
                                                
53
 ??C?mo crec?an los grupos? Ambos grupos. Entonces a medida que iban presentando bajas o muertes, 
entonces se volcaba el grupo con los familiares del muerto, por las venganzas. Entonces se iba creciendo el 
grupo: ??Ha que me mataron mi hermanito? Entonces taque, hay empezaba a ser parte del grupo.? ? Con 
el sistema de vacunas, considero que si cualquiera, porque a ra?z del conflicto era lo que ellos utilizaban 
como artima?as ??hombre, necesitamos comprar armas, necesitamos aumentar el pie de fuerza??, 
entonces pasaban recogiendo plata a las comunidades. Entonces el que no diera, entonces, ya.? ?Ernesto,? 
Interview No. 14. 
54
 ?Eran atroces en la acci?n y en la intimidaci?n de la comunidad. Con el mero ruido, la gente el temor y el 
miedo abundaban, de tal manera que cuando iba a ocurir un enfrentamiento, al menor ruido, o al menor 
termino de ver que se estaban acercando, todo el mundo ya ceraba sus puertas, sus ventanas y buscaba los 
puntos de proteci?n.? Ibid. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 147 
One day a guy hears that they?re shooting, like bang!, one shot: ?That was 
a gunshot, right?? The other guy says, ?Now those people, now there?s 
going to be, now there?s going to be a shootout.? There are times when 
there?s no response. Then it?s, ?Oh, for chrisakes, why haven?t they 
responded!? Like that, something?s hatching out there, and you just know, 
it?s coming. ? There are things that are provoked, right? They?re usualy 
provoked. I don?t know about now, [but] before, it was very, very 
common. There was like a dialog before the shooting began. So you?d 
hear, you?d hear a word, that things are being shouted, they?re being 
whistled, they?re being whistled, so now they know they have to get out 
their toys because they?re gonna start. They go shouting ?, ?Come on up, 
you fuckers!? Bam! It?s on! And so there are like these preludes [that] 
appear to be like a war ritual.
55
 
The sounds got interpreted beforehand, as people found a way to predict what was about 
to happen and when, so they could take cover when needed, and gossip circulated 
afterwards as people sought an explanation, or perhaps tried to put a spin on what they 
had heard to make it easier to bear. ?The people who circulate these mesages, the way 
that they do it, even the place where [the mesages] originate, al contribute to people?s 
perception of what the source of the threat is, which determines whether the mesage is 
acepted somehow as ?legitimate? and so therefore should be paid atention to, and 
appropriate measures taken,? wrote Juan Diego Alzate Giraldo, the anthropology student, 
who for his undergraduate thesis interviewed people in Caicedo about fear amid 
conflict.
56
 Colombia is not a country where solidarity and social capital have ever been 
                                                
55
 ?Casi siempre hay, hay como los anuncios. .. Entonces uno escucha un d?a, que disparan como ?tran!, un 
tiro. ??Eso fue un tiro, cierto?? Otro: ?Ya est? gente ya, ahorita va a haber, ahora va a haber balacera.? Hay 
veces que no responden, entonces uno: ??Eh, juemadre! ?Porque no habr?n respondido!? Eso, ah? se va 
como incubando algo y uno sabe que viene. ? Hay cosas se provocan, ?cierto? Se provocan, usualmente. 
Yo no s? ahora, antes era muy, muy com?n. Hab?a c?mo un di?logo antes de disparar. Entonces se 
escuchaba, se escuchaba una palabra, que se gritan cosas, se silban, se silban, entonces ya saben que tienen 
que ir sacando sus juetes, porque van a empezar. Se van a gritar ? ??S?base pirovo!? [sic] Ta! Y 
empiezan. Y ah? hay como esos pre?mbulos, aparece como un ritual de la guerra.? Punctuation edited for 
clarity. Quoted in Juan Diego Alzate Giraldo, ?Alg?n D?a Recuperaremos la Noche,? 38. 
56
 ?Desde quienes emiten dichos mensajes y la manera c?mo lo hacen, hasta los lugares de origen de los 
mismos, responden a esa percepci?n de la fuente de amenaza, que hace que el mensaje sea, en cierta forma, 
asumido como ?legitimo? y por lo tanto se le preste atenci?n y se tomen medidas al respecto.? Ibid., 39. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 148 
strong, Medel?n least among its cities in that respect.
57
 But Alzate Giraldo found in 
Caicedo a ?tacit solidarity,? an ?emotional community? built around fear, that had 
emerged as a way to solve the practical problem of what to do when you mised the 
subtle warnings and you suddenly found yourself in the middle of gangsters shooting al 
around you. He quoted one resident: 
At any house, at any neighbor?s house, we?d stay there. If not, we?d stay at 
a store, at a grocery, then. And always ? I know, we?re not very talkative, 
we don?t talk to anybody at the house, we wouldn?t even gret them. But 
when these things would happen, you at least greted them as you went 
inside, like, their house. And then the smal-talk started.
58
 
It?s dificult to be ?talkative? in a place where saying the wrong thing could so easily get 
to the wrong ears and so easily get you kiled, where misplaced gossip could cause 
needles fear or foment a needles batle, where phrasing something wrong to the wrong 
person on the stret could be interpreted as a sign of disrespect and get you kiled, where 
conflicts were so often resolved by resort to arms, and not only by gangsters. Talking can 
be scary when talking can get you kiled. 
But despite their suffering, the people in Caicedo La Siera found a way to live, to 
play, to dance, to have block parties, to hang out in the strets, to interact with their 
neighbors, even when they knew the police would not resolve any conflicts that might 
arise, even though security often had to be provided by criminals and sometimes 
negotiated with waring gangs. ?Young people here have energy,? Alicia, the school 
administrator told us at the school one day, ?and we?re trying to channel that energy. We 
                                                
57
 Francisco E. Thoumi, ?The Colombian Competitive Advantage in Ilegal Drugs.? 
58
 ?En alguna casa, en alguna casa de vecinos, nos qued?bamos. Si no, nos qued?bamos en una tienda, en 
una revuelter?a, pues. Y siempre ?uno conoce, nosotras somos muy caladas y no hablamos con nadie por 
la casa ni salud?bamos a nadie. Pero cuando pasaban esas cosas al que menos usted saludaba se le met?a, 
pues, a la casa. Y entonces empezaron los comentarios.? Punctuation edited for clarity. Quoted in Juan 
Diego Alzate Giraldo, ?Alg?n D?a Recuperaremos la Noche,? 41. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 149 
tel people that we?re rich in energy, rich in our desire to make progres. I never use the 
word poverty. We don?t have a lot of cash, but we consider ourselves wealthy.?
59
 
4.2.3. ?Mama Luz? and the Community Peace Talks 
Just about everyone in Caicedo La Siera knows the story of Mama Luz, so she 
dismisively rejected my offer to conceal her identity in print, as I had done for other 
interview subjects. There would be no point: she is a legend in Caicedo La Siera. A 
single black woman from the coast, Luz Edna Garc?a Copete arived in Medel?n in 1958, 
moved to the city center, and worked for diferent charities and for the city until moving 
to the United States to work with the humanitarian organization CARE for five years. 
When she returned to Colombia, she worked in comunity health until a change in 
management forced her out of her job. Looking to live near family, she bought a two-
storey house in Lower Vila Liliam in 1973 and moved in. ?This bario was shocking,? 
she said, siting in the living room of that same house, cluttered with books and pictures, 
including some of herself as a visibly confident young woman (who never maried, she 
said, because ?I didn?t want any man teling me what to do?). ?The first few months I 
was here it gave me seizures, because I?d never had to live in a place like this before.? It 
was a place where most people were very, very poor, so much so that, even with the 
modest means she had, she was considered the rich woman of the bario.
60
 
                                                
59
 ?Alicia,? Interview No. 7. 
60
 ?Este bario era impresionante, a mi me daban convulsiones los primeros meses que viv? aqu?, porque 
nunca me hab?a tocado vivir en un lugar como este.? Information and quotes in this section are from 
?Mama Luz,? Interview No. 1, and personal comunication (Interview No. 17), 7 April 209, unles 
otherwise noted. The basic outlines of her story have been verified through other sources, including Ramiro 
Alberto V?lez Rivera, et al., Governabilidad Local en Medel?n: Configuraci?n de Teritorialidades, 
Conflictos y Ciudad [Local Governance in Medel?n: The Configuration of Teritorialities, Conflicts, and 
the City] (Medel?n, Colombia: Escuela Superior de Administraci?n P?blica?ESAP, 204); Mauricio Ortiz, 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 150 
Having worked with charities and community-health services for most of her life, 
she was able to recognize that the bario needed help, and enough people in the local 
government recognized her that she was able to be a conduit for city services. But she 
started smal. With the asistance of some nieces, she drew up some signs and hung one 
on the door of her house: ?Fre injections, telephone at your service, movies for children 
at your service.? ?Wel, the movie was that I had a big television, and around here 
nobody had a television, or a telephone, or anything, so I started to provide that service to 
the community for fre,? she said. Eight months later, she decided the community needed 
a preschool to beter prepare the children for elementary school, so she went to the city?s 
secretary of education, got some training and resources, and started teling the mothers 
she would charge a nominal fe for the service. ?So the first month I opened it with 10 
children, the first year 10 children, the second there were already 22, the third 36, and 
then in the fifth year there were 50-something, and after that it never went below 60- or 
50-something.?
61
 
The year after opening the preschool she saw some men arive at the empty lot 
across the stret from her house one day and asked them what they were doing. They told 
her they were from the community and they wanted to build a health center there, since 
there hadn?t been one in the bario. She offered to help in any way she could, and with 
the help of volunters and donated materials, her contacts in the city government, plus 
                                                                                                                                            
personal communication (Interview No. 12), 1 April 2009; and informal conversations with other bario 
residents. 
61
 ?? inyectologia gratis, tel?fono a la orden y cine para los ni?os a la orden. Entonces el cine era que yo 
ten?a un televisor grande; y por aqu? nadie tenia televisor, ni tel?fono, ni nada. Entonces yo empec? a 
prestarle el servicio a la comunidad gratuitamente. ? [Abr? el prescolar] el primer mes con diez ni?os, el 
primer a?o 10 ni?os, el segundo ya fueron 2, el tercero 36, y ya el cuarto fueron 50 y pico. Y ya desde ah? 
no bajaba de 60 y 50 y pico.? ?Mama Luz,? Interview No. 11. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 151 
money they raised making and seling snacks, a comunity building was beginning to 
rise from the ground. But they had failed to check whether anyone had actualy owned 
that ground, and in fact one day a man showed up and asked them what it was that they 
thought they were building on his property, on which he had intended to build a house. 
After Mama Luz invited him for coffe and spoke with him a few times, he ended up 
donating part of the lot for the community building, and Mama Luz helped him sel the 
rest. She and the others who had led the efort then went to the city and asked that a 
Community Action Board (JAC, Junta de Aci?n Comunal) be formed for their bario as 
a conduit for city services and resources. With that achieved, she became the first Vice 
President of the JAC for Lower Vila Liliam and ended up working with the JAC for the 
next 26 years. The board continues to be housed in the same building across the stret 
from her house. 
During the 1980s, as violence was growing in the bario, she took it upon herself 
to cal the police and report delinquent activity. The JAC worked hard to get the city to 
instal a police station there, and although the city finaly relented, it required that the 
community itself furnish it with desks, beds, and ofice supplies, since the city was not 
going to budget enough to keep it running. So the community donated what it could and 
the police began making rounds in the bario. But then something strange happened: they 
closed the doors and left one day, abandoning the barios to the violent gangs that were 
stil active. When Mama Luz went downtown to ask what had happened, a city oficial 
showed her the stack of leters he had received from people in the community 
complaining about the police presence. She believes the campaign had been organized by 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 152 
the bario?s criminal elements, because the police were geting in the way of their profits. 
The community was left to face the first and second gang wars on its own. 
Mama Luz earned her nickname during the third gang war of Caicedo La Siera, 
the one that started out as a fight betwen children around 1996 and grew into a war so 
violent that people from one bario couldn?t visit people in the next bario over if it was 
controlled by a diferent gang (? 4.2.1). ?You couldn?t go down one side, you couldn?t go 
down the other side, nor could other people come up the other, and it was just shocking.? 
But I had an advantage, and this is what they said in the administration, 
that since I had the preschool, almost everyone al around me knew me 
through their children. ? So some were the children?s relatives, others 
had been students of mine, others respected me because I had helped them 
out and served them and had worked so hard [on community-infrastructure 
projects]. So I had the advantage that many of the young people involved 
in this violence had a litle bit of respect for me, yeah. So when this 
violence broke out, they would send someone to warn me to go inside with 
the kids because they were going to start shooting. Then I?d go down and 
go where they were and talk to them. I?d tel them to think about their 
lives, and I started to talk to them, to the muchachos in the gangs. ? ?I?m 
not God to judge you,? I told them. ?I?m a friend to give advice, to nag 
you. I?m like a mom, I?m taking the place of your moms because they 
don?t have time to be with you because they have to work.? ? Then they 
said to me, ?So you?re Mam? Luz, our nag.? And since that day, it was a 
Sunday, ? everybody everywhere cals me Mama Luz.
62
 
She again tried to enlist the city?s help, warning them that another gang war was breaking 
out, but the people she talked to, even though they had known her for so many years, 
                                                
62
 ?? no pod?an bajar de un lado, no pod?an bajar al otro lado, ni los otros subir al otro lado, y eso fue 
impresionante. Pero yo tenia una ventaja, y eso lo dicen al? en la administraci?n, de que como ten?a el 
prescolar, casi toda esta gente alrededor me conoc?a por los ni?os. ? Entonces unos eran familia de los 
ni?os, otros hab?an sido alumnos m?os, otros me estimaban por lo que les colaboraba y les serv?a y porque 
estaba trabajando muy fuerte [en los proyectos] ?. Entonces ten?a muchas ventajas que muchos j?venes 
que estaban en esa violencia tuvieran un poco de respeto hacia m?, si. Entonces cuando ya se recrudeci? la 
violencia, entonces ya elos mandaban a decir que me encerara con los ni?os porque iban a empezar a 
disparar. Entonces yo, ya bajaba y me iba hacia donde estaban elos y hablaba con elos, le dec?a que, que 
pensaran elos de la vida, y empezaba a hablar con elos, con los muchachos de las bandas. ?Yo no soy Dios 
para juzgarlos,? les dec?a, ?soy una amiga para darles consejos, para rega?arlos, soy como una mam?, que 
yo les remplazo a ustedes la mam?, porque no tienen tiempo de estar con ustedes por estar trabajando.? ? 
Entonces al? me dijeron, ?Entonces eres Mam? Luz la rega?ona, para nosotros.? Y desde ese d?a, fue un 
domingo, entonces ya me pusieron ? todos Mam? Luz por todas partes.? ?Mama Luz,? Interview No. 1. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 153 
dismised her warnings as overblown fears and refused to send extra help. So she decided 
to invite them up for lunch: since it?s such a safe place, why not come up for a visit? 
It was a big deal: two members of the mayor?s cabinet came with their wives, 
along with other officials and guests, about 25 people in total. Mama Luz and others in 
the community had set up tables outside and had prepared lots of food. But then, just 
before one o?clock, just as they were about to serve the food, the lunch was interupted. 
?Then this idle boy ? when he saw this smal number of cars here and these few people 
? what a sin, this boy fired a shot from over there. And after this boy fired from over 
there, the boys from over here fired back?: 
And then al those people who were here from over at [city hal] didn?t 
know what to do, they didn?t know where to take cover, and right there, 
caling and caling [to se if they could get] the army and the police. So 
right there they learned first hand that, yes, it was true what I had been 
teling them, yeah. 
So Mama Luz took a young man who had been working in the community and they went 
toward where the first shot had been fired. They asked the ?idle boy? what the hel he 
thought he was doing. ?So he said to me, ?But Mama Luz, why you didn?t tel me you 
having that meting? I thought it was people coming to atack you there.? And I told him, 
?Because I didn?t have any reason to ask your permision. This is going to cost you 
dearly, because look what you?ve done, you did bad.??
63
 She went back to the tables ? 
somebody mentioned something about ?bullets for lunch? ? and finished serving the 
                                                
63
 ?Entonces un ni?o ocioso ? cuando vio ese poco de carros aqu? y ese poco de gente ? el ni?o, que 
pecado, hizo un disparo de al?. A lo que hizo el disparo de al?, los de ac? le respondieron. Y entonces toda 
esa gente que estaba aqu? de all? de la Alpujarra no sab?a que hacer, no sab?an donde se met?an, y all? 
mismo a lamar y a lamar y que si en un momentico estuvo esto leno de ejercito y de polic?a. Entonces ah? 
se dieron cuenta en carne propia, que si era verdad lo que yo estaba diciendo, si. ? Entonces me dijo, ?Pero 
Mam? Luz, porque no me dijites que ten?as esa reuni?n. Yo cre? que era gente ven?a atacarte ah? a vos.? Y 
le dije, ?Es que yo no tenia porque pedirle permiso a usted, y esto te va a salir caro a vos porque mira lo que 
hiciste y estuvo mal hecho.?? Ibid. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 154 
food, which everyone ate politely but quickly before high-tailing it out of there. Very 
soon thereafter, the chief of police inspectors for the department of Antioquia told Mama 
Luz that a police inspector was being asigned to the area to take citizen complaints, keep 
the peace, and enforce the law ? basicaly, the job that until then had been the domain of 
the ?muchachos? who controlled the area. It wasn?t a police station like they?d had 
before, but it was something. 
But the city also began training 35 people from the area as ?peace managers? 
(gerentes de paz). The peace managers, Mama Luz among them, spent six months 
studying laws and methods for working in violent communities. Members of the 
community, with some support from the city and non-governmental organizations 
(NGOs), began reaching out to the gangsters to try to get them, eventualy, to agre to a 
cease fire and some rules of engagement to protect the community. At the time, the city 
was offering communities resources and training, through a program it had initiated with 
the help of the Catholic Church, to mediate ?peaceful coexistence pacts? (pactos de 
convivencia) betwen gangs. These micro-pacts were ?not writen down, but rather were 
based on the word of one gang to another?: 
Within the internal laws of the gangs, the respect for the spoken word has 
a special value. ?La palabra de hombre, the word of a man, that is what it 
is about.? The effect of the first pactos de convivencia was an imediate 
decrease in the number of violent deaths, especialy among youth. For this 
reason the municipal authorities decided to back the initiative through an 
Oficina de Paz y Convivencia (Office for Peace and Coexistence) that 
took part in the negotiations and provided some financial support. ? [It] 
is believed that the pacts saved many lives of youth in the 
neighbourhoods.
64
 
                                                
64
 Ralph Rozema, ?Urban DR-Proceses? (references omited). 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 155 
With renewed support, Mama Luz doubled her own eforts, approaching the 
gangsters, as she had in the past, as their favorite nag. She worked with those from La 
Ca?ada (led by Alberto), which controlled the teritory where she lived and worked; M-
6&7 (led by Hugo), which controlled a large teritory above; and La Arenera (led by 
Gamuza) and Las Mirlas (led by someone whose name she could not recal), both in 
teritory adjoining that of La Ca?ada. She had no contact with the gang at Santa Luc?a, 
possibly because, in her teling (those in Santa Luc?a would probably tel it diferently), 
the third gang war had been started when a child in Santa Luc?a hit a child in her 
neighborhood; she said al she knew about them was that they came out to fire their guns 
then went back, and that some of them had once been students of hers. 
Stil, the JAC she worked with was throwing monthly block parties, with live 
music and dancing, food and drink, and they always invited people from the surrounding 
barios to atend as a way to improve trust and dialog. ?But before doing this, I went out 
to where those four groups were. ? I?d go out one or two days beforehand, [to tel them]: 
?I want to have a party and al of you are invited, with one condition: that nobody tokes 
up on me, that nobody starts any trouble on me, because you know I?l bring in the police, 
and I won?t even defend you.? So now that I had these groups? consent, I?d go and put the 
party together.?
65
 At one of these parties, on 15 December 1997, people were dancing 
and enjoying themselves when some guys from La Ca?ada started walking over to where 
some guys from La Arenera were standing. Everyone at the party nearly panicked, 
                                                
65
 ?Pero antes de yo hacer eso, me iba para donde los cuatro grupos, porque yo trabajaba con cuatro grupos, 
me iba uno o dos d?as antes: quiero hacer una fiesta y al? est?n invitados ustedes, con una condici?n: que 
nadie me fuma un cacho, ni nadie me forma un desorden, porque ustedes sabes que les traigo la polic?a y no 
los defiendo. Entonces ya con ese consentimiento de los cuatro grupos, yo hacia esa fiesta.? ?Mama Luz,? 
Interview No. 1. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 156 
thinking there was about to be a confrontation. So Mama Luz, hearing some commotion, 
ran over to where the gangsters were heading toward each other, put herself physicaly in 
betwen them, and demanded to know what was going on. But one of them told her not 
to worry: they just wanted to make peace. Seing that there wasn?t going to be any 
trouble, Mama Luz told everyone that everything was fine, they could keep dancing. But 
they didn?t make peace that day. 
Four days later, however, at the end of the day, after she had just crawled in bed, 
she heard someone caling her name: ???Mama Luz, Mama Luz!? I said, ?God, what?s 
happening now!? [They shouted,] ?Come out, Gamuza is asking for you, he needs you, 
it?s urgent.?? At the time, two police officers were renting out the upstairs rooms in her 
house and asked her what was going on. She responded: ??Nothing, Gamuza?s caling for 
me.? ?Be safe,? [the policemen said.] We?re right here.? I got ready and left.? She 
continued the story: 
I said to [the people waiting outside], ?What?s going on now?? 
?It?s just that the muchachos want to make peace, so let?s se. What are 
you going to do? Because we have a few former gangsters who left 
Hugo?s group and are now with Alberto?s. So what?s going to happen to 
their families, or to mine?? Gamuza told me. 
?Let?s se what we can do,? I said. We headed up to where there?s a park, 
? found two kids, and told them to go get Hugo, who was at La Ramada, 
this place where al of them from up here hang out [in Upper Vila 
Liliam]. ? ?I need him, it?s urgent.? And he comes down with five men. 
?What?s up, Mama Luz? What?s going on?? 
?Come on, let?s talk. The guys from La Arenera want to make peace.?
66
 
                                                
66
 ???Mam? Luz, Mam? Luz!? Dije, ??Dios! ?Qu? pas? ahora?? ?Que vaya que Gamuza la lama que la 
necesita urgente!? Entonces me dijeron los agentes de polic?a, ??Qu? pas? Mama Luz?? ?No, que Gamuza 
me mand? a lamar.? ?Vaya tranquila, que aqu? estamos nosotros.? Aranque y fui. Y les dije, ??Qu? pas? 
ahora?? ?No, es que los muchachos quieren hacer la paz, entonces a ver. ?Qu? va a hacer usted? Porque 
aqu? hay unos reinsertados que se salieron de donde Hugo y est?n donde Alberto, entonces ?qu? va a pasar 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 157 
They talked on the second floor of a dark house, with Hugo?s men standing guard 
out front and two old women saying the rosary, praying for Mama Luz?s safety. Hugo 
told her to go talk to Alberto and said that he?d talk to his own men, so that everybody 
could then sit down together. She left Hugo and Gamuza and in the dark of night went 
back down to her own neighborhood looking for Alberto, asking people on the stret 
where he was, and finaly found him. 
So he came out himself and he says to me, ?Mama Luz, what are you 
doing out here at this hour? What do you need?? 
?I?m just coming here to talk with you, because the guys from La Arenera 
and Hugo?s guys want to make peace. Hugo told me he doesn?t want even 
a single shot to be fired anymore, and he wants you to come so he can talk 
to you.? 
He said to me, ?OK, Mama Luz, that?s what we want, peace, peace in the 
bario. So go [and tel him] I?m going to talk with al the guys.?
67
 
It?s prety clear that Mama Luz has idealized and dramatized much of the dialog. 
But she?s rightfully proud of her acomplishment: she helped get most of the gang and 
militia leaders in Caicedo La Siera together in the middle of the night and managed to 
mediate talks that led to a cease-fire and non-aggresion pact. The agrement among the 
gangs was that they would not do drugs where children and members of the community 
could se them; they would respect each other?s teritory as sovereign; they would keep 
children out of the war; they would not alow personal conflicts to devolve into gang 
                                                                                                                                            
ahora con su familia y la m?a?? me dijo el Gamuza. ?Vamos a ver que hacemos,? le dije. Arancamos hacia 
arriba que hay un parque, que yo les constru? a los muchachos, busque a dos pelados y le dije vaya l?meme 
a Hugo que esta en La Ramada, un sitio donde se reun?an todos elos los de ariba. lo necesito urgente.? Y 
hay mismo el bajo con cinco hombres. ?A ver, Mam? Luz. ?Qu? pas??.? ?Venga, vamos a hablar que los 
muchachos de la Arenera quieren la paz.?? Ibid. 
67
 ?Entonces hay mismo sali? ?l y me dice, ?Mam? Luz, ?qu? hace por aqu? ha estas horas? ?Qu? quiere?? 
?Es que vengo hablar con usted, porque los muchacho de la Arenera y los muchachos de Hugo quieren la 
paz, y Hugo me dijo que no quiere que disparen un solo tiro, y que viniera a hablar con usted.? Me dijo, 
?Listo, Mam? Luz, eso es lo que queremos, la paz, la paz en el bario. Entonces vaya usted, que est? metida 
en todo esto, arranque que yo voy a hablar con todo los muchachos.? Ibid. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 158 
wars; and they would atend training programs hosted by the mayor?s office. To protect 
the communities, they agred to hold their fire on al public holidays and alow fre 
movement by civilians through the barios.
68
 
With the agrement in place that night, the gangs al decided to celebrate together. 
Alberto and his men got in their cars and rounded up some people to party; Hugo got 
some people to come down from La Siera to join them. Mama Luz woke the neighbors 
to borrow music and speakers and tables and chairs and plates and spoons, and people 
started dancing and drinking even before they started cooking. Mama Luz grabbed a 
couple of young women and asked them to stop dancing and help the neighbors prepare 
sancocho, a type of stew traditionaly made at community gatherings, and they al went 
and found some shopkeepers to supply the makings for it. ?At thre in the morning we 
had thre ketles of sancocho ready,? Mama Luz said proudly, adding that even though 
she had recently had surgery, she was so happy that night that she felt no pain. If anybody 
got too drunk, Alberto had his men with cars ready to drive them home safely. The party 
was heard throughout Caicedo La Siera, and soon people from al over the sector started 
showing up ? even some from Santa Luc?a joined the celebration, and others from 
across the crek in Vilatina. Somebody caled the television news stations to tel them 
about this miracle that was happening around them, but none ever sent a crew: they didn?t 
dare enter the war zone in the middle of the night. The police inspector asked what was 
going on, but he didn?t stop the party, nor did he join in. In fact, nobody from outside of 
the waring communities was there, and the only authorities who ever showed up that 
                                                
68
 Ramiro Alberto V?lez Rivera, et al., Governabilidad Local en Medel?n: Configuraci?n de 
Teritorialidades, Conflictos y Ciudad, 158. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 159 
night, Mama Luz said, were ?God, who was helping me; and the [gang] leaders, who 
were there with their muchachos; and the muchachos, who wanted peace.?
69
 
I wish I didn?t have to write the next paragraph, but anyone who has read this far 
can probably se what?s coming. The midnight peace talks, the unexpected truce, the 
spontaneous al-night party that brought divided comunities and waring gangs together 
for one night of extravagantly unjustified hope, took place on 19 December 1997. With 
even the most cursory look at the time series for homicides (the only violence data 
available for the sector), anyone could se that the third gang war was only just 
beginning, the worst of its violence stil to come. And the seds for the fourth war were, 
in a way, being planted that very night: some of the gangs from ?below? ? La Ca?ada, 
La Arenera, Las Mirlas, and Santa Luc?a ? would end up not just sticking to various 
non-aggresion pacts (broken a few times along the way) but within two years would 
actualy join forces in an even higher-stakes war against the militia from ?above?: M-
6&7.
70
 
Despite the eforts of Mama Luz and many other community leaders, peace 
managers, and outsiders, the cease-fires, truces, and micro-pacts they negotiated never 
led to lasting peace there. Throughout the city, in fact, the peaceful coexistence pacts 
sponsored by the city, the Church, and the peace managers had limited efects; they 
                                                
69
 ?Eso quiere decir que a las tres de ma?ana ya ten?amos tres olas de sancocho listas. ? Solamente lo 
estaba enfrentando Dios, que me estaba ayudando; y los l?deres que estaban con los muchachos; y los 
muchachos que quer?an la paz.? ?Mama Luz,? Interview No. 1. 
70
 There is some question about the exact date of the peace agrement in Caicedo La Siera. Mama Luz said 
her party took place on 19 December 1997, but local media were reporting that another party ? celebrating 
a peace pact between M-6&7 and La Ca?ada ? tok place thre months later, on 22 March 198. It is 
possible Mama Luz got her dates wrong; it is equaly possible, however, that Mama Luz?s party was a 
spontaneous celebration of an informal agrement that was then formalized, and celebrated again, a few 
months later. Whatever the case, the exact timing of the parties does not change the analysis: the pact or 
pacts did not last very long and the wars continued in 1998. Se El Tiempo, ?Fiesta por la Paz en Barrio de 
Medel?n,? 23 March 198. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 160 
certainly saved a lot of lives, but the pacts themselves never lasted more than few 
months, and the city suspected the program in 1998. Ralph Rozema has noted the 
program?s shortcomings: 
A fundamental problem for the local government was that it was 
negotiating with criminal groups without a political or legal framework 
within which to do so. Many of the gang members had commited crimes 
that were not prosecuted and were even able to continue criminal 
activities, although no longer within their own neighbourhood. Moreover, 
they stil possesed their weapons, as the pacts did not involve 
disarmament of the groups. ? Without writen commitments the proces 
remained particularly vulnerable. ? [To addres] the lack of opportunities 
for youth ? the pacts [had] included strategies for education and the 
establishment of micro-busineses ? but the initiatives were few due to a 
lack of funds.
71
 
The micro-pacts had been short-term solutions to long-term problems. As the 
milennium came to a close, the problems were only going to get worse: not just the 
behavior of the militants and gangsters toward the communities they lived in, not just the 
intervention of the paras that was alternately welcomed and feared, and not just the 
growing involvement of the security forces in the batles against the guerilas and 
militants, but al of it, the violence, the fear, the stigmatization and marginalization of the 
periphery, the displaced families, the broken lives, the casualties of the wars that semed 
like they?d never end: al would only get worse before they would get beter. 
And then they would get worse again. 
4.2.4. The Ilegitimation of the Militias 
It was the day after the Feast of the Imaculate Conception, and Elizabeth was 
crouched in front of her house scraping the candle wax off the ground from her vigil the 
                                                
71
 Ralph Rozema, ?Urban DR-Proceses? (references omited). 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 161 
night before. She had just sent her children around the corner to fetch some groceries 
when she heard gunfire directly in front of her house. She looked up. Thre men were in 
the stret. One was on the ground, another kneeling over him. The one on the ground was 
bleding. The one who was kneeling stood up and, seing Elizabeth watching them, 
walked over to her and stuck his pistol to her head. The third man came over and shouted 
at him: ?We deal with what?s important. You?re not going to kil her.? The one with the 
pistol looked at her hard. ?I remember he had a very ugly scar on his face. ? He threw 
me into the house, and I don?t even know how I closed the door. He left me smeared with 
blood. I just about went crazy.? She heard two more shots ? the man on the ground was 
certainly dead now ? and soon her children returned, hysterical, to find their mother 
covered in blood, herself half crazy. They thought she?d been shot. 
Some days, maybe weks, later, the third man, the one who had saved her from 
the man with the scar, knocked on Elizabeth?s door. ?Ma?am, it?s best that you leave,? he 
told her. ?That man?s waiting to kil you. He?s got himself in the coffe field keeping a 
lookout for you. I suggest you get out.? She refused to leave her house, her bario: this is 
my home. But he returned a wek later. ?Today?s the day they?re going to kil you. Get 
out of here.? Afraid, she left that night, taking her kids with her to their father?s house in 
Itag??, south of the city. She found out later that six men had shown up at her house, 
kicked in the door, and shot the place up. A wek later they moved in, and as the gang 
wars heated up in the next months and over the next years, Elizabeth kept hearing 
through neighbors that they were turning her house to uses that horrified her: ?They raped 
there, they hid weapons, they brought in people tied up.?
72
 
                                                
72
 ?? ?nos metemos en la grande, no la vas a matar.? ? Recuerdo que ten?a una cicatriz muy fea en la cara. 
El tipo me tir? para adentro de la casa y no se c?mo cer? la puerta. Me dej? untada de sangre. Yo me puse 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 162 
It was January 1998. The men who took over Elizabeth?s home in bario La Siera 
were from M-6&7, the ones who used to be the good guys, who had cleaned up the 
bario, shut down the gangs and drug dealers, did social work, moved the drug use of the 
corner so children wouldn?t se it, and most of al protected the community from people 
like the men who displaced Elizabeth from her home. They had changed, become what 
they?d once opposed. An M-6&7 commander named Fernando told Arleison Arcos Rivas 
in 1998 that the social cleansing campaigns had been a mistake: 
I think that for us it was a mistake for us to dispose of other people?s lives. 
Our dispute wasn?t realy with those kids. You have to give them a 
goddamned chance, everybody, a chance to live. One problem, there?s no 
jobs, there?s no upbringing, there?s no motivation.
73
 
Hugo, the founder of M-6&7, expresed les regret to the same interviewer, suggesting 
that the state and the paramilitaries were doing the same thing. But even with that he 
recognized that social cleansing was not sustainable. ?I think that, as an armed project 
and as a political structure, we can?t keep doing that sort of work, but we also can?t just 
let anyone do whatever they want to the comunities day in and day out.?
74
 Yet by the 
mid-1990s, their victims were including homosexuals, drunks, drug addicts, bad fathers 
? people who were sons and daughters and cousins of families who lived in the 
                                                                                                                                            
para enloquecerme. ? [El otro hombre la dijo:] ?Se?ora, es mejor que se vaya. El hombre ese la est? 
esperando para matarla, ?l se mantiene metido en el cafetal vigil?ndola. Yo le aconsejo que se vaya. ? Hoy 
es el d?a que la van a matar. V?yase.? ? Al? [en la casa] violaban, escond?an armas, levaban gente 
amarrada.? Elizabeth P?rez, ?Dos Muertos Que Marcaron Mi Vida,? [?Two Deaths that Marked Me for 
Life,?] in Jam?s Olvidar? Tu Nombre, ed. Patricia Nieto (edel?n, Colombia: Alcald?a de Medell?n, 
2006), 20. 
73
 ?Ah, yo creo que eso para nosotros es un eror que nosotros dispongamos de la vida de otros. La pelea no 
es contra esos pelaos. Hay que darle la hijo de puta oportunidad, a todo el mundo le dieron la oportunidad 
de vivir. Un problema, no hay empleo, no hay educaci?n, no hay motivaci?n.? Punctuation edited for 
clarity. Quoted in Arleison Arcos Rivas, ?Ciudadan?a Armada,? 203. 
74
 ?Yo pienso que nosotros, como proyectos armados y como estructura pol?tica, no podemos seguir 
haciendo ese tipo de trabajos. Pero tampoco podemos dejar a las comunidades hoy por hoy a que cualquiera 
haga con elos lo que quiera.? Punctuation edited for clarity. Quoted in Ibid., 208. 
Lamb, Robert D. | Chapter 4. Gangs and the Coruption of Guerila Militias: 192-1998 163 
communities. It?s not that people approved of or even realy wanted to tolerate the 
behavior or the lifestyles of the social-cleansing victims, but they were family, and 
nobody wanted a family member kiled just because she was a lesbian, or a prostitute, or 
just because he was a drunk, or a drug dealer: among the potential responses to social 
disapproval, there were alternatives to the death penalty. Moreover, for some of the 
militants, as Elizabeth?s experience demonstrates, the power had simply gone to their 
heads. They were out of control. The militias were doing practicaly whatever they 
wanted to anyone they wanted ? in Hugo?s words: ?day in and day out.? 
By the end of the 1990s, the people who lived in M-6&7?s teritory had had 
enough. The militias had overeached. They had abused their power and transgresed the 
boundaries of what was tolerable and aceptable to the communities that had once 
embraced them as local heroes. The militants ? as wel as the community ? would end 
up paying a high price for their ilegitimacy: By doing what they did, by becoming what 
they had become, they had inadvertently unlocked the doors to the bario just as the paras 
were about to come knocking. 
 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 164 
Chapter 5. Mafias and the Emergence of Paramilitaries: 1998-2002 
Diego Fernando Murilo Bejarano had a job to do. A group of about 30 disident 
student radicals had recently broken away from the insurgent group the People?s 
Liberation Army
1
 (EPL: Ej?rcito Popular de Liberaci?n) to form their own group, which 
they named Red Star (Estrela Roja). The first thing they needed was money to finance 
their organization, and so one of their first acts was to rob a supermarket in Itag??. The 
operation was a succes: they got their money. But the students apparently hadn?t done 
their homework. If they had, they probably would not have chosen that particular 
busines to rob, not if they had known what was good for them. The supermarket was 
owned by Fernando and Mario Galeano. The Galeano brothers, along with the Moncada 
brothers, Gerardo and Wiliam Julio, operated a powerful drug-traficking organization 
headquartered just south of Medel?n that was part of Pablo Escobar?s Medel?n Cartel. 
Diego Murilo [?Moo-REE-jhoe?] worked for Fernando Galeano as a driver and 
bodyguard, and he was given the job of finding the students and making sure they would 
never be able to take money from the Galeano family again.
2
 
                                                
1
 Sometimes translated as the Popular Liberation Army. 
2
 Biographical information about Diego Murilo comes from: El Tiempo, ??Qui?n Es ?Don Berna? o 
?Adolfo Paz,? Inspector General de las Autodefensas Unidas de Colombia?? [?Who Is ?Don Berna? or 
?Adolfo Paz,? Inspector General of the United Self-Defense Forces of Colombia??] 20 September 203; 
Semana, ?Los Secretos de ?Don Berna?,? [?The Secrets of ?Don Berna?,?] 14 Julio 207, 
http:/ww.semana.com/wf_InfoArticulo.aspx?IdArt=104961 (acesed 13 April 2009); 
VerdadAbierta.com, ??Don Berna,? Diego Fernando Murilo Bejarano,? Paramilitares y Conflicto Armado 
en Colombia [Paramilitaries and Armed Conflict in Colombia], published electronicaly by Fundaci?n 
Ideas para la Paz (FIP) and Semana, htp:/ww.verdadabierta.com/web3/victimarios/los-jefes (acesed 13 
April 209); comon knowledge in Medel?n; and other sources as cited. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 165 
It was 1984, and Murilo was already familiar with the students? outlook: he 
himself had joined the EPL as a tenager in the department of Vale del Cauca, in the 
west of Colombia, in the late 1970s. He was a member for only a few years before 
everyone else in his unit was kiled by narcotrafickers out of revenge for their having 
kidnapped a colleague. He should have been kiled, too, but he escaped their fate and 
moved to Medel?n instead. There, he met Galeano, earned his trust, and got the security 
job. He was smart and relentles, and soon the Red Star students knew he was after them. 
They decided on a premptive atack. They sent a few of their members to set up an 
ambush and, with orders to make sure Murilo would not survive, the gunmen opened 
fire. Within seconds, Murilo was lying on the ground, blood pouring out of the holes 
bored into his body by 17 bullets. He was a dead man as far as his kilers knew: who 
could survive 17 bullets? It turns out that Diego Murilo could. He lost a leg, but he kept 
his life, cheating death for a second time. The Red Stars were not so lucky: by the time 
the Galeanos were done with them, there was nothing left of the organization. The 
disident group?s leader was riddled with bullets at a University of Antioquia cafeteria 
before being dragged into the stret and publicly mutilated under the tires of a Galeano 
vehicle. At least 12 others were likewise asasinated by the end of 1985; some reports 
suggest every Red Star member was hunted down and that none survived. A rising star in 
the Galeano-Moncada organization, Murilo recovered, his own survival, and the loss of 
his leg, only enhancing his reputation as a man to be feared. 
Murilo?s bosses were asasinated by Pablo Escobar during the drug lord?s time 
at the Cathedral (his luxury prison) in a dispute over money (se ? 2.1.3). Escobar had 
been the Medel?n Cartel?s enforcer, and al member organizations paid him a ?war tax? 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 166 
to, among other things, maintain a favorable political and legal climate in Colombia. In 
his view, he had sacrificed his fredom to get Colombia?s political establishment to 
outlaw extraditions to the United States, to the benefit of al cartel members, and Escobar 
wanted more money from the cartel in return. The Galeanos and Moncadas agred to 
increase their monthly payments from $200,000 to $1 milion. But Escobar wanted even 
more and ordered his men to steal about $20 milion he knew the two families had in 
storage. When the brothers complained, Escobar had them tortured, kiled, and 
mutilated.
3
 Fernando Galeano and Gerardo Moncada were executed when they went to 
the Cathedral; their brothers were kidnapped and kiled later. As Fernando?s bodyguard, 
Diego Murilo should have gone to the Cathedral with him; if he had, he, too, surely 
would have been kiled. But Fernando had asked him to look after Mrs. Galeano that day 
(legend has it that he was asked to take her to the beauty salon). And so, for the third time 
in his life, Murilo had cheated death. 
When Escobar escaped from prison in July 1992, Murilo, feling unsafe in 
Medel?n, returned to his roots in the department of Vale del Cauca, where he had been 
born in 1961, to se if he could find work with drug trafickers in that region, maybe the 
cartel that operated out of the city of Cali in the southern part of the department, or the 
North Valey Cartel that operated in the Norte del Vale region up north. There he met the 
paramilitary leader Fidel Casta?o ? a former friend of Escobar?s who also had been 
invited to go to the Cathedral on that fateful day, but had wisely declined the invitation 
? and with Casta?o and some members of the Cali Cartel, Murilo co-founded the Pepes, 
the death squad that liquidated Escobar?s organizations, asociates, possesions, and life. 
                                                
3
 Patrick Clawson and Renselaer W. Le, The Andean Cocaine Industry. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 167 
To maintain confidentiality while communicating with underworld figures and the 
government?s Search Bloc, members of the Pepes adopted aliases. Murilo chose ?Berna? 
as his nom de guere, probably a phonetic variation of his middle name, Fernando. As 
Berna?s power rose in Medel?n, people refered to him with the Spanish honorific don 
and despite his later eforts to get people to cal him ?Adolfo Paz? (Adolph Peacemaker) 
he always was, and stil is, popularly known as ?Don Berna.? 
After Escobar?s death on the rooftop in December 1993, ?Don Berna? Murilo 
began rebuilding the empire that Escobar had left behind, and within a few years he 
would succed ? and then join forces with the surviving Casta?o brothers in a grand 
narco-paramilitary aliance. Fidel Casta?o died sometime in 1994 under sketchy 
circumstances; his brother Carlos claims he was kiled in a batle against guerilas in 
January of that year, yet in May the Colombian newswekly Semana published an 
interview ith Fidel that it said had been taken ?a few days ago.?
4
 Two other acounts, 
more conspiratorial in tone, are that Carlos kiled his brother so he could take control of 
the paramilitaries himself, or that Fidel faked his own death and even today is living 
abroad somewhere. In any event, Fidel disappeared, and Carlos took over leadership of 
the family?s paramilitary empire, known as La Casa Casta?o (the Casta?o House). As 
Carlos Casta?o expanded the reach of that empire in the countryside, ?Don Berna? 
Murilo began building his own empire. 
He began by returning to Medel?n and, symbolicaly, seting up shop in 
Envigado, Escobar?s home town just south of the city and just across the river from the 
town of Itag??, where the Galeano-Moncada organization had been based. He wasted no 
                                                
4
 Semana, ?Yo Fui el Creador de los Pepes,? [?I Was the Creator of the Pepes,?] 31 May 194, 38. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 168 
time approaching the remnants of Escobar?s best enforcement organizations and 
traficking networks and informing them that they now worked for him. He did not 
operate appreciably diferently from Escobar in this regard, except that he kept a much 
lower profile (he did not get involved in politics or social work, for example) and he 
structured his organization more loosely than the Medel?n Cartel, as what locals refered 
to as an oficina (busines office) structure, operating as a clearinghouse for the 
Colombian underworld by connecting customers to contractors and taking a cut of the 
revenue. This was considered a necesary change from the way the cartels had operated. 
After Escobar?s death, ?groups involved in the ilicit trade in drugs became atomized, but 
in a move typical of busineses that can?t sustain themselves as a microenterprises, they 
coordinated themselves in les hierarchical and les visible networks whose purpose was 
to guarante the efectivenes of their operations, whether comercial or violent.? In the 
mid-1990s, in other words, the underworld reorganized itself into a network of 
independent nodes, albeit with some nodes having greater influence than others: 
Now these gangs are no longer mere appendages to mafia structures, but 
begin to operate as armed microenterprises with the capacity to sel their 
services to the highest bidder. [Alongside them] are the large organized 
crime structures with the capacity to operate as intermediaries betwen the 
world of the oficinas and the world of the gangs. The Terace Gang, La 
Ca?ada, the Triana Gang, Frank?s Gang, and hit squads like the Chiquis 
are the most significant examples.
5
 
                                                
5
 ?.. los grupos dedicados al negocio ilegal de las drogas se atomizaron pero, en una aci?n propia de un 
negocio que no puede sostenerse con formas microempresariales, tambi?n se coordinaron en redes menos 
jer?rquicas y visibles que ten?an como finalidad garantizar la eficacia de la aci?n comercial y violenta. .. 
?stas [bandas] ya no figuran m?s como ap?ndices de las estructuras mafiosas, sino que comienzan a operar 
como microempresas armadas con capacidad de vender sus servicios al mejor postor. El segundo aspecto es 
la aparici?n de grandes estructuras del crimen organizado con capacidad de operar como intermediarios 
entre el mundo de las oficinas y el mundo de las bandas. La Terraza, La Ca?ada, la banda de los Triana, la 
banda de Frank y grupos de sicarios, como los Chiquis, constituyen los ejemplos m?s significativos de este 
tipo de estructuras armadas.? Manuel A. Alonso Espinal, Jorge Giraldo Ram?rez, and Diego Jorge Siera, 
?Medell?n: El Complejo Camino de la Competencia Armada,? [?Medell?n: The Complex Course of Armed 
Competition,?] in Parapol?tica: La Ruta de la Expansi?n Paramilitar y los Acuerdos Pol?ticos, ed. 
Mauricio Romero (Bogot?: Intermedio/Corporaci?n Nuevo Arco Iris, 207), 121. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 169 
Murilo?s organization became known as the Envigado Ofice (Oficina de Envigado) and 
everyone knew Murilo was positioning it as the succesor to the Medel?n Cartel. Most 
of the contractors he worked with ? growers, producers, transporters, hit squads, bank 
robbers, protection racketers, etc. ? were those independently owned and operated 
microenterprises, but Murilo exercised great influence over their activities nonetheles, 
nowhere more so than in Medel?n: one report claimed, with clear exaggeration, that ?not 
one asault, robbery, mugging, or murder was commited in Medel?n without his 
consent.?
6
 Anyone who got on his wrong side ? who failed to fulfil their contractual 
obligations, pay him his cut of their criminal proceds, or comply with his extortionist 
demands ? risked death at the hands of one of the most powerful organized crime 
groups in the city, whom he used as his main enforcers: the Terace Gang (La Teraza). 
The Terace Gang, headed by a man named Elkin S?nchez Mena (known as ?El 
Negro Elkin?),
7
 was an extremely skiled and disciplined criminal organization that got its 
start as a gang in bario Manrique, in Medel?n?s tough Northeastern zone. It got its name 
from an ice-cream shop where a masacre had taken place, worked for Pablo Escobar in 
the late 1980s and then against him in the early 1990s, and by 1997 had grown into one 
of the most feared organized crime structures in the country.
8
 With about 200 members, 
they caried out high-profile asasinations, kidnappings, car bombings, and bank and 
armored-truck robberies, not only in Medel?n but in Bogot? and elsewhere, and not only 
                                                
6
 ?? ning?n asalto, robo, atraco, o asesinato se comet?a en Medel?n sin su consentimiento.? Quoted in 
Semana, ?Contra la oficina del teror,? [?Against the Ofice of Teror,?] 1 June 205. 
7
 His name is sometimes published as Luis S?nchez Mena, but I have ben unable to find a definitive 
source for his birth name, so the most comon version is used here. 
8
 C?sar Mendoza Gonz?lez, Interview No. 6; Marta In?s Vila Mart?nez, Luz Amparo S?nchez Medina, and 
Ana ar?a Jaramilo Arbel?ez, Rostros del Miedo: Una Investigaci?n Sobre los Miedos Sociales Urbanos, 
33. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 170 
for Murilo but for other customers such as Carlos Casta?o (of course, Murilo got a cut). 
It also kept Medel?n?s smaler crime gangs in line. ?Nothing moved without its 
authorization,? a former militant with underworld contacts told me, recording the 
conversation to make sure I would not misquote him. He was exaggerating, but he made 
his point: Terace was a tough gang, and ?El Negro Elkin? was a powerful mob boss.
9
 
Even with the Terace Gang on his side, however, ?Don Berna? Murilo did not 
yet have the monopoly over Medel?n?s underworld that he wanted, not by a long shot. 
The leftist militias, and the gangs that kept pretending to be leftist militias, were stil in 
control of many of the city?s peripheral barios throughout most of the 1990s, and the big 
criminal organizations ? Frank?s Gang, the Triana Gang, La Ca?ada ? and the smal 
neighborhood gangs continued to maintain their independence. The criminal groups were 
of some concern to Murilo, as they represented an opportunity for competing traficking 
networks to make inroads into his teritory. There continued to be gang wars over micro-
teritories and ilicit markets, and Murilo intervened in many of them, backing one side 
or another to be sure his influence would continue to grow. 
Of greater concern to Casta?o, however, was the continued presence of the leftist 
groups in Medel?n: his grudge against guerilas held, and he wanted to eliminate them 
wherever he could. His friendship with Murilo and the militias? own ilegitimation 
within the communities they controlled provided him the perfect opportunity to try to 
eject them from the city. In 1998, he would make his move by taking advantage of a 
structure he had created as part of his family?s grand strategy to defeat insurgent 
guerilas throughout the country (se next section). 
                                                
9
 Luis Fernando Quijano Moreno, Interview No. 5. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 171 
Litle is known about the early life of the Casta?o brothers ? Fidel, Carlos, and 
Vicente ? except that they grew up in northern Antioquia in an area near the Gulf of 
Urab? that was under the control of the Revolutionary Armed Forces of Colombia 
(FARC: Fuerzas Armadas Revolucionarios de Colombia). By the late 1970s, Fidel, the 
eldest brother, who had traveled the world, had become a wealthy landowner in the 
departments of Antioquia and C?rdoba and a traficker in stolen art and narcotics, at 
times working closely with Escobar and the Medel?n Cartel. But he became the country?s 
leading anti-guerila after the FARC kidnapped his father in 1979 and kiled him in 1980 
despite having received part of the ransom payment from the hostage?s family. Together 
with some friends, farmhands, and off-duty soldiers, the Casta?o brothers took their 
vengeance on the local guerilas they believed had been responsible, then set their sights 
on a broader anti-guerila program.
10
 Their group became known as Los Tangueros (The 
Tango Dancers, a play on the name of the Casta?os? ranch, Las Tangas). When they 
expanded and formalized this paramilitary group in 1987, they named it the Peasant Self-
Defense Forces of C?rdoba and Urab? (ACU: Autodefensas Campesinas de C?rdoba y 
Urab?), after the department of C?rdoba and the Gulf of Urab?, the regions where they 
operated. The ACU was notoriously brutal as a death squad that targeted guerilas, 
leftist sympathizers, and innocent people with the misfortune of having lived where 
guerilas had operated (Fidel Casta?o picked up the alias ?Rambo? during this period). Its 
signature tactic was the masacre, after which bodies were often left behind that had been 
beheaded or disemboweled as a warning to survivors that they should consider 
                                                
10
 VerdadAbierta.com, ??Rambo,? Fidel Casta?o Gil,? Paramilitares y Conflicto Armado en Colombia 
[Paramilitaries and Armed Conflict in Colombia], published electronicaly by Fundaci?n Ideas para la Paz 
(FIP) and Semana, htp:/ww.verdadabierta.com/web3/victimarios/los-jefes (acesed 30 July 209). 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 172 
abandoning their land. In truth, while much of the ACU?s work was truly anti-
subversive, many such atacks had nothing at al to do with guerilas: the politics was 
often used as an excuse for a coercive land grab or to setle old scores, a common 
occurrence in iregular wars throughout history.
11
 
The ACU was neither the only nor the first paramilitary group to operate in 
Colombia. While its most direct ancestors were the MAS death squad founded by 
Escobar in the early 1980s and the Pepes death squad that hunted Escobar in the early 
1990s (se Chapter 2), self-defense groups and paramilitaries have had a long history in 
the country. During the 1960s, when the guerila war began, the Colombian military 
realized that it did not have the personnel or resources to counter the guerilas where 
they operated: Colombia is a very large country with very dificult terain, including 
step mountains, thick jungles, and vast, hot lowlands. There was no way the central 
government could protect al of its citizens throughout the country, as it historicaly had 
had litle or no presence in much of it. Many vilages were left to their own devices, as 
were wealthy landowners and large oil and mining corporations, and they armed 
themselves in self-defense. At the same time, at the advice of the U.S. military ? which 
was none too hesitant to help any wiling government in the Western Hemisphere counter 
their domestic Communists ? the Colombian military recruited civilians in guerila-
controlled teritory to support the security forces as informants and guides. This approach 
was bolstered by a 1965 presidential decre encouraging citizens to organize resistance 
against guerila influence and a 1968 law that provided the legal framework through 
which citizens could form legitimate civil-defense militias. During the 1970s, wealthy 
                                                
11
 Stathis N. Kalyvas, ?The Ontology of ?Political Violence?: Action and Identity in Civil Wars,? 
Perspectives on Politics 1, no. 3 (203): 475. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 173 
landowners, including smugglers, narcotrafickers, oilmen, and miners, formed their own 
private security forces as bodyguards and to protect their property, justifying it, when 
necesary, by reference to the 1968 law. But in 1989 the Colombian Supreme Court ruled 
that that law as invalid, and the government quickly criminalized the civil-defense 
groups. But that did not stop them from continuing to operate. 
It also did not stop the government from finding other ways to make 
paramilitaries legal. A presidential decre was isued on 11 February 1994 authorizing 
and regulating the creation of nonstate organizations to provide ?special vigilance and 
private security services? (servicios especiales de vigilancia y seguridad privada) through 
a government program that later became known as Convivir (Living Together).
12
 
Convivir was meant to support the formation of community self-defense groups ? 
bodyguards, security guards, escorts, armed ?neighborhood watch?-like services, etc. ? 
in areas where state security forces were too weak to protect people from guerilas and 
criminals. Just as banks hire private security firms, and businesmen and politicians hire 
private bodyguards, supporters of the policy argued, the Convivir [cohn-vee-VEER] 
policy enabled vulnerable communities to form their own private security forces (and, not 
incidentaly, enabled the security forces to recruit informants to gather local 
inteligence).
13
 
                                                
12
 El Presidente de la Rep?blica de Colombia, ?Decreto N?mero 356 de 194 (Febrero 11),? published 
electronically by Superintendencia de Vigilancia y Seguridad Privada, htp:/ww.supervigilancia.gov.co 
(acesed 2 August 209). 
13
 Mar?a Isabel Rueda, ?Convivir en blanco y negro,? [?Convivir in black and white,?] Semana, 1 
September 197, 32. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 174 
The U.S. Embasy in Bogot? warned of the dangers of arming such groups, 
writing in a cable, following a conversation with then-Defense Minister Fernando Botero 
Zea (a few years before his arest on corruption charges for taking drug money), that the 
proposal has been met with mixed and heated reaction localy, and many 
have expresed waranted fears that without the proper supervision of the 
state, the groups could degenerate into local armed militias in the service 
of legitimate or ilegitimate economic interests with litle regard for human 
rights concerns. Botero has insisted that state supervision and 
administration of these groups wil be strict and that examples of the 
eficacy of such groups abound from the experience of other insurgent 
conflicts. We believe ? that Colombia?s protracted vaguely ideological 
conflict is quite sui generis and that there has never been an example in 
Colombia of a para-statal security group that has not ultimately operated 
with wanton disregard for human rights or ben corrupted by local 
economic interests.
14
 
Within a few years, those concerns had proven wel founded, and these words, prophetic. 
The Convivir groups, like Coosercom before them (se ? 4.1.2), were given weapons and 
technical asistance, but once formed they were poorly regulated: of the approximately 
10,000 members of Convivir, for example, fewer than 9 percent had even been subjected 
to a basic criminal background check. Oversight was poor to nonexistent, to the point that 
by 1997 government sources were unable to tel inquiring journalists how many such 
groups even existed in the country (and when they could, the numbers they cited were 
inconsistent). It was clear that, while some were operating as legitimate (acording to 
law) community self-defense groups, many others were operating in reality as 
paramilitary death squads, and stil others were being corrupted through links to 
narcotraficking and other criminal organizations.
15
 Acording to a declasified 
                                                
14
 American Embasy Bogot?, ?Botero Human Rights Leter to A/S Shatuck,? 9 December 194, cable to 
United States Department of State, published electronicaly by The National Security Archive, 
http:/ww.gwu.edu/~nsarchiv/NSAEB/NSAEB217/doc01.pdf (acesed 30 June 209) (emphasis 
added). 
15
 Mar?a Isabel Rueda, ?Convivir en blanco y negro.? 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 175 
inteligence report, a senior Colombian Army officer in mid-1997 described the Convivir 
groups as ?very dificult to control?: 
Once peasants are armed and get a taste for power and easy money, it is 
hard to disarm them and keep them under tight government control. [The 
officer] said that was the concern of his government too, i.e., awarenes of 
the potential for Convivir?s [sic] to devolve into full-fledged 
paramilitaries, though the MoD [Ministry of Defense] was reluctant to 
admit it publicly.
16
 
The government finaly recognized the failure of the policy and isued a decre on 
16 December 1997 that clarified the standards of behavior to which the Convivir groups 
were legaly required to conform and expanded the authority of the Superintendent of 
Vigilance and Private Security to suspend the license of any Convivir group that violated 
those standards.
17
 Most lost their licenses and were officialy dismantled. But many of 
their members ? now wel trained and experienced in batle ? simply formed their own 
paramilitaries or joined existing ones in guerila-controlled areas. 
And yet, despite the central government?s official withdrawal of support for its 
failed Convivir policy, paramilitaries stil enjoyed a great degre of support from regional 
and local state actors, both civilian and, especialy, military, as they had for decades in 
Colombia. Colombia is a state whose authority has always been fragmented among its 
diferent regions. The dificult geography made it possible, and perhaps necesary, for 
remote regions to develop their own economic relations with the outside world, 
independent of any central authority. Like a hundred ?Wild Wests,? the remote regions 
                                                
16
 United States Defense Inteligence Agency, ?Senior Colombian Army Officer Biding His Time During 
Remainder of Samper Regime,? 15 July 197, declasified inteligence information report, published 
electronically by The National Security Archive, htp:/ww.gwu.edu/~nsarchiv/NSAEB/ 
NSAEB217/doc12.pdf (acesed 30 June 209). 
17
 El Presidente de la Rep?blica de Colombia, ?Decreto N?mero 2974 de 1997 (Diciembre 16),? published 
electronically by Superintendencia de Vigilancia y Seguridad Privada, htp:/ww.supervigilancia.gov.co 
(acesed 2 August 209). 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 176 
grew up fiercely independent, with order maintained by local strongmen acting as patrons 
to local clients (who depended on them for jobs, social standing, and political influence) 
and as the interlocutors to outsiders (such as political candidates) seking local influence 
or resources. Moreover, Colombia was setled by ?Spaniards who arived after a seven-
century war against the Arabs,? writes Francisco E. Thoumi. 
They came from one of the most medieval regions of Europe, and the 
regional isolation that they experienced in Colombia alowed them to 
maintain their mores and customs. ? Traditional premodern Spanish 
values did not encourage respect of the laws or authorities of the central 
government, and the isolation of many descendants of the conquistadors 
alowed them to maintain a significant degre of autonomy from the 
central government. At the beginning of the 20th century, Colombian 
society had strong hierarchies and the landlords had great autonomy. Their 
local power was strong, and they frequently abused it. In other words, 
their societies did not impose significant behavioral controls.
18
 
Ricardo Vargas describes this history as having cultivated an esprit mafioso, or 
mafia mindset, in Colombian culture and society, especialy outside the major cities, and 
this characterization is about right. He describes a mafia as not so much an organization 
as a ?medieval sentiment that arises from a belief that an individual can be asured the 
protection and integrity of their person and property through their own worth and 
influence, independent of the actions of the authorities or the law.? This mafia mindset, 
he continues, 
suggests that to achieve succes in life, one must have the valor to oppose 
authority and if necesary the law, or at least support those who can do so 
and not suffer formal legal consequences.
19
 
                                                
18
 Francisco E. Thoumi, ?The Colombian Competitive Advantage in Ilegal Drugs.? 
19
 Ricardo Vargas, ?State, Esprit Mafioso, and Armed Conflict in Colombia,? in Politics in the Andes: 
Identity, Conflict, Reform, ed. Jo-Marie Burt and Philip Mauceri (Pitsburgh, Pen.: University of 
Pitsburgh Pres, 204), 107, citing the conceptualization of Marie-Anne Matard-Bonuci, Histoire de la 
Mafia [A History of the Mafia] (Bruxeles: Editions Complexe, 194); Vargas translates esprit mafioso as 
mafia sentiments. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 177 
This mindset proved fertile ground for the growth of the ilicit drug economy in 
Colombia in the 1980s, making vast wealth available to men of ?valor? in places remote 
from the central authorities in Bogot?. The mafia powers of the past ? a lot of them, at 
least ? became the narcotrafickers of the 1980s and the paramilitary leaders of the 
1990s. By joining the fight against the guerilas they had previously cut deals with in 
their areas of influence, the mafiosi were able to neutralize one set of competitors to their 
drug profits, the guerilas, while simultaneously legitimizing themselves to another set of 
competitors, the state actors and foreign powers (especialy the United States) who 
otherwise might have targeted them as ?narcoguerilas? or ?narcoterorists.? 
A consequence of the state actors? implicit support for the paras, however, was an 
increased fragmentation of the Colombian state, a further consolidation of strongman 
control over remote regions, and a reinforcement of the mafia mindset, al at the expense 
of the rule of law at the local, regional, and national levels. As Vargas warned: 
By substituting for the legitimate exercise of state power the atrocities and 
cruelties caried out by private actors, a subculture is emerging that is 
based on values of order enforced through premodern private powers. ? 
[The state?s] premodern condition does not alow it to construct 
affirmative cultural references, which in turn makes the use of force the 
primary mechanism to resolve conflicts and regulate behavior. Even with 
its use of force, the state does not act to realize and consolidate a strategic 
monopoly. Rather, it acquiesces in and tolerates private violence to resolve 
conflicts, which may be efective in the short term but over time tends to 
contribute to the state?s delegitimization.
20
 
Short-term efectivenes at the expense of long-term legitimation is a recipe for 
instability. And instability was exactly what Colombia was about to get at al levels: in 
the country as a whole, especialy in its outlying regions; in the city of Medel?n (among 
others); and in that city?s Caicedo La Siera sector (among others!). 
                                                
20
 Ricardo Vargas, ?State, Esprit Mafioso, and Armed Conflict in Colombia.? 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 178 
In April 1997, having already absorbed a number of smaler nearby paramilitaries 
into the ACU, Casta?o caled a meting of para chiefs from other regions of the country 
to propose a national asociation of paramilitaries. This asociation, to be caled the 
United Self-Defense Forces of Colombia (AUC: Autodefensas Unidas de Colombia), was 
intended to be something he could present to the country, in the words of one history of 
the group, as ?an organization with a unified command, a national plan, a mechanism to 
coordinate actions across regions, and an agenda with programed aims, al with a view 
toward achieving a place at negotiations with the state and a status that would guarante 
its future recognition as a political actor?; this agenda included ?containing the expansion 
of guerilas and penetrating the zones where [the guerilas] had sources of funding, 
principaly from narcotraficking.?
21
 Casta?o appointed himself head of the AUC?s 
political wing. To overse the military wing, he selected Gabriel Salvatore Mancuso 
G?mez, caled by his second name Salvatore or his alias ?El Mono? (?The Monkey?). 
Mancuso was an old friend who had been a founding member of Los Tangueros in 
Casta?o?s youth. Over the next months and years, Mancuso and Casta?o expanded the 
asociation by incorporating existing paramilitaries from throughout the country and 
recruiting police officers, soldiers, peasants, and even some former guerilas into the 
AUC?s diferent blocs. 
                                                
21
 ?? con el prop?sito de presentarse como una organizaci?n con un mando unificado, un plan nacional, 
una coordinaci?n multiregional de las aciones y una agenda con pretensiones program?ticas, todo con 
miras a lograr un espacio en la negociaci?n con el Estado y un estatus que garantizara, a futuro, su 
reconocimiento como actor pol?tico. A partir de este momento, las autodefensas se trazan la meta de 
contener la expansi?n de la guerrila e incursionar en las zonas donde estos grupos tienen sus fuentes de 
financiamiento, principalmente del narcotr?fico.? VerdadAbierta.com, ?El Nacimiento de las Autodefensas 
Unidas de Colombia (197-2002),? [?The Expansion: Birth of the United Self-Defense Forces of Colombia 
(197-2992),?] Paramilitares y Conflicto Armado en Colombia [Paramilitaries and Armed Conflict in 
Colombia], published electronicaly by Fundaci?n Ideas para la Paz (FIP) and Semana, 
http:/ww.verdadabierta.com/web3/la-historia (acesed 31 July 209). 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 179 
Some of these blocs were true civil-defense groups, founded to defend themselves 
and their interests against extortion and kidnapping by guerilas. Others were 
ideologicaly antisubversive, para-statal organizations, formed to take the offensive 
against guerilas. Many acted more like private security forces, formed primarily to 
protect the economic interests of their founders: coca farms, cocaine production facilities, 
oil fields, banana farms, or mining interests. A distinction is sometimes made betwen the 
?political? paras and the ?narco? paras. This distinction is sometimes useful as a 
description of general tendencies. But in fact individual paras usualy operated on mixed 
motives, each depending on some combination of personal experience, political outlook, 
and economic interest. 
The organizational purpose of the AUC was to provide a means through which 
otherwise isolated units, whatever their motivation, could work together toward shared 
objectives. To the outside world, that?s what it looked like: independent paras working 
together as a federation toward shared objectives. But internaly, it was clear that Casta?o 
and Mancuso always had only a very tenuous control over the rest of the asociation?s 
members.
22
 They acted together when it was in their imediate interests to do so, and for 
the first few years of the federation, most of them considered it in their imediate 
interests to do so: by working together, they increased their numbers and the discipline of 
their troops, increased their income by protecting or monopolizing narcotraficking in 
vast regions of the country, and succeded in defeating and ejecting guerilas from many 
parts of Colombia ? something the state?s security forces had been unable to achieve for 
decades. 
                                                
22
 Gustavo Duncan, Los Se?ores de la Guera: De Paramilitares, Mafiosos y Autodefensas en Colombia 
[Warlords: On Paramilitaries, Mafiosi, and Self-Defense Forces in Colombia] (Bogot?: Planeta, 206), 14. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 180 
This had imediate benefits for the local mafia boses and their clients, but it did 
not automaticaly improve the lives of most people living in those areas. In some places it 
did, but in most others, control had merely shifted from one set of abusive and extortive 
armed actors funded by narcotrafickers and backed by local mafiosi, to another set of 
abusive and extortive armed actors funded by narcotrafickers and backed by local 
mafiosi. The only real diference was that the paras had the backing of the state as wel 
? and were even worse abusers of human rights than the guerilas and the military. 
Asasinations, torture, masacres, disappearances, and especialy, forced displacements: 
the AUC were known for their brutality against known and suspected enemies, their 
supporters, their families, their busines asociates, their neighbors. Homicides 
nationwide increased 40 percent from 1997 to 2002,
23
 not necesarily because the AUC 
murdered a lot of people (though they did), but mostly because the air of impunity ? an 
expresion and expansion of the mafia mindset ? that pervaded the country during the 
rise of the paras had lowered the moral and practical bariers to the violent realization of 
personal and profesional grudges. Most murders and masacres went uninvestigated and 
unpunished. It wasn?t just that the government lacked the capacity to investigte and 
punish; it was also that significant sectors of the government supported what the AUC 
were achieving. 
And so did most Colombians, even while they disapproved of the methods. The 
new paras semed to be the only force in the country capable of forcing the insurgents to 
negotiate peace and put an end, once and for al, to the longest civil war in the Western 
Hemisphere, then in its fourth decade. As it happened, a cease-fire would soon be 
                                                
23
 Author?s calculation based on Departamento Administrativo Nacional de Estad?sticas (DANE), 
?Estad?sticas Vitales,? http:/ww.dane.gov.co (acesed 2 August 2009). 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 181 
negotiated with the FARC. But those talks would not bring an end to the war, not in 
Colombia, and certainly not in Medel?n. 
5.1. Medel?n: ?Don Berna,? ?Double Zero,? and the Complex War 
5.1.1. ?Double Zero? Enters Medel?n 
The AUC?s brutal campaign against guerilas nationwide ? which promised 
both political and economic benefits for the AUC ? was exactly what Carlos Casta?o 
was hoping to replicate in Medel?n when he decided in 1997 to send in troops.
24
 To do 
that job, he selected his longtime military trainer, Carlos Mauricio Garc?a Fern?ndez, 
known as ?Rodrigo Franco,? ?Doblecero? (?Double Zero? or ?00?), or most often, 
?Rodrigo 00? (?Rodrigo Double Zero?).
25
 
?Rodrigo 00? Garc?a was a virulent anti-Communist who had been a junior officer 
in the Army during the Betancur administration?s negotiations with the FARC in the mid-
1980s. He retired in 1989, however, amid acusations that he and other members of his 
unit had formed a death squad to exterminate and ?disappear? members of the Patriotic 
Union political party (UP: Uni?n Patri?tica), which had been founded in May 1985 by 
                                                
24
 Information on the activities of the AUC is drawn from VerdadAbierta.com, ?Las Auc,? Paramilitares y 
Conflicto Armado en Colombia [Paramilitaries and Armed Conflict in Colombia], published electronicaly 
by Fundaci?n Ideas para la Paz (FIP) and Semana, htp:/ww.verdadabierta.com/web3/victimarios 
(acesed 31 July 209); Mauricio Romero, ed., Parapol?tica: La Ruta de la Expansi?n Paramilitar y los 
Acuerdos Pol?ticos [Parapolitics: The Course of Paramilitary Expansion and Political Acords] (Bogot?: 
Intermedio/Corporaci?n Nuevo Arco Iris, 207); Margarita Mart?nez Escal?n, personal comunication 
(Interview No. 26), 1 July 209; Luis Fernando Quijano Moreno, Interview No. 5; and other sources as 
cited. 
25
 His family name sometime apears in print as Garc?a Duque rather than Garc?a Fern?ndez, and his given 
names sometimes appear as C?sar Mauricio instead of Carlos Mauricio. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 182 
demobilized guerilas. During the investigations into those death squads, Garc?a retired 
from the Army ? and Fidel Casta?o recruited him to train ACU units in counter-
guerila tactics.
26
 After Fidel?s death, Garc?a, who had been born and raised in Medel?n 
and spent much of his youth in the surrounding countryside, continued working with the 
ACU, taking responsibility for paramilitary units in parts of Antioquia not already 
controlled by the Peasant Self-Defense Forces of Magdalena Medio (ACM, 
Autodefensas Campesinas del Magdalena Medio), which later became one of the more 
important co-founders of the AUC. 
?Rodrigo 00? Garc?a was one of the few para leaders to refuse to alow 
narcotraficking money to corrupt his work: he was the purest of the political paras. His 
units raised money instead by stealing and reseling gasoline, seling or racketering 
private security services, and extorting busineses. Despite his contempt for the narco 
paras, Garc?a agred to join the AUC to lead a new paramilitary bloc, named Metro 
(Bloque Metro), into the Medel?n metropolitan area. 
He had already been operating on the outskirts of Medel?n, having led a series of 
atacks and masacres that had cut off the FARC and the ELN from the city. His 
intimidation campaign in the countryside began with the kiling of 14 farmers in a vilage 
in eastern Antioquia on 3 May 1997; more were to follow.
27
 When his atacks were 
against guerilas instead of unarmed civilians, he tried to capture rather than kil them so 
he could convert their fighters into paras. ?His first order was to respect the lives of the 
                                                
26
 Margarita Mart?nez Escalon, ?As? Filmamos la Siera,? [?This Is How We Filmed La Sierra,?] El 
Malpensante, htp:/ww.elmalpensante.com/index.php?doc=display_contenido&id=671 (acesed 
5 August 209). 
27
 Manuel A. Alonso Espinal, Jorge Giraldo Ram?rez, and Diego Jorge Siera, ?Medel?n: El Complejo 
Camino de la Competencia Armada,? 124. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 183 
ELN combatants? whom his units captured, a non-governmental organization (NGO) 
official told journalists for a history of the AUC. ?You?d se them at the AUC?s 
headquarters al chained up, then six months later they?re military commanders for the 
AUC. I saw this one ELN commander who was already in the AUC cal his brother, who 
also was in the ELN, and [tel him] to turn himself in and join them.?
28
 To enter a town or 
vilage, Metro Bloc members tried ingratiating themselves with the communities they 
were trying to take over by engaging in ?social cleansing? campaigns like the ones the 
militias themselves had caried out during the 1980s to win community support. They 
were bloody, but in places where the guerilas had been abusive, those campaigns 
worked. ?When you?re geting harased and you?re worried about kidnappings, about 
guerila checkpoints, you look to the state, and if it doesn?t show up, you look for 
something else,? Juan Diego Restrepo E., a journalist who has covered the paras, wrote 
about Metro?s strategy. ?They offered security and they legitimized their actions, due to 
the guerilas? exceses.?
29
 It was the same approach the guerilas themselves had used to 
enter Medel?n and other cities as part of their own urban strategies a decade earlier: offer 
security against abusive strongmen, remove low-status social groups from the 
community, provide some benefits to supporters and neutral community members, and 
threaten death or displacement to anyone who offers opposition instead of compliance. 
Some of the bigger gangs in the city then copied the militias? strategy, and now the paras 
                                                
28
 ?Su orden inicial era respetar la vida de los combatientes del ELN. ? En las sedes de las AUC los ve?a 
encadenados y a los seis meses ya eran jefes militares de las AUC. Vi a un comandante eleno, ya en las 
AUC, lamando a un hermano que tambi?n estaba en el ELN, para que se entregara y se uniera a ellos.? 
Quoted in VerdadAbierta.com, ?Las Auc.? 
29
 ?Cuando uno es acosado y est? angustiado por secuestros, por controles de la guerila, busca el Estado y 
si ?ste no aparece, buscas otro. Elos (BM) ofrecieron seguridad y legitimaron sus aciones, debido a los 
excesos de la guerrila.? Quoted in Ibid. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 184 
were doing the same. ?In its basic structure, the Metro Bloc was a typicaly rural 
counterinsurgent organization that followed a strategy of teritorial conquest based on the 
guerila model, in combination with a strategy of teror against the social base of its 
armed adversaries.?
30
 With this structure, Metro would end up controlling 45 
municipalities in Antioquia at the height of its power. 
The Metro bloc entered Medel?n sometime in 1998, starting in the barios where 
the leftist militias had their deepest roots. To the strategy he employed on the outskirts of 
the city, he added the tactics of co-opting or subcontracting the talent he needed to 
operate in an urban environment. He identified stret crews, hit squads, gangsters, 
wayward militants, and community organizations already operating in the barios whom 
he could hire or coerce into joining or backing Metro in its fight against the guerila-
backed militias. He hired ?El Negro Elkin? S?nchez and the Terace Gang ? they were 
already alied with Murilo and Casta?o ? to take care of any gangs or community 
leaders who opposed him. Otherwise, he generaly applied the same strategy he had been 
using in the surrounding region. Within a year, Metro, Terace, and their alies 
throughout the city were winning control over most of the barios they were targeting, 
and ?Rodrigo 00? Garc?a was wel on his way to achieving his lifelong anti-guerila 
dreams in his own birthplace. 
But then things got complicated. 
                                                
30
 ?En su estructura b?sica, el Bloque Metro fue una organizaci?n contrainsurgente t?picamente rural que 
desaroll? una estrategia de copamiento teritorial siguiendo el modelo guerilero, en combinaci?n con una 
estrategia de terror contra la base social de sus contendores armados.? Manuel A. Alonso Espinal, Jorge 
Giraldo Ram?rez, and Diego Jorge Siera, ?Medel?n: El Complejo Camino de la Competencia Armada,? 
125. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 185 
5.1.2. ?Don Berna? Joins the Fight 
A war broke out betwen the Terace Gang and the Envigado Ofice. Somehow, 
?El Negro Elkin? S?nchez and his men got on ?Don Berna? Murilo?s wrong side during 
1999 in a dispute involving Carlos Casta?o. There are diferent acounts of how the 
dispute started: some say Casta?o was angry about a crime they had commited against 
one of his paramilitary units, or that they had commited in the name of the 
paramilitaries; others say S?nchez was angry because Murilo wasn?t leting his men keep 
enough of their hard-earned money. Whatever the reason, the dispute escalated when the 
Terace Gang kiled Murilo?s brother in November 1999, and Murilo and Casta?o 
waged a war against them that, despite nasty counteratacks involving car bombs in 
upper-clas barios, would end up wiping out the Terace Gang within a year. S?nchez 
and his top lieutenants were ambushed and kiled in August 2000, and the rest of his 
organization drifted apart, disappearing completely sometime in 2001, its members 
fleing for their lives or geting absorbed into other criminal networks ? include 
Murilo?s. 
With Terace liquidated, not only did ?Rodrigo 00? Garc?a lose his most powerful 
aly in the city; he also gained a new, even more powerful competitor. During the war 
with S?nchez, Murilo had fled Medel?n after his brother?s murder, seking protection 
from Casta?o in the Urab? region in northern Antioquia. There, he paid the AUC a 
sizeable sum of money to formaly join the paramilitary aliance. In return, Casta?o 
named him Inspector General of the AUC, responsible for overseing the AUC?s 
traficking activities, and gave him his own paramilitary unit, which he named the Bloque 
Cacique Nutibara. The Cacique Nutibara Bloc was born directly into adolescence: 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 186 
Murilo did not have to grow the bloc from nothing and nurse it to adulthood by 
recruiting troops, but rather simply informed the drug gangs and hit squads who already 
worked for him that they were now to cal themselves ?Bloque Cacique Nutibara? of the 
AUC. They complied. It was 2000, the year of the new milennium, and with this one act 
the AUC now controlled (by proxy) most of Medel?n?s peripheral barios and almost al 
of its ilicit markets. And Garc?a, the purest of the political paras, was not at al happy 
about having to aly himself so directly with Murilo, the newest and purest of the narco 
paras: Nobody expected that Murilo would be satisfied to share power with anyone once 
they ceased to be useful to him. For now, Garc?a and his Metro Bloc were useful to him. 
They al had a war to win. 
5.1.3. Cacique Nutibara Plays ?Monopoly? 
Cacique Nutibara was neither a hierarchical criminal organization nor a federation 
of gangs. Acording to a recent case study of the group, it is best characterized as a 
network, one whose nodes (units) had emerged from any one of four ?complex routes? to 
membership: (1) the route taken by the self-defense groups that had emerged in Medel?n 
to defend their barios from crime, many of which themselves then turned to crime and 
were later absorbed into Cacique Nutibara; (2) the route taken by narcotraficking groups 
as they restructured amid the fal of the Medel?n and Cali cartels into networks of 
microenterprises and oficinas, as discussed in the introduction to this chapter; (3) the 
route taken by the stret crews and gangs that controlled micro-teritories in the 
peripheral barios and later evolved or were co-opted into the hit squads, drug gangs, and 
other organized criminal structures that were coming together under Murilo?s umbrela; 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 187 
and (4) the route taken by the mafias and paramilitaries of the AUC who brought their 
counterinsurgency from the countryside into the city.
31
 The franchises that made up the 
Cacique Nutibara network came from al of these types of groups, ariving via many 
diferent paths. They were given protection and the right to use the Cacique Nutibara 
name in exchange for their compliance with Murilo?s orders and those of his field 
officers. 
In a few short and very violent years, the Cacique Nutibara network and its 
hesitant alies in Metro?s network took over most of the city, gang by gang and bario by 
bario. The defeated Terace Gang had not been the only large criminal organization in 
the city, nor was it the only one to be defeated by the AUC: Frank?s Gang (La Banda de 
Frank) and the Triana Gang (Los Triana) had both been nearly as powerful as Terace, 
and like Terace both had been liquidated by mid-2001, their survivors chased from the 
city or absorbed into Cacique Nutibara. Metro took a similar approach with militia 
groups: fight then negotiate with the militias to try to turn them to his side. In Caicedo La 
Siera, for example, M-6&7 switched sides and become a unit of Metro (se ? 5.2). Over 
time, other AUC blocs joined the fight as wel: ACM sent some troops into Comuna 13 
in the city?s West Central zone, and Bloque Central Bol?var joined the fight for the 
AUC?s final push toward a monopoly over the whole city. The number of players on the 
board was shrinking fast: by mid-2002, this handful of AUC blocs, with Cacique 
Nutibara the most ambitious, had taken control of every bario formerly controled by a 
leftist militia or a competing criminal organization, except for a few in Comuna 13 and a 
                                                
31
 Ibid. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 188 
few in East Central Medel?n (including Caicedo La Siera). To eject those would require 
just a litle more time ? and some help from the local and national governments. 
5.1.4. The Battle for the West 
Comuna 13 was one of those areas in the city that no outsider could ever hope to 
understand. Every few blocks, it sometimes semed, a diferent group was in control. In 
the 1980s it was al the stret crews, gangs, chichipatos, asasins, and others that have 
been described elsewhere (se ?? 2.1.1?2.1.2), and that atracted the urban militias of the 
late 1980s. For example, in 1990, the ELN founded the Fre-America Militia (Milicias 
Am?rica Libre) to take control in some barios, and sent other urban fronts to deal with 
others. The FARC and its urban fronts arived beginning in 1994, and the People?s 
Armed Command (CAP: Comandos Armados del Pueblo), which was formed by 
militants in 1996 or 1997 in the aftermath of the failed Coosercom policy (se ? 4.1.2), 
showed up soon thereafter. The paras started ariving around 1998, eventualy sending 
units from Metro, Cacique Nutibara, Central Bol?var, and ACM. In response, a 
disident ELN faction founded the Anti-Paramilitary Revolutionary Front (FRAP: Frente 
Revolucionario Anti-Paramilitar), and the FARC activated more urban fronts from units 
it stil had stationed nearby as wel. By the time the war ended, many dozens of ilicit 
groups had appeared and disappeared in the area, as in the rest of the city, including some 
uncounted number of smal gangs and tiny stret crews that the outside world has never 
documented and wil probably never know about. The paras were able to defeat some of 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 189 
these groups, but the guerilas were wel entrenched in these communities and the AUC 
could neither co-opt, nor coerce, nor defeat the last of them.
32
 
So far in this chapter litle has been said about the role of the state security forces 
in the batles over Medel?n, which might have given the false impresion that they were 
not involved. In fact, in many barios the community police kept the peace, but in others, 
especialy those on the periphery, their presence was weak to non-existent, limited mainly 
to occasional raids to arest or kil militants and gangsters, which sometimes led to 
civilian casualties, sometimes a lot of them. It is also clear, however, by most acounts, 
that the police and the military were covertly supporting AUC units and some of the 
larger criminal organizations in the city ? many of-duty or retired police officers and 
soldiers even joined the ilicit groups directly ? or they used inteligence provided by 
such groups for their own raids, an arangement similar to that of the Search Bloc in 
1993, where regular and iregular forces, cops and robbers, worked together to find Pablo 
Escobar (se ? 2.1.3). 
But it wasn?t until 21 May 2002 that the national government stepped in to 
organize a large-scale, joint raid by police, military, and inteligence units designed to 
take Comuna 13 from the remaining guerilas by force. Operation Mariscal ? so named 
only because mariscal, which means marshal, begins with the leter m, and the 
Colombian military?s operations conventionaly began with the same leter as the month 
in which they were launched, in this case, May ? lasted only a few hours, and the 
                                                
32
 Pablo Emilio Angarita Ca?as, H?ctor Galo, and Blanca In?s Jim?nez Zuluaga, Din?micas de Guera y 
Construci?n de Paz; Yoni Alexander Rend?n Rend?n, Comuna 13 de Medel?n: El Drama del Conflicto 
Armado [Comuna 13 in Medel?n: The Drama of the Armed Conflict] (edel?n, Colombia: Hombre Nuevo 
Editores, 207); Ricardo Aricapa Ardila, Comuna 13: Cr?nica de Una Guera Urbana. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 190 
security forces failed to make inroads into the community, being quickly forced to retreat 
in batle. But they regrouped and added troops so that the next operation would not fail. 
Operation Orion, so named because it was to begin in October, involved more 
than a thousand troops from the National Police, Army, Administrative Security 
Department (DAS: Departamento Administrativo de Seguridad, Colombia?s equivalent to 
the FBI), and other agencies. It started in the middle of the night, on 16 October 2002, 
when residents started hearing a lot more gunfire than usual, followed by the sound of 
helicopters. Combat operations lasted nearly two days, and city and national officials 
took credit for the victory, the first of the new presidential administration of ?lvaro 
Uribe. It sent a signal to most Colombians that their government now had real credibility 
in the fight against subversives: the war against the guerilas would be won with a strong 
fist. Most observers agre, however, although it continues to be oficialy denied, that 
some members of the government had quietly enlisted ?Don Berna? Murilo?s fist to help 
prepare for and cary out Orion.
33
 (Some say it was Murilo who had originaly enlisted 
the help of the state security forces to initiate Mariscal in the first place, although, again, 
this theory is oficialy denied and has litle public evidence to support it.) One can state 
with reasonable confidence that Murilo quietly provided inteligence and tactical 
asistance from Cacique Nutibara, and that some elements of the security forces returned 
the favor with inteligence the paras could use during or following the operation. 
Speculation aside, what is certain is that Cacique Nutibara had fre reign of Comuna 13 
as soon as combat operations were completed: roadblocks were common through at least 
2004, a local children?s zoo had been named after ?Don Berna? Murilo, and the paras 
                                                
33
 Paul Richter and Greg Miler, ?Colombia Army Chief Linked to Outlaw Militias,? Los Angeles Times, 
25 March 207. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 191 
were acused of a range of abuses against residents, including at least 46 
disappearances.
34
 Their presence was not welcomed, as one resident explained: 
In this bario, with the paramilitaries, the law of silence reigns. You could 
say we were beter off with the guerila, because they acted openly. You 
could get kiled in a shooting, but at least you knew here the bullets were 
coming from. With the paramilitaries everything happens in secret. We do 
not know exactly who is responsible for the disappearances.
35
 
In August 2003 a mas grave was found nearby with the bodies of 11 people, including 
some who had disappeared during Orion.
36
 
With the completion of Operation Orion, the FARC had been completely driven 
out of Medel?n, and the only place left in the city where the ELN and its alies were stil 
operating was in the uppermost area of Comunas 8 and 9 in the city?s East Central zone, 
primarily in the Ocho de Marzo bario, just across Santa Elena crek from Caicedo La 
Siera and just above the bario Pablo Escobar had built twenty years earlier. Cacique 
Nutibara teamed up with Metro and the other AUC blocs in the city for one final asault. 
                                                
34
 Ralph Rozema, ?Urban DR-Proceses: Paramilitaries and Criminal Networks in Medel?n, Colombia?; 
Comuna 13, la otra Versi?n: Caso Tipo No. 2, Banco de Datos de Violencia Pol?tica [Comuna 13, the 
Other Version: Case No. 2, Political Violence Data Bank], Noche Niebla: Panorama de Derechos 
Humanos y Violencia Pol?tica en Colombia (Bogot?: Centro de Investigaci?n y Educaci?n Popular?CINEP 
and Justicia y Paz, 203), htp:/ww.nocheyniebla.org/node/46 (accesed 2 April 209); Amnesty 
International, Colombia: The Paramilitaries in Medell?n: Demobilization Or Legalization? (London: 
Amnesty International, 31 August 205), htp:/www.amnesty.org/en/library/info/AMR23/019/205. 
35
 Quoted in Ralph Rozema, ?Urban DDR-Proceses.? 
36
 Amnesty International, Colombia: The Paramilitaries in Medell?n: Demobilization Or Legalization? 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 192 
5.2. Caicedo La Siera: ?Job,? ?Scab,? ?Doll,? and the Final Battle 
5.2.1. Metro Takes Over M-6&7 
Just as the bad behavior of gangs in the 1980s had created the conditions that 
enabled the militias to take over the neighborhood, proposing to bring order, the bad 
behavior of the militias in the 1990s was creating the conditions that would enable the 
paramilitaries to take over the neighborhood, proposing to bring order. The takeover 
happened in two stages, one with a source internal to the bario, the other external. 
The internal source was a neighborhood crew in La Siera, led by a young man 
named Jason, known as ?Cascarita? (?Scab?), a ?lost soul,? in the description of a 
journalist who knew him wel.
37
 His parents were said to have abandoned him, leaving 
him to live with an aunt, and he grew up lonely and serious. He joined a neighborhood 
crew, which was later recruited into M-6&7, which controlled his bario. But like many 
others in the community, he didn?t like what a lot of the militants were doing to their 
neighbors, friends, and families: despite his own reputation as a harsh, sometimes brutal, 
person, even Jason thought the M-6&7 had gotten out of control. Hugo, the man who had 
founded the militia a decade earlier, was asasinated sometime in 2000. Acounts of the 
asasination difer: some say it was done by Jason and others in Jason?s combo who 
were sick of his abuses; others, that Hugo was kiled in a police shootout after a felow 
militant left to become a para and a police informant; stil others, that a felow militant 
named Fredy kiled Hugo because Hugo had taken his girlfriend. It?s possible that there 
are elements of truth to two or al thre of these acounts, since it was not uncommon for 
                                                
37
 Information for this and the next two paragraphs are based on Margarita Mart?nez Escal?n, Interview 
No. 26; ?Danilo,? Interview No. 10; and Arleison Arcos Rivas, ?Ciudadan?a Armada,? 18. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 193 
the police or the army to ilegaly help local gangs target leftist militants, nor for the local 
gangs to provide inteligence to help the police or the army target rival gangsters, nor for 
groups of gangsters to participate in a kiling, each with his own motive. Whatever the 
case, Hugo was dead, and Jason and his crew took over M-6&7. 
The external source of the paramilitaries? entry into Caicedo La Siera was a 
prison conversation. One journalist told me, regarding how the underworld did busines, 
that ?al the deals were made on Saturdays in prison: who?s going to get kiled, who?s 
going to get paid.?
38
 Stil, it is dificult to say exactly what happened in prison that day, 
since the principals are dead or their identities unknown, and the recollections of diferent 
people afected by that conversation are contradictory. 
What can be pieced together, however, is that in 1999 or 2000, sometime before 
Hugo was kiled, he or somebody very close to him (possibly his brother) was arested 
and spent some time in Medel?n?s notorious Belavista prison. There he encountered a 
former colleague named Severo Antonio L?pez, known by everyone as ?Job.? L?pez had 
started out as a militant in M-6&7 years before, and there he had built himself a 
reputation as a political strategist of the first order. But he got into a dispute with some of 
his felow militants over some cash they had robbed and that he apparently was supposed 
to guard but spent instead. Going down the hilside seking protection, he joined La 
Ca?ada. ?He left,? one of his former colleagues, Danilo, told me, ?and since he was very 
strategic about organizing things with gangs and things like the politics of organizations, 
he was very useful? to them.
39
 ?Job? L?pez had a lot of information about M-6&7 
                                                
38
 Margarita Mart?nez Escal?n, Interview No. 26. 
39
 ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 194 
members, such as their identities and where they lived, and he did not hesitate to share it 
with La Ca?ada or even with the police when it was mutualy convenient. A few years 
later, he was captured during a bank robbery (acording to one acount) and sent to 
Belavista. At the time, La Ca?ada was doing a lot of busines with the Envigado Ofice, 
which itself was cooperating with the AUC in its eforts to rid Medel?n of militias such 
as M-6&7. So L?pez was wel placed to try to get M-6&7 to switch sides, neutralizing 
them so the paras would not have to fight them. He must have gotten into a discussion 
about this with Hugo (or whomever it was from M-6&7 he met in prison), and at some 
point, L?pez offered M-6&7 a chance to avoid geting into a war with the paras by 
switching sides and joining the Metro Bloc of the AUC. L?pez and his men ?offered 
them oney, they offered them everything,? Danilo said. ?Imagine when they come in, 
mafia-style: they come in, they negotiate, and they give them lots of cash. You went in 
there, and they already have gold chains, gold earings, gold rings. These were not 
hippies.?
40
 Whether it was for money or from a prediction that the paras would ultimately 
prevail in the war, the choice was made: M-6&7 would join Metro. 
The membership of M-6&7 was both stunned and divided: it stil had some 
members who were old-school leftists, political militants, in the movement for the good 
of the people. The cal for them to join up with the mortal enemies of the guerila, to join 
the worst human-rights violators in the country, was not something they could easily 
stomach. Moreover, as Danilo explained, they had been at war with La Ca?ada (refered 
                                                
40
 ?? les ofrecen plata les ofrecen de todo. Imaginencen cuando entran al estilo mafioso, entran, negocian 
all? y les dan mucha plata. Usted iba all? y ellos ya eran con cadenas de oro, aretas de oro, anilos de oro. 
No eran hipies.? Ibid. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 195 
to as ?those people from below?) for over a decade, so the thought of alying with them 
was anathema: 
They caled us and told us that we now have to be Metro. ? And we ? 
some of us, said, hel no. I was afraid to become one of those people from 
below. So I said to them, no, screw that. You fight with these dudes from 
down there and now it?s like we?re friends? Things aren?t like that, 
because there were grudges and hurts, and to say that it?s like now we?re 
friends and like nothing?s happened! So in that space of time, that?s what 
they negotiated in prison from Belavista.
41
 
Many chose to leave the bario rather than join their enemies. Some had to leave the 
country. Those who refused to leave or switch sides risked death. Others found sanctuary 
with the FARC or ELN militias that stil held teritory in the city. Danilo quit entirely, 
leaving the bario and the war behind and not returning until years later, after there was 
peace. Those who stayed did so for many reasons, chief among them that most of the 
current members had never had any grand political agenda in the first place, and things 
were geting dangerous: 
The urge almost always was to have a motorcycle, to get girls, to be in a 
gang. And that?s what it was [for a lot of those who stayed]. And also 
because the state was hiting hard. There was also a lot of presure from 
the state. Even the Metro Bloc made aliances with the Army to finish off 
the [ELN], because, let?s just say they had information and details. And 
the Metro Bloc did their dirty work, the violations of human rights and al 
that, so that the state could say that it was the Metro Bloc and not them.
42
 
                                                
41
 ?A nosotros nos laman [sic] y nos dec?an que ten?amos que ser ya Bloque Metro ?. Y nosotros ? unos 
dec?amos, no, juepucha. A mi me daba mucho temor ser de esa gente de abajo. Entonces yo les dec?a, no, 
?las pelotas! Uno pelea con estos manes de abajo y ?ahora es que ya son los amigos? Eso no es as?, porque 
hab?a rencores y dolidos, y ?decir es que ya somos amigos y que ya no pasa nada! Entonces en ese lapso de 
ese tiempo, eso lo negociaron en la c?rcel desde Bellavista.? Ibid. 
42
 ?El anhelo casi siempre era tener una moto, tener muchachas, estar embandados. Y as? era. Y tambi?n 
porque el estado estaba golpeando muy duro. Hab?a mucha presi?n del estado tambi?n. Incluso el Bloque 
Metro tuvo alianzas con el ej?rcito para acabar los Elenos, porque elos hac?an la parte digamos 
informaci?n y detalles, y el Bloque Metro hacia la parte sucia, la violaci?n de derechos humanos y todo 
eso, pa? que el estado dijera que fue el Bloque Metro y no ellos.? Ibid. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 196 
Jason and his crew stayed, taking over as soon as Hugo had been kiled, and went to war 
? the fourth, and the most complex, of Caicedo La Siera?s gang wars ? against the 
ELN guerilas, their former alies, whose teritory was just across the river from them in 
Comuna 9, centered in a bario caled Ocho de Marzo. Some of those who remained in 
Metro, however, did not leave behind al their grudges against La Ca?ada. 
5.2.2. ?Job? Takes Over Cacique Nutibara 
After ?Job? L?pez left prison, he returned to La Ca?ada, and soon became a key 
figure in Murilo?s network, first in Comuna 8, since he was from there and was the one 
responsible for neutralizing M-6&7, and then in Medel?n as a whole, and eventualy as a 
national spokesman for the paras after their demobilization (se ? 6.1). With his strategic 
mind and political skils, he came to ?Don Berna? Murilo?s atention and quickly rose to 
a high position in Murilo?s network. Diferent sources describe him as having become 
Murilo?s ?right-hand man,? his ?No. 2,? his adviser, and his political spokesman. 
Whatever his position, when Murilo founded Cacique Nutibara, he put L?pez in charge 
of its day-to-day operations, and they recruited Alberto, founder of La Ca?ada (se ?? 3.2 
and 4.2), to join their network.
43
 M-6&7 was now part of the Metro Bloc, but Cacique 
Nutibara owned La Ca?ada and soon recruited under its own banner the rest of the gangs 
in lower Caicedo La Siera ? including those that a couple of years earlier had made 
peace with La Ca?ada in negotiations facilitated by Mama Luz (se ? 4.2.3). 
Since Metro?s chief ?Rodrigo 00? Garc?a had an ideological bias against drug 
money (although his crews in La Siera certainly did not), and since Metro and Cacique 
                                                
43
 Margarita Mart?nez Escal?n, Interview No. 26. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 197 
Nutibara were technicaly in an aliance, ?Job? L?pez was now in a position to control 
most of the ilicit markets in Comuna 8, in both upper and lower Caicedo La Siera. He 
became the de facto ward boss, the man people went to when they had problems to solve 
and conflicts to resolve. Diego R?os, who grew up there, told the story of a city 
investigator who came to Comuna 8 to respond to a citizen?s complaint: when the 
investigator met the woman who had caled, the first thing he asked her was whether she 
had brought it up with ?Job.? ?Imagine that,? R?os said: ?A city detective asking a citizen 
if she?d taken her complaint to the criminal! Who?s realy in charge here??
44
 
5.2.3. The Battle for the East 
Margarita Mart?nez Escal?n was a Colombian reporter for the Asociated Pres 
(AP) who had interviewed ?Rodrigo 00? Garc?a on several occasions and had earned his 
trust as a journalist. Scott Dalton was a photojournalist and cameraman from Texas who 
had worked with Mart?nez on several stories about Colombia?s war and its peace 
negotiations. He had recently left his job at AP?s Bogot? office when he saw some 
photographs that had been published with a story Mart?nez had writen about the Metro 
Bloc?s conquest of Medel?n?s peripheral barios, including La Siera. Interested in the 
possibilities there, he asked her for some contacts so he could get into one of the war 
zones to do a frelance photojournalism project. It was the summer of 2002. The AUC 
was wel on its way to defeating the city?s guerilas and militias, Operation Mariscal had 
already come and gone, and preparations for Operation Orion were then under way. In 
short, the war was about to reach its climax, and Dalton, who had covered wars in Latin 
                                                
44
 Diego R?os, Interview No. 4. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 198 
America in the past and had recently bought a new videocamera, thought it would be 
interesting to film a documentary about it. He talked to Mart?nez about the idea: Would 
she be interested in working with him on the project? At first reluctant, she ultimately 
agred, picked up her phone, and caled Garc?a to ask his permision to film one of his 
Metro Bloc units. ?We asked for his authorization to do a documentary, making it clear 
that we had no experience, no financing, and that we didn?t even know if it was going to 
end up being a fiasco.?
45
 Permision granted, they jumped in a taxi and headed to La 
Siera, the uppermost bario of Comuna 8 where some of the most intense fighting was 
taking place.
46
 
At the appointed hour, they were met by a group of young men in ski masks and 
camouflage, who took them around the neighborhood, marched in formation with their 
military-grade weapons, told them about how the paras had imposed order in the bario, 
and posed for the camera like batle-hardened soldiers, at the top of the hil, with the 
panorama of Medel?n making for a dramatic backdrop. The oldest in the group was 22. 
At the end of the tour, they thought they were finished and, as Mart?nez described it, 
?they wanted to leave and wanted Scott and me to go.? 
We, however, had more ambitious plans and, much to their consternation, 
we told them we wanted to se them the next day. They just looked at each 
other. I think they were figuring out what they were going to do, since 
they had already done everything they had prepared for the pres. They 
couldn?t refuse, since their boss, ?[Rodrigo] Double Zero,? had authorized 
us ?. ?What for?? they asked us. Obviously they had done their 
performance and they didn?t have anything else to show. ? ?Our project 
                                                
45
 ??le ped? autorizaci?n para hacer un documental, dej?ndole claro que no ten?amos experiencia, ni 
financiaci?n, y que no sab?amos si iba a resultar un fiasco.? Margarita Mart?nez Escalon, ?As? Filmamos la 
Siera.? 
46
 Margarita Mart?nez Escal?n, Interview No. 26. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 199 
is a litle longer,? we told them. They stood there in silence for a bit, and 
then told us to return the next day.
47
 
Their original plan had been to follow around the leader of Metro?s La Siera unit, 
a young man named Jason, alias ?Cascarito? (?Scab?), the former crew leader who had 
joined M-6&7 then took it over after the asasination of its founder, Hugo (se ? 5.2.1). 
But Jason was serious and camera-shy, and he didn?t show up for the second day of 
filming. His top deputy arived instead, a charismatic young man named ?dison 
Alejandro Fl?rez Ocampo, known as ?La Mu?eca? (?Doll?). In addition to being a top 
leader of Metro?s La Siera unit, Fl?rez did a lot of work in the community, acting as de 
facto mayor of the bario and therefore as Metro?s de facto director of community afairs 
and outreach. A native son, he was wel liked and respected and had a reputation for 
usualy being fair in his judgments. More importantly, from the filmakers? perspective: 
?He wanted to be filmed,? Mart?nez said. ?I think he wanted to be something more. He 
knew he would die young, so this was his way of maybe having a longer life.?
48
 
Mart?nez and Dalton filmed ?dison and others for nearly a year, beginning in 
January 2003, when the area around Caicedo La Siera was the last place in the city 
where guerilas were fighting: they had lost everywhere else, and these batles were their 
last stand. ?dison, Jason, and crew intended to defeat them, and they had powerful people 
backing them: ?Rodrigo 00? Garc?a, head of al Metro Bloc units in the country; ?Don 
                                                
47
 ?? ellos quer?an irse y que Scot Dalton y yo nos fu?ramos. Nosotros, sin embargo, ten?amos planes m?s 
ambiciosos y, para su desconcierto, les dijimos que quer?amos verlos al otro d?a. Se miraron, pienso que 
calculando qu? iban a hacer si ya hab?an hecho lo que ten?an preparado para la prensa. No pod?an negarse, 
pues nos hab?a autorizado su jefe ?Doblecero? ?. ?Para qu??, nos preguntaron. Obviamente ellos ya hab?an 
hecho su espect?culo y no ten?an nada m?s que mostrar. ? Nuestro proyecto es un poco m?s largo, les 
dijimos. Se quedaron en silencio un instante y luego dijeron que volvi?ramos a la misma hora al d?a 
siguiente.? Margarita Mart?nez Escalon, ?As? Filmamos la Siera.? 
48
 Margarita Mart?nez Escall?n, Interview No. 26. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 200 
Berna? Murilo, head of Cacique Nutibara and proprietor of the Envigado Ofice, one of 
the wealthiest drug-traficking networks in the world; and, unofficialy, the state security 
forces, who were unofficialy on the side of whomever was against the guerilas. The 
fourth gang war of Caicedo La Siera was reaching its climax. 
The fight did not last much longer, barely two months into 2003. In February, 
Metro and Cacique Nutibara decided on a joint operation. ?We decided we had to risk 
everything and go in,? ?dison explained. ?The first group went in and atacked. Al of 
[the guerilas] who were in that area over there [across the crek] came out to respond, 
but since we were above them it was easy to blow them away like this.? Cacique 
Nutibara went in from below, Metro went in from above. The guerilas put up a fight, 
then the survivors fled into the mountains, from where they would occasionaly launch 
brief retaliatory atacks into Caicedo La Siera. But from the AUC?s perspective, the 
batle was won: the war was over. 
Medel?n belonged to the paras now. 
In Caicedo La Siera, Metro and Cacique Nutibara divided the teritory more or 
les acording to their traditional areas of influence, with Metro in control of upper 
Caicedo La Siera, but now al the way south to the crek, and Cacique Nutibara in 
control of lower Caicedo La Siera and the teritory on the other side of Santa Elena crek 
in Comuna 9. Together the two blocs reached out to the communities that had been under 
ELN control for so long, with ?dison continuing to play his traditional community-
outreach role, but this time in hostile teritory, as he explained in Dalton and Mart?nez?s 
film, La Siera: Urban Warfare in the Barrios of Medel?n, Colombia: 
? our relationship with the people [across the crek] is that we?re trying 
to take it slow, because we have a dirty reputation with that community. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 201 
They say the Metro Bloc of La Siera is bad, realy ruthles. We?re trying 
to integrate the people from both neighborhoods. For example, tomorrow 
there?s a festival where we?re going to try to integrate the people, to show 
them that we?re people, too, and not just war machines.
49
 
The film ? which in 2005 won an honorable mention at the Slamdance film festival and 
was chosen as best documentary at the Miami International Film Festival ? shows the 
block party briefly, with music and dancing, food and drink, and the filmakers reported 
no imediate outbreak of hostilities. By al acounts, everybody was happy that, if 
nothing else, the shootouts and the stray bullets were a thing of the past. 
But ?dison was realistic about the prospects for lasting peace. ?For now we?re 
living wel, because they have theirs and we have ours,? he said, refering to the division 
of teritory betwen Metro and Cacique Nutibara. ?But they ? I know that later they?re 
going to want al of it.?
50
 
5.2.4. Cacique Nutibara Targets Metro 
In late 2002, Jes?s Alberto Mart?nez, ?El Mocho,? was building a homemade 
grenade when it acidentaly detonated. He lost his left hand in the explosion. He was a 
member of Metro under Jason and ?dison?s command, and was one of the documentary 
film La Siera?s main protagonists. In the film?s first scene with him, Jes?s is snorting 
cocaine, and he is shown throughout the film smoking pot and snorting coke with his 
friends. In most of his on-scren interviews, he speaks deliberating and lazily, as if he 
were high. In one, he spoke about one of the rules the Metro Bloc imposed on the 
                                                
49
 La Siera: Urban Warfare in the Barios of Medel?n, Colombia, DVD, documentary film writen and 
directed by Scot Dalton and Margarita Martinez (Medel?n, Colombia: Human Rights Watch, 2005). 
50
 Ibid. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 202 
community: You have to smoke marijuana only where people can?t se you. ?Not like 
those other barios,? he added with ironic contempt.
51
 
?A lot of mistreatment? is how Jes?s described life under M-6&7, ?a lot of 
mistreatment by the gangs that roamed around here.? He explained that he and some 
friends decided that things needed to change, so they put together a crew and, just as the 
story has been told already (se ? 5.2.2), took control of the area. There were many wars 
? this was the fourth gang war in the sector, but each war had many batles separated by 
tense periods of calm ? and he had clearly become resigned to his fate. ?We might 
survive [one war, but] another wil come, another wil come, another wil come, until 
eventualy you get kiled ? maybe not in this war, but there wil always be more wars. 
After one comes, another comes.? The peace that setled on East Central Medel?n after 
the joint operation against the ELN by Cacique Nutibara and Metro did not last more than 
a couple months. As ?dison had predicted, and as Jes?s explained, Cacique Nutibara now 
wanted to take over Metro?s teritory. ?And they?re narcos!? Jes?s exclaimed, taking a hit 
of wed. ?This neighborhood is our life, and we?l defend it with our lives. As long as 
we?re here, as long as we?re alive, they won?t get us out of here.?
 52
 
?Rodrigo 00? Garc?a was, by that point, already at war with the AUC in the rest of 
the country, and he was losing: since he did not have the vast resources of the narco-
backed paramilitaries, he could only fight, as one source put it, until his amunition ran 
                                                
51
 Ibid. 
52
 Ibid. The scene this quote is taken from apears only in the pirated version of the film. The filmakers 
had shown an unedited cut of their documentary to the community, and had left a copy of it with the main 
protagonists. Within days, it was being sold on the black market in Medel?n. Jes?s told Mart?nez that, 
when he went downtown to sel his copy to the video pirates, ?somebody had already beaten him to it.? The 
version of the film available in the United States is the shorter, edited edition. Margarita Mart?nez Escalon, 
?As? Filmamos la Sierra?; Margarita Mart?nez Escal?n, Interview No. 26. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 203 
out. The state?s security forces unofficialy aligned themselves with Cacique Nutibara, 
targeting Metro in raids as they had done to M-6&7 before they had become paras. They 
never knew if the next atack was going to come from the paras who controlled the 
peripheral barios, from the guerilas who had fled to the mountains, or from the security 
forces stationed downtown. They were under atack from al sides. Metro was running 
out of amunition. 
On Saturday, 24 May 2003, ?dison bought a new shirt. It was Jason?s birthday, 
and he wanted to wear it that night when they went out to celebrate. The television news 
reports said that he was kiled in a shootout with the ELN. In fact, sometime after the 
party, he and a friend found themselves under atack by the Army?s special forces.
53
 In 
the documentary of the war, a young boy nicknamed ?Pirulu,? not yet a tenager, is 
shown throughout the film following Jason?s crew around, drinking beer with them, 
acting as a lookout, carying their amunition. Jason appears (uncredited) only once in 
the film, and that is in the scene when young ?Pirulu,? who had witnesed the batle, was 
teling Jason what had happened
54
: 
?dison came up first, with [another Metro member] behind him. So 
?dison hid behind that [broken-down car]. .. That?s when they started 
shooting. Blam, blam, blam, blam, blam, blam! So ?dison came down 
here and he got hit in the leg, and they caught him, dragged him over 
there, and shot him in the head.
55
 
He was buried two days later, his funeral atended by an outpouring of community 
members, including the seven tenage girls who were the mothers or mothers-to-be of his 
                                                
53
 Ibid. 
54
 Margarita Mart?nez Escal?n, Interview No. 26. 
55
 La Siera. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 204 
children. ??dison was an amalgam of complexities, of beauty and darknes,? Mart?nez, 
the filmaker, would write about him later, ?a born leader with hopes of doing 
something more with his life, but who, in his misfortune, came from a world marked by 
violence.? That violence marked his youth, but he managed to transform his lot into 
something beneficial to many whose lives were also marked by violence, to become a 
respected community leader. Yet he made his mark on the world with no smal amount of 
violence himself. His public persona hid a dark side. ?His disregard for life was 
shocking.? Mart?nez wrote. ?[During] shootouts with rival factions, he would just fire 
into the void, without any apparent target and without the slightest precaution against 
wounding people who weren?t involved. In that, he was like everyone else.?
56
 
Jason had already been geting cals from prison, from men offering him oney 
and power if Metro would just give up the fight and join Cacique Nutibara. But he had 
resisted their offers, his men afraid of what would happen once Cacique Nutibara took 
over the neighborhoods they had grown up in. War had broken out betwen them ? not 
at al beneficial to the neighborhoods they had grown up in ? and now his top deputy, 
the man whose charisma had kept the unit?s morale high and its relations with the 
community warm, was dead. A few months later, Jason got another cal from prison. Was 
he ready to switch? This time the pot had been swetened: the government was now 
offering the paras amnesty from prosecution and paid jobs in exchange for their agreing 
to lay down their weapons and renter civilian society (se next chapter). ?Rodrigo 00? 
                                                
56
 ??dison era una amalgama de complejidades, de bellezas y oscuridades, un l?der nato con ganas de hacer 
algo con su vida, pero que, para su infortunio, ven?a de un mundo marcado por la violencia. ? Su 
desprecio por la vida era impresionante. Por ejemplo, en las plomaceras con las faciones rivales disparaba 
al vac?o, sin objetivo aparente y sin la menor precauci?n de herir a gente no involucrada. En eso era como 
todos.? Margarita Mart?nez Escalon, ?As? Filmamos la Siera.? 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 205 
Garc?a, whose Metro Bloc had by then been reduced to les than half of the 
municipalities he had controlled just a couple of years earlier, was having none of it. He 
wanted the state to negotiate with the ?political? paras, such as Metro, separate from the 
?narco? paras, and until the state agred to that, he refused to demobilize what remained 
of Metro, including what was left of his urban units. Jason was in charge of one of those 
urban units, the last Metro holdout in Medel?n, and he was ready for the whole thing to 
be over. So were most of his crew. In Caicedo La Siera, they laid down their weapons, 
and Metro became Cacique Nutibara. 
Cacique Nutibara and its boss, ?Don Berna? Murilo, now had, for al practical 
purposes, a monopoly on Medel?n?s underworld. There was noone left for them to fight. 
Jes?s, the Metro soldier who had lost his hand, was talking with another friend 
about the demobilization, about what kind of job they might get, about earning a normal 
salary. ?My family?s very happy about this,? Jes?s told his friend. ?It?s a good idea, isn?t 
it?? His friend responded, ?To leave the weapons behind.? Jes?s: ?And the drugs.? His 
friend smiled: ?The drugs [we keep] for ourselves.? They laughed, probably not realizing 
how profoundly their joke foretold of more violence to come in the wake of the paras? 
demobilizations. Jes?s turned out not to be eligible for the first round of demobilizations: 
he had lost his identification card.
57
 Things would change, and he later demobilized, but 
like many of those who laid down their weapons in the years after the end of the war, he 
did not stay demobilized for long. His return to a life of crime continues in the next 
chapter. Jason, the leader of M-6&7 and then Metro?s Caicedo La Siera unit, did 
demobilize with Cacique Nutibara. But, like many others in his position, he was later 
                                                
57
 La Siera. 
Lamb, Robert D. | Chapter 5. Mafias and the Emergence of Paramilitaries: 198-2002 206 
kiled in a dispute with one of his former brothers in arms. ?Pirulu,? the young boy who in 
the documentary of their lives was the one teling Jason how ?dison had been kiled, the 
boy who was a lookout and a munitions porter, was himself kiled in the bario a few 
short years after the war had ended. He was 17. 
Even after the wars, there was stil violence in Caicedo La Siera; it had just 
changed form. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 207 
Chapter 6. Traffickers and the Corruption of Paramilitaries: 2002-2007 
The rise of Colombia?s national paramilitary asociation, discussed in the last 
chapter, took place in a national context of growing public frustration with the 
government and the guerilas who were fighting them. The cumulative result of a long 
string of government failures in that fight over the previous 20 years was a public who 
were growing cynical about politicians and political parties and their ability to bring 
peace ? a public who therefore were wiling to back even something as ruthles as the 
United Self-Defense Forces of Colombia (AUC: Autodefensas Unidas de Colombia) if it 
could raise their hopes at al. But even the AUC would dash their hopes, and by 2002 the 
country was very much ready for a change. What follows in this introduction is a brief 
discussion of the evolution of public atitudes about the guerilas, the paras, and the 
state, by way of seting the context for the narco paras? takeover of the AUC. 
During the presidential administration of Julio C?sar Turbay Ayala (Liberal Party, 
1978-1982), the civil war took a turn for the worse, driven in large part by a worldwide 
economic recesion. President Belisario Betancur Cuartas (Conservative Party, 1982-
1986) tried to cool it off through an important national strategy that involved negotiating 
with the guerilas, but after the failure of peace talks and the disastrous takeover of the 
Palace of Justice by the M-19 guerilas on 6-7 November 1985 ? which ended in the 
deaths of al the guerilas, half of the Colombian Supreme Court, and a hundred other 
hostages ? the conflict only heated up further. Virgilio Barco Vargas (Liberal, 1986-
1990) did manage to negotiate the demobilization of the M-19 ? Colombia?s only 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 208 
succesful demobilization ? but he also ended up presiding over a period of the greatest 
increase in violence in Colombia?s history, during which hundreds of narcotraficking 
and paramilitary groups appeared in force throughout the country, adding their violence 
to that already perpetrated by the guerilas and the state?s own security forces. 
Presidential candidate Luis Carlos Gal?n of the New Liberalism faction of the 
Liberal Party was asasinated by drug trafickers in 1989 (se ? 2.1.3), and C?sar 
Augusto Gaviria Trujilo, a Gal?n aide, won the election in his place (Liberal, 1990-
1994). Now very personaly aware of the risks his country?s violence posed and hoping to 
stem its tide, Gaviria tried opening up the political system by appointing a demobilized 
M-19 guerila to his cabinet, supporting the establishment of the 1991 Constitution, 
expanding government programs to addres social grievances, and negotiating the 
surrender of narcotrafickers and the demobilization of guerilas. Homicide levels and 
other forms of violence did fal, and by the end of his presidency, he had taken down the 
Medel?n Cartel, was starting on the Cali Cartel, and had succeded in demobilizing most 
of the guerila group EPL (Ej?rcito Popular de Liberaci?n, People?s Liberation Army). 
He ended his presidency with a majority of Colombians approving of his performance.
1
 
Yet his presidency was the period when the guerilas lost their patrons in the Soviet 
Union, which collapsed in 1991, and turned increasingly to the drug trade to finance their 
operations ? and the increase in income gave them aces to higher-quality recruits and 
more lethal weapons, which made them uch more formidable against the demoralized, 
underfunded, and mostly conscripted armed services. 
                                                
1
 All public-opinion polling data in this section are taken from Galup Colombia Ltda., Galup Pol 
Bimestral: Pol 62 (Bogot?: Galup Colombia, 208). 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 209 
But what realy drove the growing cynicism in Colombia, and the growing 
support for the paras, happened during the next two administrations. The Liberal Party 
that Gaviria?s presidency had helped to renew suffered fatal damage during the term of 
his succesor, Ernesto Samper Pizano (1994-1998). The scandal began during the 
campaign of 1994, when rumors surfaced that the Cali Cartel had donated large sums of 
money to support Samper?s candidacy. He won the election anyway. But after the 
election, casetes surfaced of conversations betwen a Cali Cartel spokesman and 
Samper?s campaign staf. The subsequent investigation, which uncovered what came to 
be known as the ?Proces 8000? scandal (named after the case number on the files of the 
official investigation), and led to the arests of dozens of mostly Liberal Party oficials ? 
including Samper?s Minister of Defense, Fernando Botero, and other ministers ? and to 
the public revelation of pervasive corruption throughout elite Colombian society. 
Disgusted, the U.S. State Department canceled Samper?s visa. Domestic pesimism grew 
as wel. Whereas at the end of the Gaviria administration only 31 percent of Colombians 
polled by Galup believed that things in the country were ?geting worse? (38 percent 
thought things were ?geting beter?), that figure rose to 44 percent at the beginning of the 
Samper administration, to 70 percent by October 1995, and to 78 percent by October 
1996 (he would end his presidency with a just a 30 percent favorability rating). To 
distance himself from the budding Proces 8000 scandal, Samper went hard after the Cali 
Cartel and finished dismantling it (most of its leaders had been arested by July 1995), 
amended the 1991 Constitution to alow for the extradition of Colombian nationals, and 
sponsored other reforms that strengthened the government?s hand against 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 210 
narcotrafickers.
2
 But those succeses hardly made a dent in the drug trade. It succeded 
mostly in fragmenting the ilicit economy: after the fal of the Cali Cartel, drug 
traficking organizations, such as ?Don Berna? Murilo?s, shifted from cartel to more 
difuse oficina arangements, making it rather more dificult for police to trace links 
among trafickers. Moreover, with so much drug money at its disposal, the Revolutionary 
Armed Forces of Colombia (FARC: Fuerzas Armadas Revolucionarias de Colombia) 
became a wealthy, nearly unbeatable insurgent force, while the number of paramilitaries 
fighting them grew from nearly 3,000 combatants in 1995 to almost 4,000 in 1997, the 
year Casta?o organized the paras into the AUC.
3
 Samper made some eforts to stem their 
growth, including some military actions against them. But in the view of many in the 
security forces and the Colombian public more generaly, the state simply wasn?t capable 
of beating the guerilas in batle: at least the paras were giving the guerilas a reason to 
negotiate a peace agrement. 
Peace became a public mandate during the 1998 election campaign, which was 
won by Andr?s Pastrana Arango (Conservative, 1998-2002). The FARC had been 
demanding for some years a despeje (demilitarized zone) where it could operate frely, as 
a precondition for negotiations with the government. Initialy the request was for one 
municipality, then in 1996 it was for four municipalities. By the 1998 campaign, the 
guerilas were demanding five and the demand kept growing. Pastrana, who would take 
office in August 1998, announced he would grant the despeje so that negotiations could 
begin. By the time the last of the armed forces, police officers, judges, prosecutors, and 
                                                
2
 Alfonso Monsalve Sol?rzano, Legitimidad y Soberan?a en Colombia 1958-2003, 194. 
3
 Colombian Ministry of Defense figures cited in Ibid., 197. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 211 
all other state oficials except for mayors were ordered out of the zone in December of 
that year, the area in the southern region of the country that had been designated and 
granted to the FARC as a demilitarized zone had grown to more than 42,000 square 
kilometers ? the FARC now controlled and governed a statelet within Colombia that 
was as large as the entire country of Switzerland. 
But as early as the negotiations with the state over the agenda of the peace talks, 
the FARC began using that statelet to rearm and regroup: ?they planned actions, trained 
their troops, hid hostages and negotiated their release, stored stolen vehicles, acumulated 
explosives, [and] maintained and expanded their ilicit coca crops and procesing 
laboratories.?
4
 But Pastrana continued the negotiations over the negotiations, making 
concesion after concesion to the guerilas, until finaly an agenda was agred upon, on 
6 May 1999, containing 12 points with more than 100 subpoints intended to addres, in 
one acount, ?fundamental aspects of the economic, political, and social life of the 
country.?
5
 The public was already angry about the deal: once the agenda was signed, 80 
percent of Colombians polled by Galup said they disapproved of how Pastrana was 
managing the guerilas, and that disapproval held at more or les that level until the end 
of 2001. Yet even as the guerilas repeatedly broke their side of the bargain, even as they 
continued kidnapping and kiling and seting off car bombs and laying landmines, 
Pastrana kept taking them at their word, ofering them what they asked for ? national 
                                                
4
 ?? planeaban sus aciones, entrenaban sus tropas, escond?an secuestrados y negociaban sus rescates, 
guardaban caros robados, acumulaban explosivos, manten?an y aumentaban sus cultivos il?citos y 
laboratorios de procesamiento de coca ?.? Ibid. 
5
 ?? aspectos fundamentales de la vida econ?mica, pol?tica y social del pa?s ?.? Ibid., 19. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 212 
recognition, international recognition, prisoner exchanges, etc. ? as if he believed they 
would at some point begin to make good on their word. 
They never did. Word was beginning to leak out of the demilitarized zone about 
narcotraficking activities and human-rights violations, such as summary executions and 
the persecution of religious minorities. It continued atacking the country?s infrastructure 
and laying land mines and seting off car bombs. In late 2001, the group asasinated a 
former cabinet minister, an incumbent senator, and the senator?s family. In February 
2002, it hijacked an airplane containing another senator who was also a human rights 
commisioner ? then bombed a bridge that took down with it an ambulance carying a 
pregnant woman and two other women, kiling them al. The kiling of a pregnant woman 
in an ambulance was one atrocity too many: the Colombian public had had enough, and 
so had Pastrana. He caled of the negotiations and ordered the military to retake the 
zone. (The National Liberation Army [ELN: Ej?rcito de Liberaci?n Nacional] had also 
been given a demilitarized zone for separate negotiations with the state, but the locals 
there formed self-defense groups to fight the guerilas themselves.
6
) By then, Pastrana 
had already become one of the most hated presidents in recent history. Galup found his 
approval rating at just 17 percent in December 2001, with 74 percent of those polled 
reporting an unfavorable view of him, the same percentage of people who thought things 
were geting worse in the country (only 10 percent thought things were improving); 
disapproval of his handling of the negotiations was up to 87 percent. 
Stil, while his administration?s most important policy was a complete failure in 
terms of its stated objective ? a negotiated setlement with the FARC ? it did 
                                                
6
 Ibid., 197. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 213 
acomplish something else that would turn out to be at least as significant, something 
Pastrana is rarely given credit for: the definitive ilegitimation of the insurgent project in 
Colombia. By publicly taking the guerilas at their word and offering them practicaly al 
the concesions they had been publicly saying they required before beginning peace talks, 
Pastrana was unmasking the FARC for what, by al evidence, they had become: part 
narcotraficking organization masquerading as a people?s insurgency, part extremist 
organization masquerading as a political movement wiling to negotiate peace for justice. 
Once exposed, they lost almost al of their remaining support within the population: in 
February 2000, the FARC?s favorability rating was stil 3 percent of the population; by 
April 2002, it was down to just 1 percent; even given that both figures are within 
Galup?s margin of eror and that its method likely failed to acount for the views of rural 
vilagers in FARC-controlled teritory, where its traditional base of support had always 
been, that is stil an extraordinarily low level of public support, and it indicates not just 
that the FARC had failed to legitimize itself beyond a tiny constituency but that it had 
actualy ilegitimized itself as a result of its response (narcotraficking) to the loss of its 
foreign patron (the Soviet Union) during the 1990s, wiping out most of its remaining base 
of support in the country. 
In other words, Pastrana may have left office a hated man, but he had 
acomplished something that no president before him had been able to do: He had caled 
the FARC?s bluff. It worked. 
Halfway through his presidential term, in June 2000, about half of the country stil 
believed that guerilas were capable of taking power in the country by force and that the 
Colombian military was incapable of defeating them ilitarily, about 7 or 8 points higher 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 214 
than those who believed otherwise. By the end of Pastrana?s term, things had changed: 61 
percent believed the guerilas were no longer capable of winning and 62 percent 
believed the Colombian military could defeat them; on both questions, only 32 percent 
thought otherwise. Wishful thinking or no, the tide had turned: since the guerilas ? al 
of them had been tainted by the failure of the FARC peace talks ? had demonstrated 
such bad faith, had demonstrated they never intended to negotiate in good faith, most of 
the public now believed (51 percent to 41 percent in June 2002) that the best option for 
dealing with them was not negotiation but defeat by the Colombian military. 
I say ?military? and not ?paramilitaries? because the paras had lost most of their 
public support as wel. Their atrocities during the Pastrana administration had become 
impossible to keep rationalizing, even by a public fed up with the nominal targets of the 
paras? growing campaigns of death and increasingly cynical about the government for 
whose counterinsurgency eforts the paras were claiming to be a proxy. The paras were 
the subject of powerful denunciations by domestic and international human rights 
monitoring organizations and the government was coming under harsh criticism from 
intergovernmental organizations for its failure to distance itself from their actions. 
Domesticaly, 63 percent of Colombians polled by Galup a few months into Pastrana?s 
term in December 1998 disapproved of the president?s handling of the paramilitaries; it 
was up to 82 percent by the end. Galup didn?t start asking about the paras? favorability 
ratings directly until February 2000, but by then 77 percent were reporting an unfavorable 
view, with just 9 percent in favor; the August 2000 poll did find 16 percent favorable and 
74 percent unfavorable ? but that was the best the paras would ever do again. At the end 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 215 
of the Pastrana administration the figures for public opinion of the paras were 6 percent 
favorable and 86 percent unfavorable. They had hardly more support than the guerilas. 
What was left, then, was the government, and it was scarcely doing beter in the 
public?s view than the nonstate actors. Samper had damaged the public?s confidence in 
the Liberal Party, and Pastrana had done the same to the Conservative Party. At the 
beginning of both of their administrations, for example, more people approved of their 
handling of coruption in the country than disapproved, but both ended their terms with 
the same number, 72 percent of the country, believing that they were not adequately 
addresing corruption. A lot of Colombians I spoke with informaly talked about the 
incredibly low morale in Colombia before the 2002 presidential campaign, the high level 
of cynicism, and the fear that, not only was the government ilegitimate, but so were all 
of the alternatives: the guerilas and the paras. Fern?n Gonz?lez, too, described the 
climate of fear that was overtaking the country at the beginning of the 2002 campaign, 
amid ?the expansion of guerila activity toward the more central and integrated zones of 
the country, which ended up afecting the outskirts of the biggest cities and the roads 
connecting them, [and which] produced as a response the organization and expansion of 
right-wing paramilitary groups?: 
This change in teritorial logic caused the war among the guerilas, the 
paramilitary groups, and the state to stop being sen as something that was 
taking place in peripheral regions, far from the biggest cities and the 
centers of economic production, and to become perceived as a threat to the 
development and daily life of al Colombians, including city-dwelers. 
Many sectors of the population thought that they could be kidnapped or 
extorted at any moment ? [and they] felt trapped in their towns and cities 
?. Colombians believed ? that the nation?s woes were due in large part 
to the incompetence of their leaders and representatives, together with the 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 216 
corruption and political maneuvering invading state agencies, Congres, 
and political activity in general.
7
 
In short, the state was sen as increasingly weak ? the terms of art at the time were 
?failing? or ?collapsing? ? and the political parties and government bureaucracies, and 
often even the Colombian state itself, as les and les legitimate.
8
 If anything, however, 
people feared the nonstate actors more than the state actors; some even held out hope that 
their country might someday start acting like a state. If some political candidate could 
come along and fed that hope, could give Colombians a reason to believe that things 
realy could get beter, he would be a shoe-in for the presidency. To put it another way, 
what the country needed, in the popular view of the time, was somebody independent 
from the traditional parties, somebody wiling and able to defeat the guerilas militarily, 
and somebody who could set a good example for the country?s political establishment. 
?lvaro Uribe V?lez had been a former senator, governor of Antioquia, and mayor 
of Medel?n, al as a member of the Liberal Party. But when he ran for president he did so 
as an independent. He never technicaly left his party, but he did manage to distance 
                                                
7
 ?? la expansi?n de la actividad de la guerilla hacia zonas m?s centrales e integradas del pa?s, que lleg? a 
afectar los alrededores de las ciudades m?s grandes y las v?as de comunicaci?n entre elas, produjo como 
respuesta la organizaci?n y expansi?n de grupos paramilitares de derecha. Este cambio de l?gica territorial 
hizo que la guera entre las guerilas, los grupos paramilitares y el Estado dejara de ser vista como algo 
que pasaba en las regiones perif?ricas, lejos de las ciudades importantes y los centros de producci?n 
econ?micos, para pasar a ser percibida como una amenaza para el desarrolo y la vida cotidiana de todos los 
colombianos, incluso los citadinos. Muchos sectores de poblaci?n pensaban que pod?an ser secuestrados o 
extorsionados en cualquier momento ? [y] se sent?an encerados en sus pueblos y ciudades ?. [Los] 
colombianos consideraban ? que los males de la naci?n se deben en gran parte a la incompetencia de sus 
l?deres y representantes, junto con la corupci?n y politiquer?a que invaden las agencias estatales, el 
Congreso y la actividad pol?tica en general.? Fern?n Gonz?lez, ?El Fen?meno Pol?tico de ?lvaro Uribe 
V?lez: ?De D?nde Proviene la Legitimidad de Este L?der Elegido por Segunda Vez Como Presidente?,? 
[?The Political Phenomenon of ?lvaro Uribe V?lez: Elected President for the Second Time, Where Does 
This Leader?s Legitimacy Come From?,?] Instituto de Investigaci?n y Debate sobre la Gobernanza 
[Institute for Research and Debate on Governance], published electronically by Equipo de Investigadores 
del Centro de Investigaciones y Educaci?n Popular (CINEP), http:/ww3.institut-
gouvernance.org/es/analyse/fiche-analyse-245.html (acesed 8 February 2009). 
8
 Eduardo Posada Carb?, ?Ilegitimidad? del Estado en Colombia. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 217 
himself from it. A dark-horse candidate for president until late in the Pastrana 
administration (amid the failure of the peace talks), he campaigned on the theme of ?firm 
hand, big heart? and proposed a series of reforms intended to improve both the country?s 
security and its democracy: to strengthen the military and police to fight trafickers and 
guerilas, to treat subversives not as political actors worthy of recognition but as 
narcoterorists worthy of defeat, to clear the highways of guerila and paramilitary 
checkpoints, to boost economic activity and investment, and to restablish the state?s 
presence in al municipalities of the country. He was a notorious workaholic, the type of 
person who needed only a few hours of slep a night, and his tirelesnes and self-
discipline were asets to the campaign, something that most Colombians considered to be 
admirable and perhaps worthy of imitation. He won the election with 53 percent of the 
popular vote and took office in August 2002 with a 69 percent approval rating. He turned 
his campaign promises into an overal strategy he caled Democratic Security. In fact, he 
was so succesful in implementing this strategy and so popular during his first term in 
office, that he was able to persuade Congres, with broad public support, to amend the 
Constitution to alow him to run for a second term, which he won easily in 2006. His 
second-term strategy focused on what his administration caled the ?consolidation? of 
democratic security.
9
 
Operation Orion took place a few months into Uribe?s first term. The joint Army-
Police raid of Comuna 13 in Medel?n succeded in ejecting the last of the FARC from 
the city in October 2002, an operation that stood in stark contrast to an identical efort 
                                                
9
 Ministerio de Defensa Nacional, PSCD: Pol?tica de Consolidaci?n de la Seguridad Democr?tica [Policy 
for the Consolidation of Democratic Security] (Bogot?: Ministerio de Defensa Nacional, Rep?blica de 
Colombia, 207). 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 218 
just a few months earlier, at the end of the Pastrana administration. Whether its succes 
had anything to do with Uribe?s leadership, it nonetheles signaled a major change in the 
country ? and in the city of Medel?n. Before Uribe?s election, in July 2002, 49 percent 
of people polled in Medel?n said they believed things were geting worse in their city; in 
September 2002, after his election, that fel to 38 percent; at the next Galup poll in 
January 2003, a few months after Orion, it dropped to 10 percent. At the same time, the 
percentage of people saying things were geting beter jumped from 27 percent before the 
election to 42 percent after the election to 79 percent after Orion. 
Uribe complied with his campaign promise to take back the highways, and did so 
in dramatic fashion. On national television, caravans of security vehicles were shown on 
the country?s highways escorting scores of Colombians traveling betwen cities in their 
own vehicles, something that had long been simply too dangerous even to consider. Soon 
enough, the caravans disappeared as Army posts were stationed every few kilometers on 
the highways, with bilboards appearing al over the country trumpeting the heroism of 
the soldiers protecting the nation?s highways. Travel betwen cities became safe for the 
first time in many years ? an enormous boost to the morale of a people who, in 
Gonz?lez?s words, had previously ?felt trapped in their towns and cities.?
10
 
At least as significant was his ability to propose and succesfully implement a 
one-time tax on wealthy Colombians as a way to fund his policies; in Colombia, where it 
had long been said that tax evasion was the national sport, it is nothing short of 
astonishing that so many actualy paid the tax. He pushed Congres to increase spending 
on the military and police forces, building on Pastrana?s increase in security spending, to 
                                                
10
 Fern?n Gonz?lez, ?El Fen?meno Pol?tico de ?lvaro Uribe V?lez.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 219 
further strengthen the foundation on which military investment could one day be 
sustained by fully domestic sources (the United States had been giving Colombia bilions 
of dollars a year to fund Pastrana?s Plan Colombia initiative). It did not hurt that during 
Uribe?s first term the economy was growing at a healthy pace, after the recesion years of 
his predecesor. 
Uribe continued to use his personal tirelesnes as an aset as wel. Every single 
wekend he left Bogot? and took one or more of his cabinet ministers and a few staf 
with him to a diferent municipality in a diferent part of the country, the more remote the 
beter, to engage in lengthy discussions with citizens in town-hal-style metings that 
often lasted many hours. Most previous presidents had started their political carers in 
Bogot? and spent most of their time in that capital city; Uribe had come from Medel?n 
and recognized the political value of leaving the capital regularly. At these community 
forums, he would hear people out as they expresed their opinions and told him about 
their problems, and he would imediately asign his staf or ministers to take care of 
those problems, even handing out his own cel-phone number to locals so they could 
follow-up as needed. ?With his recognized slogan of work, work, and work, Uribe 
fascinated citizens who had had the idea that the people who govern steal much and work 
litle. ? These [community forums] helped him project an image of a man who was 
diligent and efective in resolving problems, sensitive and acesible to citizens, and 
talented in leadership.?
11
 
                                                
11
 ?Con su reconocido lema de trabajar, trabajar y trabajar, Uribe fascin? a una ciudadan?a que ten?a la 
imagen de que los gobernantes robaban mucho y trabajaban poco. ? Estos espacios de gobierno le 
sirvieron para proyectar un perfil de diligente y eficaz en la resoluci?n de las necesidades, sensible y 
asequible a los ciudadanos y poseedor de un don de mando.? Ibid. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 220 
In truth, he also benefited from a lot of policies that had been put in place by 
Pastrana, the most important being Plan Colombia, which, with bilions of dollars in aid 
from the United States, had strengthened the security forces (with beter training and 
useful equipment, including helicopters) and had initiated important reforms in the 
judicial system. The military under Pastrana also had quietly begun to revise its doctrines 
and improve its training for counterinsurgency, using money and advice from the United 
States to profesionalize its forces, require human rights training, offer beter pay, and 
shift its emphasis to volunters so it could decrease its dependence on conscripts.
12
 Even 
under Samper the military had recognized it had a problem with human rights. In 1994, 
Minister of Defense Fernando Botero laid out four ?long-term strategic goals?: (1) 
?promoting and strengthening a human rights culture and ethic within the military and 
police forces?; (2) ?creating the necesary tools and mechanism [sic] to make the military 
criminal justice system ? an eficient instrument?; (3) ?resolving situations of human 
rights violation in which members of the public may be invovled?; and (4) ?strengthening 
the legitimacy and credibility of the military forces both at the national and international 
levels concerning human rights. ? The state should atempt to not only recover its 
monopoly over armed forces, but to guarante its legitimate enforcement.? To those ends, 
the Ministry of Defense (MoD) created an ofice of human rights and political afairs, 
required al military and police units to establish human rights offices, and increased 
human rights training, among other eforts.
13
 Nearly a decade would pas, however, 
                                                
12
 Peter DeShazo, Tanya Primiani, and Philip McLean, Back From the Brink: Evaluating Progres in 
Colombia, 199-2007, A Report of the Americas Program, Center for Strategic and International Studies 
(Washington: CSIS, 207), 10. 
13
 American Embasy Bogot?, ?Botero Human Rights Leter to A/S Shatuck,? 9 December 194, cable to 
United States Department of State, published electronicaly by The National Security Archive, 
http:/ww.gwu.edu/~nsarchiv/NSAEB/NSAEB217/doc01.pdf (acesed 30 June 2009). 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 221 
before these internal reforms started showing real results. (Botero himself was later 
arested on charges of coruption.) 
Finaly, Uribe imediately began talks with the AUC over the possibility of their 
demobilization, disarmament, and reintegration (DR) into society. This, of course, had 
not been the first of Colombia?s demobilization atempts. ?Colombians have acquired the 
status of serial demobilizers,? Douglas Porch and Mar?a Jos? Rasmusen have writen. 
?Colombian presidents have isued amnesties so often that most could dictate them in 
their slep.? 
The proces has become ritualized: cease-fires are declared and 
negotiators asemble in jungle clearings. After violations caried out by 
renegade elements on both sides, insurgents eventualy mas in designated 
areas. Amid proclamations before an asembled international pres that 
Colombia has, at last, turned the corner in its long history of armed 
violence, weapons are surrendered and peace declared. So far ? [this 
history] has merely transitioned, rather than terminated, violence ? for 
the same reasons: the state lacks the power and legitimacy to enforce the 
agrements and the resources to integrate demobilized fighters into the 
legitimate economy. As a result, enough bad actors remain to cary the 
violence forward.
14
 
The story would be scarcely diferent with the demobilization of the AUC. After the 
military victory of Operation Orion in Medel?n, folowed a few months later by the 
victory of the AUC blocs against the city?s last remaining guerilas in East Central 
Medel?n in early 2003 (se Chapter 5), Uribe and the AUC decided on a pilot 
demobilization involving the Cacique Nutibara Bloc, which, if succesful, would be 
followed by succesive demobilizations of the rest of the AUC nationwide ? or at least 
of those that agred to participate. 
                                                
14
 Douglas Porch and Mar?a Jos? Rasmusen, ?Demobilization of Paramilitaries in Colombia: 
Transformation Or Transition?,? Studies in Conflict & Terorism 31, no. 6 (208): 520. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 222 
The period during which the AUC were demobilizing was a period of great hope 
throughout Colombia, nowhere more so than in Medel?n. The city?s homicide rate 
plummeted, from among the highest in the world to levels lower than the rates in 
American cities such as Detroit, Miami, Baltimore, and Washington. The same drop was 
sen at the national level, and in Caicedo La Siera as wel. Common crime in Medel?n 
also fel: asault and batery, robberies of delivery vehicles, kidnappings, and reports of 
extortion.
15
 By 2007, police were patrolling the strets and aleys of barios formerly 
controlled by militants, gangs, and paras.
16
 People were starting to go out dancing more, 
visit friends more, hang out downtown more, and take beter advantage of public spaces 
that had once been considered too dangerous. The mayor initiated a masive development 
efort, building two aerial cable-car systems to connect the hilside slums to the public-
transportation rail system, and building parks, soccer fields, community centers, and big, 
modern public libraries in the peripheral barios. The city began touting its turnaround as 
the ?Medel?n model of good governance and comprehensive social development? and 
invited foreign officials to come and visit the ?Medel?n Laboratory? to help the city 
experiment with new approaches to managing complex problems or simply to learn how 
they, too, could bring peace amid the complex violence of their own cities. Tourism was 
up, foreign investment was up, self-confidence was up, and fear and violence were 
down.
17
 How did this miracle happen? How long could it last? 
                                                
15
 Polic?a Metropolitana del Vale de Abur?. 
16
 Ralph Rozema, ?Urban DR-Proceses.? 
17
 Gerard Martin, et al., Medel?n: Transformaci?n de Una Ciudad; Medel?n en Cifras; Medel?n en Cifras 
[Medel?n by the Numbers] (Medel?n: Alcald?a de Medel?n, 206); Medel?n en Cifras [Medel?n by the 
Numbers] (Medel?n: Alcald?a de Medel?n, 205). 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 223 
6.1. Medel?n During the ?Miracle? Period 
6.1.1. The Fall of the ?Political? Paras 
On 10 September 2001, a day nobody yet knew ould be sen as the end of an 
era, the U.S. Department of State announced a revision to its list of foreign terorist 
organizations (FTOs). Already on that list had been Colombia?s two largest insurgent 
groups, the FARC and the ELN. To these were now added (efective 5 October 2001) the 
county?s paramilitary network, the AUC.
18
 The AUC and its member organizations had 
long been the subject of strong domestic and international criticism for their human rights 
abuses ? including in the State Department?s annual human rights reports ? but 
succesive governments in both countries had variously supported in secret or turned a 
blind eye to those abuses because, in addition to the harm caused to innocents and mere 
sympathizers, the paras had been the only forces in the country that consistently kiled 
real guerilas as wel: they were considered in many parts of the country and among 
many elites to be a necesary evil, distasteful but efective, and therefore les ilegitimate 
than the guerilas whose war, more and more people were feling, had destabilized the 
country long enough. But the FTO declaration held legal implications for U.S. foreign 
aid, which would make implicit support for the paramilitaries much more dificult. The 
move imediately legitimized the views of the paramilitaries? critics. The next day, when 
the United States suddenly found itself under atack, it declared war not merely against 
the terorists who had atacked it, but against ?terorism? itself. Imediately, as the 
                                                
18
 State Department Daily Briefing, 10 September 201; United States Department of State, 
?Redesignation of Foreign Terorist Organizations,? Federal Register 66, no. 194 (5 
October 2001). 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 224 
Colombian newswekly Semana wrote, ?sectors of the [Colombian] establishment who 
had once looked favorably [upon the paras] ? began to distance themselves now that the 
United States was painting the AUC as terorists.?
19
 
The AUC?s leader, Carlos Casta?o, foresaw greater opposition and quickly 
announced a strategy for the ?conversion? of the asociation from a paramilitary 
organization with political ambitions to a political organization with paramilitary 
backing. This strategy ? which involved lowering the profile of its violent activities and 
coercing aces to political offices at al levels of government, an approach that came to 
be known as ?parapolitics?
20
 ? would within a few short years exacerbate existing 
divisions within the asociation, spark a major political scandal, and lead to the break-up 
of the AUC and the asasination of many of its key commanders ? including Casta?o 
himself. 
One of the first things Casta?o wanted to acomplish was to lower the AUC?s 
profile as much as possible, and to that end he announced that there would be no more 
masacres, which in his judgment brought too much negative pres and unwanted 
atention. Ministry of Defense figures show that paramilitaries had been responsible for 
281 deaths in 42 masacres in 2001, a figure that fel to 54 deaths in 11 masacres once 
the conversion strategy was in place the following year
21
; the Conflict Analysis Resource 
Center (CERAC, Centro de Recursos para el An?lisis de Conflictos) recorded 132 
masacres commited by paras in 2001, 88 in 2002, and 44 in 2003, faling eventualy to 
                                                
19
 ?Sectores del Establecimiento que ve?an con buenos ojos ? comenzaron a marcar distancia ahora que 
Estados Unidos rotulaba a las AUC como teroristas.? Juanita Le?n, ?La metamorfosis de las AUC,? [?The 
Metamorphosis of the AUC,?] Semana, 23 December 202, 58-59. 
20
 Mauricio Romero, Parapol?tica. 
21
 Le?n, ?La metamorfosis de las AUC.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 225 
six in 2007.
22
 This decline in masacres was a direct result of the AUC?s strategic 
response to their marginalization by the United States. The figures, however, masked the 
fact that this was not, at the outset, a strategy of non-violence but a strategy of deception: 
Nevertheles, in the zones where the self-defense forces were disputing 
control, as they were in Medel?n, C?cuta, and Vale de Cauca, they 
caried out homicides individualy. What makes one think that this had the 
efect in the media of lowering the profile of the AUC?s military actions 
? since kiling one person a day for 15 days does not cal as much 
atention as kiling 15 people al at once ? is that atacks by this 
organization against the civil population continued during 2002.
23
 
Part of the strategy was to infiltrate the government at al levels, identifying 
candidates who were friendly to the paras? interests and threatening their political 
opponents and intimidating voters to ensure that their slates would win during the 2002 
elections. Mayors, town councils, and members of congres were elected in al regions of 
the country with the backing of local mafia bosses and narcotrafickers asociated with 
the AUC. With friendly officials in office, the AUC bosses expected certain benefits, 
ranging from government contracts for their legal busineses to favorable laws and 
regulations regarding the political standing of their members. With friends in high places, 
they believed, they could negotiate a demobilization without having to wory that they 
would be required to surrender their wealth, pay reparations to their victims, or spend 
lengthy terms in prison. In other words, their people on the inside would help them wipe 
the slate clean so that they could retire without worry ? or keep earning drug profits 
                                                
22
 Centro de Recursos para el An?lisis de Conflictos (CERAC) Base de Datos de Conflicto (CCDB-
CERAC V.8; acesed 11 June 2008). 
23
 ?No obstante, en las zonas donde las autodefensas disputaban el control, como en Medel?n, C?cuta y 
Vale del Cauca, se dispararon los homicidios individuales. Lo que hace pensar que se logr? el efecto 
medi?tico de bajarle el perfil a las aciones militares de las AUC, ya que matar de uno en uno cada d?a 
durante 15 d?as no lama tanto la atenci?n como asesinar a 15 personas de un tajo, los ataques de esta 
organizicaci?n contra la poblaci?n civil continuaron durante 2002.? Le?n, ?La metamorfosis de las AUC.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 226 
without interference. What they offered to their government representatives in return was 
electoral support, inside influence, and an opportunity to take credit for stabilizing the 
country. 
But Casta?o was concerned that the growing influence of narcotrafickers within 
the AUC would threaten the ultimate succes of his conversion strategy. Despite his own 
narcotraficking wealth, his position on the purpose of the AUC was farther away from 
that of the narco para ?Don Berna? Murilo, whose Cacique Nutibara Bloc was trying to 
monopolize the ilicit economy of Medel?n, and closer to that of the AUC?s purest 
political para, ?Rodrigo 00? Garc?a, whose Metro Bloc was batling Medel?n?s guerila-
backed militias. Narcotraficking was the asociation?s main source of income but it was 
also an important source of tension: the political paras (some of them, at least) considered 
drug money a necesary evil, but feared its corupting influence would prevent the 
achievement of their broader political goals and began to believe it was time to cut the 
links; the narco paras, on the other hand, if given a choice betwen the political project 
and the money they were making off the drug trade, would have chosen the drug money. 
Casta?o gave them that choice. He wanted to weaken the narco wing before it was 
too late. So in 2002 he offered himself ? both in an open leter and in secret talks with 
the U.S. Drug Enforcement Agency (DEA) ? as a mediator betwen the United States 
and the most powerful narcotrafickers in Colombia, most of whom had already 
purchased their membership in the AUC, having recognized that joining the paras was an 
excelent way to win political imunity for their crimes (and thereby avoid extradition); 
the United States wanted a lot of those trafickers extradited and jailed in the United 
States. Casta?o caled a meting of the hundred most important trafickers in the country 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 227 
and persuaded half of them to sign a leter to the U.S. State Department offering to 
negotiate a setlement of some sort, possibly sacrificing their drug asets and entering the 
legal economy in exchange for non-extradition. Most of those who didn?t sign were 
already enjoying the benefits of membership in the paramilitary asociation ? national 
coordination, political status, etc. ? without even concerning themselves with working 
toward the AUC?s original mision of counterinsurgency, unles doing so happened to 
benefit their trade. They were beting that they could beat the political paras at their own 
game and keep winning drug profits indefinitely. 
Casta?o, his military chief Salvatore Mancuso, and ?Rodrigo 00? Garc?a, now one 
of his strongest alies, were disgusted. Even though both of them enjoyed vast wealth 
from their own asets, Casta?o and Mancuso considered themselves counterinsurgents 
first, with narcotraficking mainly as a means to that end, and they publicly denounced 
the rest of the AUC in August 2002, the month president ?lvaro Uribe V?lez took office. 
?We find ourselves with a series of groups, fragmented and deeply penetrated by 
narcotraficking, that in many cases went from confederation to anarchy or lost their 
principles.?
24
 A month later, the United States, apparently unsatisfied with the outcome 
of its secret talks with Casta?o, submited a request directly to the new president for 
Casta?o?s extradition ? on charges of drug traficking. 
Stil, the parapolitical component of his conversion strategy was working, at least 
through the end of 2002. The AUC?s candidates had won handily in elections throughout 
the country. After Operation Orion, Uribe?s High Commisioner for Peace, Luis Carlos 
                                                
24
 ?Nos encontramos con una serie de grupos atomizados y altamente penetrados por el narcotr?fico que, en 
muchos casos, pasaron de la confederaci?n a la anarqu?a o perdieron sus principios.? Quoted in Semana, 
?La Maldici?n de Ca?n: Qu? Llev? a Vicente Casta?o a Mandar Matar a Su Hermano Carlos?,? [?The Mark 
of Cain: What Led Vicente Casta?o to Order the Kiling of His Brother Carlos?,?] 26 Agosto 2006. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 228 
Restrepo, approached the AUC to suggest a unilateral ceasefire on the part of the paras so 
that they might begin DR talks. The AUC agred, announcing its ceasefire on 
1 December 2002 and promising an end to drug traficking and atacks against civilians 
while the framework for the talks was being negotiated. On 15 July 2003, AUC and 
government negotiators met at Santa Fe de Ralito in the department of C?rdoba, just 
north of Antioquia and close to Salvatore Mancuso?s ranch, and signed the framework for 
the negotiations. AUC leaders agred to maintain their ceasefire, support government 
counternarcotics eforts, and concentrate their troops in designated locations for the 
duration of the talks, the goal of which, acording to what became known as the First 
Santa Fe de Ralito Acord, was to demobilize al AUC combatants by the end of 2005, 
starting with a pilot demobilization of more than 850 members of the Cacique Nutibara 
Bloc in Medel?n on 25 November 2003. With the framework in place, the talks began in 
earnest. 
But without ?Rodrigo 00? Garc?a. He refused to participate in any proces that 
would legitimize narcotrafickers as political actors. In mid-2003, he proposed paralel 
talks in which he and the other political paras could hope to achieve the same objectives 
proposed at Ralito but without the taint of drug money. His proposals were ignored. The 
war betwen Metro and Cacique Nutibara that was taking place in Medel?n at the time 
was also taking place in macrocosm at the national level. Garc?a had been speaking out 
publicly against the narco paras for years. He?d found the emergence of ?Don Berna? 
Murilo as the AUC?s inspector general to be particularly grating, and a personal rivalry 
had grown betwen them, even as their urban units cooperated in anti-militia operations 
in Medel?n. Back in September 2002, as Operation Orion was about to get under way, 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 229 
Garc?a had started distancing the Metro Bloc from the AUC. And now, in May 2003, 
after the guerilas had been forced out of Medel?n and Cacique Nutibara had gone on 
the offensive against his men, he declared war against the AUC.
25
 
Within a few months, that war was over. Metro was defeated in Medel?n and had 
lost almost half of the other municipalities it had once controlled. The narco paras were 
winning. 
During the talks with the government, the paras were supposed to stay 
concentrated in certain designated zones of the country, which generaly meant they 
could travel betwen Ralito where the talks were being held and whichever of their 
ranches they chose to stay in. Almost completely marginalized within the organization he 
had founded with the help of his older brother Vicente, Carlos Casta?o was keeping 
himself in the location he considered safest, in a region his family had long dominated, in 
a ranch that abutted Vicente?s ranch. Known as ?El Profe? (?Profesor?) for his 
inteligence, Vicente, far more than Carlos wanted to admit, had no problem with the 
presence of the narco paras in the AUC, having recruited many of them into the 
asociation himself. But Carlos had always refused to denounce his brother during his 
own campaign against the narcos. Vicente was the quiet brother, the one whose image 
never appeared on television or in newspapers, who never spoke in public, who had never 
commanded a bloc. But he was very much respected among the country?s paras, having 
been the one who had quietly asisted many of them in building their own blocs and then 
introducing them to the broader community of paras. While Carlos?s star was faling, it 
was Vicente, along with Mancuso, who had succeded in persuading almost al of the 
                                                
25
 Semana, ?La Cacer?a de ?Doblecero?,? [?The Hunt for ?Double Zero?,?] 29 September 2003, 11. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 230 
country?s paras to enter talks with the state. He was always the problem-solver, the 
conflict-resolver, the counselor, and most importantly the strategist: Vicente, until then a 
businesman, joined his brother Carlos in 1995 after Fidel?s death, and he designed the 
paras? strategy of national expansion, which included forming the national asociation, 
the AUC, and cutting deals with whomever he needed to achieve national coverage: 
where paras were not available, he approached narcotrafickers to do the job. They had 
then grown to become dominant force among the paras, and Vicente was unashamedly on 
their side.
26
 
Vicente?s chief of security and most trusted aid was a pale-skinned paisa named 
Jes?s Ignacio Rold?n P?rez, alias ?Monoleche? (?Milky White?). In early March of 2004, 
Rold?n invited the best troops of the Peasant Self-Defense Forces of C?rdoba and Urab? 
(ACU: Autodefensas Campesinas de C?rdoba y Urab?) ? which had been founded by 
the Casta?o brothers in the years before the AUC ? to come to Vicente?s ranch. For a 
month, Rold?n trained them. He brought them new uniforms. He brought them new 
weapons. He had them escort Vicente betwen his ranch and the talks at Ralito. He was 
preparing them, they al could tel, for an important military operation ? the ceasefire 
agrement be damned ? but they didn?t know what it was going to be, or when. On 
16 April 2004, Rold?n arived on the ranch with enough trucks to transport 30 fully 
armed men. He did not tel them where they were going or what their mision was. But 
they didn?t travel very far along the rural highways. At 2:00 in the afternoon, the trucks 
turned in to visit the ranch next door. 
                                                
26
 Information for this and the next two paragraphs is taken from: Semana, ?Habla Vicente Casta?o,? 
6 June 205; Semana, ?La Maldici?n de Ca?n?; Gustavo Duncan, Los Se?ores de la Guera. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 231 
Carlos Casta?o was in his kitchen, his bodyguards outside. In a storm of gunfire, 
the bodyguards were dispatched quickly, and Carlos was told to come outside. No 
response. Two men were sent in to get him. They found him hiding in the refrigerator. 
Out of bullets, he surrendered without yet knowing who it was who had sent these men to 
kil him: which of his enemies among the narco paras was it? When he stepped outside, 
his arms tied up, the answer was imediately clear. But he asked anyway: ?Who ordered 
this?? he demanded of his brother?s most trusted aid. ?Monoleche? Rold?n?s response ? 
??El Profe.?? ? was followed imediately by a voley of twelve 9 milimeter rounds into 
Casta?o?s white shirt. His body was loaded into one of the trucks, covered with plantain 
leaves, taken to Vicente?s ranch, and hidden ? where, exactly, it would not be known 
until August 2006, when ?Monoleche? confesed the crime. Until then, one rumor had it 
that Carlos Casta?o was stil alive but had escaped to live a secret life outside of the 
country, the same rumor that had followed their brother Fidel?s disappearance ten years 
earlier. DNA evidence in September 2006 dispatched the more recent rumor. Carlos 
Casta?o was dead. 
?Rodrigo 00? Garc?a knew that ?Don Berna? Murilo and the rest of the narco 
paras wanted him dead, too. Aside from his own bodyguards, Carlos Casta?o had been 
the only thing standing betwen them and his own life. With Casta?o gone, he gave up 
the fight, disbanded most of the Metro Bloc, and went into hiding.
27
 
In his last e-mail to Margarita Mart?nez, the AP correspondent and documentary 
filmaker to whom he had given permision to film his men in La Siera, he expresed 
no fear of speaking frely about the AUC any longer. ?What can happen to me?? he 
                                                
27
 Amnesty International, Colombia: The Paramilitaries in Medell?n: Demobilization Or Legalization? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 232 
wrote. ?What, they want to kil me more times, or more intensely, or with a bigger 
weapon?? On 30 May 2004, possibly a day or so earlier, Garc?a was on Colombia?s 
Caribbean coast, in a port town caled Santa Marta, when he and a woman were walking 
out of a supermarket together. Five bullets to the head.
28
 He was 39 years old, and in his 
carer as a military and then a paramilitary commander he was never tempted nor 
corrupted by drug money; his fight was always to defeat the guerilas and end his 
country?s civil war. But after six years as commander of the Metro Bloc, the oficial 
records show, he left behind a very diferent legacy, far les noble: the number of people 
registered as victims of violence perpetrated by his units exceded 12,000.
29
 
6.1.2. The Fall of the ?Narco? Paras 
The first of the AUC units to demobilize was ?Don Berna? Murilo?s Cacique 
Nutibara Bloc. The ceremony took place on 25 November 2003 and was televised to a 
live national audience, as more than 850 former combatants handed in their (mostly 
broken and least favorite) weapons and pledged to renter society as peaceful civilians 
and productive citizens. Then they were carted of to the municipality of La Ceja, 
Antioquia, where they received training on the legal and social isues they were expected 
to face once released, and got whatever medical, legal, and psychological help the 
authorities thought they needed. ?During this time their criminal records were checked by 
the judicial authorities to ensure there were no criminal investigations pending against 
                                                
28
 Margarita Mart?nez Escalon (AP), ?Asesinan a Fundador del Paramilitarismo en Colombia,? La Prensa: 
El Diario de los Nicarag?enses, 31 May 202.  
29
 Sistema de Informaci?n de Justicia y Paz (SIJYP), [Justice and Peace Information System,] cited in 
VerdadAbierta.com, ?Las Auc.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 233 
them, before they received a de facto amnesty under Decre 128 and alowed to return 
home,? read one report on the demobilization. Decre 128 was a legal instrument isued 
on 22 January 2003 to regulate what benefits the government could bestow upon the 
demobilizing paras and the circumstances under which those benefits must be denied. 
The report continued: 
But the proces raised serious concerns over whether combatants were 
efectively being removed from or ?recycled? into the conflict. The fact 
that most demobilized paramilitaries would simply be alowed to return to 
their homes following a short rehabilitation course heightened concerns 
that paramilitaries would continue criminal operations on their return. ? 
The short space of time which judicial authorities were given to verify the 
criminal record of each demobilized combatant meant that it was unlikely 
that each could be subjected to a full and impartial judicial review. Many 
of those who participated in the demobilization ceremony ?, although 
possibly responsible for serious offences ?, would thus not be subject to 
criminal procedings.
30
 
A wek before Cacique Nutibara?s demobilization in November 2003, the Uribe 
administration also isued Decre 3360, which among other things removed the 
verification procedure intended to prove that those who had wished to demobilize with a 
paramilitary bloc had actualy been a member of that bloc; instead, the bloc?s leader 
needed only to submit a list of names. As a consequence, some unknown number of 
common criminals ?joined? Cacique Nutibara just a few days before demobilizing, 
thereby wiping clean their criminal records before returning to their barios a few eks 
later. 
Because many of the para leaders would not be eligible for benefits under either 
Decre 128 or Decre 3360, they agred to what came to be caled the Second Santa Fe 
de Ralito Acord on 13 May 2004. The government designated Santa Fe de Ralito as a 
                                                
30
 Amnesty International, Colombia: The Paramilitaries in Medel?n: Demobilization Or Legalization?, 38. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 234 
?placement zone? in which any paramilitary leader who wanted to demobilize would 
have to remain until a legal framework could be worked out for the demobilization. The 
administration then spent two years trying to get relevant legislation pased. The result 
was the Justice and Peace Law, ratified on 22 July 2005, which granted the paras just 
about everything they had set out to achieve when Casta?o first proposed his conversion 
strategy: It granted political status to paramilitary activities, thereby protecting them from 
extradition; provided for minimal prison sentences, some of which could technicaly be 
served out at their private ranches; did not technicaly require the dismantling of existing 
paramilitary structures, alowing instead for individual demobilizations, making it 
possible for the larger structure to continue its ilegal activities; protected most of their 
asets from repatriation or reparation; and did not even require a full confesion in 
exchange for pardon. Victims advocates were furious: the language recognized no right 
to truth, justice, or reparation and nearly guaranted impunity for the paras. The paras 
couldn?t have been happier.
31
 
Murilo was among the para leaders who had relocated to Ralito. His fighters 
from the Cacique Nutibara Bloc, however, had returned to their vilages and barios on 
16 December 2003, except for those who had reason to believe that doing so would be 
too dangerous. In those cases, the government was helping them relocate and start a new 
life. Al of the demobilized fighters from Cacique Nutibara (and subsequently from the 
rest of the demobilized blocs) became members of an organization formed for the explicit 
                                                
31
 Rodrigo Uprimny Yepes and Mar?a Paula Safon San?n, ?La Ley de ?Justicia y Paz?: ?Una Garant?a de 
Justicia y Paz y de No Repetici?n de las Atrocidades?,? [?The ?Justice and Peace? Law: A Guarante of 
Justice and Peace and No Recurence of Atrocities?,?] Revista Foro No. 55 (205): 49; Amnesty 
International, Colombia: The Paramilitaries in Medell?n: Demobilization Or Legalization?, 21; Herman 
Eduardo Nore?a Betancur, ?Los Paramilitares en edel?n.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 235 
purpose of representing their needs and rights to the government. Its status was officialy 
recognized in the formal peace agrement betwen Cacique Nutibara and the national 
government, which was signed at La Ceja, where the reducation was taking place, on 
10 December 2003, about a wek before the first of the demobilized paras went home: 
The government recognizes the non-governmental organization named 
?Corporaci?n Democracia? as the corporate representative of the 
reincorporated forces of the Cacique Nutibara Bloc, with whom wil be 
maintained a permanent dialog for the monitoring, development, and 
support of the proces of reincorporation. The national government, the 
office of the mayor of Medel?n, and Corporaci?n Democracia
 
 wil design 
monitoring, development, and support programs for the reinsertion 
proces.
32
 
As 13 more blocs demobilized betwen November 2004 and August 2005, many former 
paramilitary commanders became officers of Corporaci?n Democracia. Within a few 
years, almost al of those officers would be dead or in prison. 
During the reinsertion proces, a lot of paras were not able to find productive 
work and returned to crime. Others had never intended to distance themselves from 
ilegal activities in the first place, using the DR proces mainly as an easy way to wipe 
clean their legal record. One mechanism the ex-paras-now-criminals used to make money 
was the participatory budgeting proces that was put in place by Mayor Sergio Fajardo 
Valderama (2004-2007) as a way to improve citizen aces to city services. That proces 
gave the bario-level JACs substantial influence over the awarding of city contracts 
                                                
32
 ?El Gobierno nacional reconoce a la Organizaci?n No gubernamental denominada ?Corporaci?n 
Democracia? como la organizaci?n representante de los reincorporados del Bloque Cacique Nutibara, con 
quien se mantendr? una interlocuci?n permanente para el seguimiento, desarollo y apoyo del proceso de 
reincorporaci?n. El Gobierno nacional, la Alcald?a de Medel?n y la Corporaci?n Democracia dise?ar?n los 
programas de seguimiento, desarolo y apoyo al proceso de reinserci?n.? Quoted in Agencia de Prensa 
IPC, ?Nueva Captura en la Corporaci?n Democracia,? [?Another Arrest in Corporaci?n Democracia,?] 
VerdadAbierta.com, published electronicaly by Fundaci?n Ideas para la Paz (FIP) and Semana, 
http:/ww.verdadabierta.com/web3/conflicto-hoy/50-rearmados/1371-la-corporacion-democracia-
acoralada-por-la-justicia (acesed 2 August 209). 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 236 
(below a certain value) to local (bario-level) busineses, gave legal smal busineses 
beter aces to city money, and a gave citizens more say in how city money was spent. 
But it also posed an opportunity for the ex-paras. Some of them started approaching JAC 
officers and teling them to award the city contracts to the companies that some of the ex-
paras had formed as part of the demobilization proces. By threatening JAC officers who 
didn?t comply, kiling or forcing them out of the bario, the ex-paras could get easy 
aces to city money without necesarily having to then do the actual work the contract 
required. Later, when election time came, some ex-paras implemented a local version of 
Casta?o?s parapolitical strategy ? which was stil being implemented, right through the 
national elections of 2006 ? by cutting out the middle man (the JAC officers) and 
running for ofice directly. By threatening or kiling their political opponents and 
intimidating their opponents? supporters, they ensured that their own candidates would 
win the JAC elections.
33
 
It was political violence, but it was political violence ?corrupted? by economic 
violence. Or beter put: it was economic violence disguised as political violence. It was 
the most recent incarnation of a longer-term trend in Medel?n as wel as the country as a 
whole. Fiften years earlier, the militias in the peripheral barios had used political 
violence disguised as social violence, their social-cleansing campaigns intended as a 
means to win community support for the guerilas? broader political struggle. Later, they 
used economic violence disguised as political and social violence, their revolutionary or 
communitarian rhetoric only masking the fact that they had become drug trafickers and 
hit squads. Similarly, anti-union violence had once been part of a mainly political 
                                                
33
 This conclusion is based on many informal conversations, pres reports, and what amounts to comon 
knowledge in Medel?n, as wel as Ibid. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 237 
struggle, under the asumption that labor had an incurable afinity for Communism and 
was therefore by definition subversive. But hidden beneath the antisubversive rhetoric of 
the self-defense groups and paramilitaries was an undeniable economic motive: union 
organizers, after al, were a threat to the profits of businesmen and regional mafias. 
Again, economic violence was disguised as political violence. But as the guerilas 
themselves turned increasingly to economic violence disguised as political violence ? as 
they, too, were corupted by the influence of drug profits ? the political rationale for the 
anti-union forces gradualy eroded as wel, to the point where anti-union violence today 
cannot be said to have any conceivable political motive: it is, once again, economic 
violence disguised as political violence. And of course the takeover of the AUC by the 
narco paras was just one more example of the turn from political violence to economic 
violence. 
The self-defense groups, the guerilas, the militias, the paras, and now the 
demobilized paras ? al had been corrupted by the lure of easy money. Whatever 
collective violence remained in Colombia by the early 2000s could not be said to be 
principaly political or social by nature. Even where motives truly were mixed ? as in 
the case of the political paras who had considered drug money an easy means to anti-
guerila ends, or the political guerilas who had considered drug money an easy means 
to revolutionary ends ? the mix had an increasingly economic flavor. Colombia?s 
complex violence was becoming simpler and simpler with each pasing year: les 
political, les social, more economic. It could be argued, in fact, that political and social 
violence in Colombia ? including Medel?n ? has suffered a gradual delegitimation 
during the 1990s and 2000s as it has become increasingly ?corrupted? by economic 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 238 
motives. Support for and participation in subversive, anti-subversive, and social-
cleansing groups has been experiencing a downward trend for at least a decade and 
possibly longer. That trend had begun earnest at the city level after the ilegitimation of 
the militias and during the para-militia wars that folowed, and at the country level after 
Pastrana caled the FARC?s bluff and Uribe started taking back the highways. It has only 
acelerated since, as the generaly favorable impresions of Mayor Fajardo in Medel?n?s 
peripheral barios and of President Uribe countrywide started to win back for the state 
some of the legitimacy it had lost years earlier to the nonstate armed actors who later 
crushed the country?s hopes for peace. 
As ?Don Berna? Murilo was holed up in Santa Fe de Ralito waiting for the Justice 
and Peace Law to pas so he could receive imunity for his crimes, he continued 
commiting those crimes through what had become one of the most powerful and 
profitable international drug-traficking networks in history: the Envigado Ofice was the 
clear succesor to Pablo Escobar?s Medel?n Cartel. The word governance ? meaning 
policy-making, public-goods delivery, institution-building, and network management (se 
Appendix A) ? is sometimes translated into Spanish as gobernabilidad. It has not gone 
unnoticed in Medel?n that this word rhymes with donbernabilidad, or, to se it more 
clearly, DonBerna-bilidad. Who managed vast ilicit networks in Medel?n? ?Don Berna.? 
Who controlled scores of community organizations, JACs, judges, police officers, 
prosecutors, bus and cab companies, private security firms, even pharmacies and 
bakeries, and any number of other organizations and busineses that provided community 
services or kept the peace in Medel?n?s peripheral barios? ?Don Berna.? So who realy 
set policy in the city? Among those who looked with skepticism upon the increasingly 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 239 
popular idea that the city was coming into its own after the paras? demobilization, the 
answer was that the city stil operated at Murilo?s command and that city officials stil 
served at Murilo?s pleasure. 
?Don Berna,? wrote Mauricio Romero in the preface to his edited volume about 
the origins of the parapolitics scandal ? the AUC?s corruption of elections betwen 2002 
and 2006 came to light in late 2006 ? had ?won many confrontations?: 
against the FARC and the ELN, against the Terace Gang, and finaly, 
against the Metro Bloc. In 2001 the city ended up reaching 220 murders 
for every 100,000 residents, the highest rate in Latin America, which can 
be explained only by a veritable situation of war. He thus established 
control over security in the city and a decisive influence over electoral 
niches in the city wards and the municipalities of the metropolitan area, 
such as Envigado and Belo.
34
 
With so much power supposedly at his disposal in and around Medel?n, why did 
Murilo demobilize Cacique Nutibara and the other blocs he controled? Juan Carlos 
Palou has given thre reasons. First, because the paras simply weren?t needed anymore: 
the guerilas were out of Medel?n and no longer a threat, and the state had built up the 
strength of its military units to the point where they could be counted on to keep the 
guerilas in the jungles and out of the cities. Second, Murilo stil controlled the ilicit 
economy in Medel?n and had influence over it in most other regions of the country: he 
had been using the paras just to get certain political benefits, and now they paras were 
just dead weight on his organization. It wasn?t worth continuing to support the bario 
gangster-paras who had helped him, since now they could get job training and education 
                                                
34
 ?En Medel?n, Diego Murilo Bejarano, alias don Berna, varias confrontaciones: a las FARC y al ELN, a 
la banda La Teraza y, finalmente, al Bloque Metro. En el 201 la ciudad lleg? a tener 20 asesinatos por 
cada 10 mil habitantes, la tasa m?s alta de Am?rica Latina, s?lo explicable por una verdadera situaci?n de 
guera. Se estableci? as? un control sobre la seguridad de la ciudad y una influencia decisiva en los nichos 
electorales de las comunas y en municipios de la zona metropolitana, como Envigado y Belo.? Mauricio 
Romero, Parapol?tica, 20. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 240 
from the state; and those who weren?t interested in those benefits from the state could just 
keep working for him anyway. Third, he realy was afraid of being extradited to the 
United States, where he would certainly be unable to control his empire or enjoy his 
wealth as he could in Colombia ? even in prison if it ever came to that. The 
demobilization, therefore, was a way for him to earn both legal and moral protection from 
extradition: legaly, by gaining official political status, and moraly, by being sen as 
someone who was ?helping? the state and therefore worthy of keeping around. And the 
state kept him around, Palou argued, because at the time of his demobilization it did not 
have the military strength or the backing of the city?s marginalized citizens it would have 
needed to get into a direct confrontation with him: beter to engage Murilo in a DR 
efort where they could keep track of his activities in the short-term, then slowly chip 
away at his influence as the state built up its own strength and support in the years that 
followed.
35
 
The next section addreses the contrary view: that Murilo and city officials had 
entered a secret pact in which Murilo would control violence in the city but let Fajardo 
take credit for it, in exchange for the city?s granting him the benefits of demobilization 
but turning a blind eye to his criminal activities. Whatever the real nature of the 
relationship betwen Murilo and Fajardo, what can be said is that Murilo?s gamble did 
not pay off. Even if he had made a secret pact, the city did not end up living up to its end 
of the bargain. After a high-profile murder atributed to his men, Murilo was arested in 
October 2005. There was no search bloc, no shadowy death squad of his enemies helping 
to look for him: the police just went to Santa Fe de Ralito and seized him. On the day of 
                                                
35
 Juan Carlos Palou, personal comunication (Interview No. 3), 1 March 209. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 241 
his arest, there was a general strike of bus and cab drivers that halted most transportation 
within the city, a fact that some cite as evidence of Murilo?s vast power over Medel?n. 
But the strike lasted only a few hours; it was litle more than a symbolic protest. Escobar 
could shut the city down for weks. Murilo was stil in control of his empire, 
commanding his staf from the maximum-security prison in Itag??. But it would not last 
much longer. On 24 August 2007, the authorities, tired of hearing about his continuing 
control over the Envigado Ofice, transfered him to the maximum-security prison in 
C?mbita, in the department of Boyac?, greatly complicating his ability to communicate 
with his field commanders and thereby diminishing his influence. But things would get 
even worse for Murilo. On 13 May 2008, in the middle of the night and without warning, 
he and 13 other paramilitary leaders ? including Salvatore Mancuso ? were extradited 
to the United States. 
6.1.3. The Rise of the ?City of Eternal Spring? 
?Some people say that Medel?n today is a city of peace, a city of calm, a city of 
rights for everyone. But I have another view. It?s al bullshit. It?s a lie. It?s a city of 
paramilitaries.? The short man with the tailored suit and huge personality continued in 
this vein for an hour, pacing the room as he spoke: ?When one group is in control, 
obviously the crime statistics are going to fal. But extortion, prostitution, arms 
traficking, and drug sales al continue.? This is not a phenomenon of common-crime 
gangs, or of organized-crime gangs, but of paramilitary-crime gangs. The murder rate is 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 242 
down because the paramilitaries control the rural land surrounding the city where there 
are mas graves: ?If there are no bodies, there?s no crime.?
36
 
Luis Fernando Quijano Moreno was explaining to me his theory of the Medel?n 
Miracle. He spoke quickly, interupting the conversation frequently to check something 
in the other room or look up the name of a book he was thinking about. Antioquians wil 
make a deal with anyone if it wil benefit them, he said. It?s al busines. Militias have 
made deals with the government. Mafias have made deals with the government. Drug 
trafickers have made deals with the government. And the government has made deals 
with paramilitaries. Today they are caling the gangs of ex-paras bacrim, or bandas 
criminales emergentes (emerging crime gangs). But these bacrim, he said, represent 
nothing more than ?neoparamilitarism? in the city: it?s a ?paramafioso? project. He then 
proceded to lay out a series of 20 hypotheses that he said his organization was 
investigating regarding Medel?n?s neoparamilitarism. 
Quijano, head of an organization founded more than a decade ago by demobilized 
guerilas, ofered the most developed of what I would characterize as the conspiracy-
theory explanation of Medel?n?s peaceful period after 2003. Caling it a conspiracy 
theory is not at al meant to disparage his acount. Acording to my own findings, the 
conspiracy acount turns out to be more than half right. But it is a hard-line, 
antiestablishment position that sems to be based on a reading of the facts that overlooks 
evidence for any interpretation that would tend to cast a favorable light on the motives or 
actions of government oficials. It is a position shared by a pasionate minority in 
Medel?n, particularly among a subset of the city?s human rights and labor activists, 
                                                
36
 The quotes in this section are from Luis Fernando Quijano Moreno, Interview No. 5. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 243 
advocates and social workers for victims, and survivors of atrocities commited by 
paramilitaries and state security forces. It is a position that admits of litle possibility for 
government legitimacy, especialy that of the current president, ?lvaro Uribe V?lez, 
whom they consider, acording to a number of informal conversations I have had in 
Medel?n, to be the among worst presidents in the country?s history. Their disdain for 
Uribe is understandable: he was governor of Antioquia during the mid-1990s, a place and 
a time in which some of the worst paramilitary atrocities in the country?s history took 
place, leaving thousands of victims dead, displaced, and traumatized. That level of 
violence is dificult to forgive, and impossible to forget. Most people who are pasionate 
in their defense of victims and their promotion of human rights and humanitarian norms 
in Colombia, I should make clear, do not take such an extreme conspiratorial position. 
They recognize that there have been enough real examples of impunity and conspiracy in 
Colombia to keep them busy for years to come: they do not fel the need to manufacture 
any more. 
With Quijano, I did not engage in a debate about such isues: for my formal 
interviews, I was there only to ask questions and listen. Quijano, who was once 
asociated with the guerilas but today runs a research and advocacy NGO in Medel?n, 
has kept in contact with what he ambiguously cals the ?underworld? so that he can get 
information when he needs it for his research. To be sure I would not mischaracterize his 
comments, he recorded our conversation (my quotations, however, are based on hand-
writen notes that my research asistant and I took during the conversation). 
He was trying to argue, as others have, that the peace that descended upon 
Medel?n after 2003 was a result of an explicit but secret strategy by the government to 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 244 
outsource the city?s stability to the ?paramafiosi.? ?Medel?n is the model for the 
Democratic Security policy,? he said, explaining why the Uribe government has been so 
concerned about keeping the homicide rate low there. ?If it fails here, the policy fails 
nationaly.? The deals they had cut with the paras to keep the murder rate low had to be 
kept hidden, he argued, lest the policy be tainted by the scandal that would surely result. 
He wanted to explain why the secret pacts were made and how they came about. 
Let me ask you a question, he told me: ?Who is it easier to negotiate with: two hundred 
gangs, or one patron? The city cuts deals to let a patron consolidate control and control 
the violence.? The strategy had been used many times in the past: he suggested, without 
saying it explicitly, that such a deal had even been cut with Pablo Escobar. And the fact 
that violence fel imediately after ?Don Berna? Murilo had managed to take control of 
ilicit activity in the city by the end of 2003, he claimed, demonstrates that the city had 
made a similar deal with Murilo. Without prompting, he imediately responded to the 
obvious objection, namely, that the state arested Murilo in 2005 and then extradited him 
to the United States in 2008 ? how does that square with the claim that there was a 
secret pact betwen him and the government? ?Here?s a point of debate,? he started. ?Is 
?Don Berna? the name of a man, or the name of a mafia society?? After a pause too brief 
to alow for an answer, he answered: ?He?s part of a mafia society. It?s impossible to 
dismantle a mafia just by removing its head: the society remains.? It was the same with 
the Italian mafias in the United States, he argued: the government kept aresting leader 
after leader, but the mafias remained. ?So who do you make the deal with: ?Don Berna? 
as a man or ?Don Berna? as a society? Obviously, with ?Don Berna? as a society.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 245 
To understand who was realy behind the secret deals with the ilegal armed 
actors, Quijano suggested a simple exercise that is the standard heuristic of conspiracy 
theorists: look at who benefits. In Medel?n, he said, the deals are betwen the ruling clas 
and the ilegal powers who run the ilegal economy; the elites use the law only when it is 
convenient. Who solves their problems, he asked rhetoricaly, the police or the 
hegemonic power? The answer was too obvious to answer. ?Antioquians make deals with 
whomever has real power. Today it?s the paras. But tomorrow if a group of anti-paras 
comes into power, [the ruling clas] wil make deals with them.? In Medel?n, he 
reiterated, there is a system of deal-making: you either arm yourself or you make pacts 
with those who are already armed. With so many people unemployed, so many people 
involved in the conflicts, it?s the best anyone can do. 
An important flaw in Quijano?s explanation is that the identities of the supposed 
deal-makers were such a moving target: Who made a pact with whom? City oficials with 
Murilo? Departmental or national oficials with Murilo? Government officials with 
??Don Berna? as a society,? as he puts it (which must be either the Envigado Ofice or the 
Cacique Nutibara Bloc)? Or betwen local elites (never identified) and the 
?paramafiosi?? Which paramafiosi? Which local elites? Who was in on the deal, and 
what mechanisms were in place to enforce it, other than the threat of violence? Were 
there other government officials who opposed such pacts? How were they marginalized 
to make sure the pacts would hold? 
But an even more significant flaw in this line of thinking is that, as part of an 
explanation for the Medel?n Miracle, these secrets pacts simply aren?t necesary: the 
public pacts, the Santa Fe de Ralito Peace Acords; the fact that Murilo had succesfully 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 246 
ejected his rivals from Medel?n; and the structure of interests that both sides in the public 
pacts had for keeping violence under control, al provide sufficient explanatory power 
without having to resort to a secret-pacts explanation. 
To be sure, there were secret pacts, and a lot of them at that. The entire 
parapolitics scandal ? the AUC?s rigging of almost every local, regional, and national 
election betwen 2002 and 2006 in an efort to infiltrate the country?s political system ? 
emerged from a series of revelations about scores of secret pacts that had been made (and 
in some cases literaly signed) betwen the paramilitary chiefs and the candidates for 
political office whom they helped get elected. Those candidates and other public officials 
in on the deals were the ones who later helped draft and pas the Justice and Peace Law 
that provided such generous benefits to the paras and almost no rights to their victims. 
The quid pro quo involved was: the paras? support for political campaigns, in exchange 
for the candidates? support for para-friendly policies once in office. That, however, is not 
what the conspiracy theory argues, or at least, that is not all that the conspiracy theory 
argues. The conspiracy theory argues that the quid pro quo was: the paras? direct 
suppresion of violence in Medel?n, in exchange for imunity from extradition and 
impunity for their crimes. That is, as long as they were safe from extradition and fre to 
run their criminal enterprises, the narco paras would keep a lid on violence and let city 
officials take credit for it. 
Of course, that did not happen. Even if, for the sake of argument, such pacts had 
been made, they must have been broken almost from the start and therefore could not 
have any real explanatory power. First, upon his victory over the guerilas and the Metro 
Bloc, ?Don Berna? Murilo changed his alias to ?Adolfo Paz? (?Adolph the Peacemaker?), 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 247 
thereby publicly taking direct credit for the Medel?n Miracle that followed. Second, this 
peace-maker?s staf continued to kil, kidnap, extort, and otherwise intimidate people 
even after the aleged secret pacts had supposedly been made. While it is true that 
homicides fel during that period, that was mainly because the war had ended: without 
war, of course there would be les collective violence; without the context of war, of 
course there would be les of the interpersonal-communal violence that always 
acompanies such wars. The idea that Murilo had explicitly ordered his network of 
criminal organizations to cut back on violence is not inconceivable, but it also is not 
necesary to the explanation. And even if it were true, there were enough violent crimes 
commited in Murilo?s name during the Miracle period that either he wasn?t carying out 
his end of the bargain or he didn?t have as much direct control over his network as the 
conspiracy theorists gave him credit for. In fact, Murilo?s arest came as a result of his 
having ordered the asasination of a legislator in the department of C?rdoba. Third, state 
actors were certainly not turning a blind eye to his crimes. Investigations continued, and 
eventualy he and dozens of other high-level paramilitary leaders were arested. If city 
and state actors had made any secret pacts with the paras, they were not in compliance. 
Finaly, in general, regardles of whatever pacts had or had not been agred to with city 
officials, or national oficials, or members of Colombia?s ruling clas, there were many 
others whose eforts had the ultimate efect of undermining them. Most significantly, the 
Justice and Peace Law that so benefited the paras was not imune to revision: 
The Constitutional Court stood in the way. In the legitimate exercise of its 
inspection of laws, it removed from the Congresionaly approved 
framework the [granting of] political status [and] mandated truth in 
confesions and more severe penalties. For its part, the Supreme Court of 
Justice and the Public Prosecutor?s Ofice kept up their investigations and 
initiated proceses against political leaders and high officials against 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 248 
whom there was evidence of collaboration with the paramilitaries. The 
United States continued insisting on the extradition of some of the 
paramilitary chiefs and presuring [the Colombian government] not to 
make any firm commitments on this isue in the case of the self-defense 
groups. Some members of the pres uncovered veritable conspiracies in 
favor of the paramilitaries and the politicians commited to them. The 
judicial uncertainty was evident and the untroubled admision of the 
paramilitaries into civilian life was not guaranted.
37
 
In the end, fourten of the narco paras ? a term that, by this point, had become a 
redundancy ? were extradited to the United States in May 2008, in a midnight operation 
that took the entire country by surprise, perhaps especialy the men who themselves were 
extradited. Among them were ?Don Berna? Murilo, who had bought his way into the 
AUC, heading the Cacique Nutibara, Heroes of Granada, and Heroes of Tolov? blocs; 
Salvatore Mancuso, the former head of the AUC?s military arm, chief of the AUC?s 
Northern Bloc, and lifelong right-hand man to Fidel, then Carlos, then Vicente Casta?o; 
Rodrigo Tovar Pupo, ?Jorge 40,? who had been Mancuso?s deputy at the Nothern Bloc; 
Juan Carlos Siera Ram?rez, ?El Tuso? (?Pockmark?), Mancuso?s chief of staf; Ramiro 
Vanoy Ram?rez, ?Cuco Vanoy? (?Pussy Vanoy?), chief of the Miners Bloc (Bloque 
Mineros) in northeastern Antioquia and a North Valey Cartel asociate; Francisco Javier 
Zuluaga Galindo, ?Gordolindo? (?Fat Cutey?), a former Medel?n Cartel asociate who 
later commanded the Pacific Bloc of the AUC and managed finances for some of the 
Casta?os? narcotraficking activities; Eduardo Enrique Vengoechea Mola, ?El Flaco? 
                                                
37
 ?Corte Constitucional se interpuso en el camino. En ejercicio leg?timo del control de las leyes, retir? del 
marco aprobado por el Congreso el estatus pol?tico, hizo obligatoria la verdad en las confesiones y m?s 
rigurosas las penas. A su vez, la Corte Suprema de Justicia y la Fiscal?a mantuvieron sus investigaciones e 
iniciaron procesos contra dirigentes pol?ticos y altos funcionarios de los cuales se ten?a indicios de que 
colaboraban con los paramilitares. Estados Unidos segu?a insistiendo en la extradici?n de algunos de los 
jefes paramilitares y presionando para que no se hiciera ning?n compromiso duradero sobre este tema para 
el caso de las autodefensas. Algunos medios de comunicaci?n impresos destapaban verdaderas 
conspiraciones para favorecer a los paramilitares y a los pol?ticos comprometidos con elos. La 
incertidumbre jur?dica era evidente y el ingreso tranquillo de los paramilitares a la vida civil no estaba 
garantizado.? Mauricio Romero, Parapol?tica, 42. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 249 
(?Thin Man?), asociated with both the AUC and a narcotraficking organization caled 
Los Melizos; Hern?n Giraldo Serna, chief of the Tayrona Resistance Bloc (Bloque 
Resistencia de Tayrona); Giraldo?s nephew Nodier Giraldo Giraldo, ?El Cabez?n? 
(?Bighead?); Mart?n Pe?aranda Osorio, ?El Burro? (?Mule?), who had demobilized with 
Tayrona Resistance; Edwin Mauricio G?mez Luna, ?Pobre Melo? (?Poor Spooky?), also 
with Tayrona Resistance; Diego Alberto Ruiz Aroyave, ?El primo? (?Cousin?), of the 
Centauros Bloc; Manuel Enrique Torregrosa Cargos, ?Chan,? who after demobilizing was 
said to head a traficking organization caled Los 40; and Guilermo P?rez Alzate, ?Pablo 
Sevilano,? who demobilized with the Liberators of the South Bloc. Scores of other narco 
paras have ben captured, imprisoned, and extradited since then.
38
 
Quijano made an observation that is very important to recognize. ?When one 
group is in control, obviously the crime statistics are going to fal. But extortion, 
prostitution, arms traficking, and drug sales al continue.? He was right that the nature of 
violence in the city did change during the Medel?n Miracle: physical and psychological 
violence both fel, but the ratio of psychological to physical rose. Of the violence that 
remained, it was more economic than social or political violence, and many of the victims 
were asociated with criminal enterprises rather than being innocent bystanders. This is 
suggestive of a strategy on the part of organized crime to keep violence at a lower profile. 
But one does not need to resort to an argument about conspiracies to explain this strategy: 
a similar leson had been learned by the Cali Cartel in the wake of the high-profile 
                                                
38
 El Tiempo, ?Extradici?n de 14 jefes ?paras? se comenz? a planear desde que ?Macaco? fue enviado a 
E.U.,? [?Planing for Extradition of 14 ?Para? Chiefs Started When ?Macaco? Was Sent to the U.S.,?] 14 
May 208; ?The Paramilitary ?Extraditables?,? Plan Colombia and Beyond, published electronicaly by the 
Center for International Policy (CIP), http:/ww.ciponline.org/colombia/blog/archives/000313.htm 
(acesed 31 August 209); ?Los Otros 10 Extraditados,? [?The Other 10 Who Were Extradited?] 
published electronicaly by El Heraldo, http://ww.elheraldo.com.co (acesed 31 August 209). 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 250 
exceses of the Medel?n Cartel. High-profile atrocities generate widespread opposition, 
which leads the public to cal for government action; low-profile violence ? 
psychological instead of physical (i.e. threats instead of atacks), murders of criminals 
instead of bystanders, etc. ? tends not to generate such opposition, enabling one to go 
about one?s busines, as they say, under the radar. Or, to put it in terms of this study?s 
findings: atrocities generate ilegitimacy, ilegitimacy raises the costs of control (in this 
case, control over the ilicit economy), and the ex-paras wanted to keep those costs low. 
In short, it is a long way from the observation that the ex-paras had engaged in a profile-
lowering strategy, to the claim that the decision of the ?paramafiosi? to keep a lower 
profile had come at the request of government oficials or local elites through secret 
pacts: it is far more likely that the ex-paras recognized that the strategic environment had 
changed such that keeping a low profile and resorting to les violence was simply in their 
best interest. 
But, again, even this observation is almost unnecesary: violence in Medel?n fel 
mainly because ?Don Berna? Murilo and his network had played a big, violent game of 
Monopoly and came out the undisputed winner of the game board, and violence 
asociated with a conflict naturaly fals once the conflict ends; to this the national 
government added real resources for security in the city, most visibly during Operation 
Orion. Together, the legal and the ilegal powers had brought real force to bear against a 
mutual enemy for reasons that were entirely consistent with their own interests or values. 
Then they negotiated a deal with each other, in public, that both sides felt were in their 
own interest: demobilization of the ilegal in exchange for political recognition by the 
legal. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 251 
The deal betwen the government and the paras was mostly a national-
government initiative. When Sergio Fajardo took ofice at the beginning of 2004, the 
Santa Fe de Ralito Acords had already been agred upon, the demobilization of Cacique 
Nutibara had already taken place, and the agrement recognizing Corporaci?n 
Democracia as the ex-paras? representative had already been signed. The mayor was 
given responsibility for implementing a reinsertion proces that he had had no say in 
developing. There was much about it that he did not agre with, but he acepted it as a 
fait accompli and incorporated it into his own agenda for city, an agenda widely viewed 
as having been implemented so succesfully that he would become a front-runner for the 
2010 presidential elections.
39
 
Fajardo, a newspaper columnist and mathematician educated in the United States, 
had gotten some friends together in 1999 to talk about what could be done to improve the 
quality of life in Medel?n. ?We realized that we could work, talk, dream, but to realy do 
anything we had to go into politics, because politicians are the ones who have power,? he 
explained in an interview ith Newsweek. ?So after many years of being outside of 
traditional politics, we built an independent civic movement.?
40
 With that foundation, the 
profesor ran for mayor in the 2003 elections as an outsider to the traditional political 
parties, with his independent party, Citizens Promise, an approach Uribe had taken in the 
presidential election the year before. A charismatic man with long, wavy hair and a 
                                                
39
 Information on Fajardo?s carer and policies is drawn from Jes?s Balb?n, personal comunication 
(Interview No. 1), 25 February 209; Gerard Martin, et al., Medel?n: Transformaci?n de Una Ciudad; 
Simon Romero, ?Mayor of Medel?n Brings Architecture to the People,? The New York Times, 15 July 
2007.; Charlie Rose, ?A Conversation with Sergio Fajardo,? television interview (20 February 2009), The 
Charlie Rose Show, htp:/ww.charlierose.com/view/interview/10098 (acesed 26 August 2009); Daniel 
Kurtz-Phelan, ?The Mathematician of Medel?n,? Newsweek, 1 November 207. 
40
 Ibid. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 252 
penchant for blue jeans, Fajardo campaigned in the city?s poorest barios, on foot, and 
?sold them a dream,? in one expert?s words, teling them not to steal, and not to kil, and 
that if he became mayor he would work with them to overcome their problems.
41
 They 
loved it. Walking the strets of barios that until very recently had been war zones and 
riding a wave of national optimism following Uribe?s election, Fajardo won his campaign 
for mayor. He entered office with a 72 percent approval rating, which generaly rose over 
the next four years, at one point reaching 95 percent; he left office at the end of 2007 with 
a favorable image among 89 percent of city residents, and in 2009 was considered the 
strongest candidate for the presidency in the 2010 elections.
42
 
As mayor, he continued walking the poorest barios and made them the focus of 
his eforts to rebuild the city. Demobilized paramilitaries and narco paras had already 
returned to many of those barios, and he wanted to be sure they would not have reason to 
return to the ilegal economy. But making sure that the system for reintegrating them into 
city life was providing them the opportunities they needed was not the only, or perhaps 
even the main, focus of his eforts. The main focus was public education, for which his 
administration increased spending to 40 percent of the city?s entire budget, more than 
$350 milion. It was not just education in the schools, however, but in the ?the whole life 
of society?: 
We went school to school, clasroom to clasroom, designing and carying 
out ?quality pacts.? We mobilized everyone ? busines leaders, 
universities, private schools ? to start working in the public education 
system. ? We also built a lot of new schools and five ?library parks? in 
the poorest neighborhoods in the city. These are not just libraries; they are 
                                                
41
 Balb?n, Interview No. 1. 
42
 Galup Colombia Ltda., Galup Pol Bimestral: Pol 62. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 253 
community centers, the new axis of the neighborhood. And we made sure 
that they were beautiful, with spectacular architecture.
43
 
The architecture drew orldwide atention, despite some local complaints that they were 
a waste of resources. Fajardo claimed that putting the city?s most beautiful buildings in 
its most destitute neighborhoods was sending a mesage of inclusion in a city long known 
for its marginalization of the poor. At least as importantly, he argued, it gave those 
communities high-quality facilities to which they had not previously had aces: libraries, 
clasrooms, meting spaces, Internet aces, auditoriums, public parks, etc. 
These were barios that the state had neglected for years, and the social 
investments were explicitly intended to addres inequality and injustice. But implicitly 
they were a clear efort to displace or prempt the rivals to the state that had existed for 
years: community self-defense groups, then militias, then paras, and now, possibly, the 
demobilized paras. Those were the entities that had provided esential services and public 
goods, set local policy in their micro-teritories, resolved conflicts, endeavored to keep 
the peace ? in short, the entities that had been governing the peripheral barios. And 
Fajardo wanted to incorporate them into the life of the city. One way to do that was to 
literaly connect them to the city. The hilside barios had step, narow strets, and while 
bus service to most of them was regular, it was not fast, sometimes taking hours just to go 
a few miles to the city center; moreover, some areas were acesible only by foot, up 
long, concrete stairs, some exceding 300 steps. Fajardo?s predecesor had initiated a 
plan to connect two of these hilside barios to Metro, the rail-based public transportation 
network, using cable cars, the kind that normaly take wealthy skiers to the top of the 
slopes at expensive resorts. The cable car system was completed during Fajardo?s 
                                                
43
 Daniel Kurtz-Phelan, ?The Mathematician of Medell?n.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 254 
administration. A trip downtown that, by bus, had taken two hours or more, would now 
take just 45 minutes by cable car ? connecting to rail at no extra cost ? making the 
commute to some jobs feasible and thereby expanding the possibilities for employment. 
It was a strategy aiming for long-term stability: 
We had to reduce violence, but every reduction in violence we had to 
follow imediately ? and imediately is a key word ? with social 
interventions. The order is important. Social interventions require time and 
resources to work, so they wil have litle efect in the midst of such 
profound violence. It is true that you must have efective social 
interventions to make sure violence does not return, but first you must do 
something about violence ? and for that we needed more police ? as 
long as they were police who respected human rights, and out of 
conviction, not just because Human Rights Watch tels them to.
 44
 
No place in the city, he said in a television interview, should be ruled by somebody other 
than the state.
45
 To pay for al this, he acepted donations from other countries ? Mexico 
donated money for one project along the city?s main thruway, Spain funded the building 
of the most architecturaly dramatic library in the city ? but he also did something his 
predecesors had not managed to acomplish: he improved the city?s tax revenue. ?We 
have improved transparency in the city?s finances, so more people are paying their taxes. 
When busineses trust that we are not stealing, and they know that we are going to use 
their money efectively, they pay,? he told Newsweek.
46
 
His chief of staf, Alonzo Salazar Jaramilo, won election for mayor in 2007, 
ensuring that the model of governance the two of them had developed during Fajardo?s 
administration would be extended for at least another four years. (Salazar had once lived 
                                                
44
 Ibid. 
45
 Charlie Rose, ?A Conversation with Sergio Fajardo.? 
46
 Daniel Kurtz-Phelan, ?The Mathematician of Medel?n.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 255 
in the Santa Luc?a bario of Caicedo La Siera for four years, during which he got to 
know some of the protagonists of the gang wars; his book of interviews with Medel?n?s 
gangsters, No Nacimos Pa? Semila, has long been a best-seler in Colombia.
47
) The 
?Medel?n Model of Good Governance and Comprehensive Social Development? 
(Modelo Medel?n de Buen Gobierno y Desarollo Social Integral), as they caled it, was 
based on their municipal development plans: Medel?n, the Most Educated, under 
Fajardo, and Medel?n Is Caring and Competitive, under Salazar. Taken together, they 
prioritized six ?management areas? (?reas de gesti?n: education, public spaces, inclusion 
and equity, arts and culture, peace and security, and a culture of learning and 
competitivenes) and six ?management mechanisms? (mecanismos de gesti?n: long-term 
planning, financial transparency, participation, and public information). Most of the 
experts and residents I interviewed, most of the people I talked to informaly, and most of 
the foreign visitors who have come to se the Medel?n Miracle first-hand generaly 
agred that the city was wel managed, and that the approaches Fajardo and Salazar were 
taking ? built on city-government best practices, science-based evaluation, and, to give 
due credit, key institutions established under their mayoral predecesor Luis P?rez 
Guti?rez ? were the right steps for establishing long-term stability in the city. The 
mayors were popular and generaly trusted, and during their administrations the quality of 
life was improving more, and more rapidly, than anybody had dared to hope. The city 
semed to be doing everything right. 
What were they mising? 
                                                
47
 Diana Marcela Ruiz Ram?rez and Diana Mar?a Gaviria Ram?rez, ?Alonso Salazar: Las Palabras Son 
Semilas,? [?Alonso Salazar: Words Are Seds,?] Sextante, published electronicaly by Fundaci?n 
Universitaria Luis Amig?, htp:/www.funlam.edu.co/sextante/edicion2/entrevista.htm (acesed 27 August 
2009). Se also: Alonso Salazar J., No Nacimos Pa? Semila. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 256 
6.2. Caicedo La Siera: ?Job,? ?Mem?n? and the Quiet War 
6.2.1. ?Mem?n? and the Death of ?Alberto Ca?ada? 
Mem?n is a comic book figure from Mexico, whose adventures are told, soap-
opera style, in a wekly magazine that has been published off and on since the 1940s. 
Despite its depiction of Mem?n and his mother in a stereotypical ?pickaninny? and 
?mamy? style ? which has caused more controversy when imported to the United 
States than it has in Latin America ? Mem?n has been and continues to be an 
enormously popular comic-book character. His Wikipedia entry describes him as ?a 
restles child, not a very good student, not for lack of inteligence, but for not being able 
to pay atention (he is surprisingly good at arithmetic). He helps his mother working in 
the stret, seling newspapers, and as a shoe shine boy. Mem?n reflects the life of a poor 
Mexican boy in Mexico City. Mem?n and his mother are the only Afro-Mexican 
characters.?
48
 
I have not been able to find out when or why John Wiliam L?pez Echavar?a took 
on the alias of Mem?n, but given Mem?n?s popularity in Colombia, L?pez Echavar?a?s 
curly hair, his dark skin, and the context of a society that does not shy away from 
nicknaming people by their looks, even by stereotypes that American liberals would 
consider racist, it would not be surprising if his alias came from the comic hero. (In fact, 
other Colombians have been nicknamed ?Mem?n? by friends explicitly citing the comic, 
                                                
48
 Wikipedia, s.v. ?Mem?n Pingu?n,? http:/en.wikipedia.org/wiki/Mem?n_Pingu?n (acesed 22 August 
2009). 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 257 
including at least one soccer player and, possibly, another paramilitary leader from the 
AUC.
49
) 
An American reporter who interviewed Mem?n in 2004 after he demobilized with 
the Cacique Nutibara Bloc described the former paramilitary leader as ?a thick-chested 
39-year-old with a crucifix hanging from a silver chain around his neck ? a father of two 
and [a self-described] model citizen in this city overun with violence.? Mem?n told the 
reporter that ?we did what was necesary to protect our community,? citing the AUC?s 
cultural projects, social work, social-cleansing campaigns, and defense against guerilas 
as the things that Mem?n claimed had won him support as one of the paramilitary leaders 
in Comuna 8, in East Central Medel?n, the ward where Caicedo La Siera is located. ?My 
community believes in me.?
50
 
After demobilizing with Cacique Nutibara, Mem?n and others from the sector 
spent thre weks of ?reducation? at La Ceja (se 6.1.2) then arived in Comuna 8 on 
17 December 2003 in the middle of the night. They were supposed to be given jobs, but 
the new mayor had only just taken office when he was told by the national government, 
which had initiated the demobilization, that the city was going to be responsible for 
implementing it. Designing such a complicated program would take time, and the jobs 
just weren?t ready yet. The paras? return to what everyone caled ?legality? was therefore 
                                                
49
 Se Santiago Hern?ndez Henao and Jaime Herera Corea, ?Cl?sico con un Solo Doliente,? El 
Colombiano, 9 May 209; and VerdadAbierta.com, ??Mem?n, Luis Arnulfo Tuberquia,? Paramilitares y 
Conflicto Armado en Colombia [Paramilitaries and Armed Conflict in Colombia], published electronicaly 
by Fundaci?n Ideas para la Paz (FIP) and Semana, htp:/ww.verdadabierta.com/web3/victimarios/los-
jefes (acesed 2 August 209). In fact, there is some confusion acros multiple published acounts of the 
paramilitaries: either there are two people asociated with the AUC who have used the alias ?Mem?n,? or 
there is one person who has used that alias but has two ?legal? names: John Wiliam L?pez Echavar?a (or 
variants: ?Jhon? or ?Wilmar?) and Luis Arnulfo Tuberquia; both ?legal? names have been asociated with 
the death squad ?guilas Negras in print. In my judgment, they are two different people who use the same 
alias. 
50
 Daniel Kurtz-Phelan, ?Mr. Nice Guy,? New Republic 230, no. 18 (204): 11. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 258 
met by unemployment; they were stil geting paid just for participating in the 
demobilization program, but they were not al given a productive way to spend their days. 
?It?s mesed up to say it but the state already hasn?t kept its promises to us,? 
Mem?n told a visiting reporter in January 2004, just a month after being ?reinserted? into 
Comuna 8. As one of the leaders of the AUC?s Cacique Nutibara Bloc, he used to patrol 
the barios al day long, solving problems in the comunities as they arose. And now as 
one of the coordinators of the demobilized paras in Comuna 8, he was stil sen within 
the community as one of its governors, or muchachos, as people always caled the 
barios? strongmen. His patrols, his command structure, and his community service, 
therefore, simply continued once the demobilization was over. ?Legality sucks, because 
you stil gotta get up early.?
51
 
Edwin Tapias, who had worked with Jason and ?dison in the Metro Bloc during 
its war with Cacique Nutibara (se ? 3.2), ended up joining Cacique Nutibara before 
demobilizing with them and returning to La Siera with Mem?n and the others as one of 
the bario?s muchachos. He had been arested at an Army checkpoint in Las Estancias 
around 1999 and was acused of being asociated with the paras, but ?at the time, he 
wasn?t with those groups,? a community leader told me: Edwin was just a leader in the 
bario?s New Generation Youth Organization (JNG: Corporaci?n J?venes Nuevo 
Generaci?n), a youth group involved in sports and cultural activities and supported by the 
Community Action Board (JAC: Junta de Aci?n Comunal) of La Siera.
52
 But the false 
                                                
51
 ?Es maluco decirlo pero el Estado a?n no nos ha cumplido. ? La legalidad es jodida porque a uno le 
toca madrugar.? Quoted in Armando Neira, ??S? Nacimos Pa? Semilla?,? [??Not Born to Die?,?] Semana, 26 
January 204. 
52
 ?Boris,? Interview No. 8. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 259 
arest ended up becoming a self-fulfiling prophecy: after geting out of prison he joined 
Metro. After the demobilization, the same reporter who interviewed Mem?n was talking 
to Edwin about his hopes for La Siera, and mentioned that Edwin sounded like a leftst, 
what with al the talk about social justice and human dignity. ?Dear God, don?t tel me 
that!? he responded. ?We?re from the Right. The Left is the enemy we defeated.? The 
reporter asked him if al the leftists had been kiled or forced out of La Siera: 
?We know that there are some profesors and comunity leaders [here] 
who sympathize with those groups. But if they don?t help the guerilas 
they can keep living here without worry.? 
?But what would happen if, for example, a legal organization of the left 
started operating here?? 
?Like what kind?? 
?I don?t know. For example an office of the Communist Party. Would you 
respect them?? [the reporter asked.] 
?Oh, no. If they?re Communists then we?d have to talk with the staf 
commanders of the AUC to se what to do about them.?
53
 
This was exactly was critics of the demobilization had feared, that the program 
had amounted to an amnesty for criminals and that the demobilized paras would go back 
to their same old tricks once they returned home, now with a clean slate as far as the state 
was concerned. In fact, after the demobilizations and a brief rest for reducation, many of 
the paras, though not al, did simply return to their barios and continue operating under 
the same command structure as before ? as indicated by Edwin?s reference to the ?staf 
                                                
53
 ??Por Dios, ni me diga eso. Nosotros somos de derecha. La izquierda es el enemigo al que vencimos.? 
??Mataron a todos? ?No queda gente de izquierda viviendo aqu??? ?Sabemos que hay algunos profesores y 
l?deres comunales que simpatizan con esos grupos, pero si no ayudan a la guerila pueden quedarse a vivir 
tranquilos.? ??Pero qu? pasar?a si por ejemplo aqu? se montara una organizaci?n legal de izquierda?? 
??C?mo cu?l?? ?No s?. Por ejemplo, una sede del Partido Comunista. ?La respetar?an?? ?Ah, no, si es 
comunista ah? s? nos tocar?a hablar con los comandantes del estado mayor de las AUC para ver qu? se hace 
con elos.?? Armando Neira, ??S? Nacimos Pa? Semila?.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 260 
commanders of the AUC? rather than to the officers of Corporaci?n Democracia. They 
no longer caried weapons and they no longer kiled people in public to make an example 
of them. But people did disappear during this period, and a lot of people were forced out 
of their homes; it was some time before sufficient evidence arose to link those crimes to 
some of the demobilized paras. A lot of them realy did want to return to civilian life, but 
for others it did not take long to return (to use the other term that became popular) to 
?ilegality? ? with a much lower profile. 
Stil, it was a time when the gang wars were finaly over and the community?s 
hopes for peace were just beginning to find fertile soil. From what most people could se, 
the demobilized paras in Comuna 8 were acting as prodigal sons, returning to bario life 
and finding ways to improve the community and kep the peace without resort to 
violence. With the return of the prodigal sons, the JAC presidents and other community 
leaders recognized the need to help them reintegrate into the barios as peaceful citizens. 
Or, as one community leader put it, they wanted to work with the ex-paras to ?repair the 
social fabric.?
54
 Betwen 2004 and 2006, some JACs in Caicedo La Siera tried helping 
the ex-paras form community organizations and start community projects. It was a 
hopeful time, he said, because a lot of people believed in the reinsertion proces. But the 
community leaders were ultimately disappointed by the response of the demobilized 
leaders. ?Some of them did want peace, but others didn?t,? the community leader told me. 
Some took the peace proces seriously and engaged with the community leaders, but 
others just went back to doing what they had been doing al along, be it drugs or 
                                                
54
 ?Boris,? Interview No. 8. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 261 
coercion. The JACs decided to just let the demobilized leaders form their own 
organizations. The social fabric was divided in two. 
The demobilized leaders and their community organizations did do productive 
work for their barios, just as the JACs and their comunity organizations had been for 
years. And as people saw that they were, in fact, acting as productive members of society, 
the ex-paras were beginning to win the afection of the community. (One resident said 
she thought that, after his demobilization, ?Job? L?pez, ?Don Berna? Murilo?s right-hand 
man and Comuna 8 boss, had done a lot to help the community, especialy his eforts to 
keep young people out of gangs.
55
) After a couple of years, they started to run for elected 
office. Edwin was rewarded for his service to the community by being elected president 
of La Siera?s JAC. Mem?n stood as candidate for president of the Local Administration 
Board (JAL: Junta Administradora Local) for al of Comuna 8 in the elections of October 
2007 and won the election, becoming the ward?s oficial representative to the city and 
giving him even greater influence over its 18 barios and 38 JACs. 
On 16 May 2008, les than a year into his term, however, Mem?n was arested. ?I 
think you?re familiar with the case of this young man ? Mem?n?? Flavia, one of only 
about 30 women who demobilized with Cacique Nutibara, asked me. ?He?s my 
supervisor. I?m part of his work staf.? Now an elected community leader in Caicedo La 
Siera herself, she had been responsible for carying out a lot of Mem?n?s community 
service and social work, she said. She knew him wel, and 
there was a lot of controversy over his case, because acording to what 
they convicted him of, wel, acording to when we went to his hearing and 
al of that stuff, they say he was the bos of Comuna 8, that nothing 
happened without his authorization, that he ordered extortions, that he 
                                                
55
 ?Cristiana,? Interview No. 9. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 262 
ordered people to commit murder. And it wasn?t like that. No. ? We?re 
human beings and we al make mistakes. But there are things that aren?t 
true, and proving otherwise is very hard because we?re in a society where 
politics influences a lot of things.
56
 
The day after his arest, Corporaci?n Democracia, the legaly recognized representative 
of the demobilized paras, released a statement saying that it would not formaly object to 
Mem?n?s arest, but it would encourage the public and the government to remember that 
Mem?n had undertaken ?productive work on behalf of his community?: 
as atested by the fact that he was recently elected head of the JAL, with a 
considerable vote that is indicative of the backing he has among citizens in 
the zone. ? [We] calmly but firmly invite the relevant authorities to 
undertake in-depth investigations and ignore the rumors and tips that the 
enemies of [the reinsertion] proces are so fond of.
57
 
Among those ?rumors and tips? were acusations that, in addition to holding elected 
office, he was one of the bosses of a criminal organization made up of demobilized paras 
operating in Comuna 8, especialy in Caicedo, and had been atending secret metings 
with other paras to decide how Comuna 8 was going to be ruled. 
?After a while I began to atend metings where it was decided who to kil for not 
supporting the demobilization proces,? ?scar Lub?n Rodr?guez Yepes said during 
testimony at the trial. Rodr?guez Yepes, an asociate of the demobilized paramilitaries, 
said those metings took place in a room on the sixth floor of Medel?n?s Hotel Dann 
Carlton in the wealthy El Poblado neighborhood. Among the atendees were Mem?n, 
                                                
56
 ?Flavia,? personal comunication (Interview No. 15), 3 April 209. 
57
 ?? una fecunda labor a favor de su comunidad, como lo atestigua el hecho de que recientemente fue 
elegido edil en la Jal, con una votaci?n importante, que es un indicativo del respaldo ciudadano con que 
cuenta en la zona. ? con serenidad pero con firmeza, invitamos a los organismos competentes, para que las 
investigaciones se hagan con profundidad, sin hacer caso de rumores o consejas a las que son tan 
aficionados los enemigos de este proceso.? Quoted in Agencia de Prensa IPC, ?Nueva Captura en la 
Corporaci?n Democracia.? 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 263 
Severo Antonio L?pez (?Job?), and other demobilized paras who lived, worked, or 
oversaw operations in Comuna 8 and held undue influence there.
58
 
At one of those metings, Rodr?guez Yepes testified, it was decided that Jairo 
Alberto Ospina Olaya ? the founder of the gang La Ca?ada who himself was nicknamed 
?Alberto Ca?ada? ? was not adequately supporting the reinsertion proces. It is not clear 
what Alberto had done that had made his former Cacique Nutibara Bloc colleagues 
decide he was not being supportive. But the penalty for being an ?enemy of the proces? 
was death, and Alberto?s death sentence was handed down in that hotel room in Poblado 
sometime in mid-2005. On 18 October 2005, Alberto was asasinated by the men he had 
helped win the batle for the East (se ? 5.2.3). Juan Carlos Para, one of ?Job? L?pez?s 
body guards, had objected to the asasination, and he was kiled a few months later, in 
the first days of January 2006. The man said to be in charge of these asasinations, 
Carlos Mario Gonz?lez Escobar, disappeared not long after that, on 4 January 2006. 
?They kiled him,? Rodr?guez Yepes testified, ?because acording to ?Job? he knew too 
much.?
59
 
During Mem?n?s trial in late 2008, a number of witneses appeared to testify 
against him. One by one, they, too, were geting kiled. On the first of October, a former 
para by the name of Jos? de Jes?s Mazo Cebalos answered the door at his house in 
Lower Vila Liliam and was met there by asasins, who shot him using pistols with 
                                                
58
 ?Despu?s de un tiempo comenc? a asistir a reuniones donde se decid?a a qui?n iban a matar ya que no 
ayudaba para el proceso de la desmovilizaci?n.? Quoted in Agencia de Prensa IPC, ?Asesinan a Otro 
Testigo en Proceso contra Alias ?Mem?n?,? [?Another Murder of a Witnes against Alias ?Mem?n?,?] Ud. 
Reportero, published electronicaly by Tera Colombia, htp:/ww.tera.com.co/noticias/articulo/ 
html/acu17137-ud-reportero-asesinan-a-otro-testigo-en-proceso-contra-alias-memin.htm (acesed 
22 August 209). 
59
 ?A ?ste lo mataron porque, seg?n ?Job,? sab?a demasiado.? Quoted in Ibid. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 264 
silencers. His nephew, Juan David Zapata Mazo, and Mauricio Londo?o Londo?o, a 
former para who had demobilized with the Heroes of Granada Bloc, were both executed 
outside of Medel?n on the 20
th
 of October. Rodr?guez Yepes became a witnes for the 
prosecution and testified in November. A few eks later, on 2 December 2008, he was 
kiled on the stret in Caicedo.
60
 
Flavia and others deny that these murders were related to their testimony in 
Mem?n?s trial. The people who have made those acusations, she said, ?did not look at 
the history [the witneses] had. They?re people who had a lot to do with the conflict in the 
past, and ? there are people who hold grudges. There are stil people who fel pain from 
what was done to their loved ones. Those are things that can?t be forgotten. So [those 
witneses] might have been kiled because [their victims] had, as they say, a score to 
setle.?
61
 She complained, moreover, that the state had been protecting those witneses 
without having investigated to se whether they were criminals who themselves were 
intimidating the community: the state wanted the community to offer proof of the 
witneses? ilegal activities before doing anything about it. ?But sometimes the 
community is afraid to inform on them, because that fear of them stil exists.? 
So the community comes to us, the leaders, and says, ?You?re the leaders, 
you?re not afraid to do it. But me, I am afraid, because I?m just a run-of-
the-mil person. I have to think about my husband, my daughters, my 
family. So I go and denounce [a witnes?s continuing involvement in 
crime], and today or tomorrow they kil me. ? Who wins, him or me? He 
                                                
60
 Ibid. 
61
 ?Pero son personas que no miraron el historial que elos ten?an. Son personas que tuvieron que ver con el 
conflicto pasado y ? hay gente que guarda rencor. Todav?a hay gente que le duele lo que le hicieron a sus 
seres queridos. Son cosas que no se olvidan. Entonces, esas personas la pudieron haber matado porque 
ten?an, como se dice, el cuento con ello.? ?Flavia,? Interview No. 15. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 265 
wins. Unfortuatley, he wins, because nobody came to investigate to se if 
the denunciation [against him] was true.?
62
 
In February 2009, despite masive witnes intimidation, and despite claims that 
the witneses were themselves crimials, Mem?n was sentenced to 22 years and eight 
months in prison for forced displacement, false imprisonment, and conspiracy to commit 
crime; he was found responsible for acts of murder, exortion, drug traficking, and voter 
intimidation, among other crimes. Alegations of voter intimidation arose during his JAL 
campaign and the campaigns of other demobilized paras and alies, many of whom had 
stood for election and won the presidencies of at least eight and possibly as many as 14 
JACs in Comuna 8 alone. One of them had been Edwin, but he later got into a dispute 
with other ex-paras and had to fle La Siera; he was no longer a JAC president and as of 
early 2009 was stil in hiding.
63
 The tension betwen the JAC presidents who had 
demobilized and the JAC presidents who had never been involved in the conflicts was 
higher than ever. 
6.2.2. An Uneasy Peace, with Hope for the Future 
Caicedo La Siera became known outside of Medel?n and Colombia in 2005 
when a documentary film about its urban wars was chosen as best documentary at the 
Miami International Film Festival: La Siera: Urban Warfare in the Barrios of Medel?n, 
                                                
62
 ?Pero es que a veces a la comunidad le da miedo denunciar, porque todav?a existe ese temor de esa 
persona. ? Entonces legan donde nosotros los lideres, y le dicen, es que ustedes son lideres, a ustedes, nos 
les da miedo hacerlo. A mi s? me da miedo, porque yo soy una persona com?n y coriente. Yo tengo que 
pensar en mi esposo, en mis hijas, en mi familia. Entonces yo voy y denuncia, y a mi hoy o ma?ana a mi 
me matan ?. ?Qui?n gan?, ?l o yo? Gan? ?l. Lastimosamente gan? ?l, porque no leg? a darse la 
investigaci?n para ver si era verdad lo que ella denunci? ?.? ?Flavia,? Interview No. 15. 
63
 Diego R?os, Interview No. 4. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 266 
Colombia (se ? 5.2.3).
64
 The problem, for those who lived there, was that by the time the 
film was released in 2005 their bario was no longer anything like how it had been 
portrayed in the film: the documentary was about the fourth gang war, the one that started 
out as a batle betwen paras and militias and ended as an internecine war among paras. 
That war had ended in mid-2003, almost two years before the film was released 
internationaly. 
Residents have a complex relationship with the film: They do sem to fel that it 
was an honest portrayal of their lives during a very specific period, and it has become 
something of a touchstone in the community, used as a point of reference for their own 
understanding of that period as wel as for those times when they are asked to explain it 
to outsiders. Yet several residents told me that they fel the film had further stigmatized 
them, making it dificult for people outside the bario to give them opportunities, to trust 
them, to respect them. Some residents of Vila Turbay, just below and informaly 
considered by many to be part of La Siera, have begun in public to take more pride in, 
and insist upon, their distinct identity: no, not La Siera ? Vila Turbay, right next door. 
Meanwhile, comunity leaders in bario La Siera have begun writing a treatment for a 
documentary that they want to make, sort of a La Siera I, showing the world what their 
bario is realy like, how things have turned out, what has changed. If they can find the 
resources to do it, their plan is to train young people in the community to use camera and 
sound equipment so they can make the documentary themselves.
65
 
How did things change after the wars ended? 
                                                
64
 La Siera. 
65
 Information in this section is based on: ?Alicia,? Interview No. 7; ?Boris,? Interview No. 8; ?Cristiana,? 
Interview No. 9; ?Danilo,? Interview No. 10; and informal conversations with other residents. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 267 
First, there was a lot les mobility after 2003. There used to be people moving in 
and out al the time due to the conflicts: displaced people moving in, residents being 
forced out. Alicia had told me that, in the past, she knew that things were heating up n the 
barios when students would stop showing up to school. ?There is more stability now,? 
she said. ?More tranquility.? 
Second, public spaces started feling safer to residents. People were now able to 
go to the cinema at night, and to play on the sports fields and hang out in the parks, 
spaces that had been controled by gangs. There was a lot les anxiety about strangers 
pasing through than there had been in the past, although the presence of outsiders was 
stil noticed imediately. 
Third, children developed more positive aspirations. In the past, when students 
were asked what they wanted to be when then grew up, they would often talk about 
gangsters or militants or guerilas, the kinds of roles that were familiar to them through 
family members. That was no longer the case: most students were talking about college, 
about technical or profesional carers, and many students who had recently graduated 
were succeding. 
Finaly, the state established a much stronger, much more constructive, presence 
in the community. Mayors Fajardo and Salazar had both invested more heavily in the 
community than ever before. ?They?ve brought in everything,? Alicia said, refering to 
school supplies. Her staf were trying to reinforce the mesage to the students the state 
was something to appreciate, since it had brought them new computers and other cool 
supplies. The participatory budgeting proces that gave residents more say in how money 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 268 
was to be spent in their barios helped bring in more social investment, and encouraged 
the community to develop a more positive image of the government as wel. 
Even more significant, however, has been the presence of the police in Caicedo 
La Siera. Before, there had been no regular police patrols, only raids; the only other time 
residents would se police was when they came up to retrieve corpses after a night of 
shootings. The community police, units of the Imediate Atention Command (CAI: 
Comandos de Atenci?n Inmediata
66
), established a station in Caicedo La Siera around 
2006. ?People didn?t want the police,? Boris told me, but bario leaders lobbied hard for 
the CAI. ?For some people it was tough to acept them,? but people got used to it. Flavia 
said people generaly prefered to deal with the CAI rather than other police units because 
they were more acesible. ?That?s not to say that the others aren?t, but they do have a 
more forceful, more rigid temperament; they?re more strict. By contrast, the community 
police try more to communicate, to engage in dialog.?
67
 When people had problems 
before, Boris said, ?it was, ?Go to the muchachos,? but now people say, ?Let?s go to the 
CAI, we?l talk to the lieutenant.??
68
 
People in the community had misgivings about the police at first. When they 
wanted to deal with ?problems betwen neighbors, domestic violence, rapes, drug 
addiction, then people would go under the table and look for the demobilized paras to 
                                                
66
 Sometimes caled Imediate Atention Centers (Centros de Atenci?n Inmediata). 
67
 ?No quiere decir que la otra no sea, pero s? tienen un temperamento mas r?gido, mas fuerte, son mas 
estrictos. En cambio la polic?a comunitaria se trata m?s de comunicar, de dialogar.? ?Flavia,? Interview No. 
15. 
68
 ?La gente no quer?a a la polic?a. ? Era, ?Ir donde los muchachos,? [pero] ahora la gente dice, ?Vamos al 
CAI, hablemos con el teniente.?? Boris, Interview No. 8. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 269 
solve the problems.?
69
 Many of the people I talked to remembered a time when it could 
be dangerous to cal the police, because there was a risk that the police might actualy 
respond ? and if there was one thing the muchachos didn?t like, it was too much police 
atention: they would remind people that they were the ones to go to when there were 
problems, not the police. Once the CAI were stationed permanently in the bario, 
however, the community police started proving themselves capable of solving problems 
as wel, slowly building up trust and credibility. Cristiana said that she herself had 
recently gone to the demobilized paras to complain about her son after he had gotten 
drunk and hit his sister, but the ex-paras told her that that was not something they would 
get involved in: take it to the CAI. ?Today, yes, it?s the police. There are no 
[paramilitary] commanders, at least not for people who aren?t one of them,? she said, 
recognizing that the paras? command structure was stil intact despite the 
demobilization.
70
 It was not clear who was actualy in control, the police or the ex-paras, 
because while both now regularly patrolled, neither side semed to be trying to prevent 
the other from doing so. But if nothing else, people were beginning to appreciate the 
presence of the police; it gave them hope that they might one day be able to fel 
completely secure and not so abandoned by the state. 
One thing that had not changed much, several residents mentioned, was 
intrafamiliar violence, both physical and sexual. It was not a highly visible problem, but 
everyone semed to know it was widespread. One resident went so far as to say that he 
                                                
69
 ?? problemas entre vecinos, violencia intrafamiliar, violaciones, drogadici?n, entonces por debajo de la 
mesa la comunidad buscaba a los desmovilizados para que les solucionar? los problemas; elos dec?an que 
no los buscar?n que elos no pod?an solucionar eso.? Boris, Interview No. 8. 
70
 ?Hoy en d?a s? es la polic?a, no hay mandos, siempre y cuando no se sea de elos.? ?Cristiana,? Interview 
No. 9. 
Lamb, Robert D. | Chapter 6. Trafickers and the Coruption of Paramilitaries: 202-2007 270 
thought violence within families had actualy increased once people started to lose their 
fear of the demobilized paras. 
Stil, the overal quality of life had undeniably improved in Caicedo La Siera 
betwen 2002 and 2007, as it had throughout the city. It was an uneasy peace, with 
control structures and governance structures both divided betwen legality and ilegality, 
police patrols and para patrols, JAC presidents and demobilized leaders, coexisting in an 
uneasy equilibrium that, given the sector?s history, was nevertheles a welcome change. 
?I?d like for people to know about the change we?ve had here, that it?s no longer 
dangerous, that people are diferent now,? one resident told me, echoing the sentiments of 
many others. ?The bario changed radicaly. Why not show the good side the way the 
documentary showed the bad side? We?ve worked hard, fighting to change.?
71
 For the 
bario?s residents, the only thing that would be worse than the outside world?s not 
learning about how things have improved, would be for fate to put the lie to their hopeful 
narative of change, for them to wake up one day to discover that the violence of the past 
had returned to their bario. My interviews in Caicedo La Siera took place in March and 
April of 2009, just after a shootout in the sector left had several people dead in the strets, 
and just before some of the rumors about mas graves and threats and disappearances and 
corruption were turning out to be true after al. 
                                                
71
 ?Me gustar?a que conocieran el cambio que tuvo, que ya no es peligroso, que ya la gente es muy 
diferente. El bario cambi? radicalmente. ?Porqu? no mostrar la parte buena como el documental que 
mostr? lo malo? ? Hemos trabajado, luchando por cambiar.? ?Cristiana,? Interview No. 9. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 271 
Chapter 7. A Window of Oportunity, Closing Quickly: 2007-2009 
The arest of ?Mem?n? L?pez was not the only thing that shifted the internal 
dynamics of Caicedo La Siera?s underworld in 2008: ?Job? L?pez (no relation to 
?Mem?n?) was asasinated on the 28
th
 of July that same year. ?Job? was the former 
militant from M-6&7 who had switched sides to La Ca?ada and then negotiated from 
prison M-6&7?s conversion to the Metro Bloc before himself becoming a key figure in 
the Cacique Nutibara Bloc and one of ?Don Berna? Murilo?s most trusted aids (se 
? 5.2). After Cacique Nutibara?s demobilization, ?Job? became a spokesman for 
Corporaci?n Democracia, the organization formed to represent the demobilized 
paramilitaries. By 2006, however, the country?s Public Prosecutor?s office was 
complaining that ?Job? ?continues to be a pacifist by day, and a ?patron? who orders 
deaths and transactions of arms and drugs by night in vehicles asigned for the 
[de]mobilization. ? He [has] become a real obstacle to the resocialization of young 
people in the comunas, since while they study and work toward a change in their lives 
and their families? lives during the day, at night they go back to being thieves and 
murderers.?
1
 A former gangster told a reporter that ?Job? had anyone who did not comply 
with his orders kiled. ?He gave the orders to asasinate a lot of the demobilized people. 
He kiled them because they were young group leaders who resisted the idea of his giving 
                                                
1
 ?? contin?a siendo un pacifista de d?a y un ?patr?n? que ordena muertes y movimiento de armas y droga 
en las noches en los veh?culos asignados para su movilizaci?n [sic]. ? Se ha convertido en un verdadero 
obst?culo para la resocializaci?n de los j?venes en las comunas ya que mientras estudian y son atendidos 
durante el d?a para un cambio de vida para elos y sus familias, en las noches vuelven a ser bandidos y 
sicarios.? Quoted in Mary Luz Avenda?o, ?En la Mira de Delincuentes,? [?In the Sights of Criminals,?] El 
Espectador, 10 August 208; and Agencia de Prensa IPC, ?Nueva Captura en la Corporaci?n Democracia.? 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 272 
them orders.? There was a backlash, the gangster said, against his having so much control 
over people?s lives and over almost al the drug markets in Comuna 8: 
?Job? started to get a lot of enemies because of how he was. He was realy 
arogant, egocentric, and he thought of himself as the sucesor to ?Don 
Berna.? Things turned bad for him when he started trouble with Julio, a 
kid from here who people had an appreciation for and imitated. ?Job? lost 
control and was left with just Las Mirlas, Las Estancias, Vila Liliam, and 
San Antonio. Al that was left was to kick him out and bump him off.
2
 
Murilo sent ?Job? to Bogot? to become the national spokesman of Corporaci?n 
Democracia. ?This was a political move to get him out of the comuna and out of the 
city,? because he had kept kiling other leaders in Comuna 8 amid power struggles, 
someone who works on development in Comuna 8 told me.
3
 When the chicken came 
home to roost in July 2008, it did not come as a great surprise to many people. He died by 
an asasin?s bullet at a restaurant in Medel?n owned by another ex-para, who had 
already been arested and extradited. 
Nor was ?Job? L?pez the only demobilized paramilitary the city was having 
problems with. The last chapter reviewed some of those problems, including the 
intimidation and kiling of JAC officers and other community leaders as an efort to win 
city contracts, their continuing involvement in the ilegal narcotics trade, and the return of 
vacunas (protection rackets). By 2008 it was clear that a lot of the narco paras had 
demobilized as paras but not as narcos. People in the drug trade have a tendency to 
resolve their disputes by kiling each other. Betwen 2004 and 2008, of the 4,200 
                                                
2
 ?A muchos de los desmovilizados los mand? matar ?l. Los asesin? porque eran j?venes con liderazgo en 
grupos y eran reacios a que ?l los mandara. ? Job comenz? a ganarse varios enemigos por su forma de ser. 
Era muy arogante, egoc?ntrico y se cre?a el sucesor de Don Berna. Las cosas se le voltearon cuando tuvo 
problemas con Julio, un muchacho de aqu? al que la gente le tiene aprecio y le copia. Job perdi? el control y 
se qued? s?lo con los barios Las Mirlas, Las Estancias, Vila Lili?n y San Antonio. No fue sino que lo 
extraditaran y lo quebraron.? Quoted in Mary Luz Avenda?o, ?En la Mira de Delincuentes.? 
3
 Diego R?os, Interview No. 4. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 273 
combatants who had demobilized from the Cacique Nutibara and Heroes of Granada 
blocs (both belonging to ?Don Berna? Murilo), 237 were kiled, 172 were arested, and 
86 were removed from the reinsertion program; another 14 died in the first few eks of 
2009.
4
 
Murilo?s arest in October 2005 and his transfer in August 2007 to the middle of 
nowhere (a maximum-security prison in Boyac?, far from his area of influence) had 
created a vacuum. And narcos abhor a vacuum. In November 2007, National Police 
commander General Oscar Naranjo acused Corporaci?n Democracia vice president 
Carlos Mario Aguilar Echeveri, known by the alias ?Rogelio,? of having taken control of 
the Envigado Ofice. But Aguilar wasn?t the only one hoping to fil the vacuum. Power 
struggles were taking place both within the Envigado Ofice and against other trafickers 
who saw in Murilo?s absence an opportunity to expand their own networks. Murilo?s 
former subordinates ?didn?t have enough power to sustain the monopoly their boss had 
achieved,? one report observed. ?There were confrontations at the end of that year, 
initialy with the North Valey Cartel and then with the structures of [Daniel Rend?n 
Herera,] alias ?Don Mario,? who was trying to position himself as the city?s new 
?patron.??
5
 Things got even worse after Murilo?s extradition in May 2008, and by the end 
of that year it was clear that new ars were being fought within, betwen, and among 
groups of demobilized narco paras and groups of narcotrafickers who had never 
                                                
4
 Peace and Reconciliation Program (Programa de Paz y Programa de Paz y Reconciliaci?n) figures, cited 
in Agencia de Prensa IPC, ?Medell?n, Sin Rumbo Claro en Materia de Seguridad,? Am?rica Latina en 
Movimiento, published electronicaly by Latin American Information Agency (ALAI: Agencia 
Latinoamericana de Informaci?n), htp:/alainet.org/active/29137 (acesed 28 August 209). 
5
 ?? quienes no tuvieron el suficiente poder para sostener el monopolio alcanzado por su jefe. A finales de 
ese a?o se registran, inicialmente, enfrentamientos con el cartel del Norte del Valle y luego con estructuras 
de alias don Mario, quien busca posicionarse como el nuevo ?patr?n? de la ciudad.? Ibid. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 274 
demobilized in the first place. In April 2009, a district atorney in Bogot? claimed 
publicly that Corporaci?n Democracia had become the ?political arm of Los Paisas, the 
rural unit of the Envigado Ofice.?
6
 
Los Paisas ? the group most closely asociated with Murilo?s narco paras ? 
was only one of at least a dozen groups of narcotrafickers from around the country vying 
to take control of Murilo?s empire. The four most powerful in the country at the time ? 
or at least the four whose chiefs had the highest rewards being offered by the national 
government for their capture ? were those operated by: ?Don Mario? Rend?n, about 
whom we wil read more in the next section; Luis Enrique Cale Serna, alias ?Comba? 
(?Jump Rope?), head of a gang asociated with the North Valey Cartel caled Los 
Rastrojos (The Leftovers), operating mainly on the Pacific Coast; Daniel Barera Barera, 
alias ?El Loco? (?Madman?), head of the narcotraficking network that controlled Bogot? 
and its surrounding area; and Pedro Oliverio Guerero Castilo, alias ?Cuchilo? (?Knife?), 
a narcotraficker operating in southern Colombia and former head of the Guaviare Bloc 
of the AUC. Of these, only Rend?n had been captured as of mid-2009. Among those who 
managed to take over significant, and often competing, portions of the Envigado Ofice?s 
Medel?n network at diferent points were: ?Rogelio? Aguilar, the former Corporaci?n 
Democracia vice president who was arested and later extradited to the United States the 
same day as Murilo; Jos? Leonardo Mu?oz Mart?nez (?Douglas?), who was captured in a 
luxury apartment building in the exclusive El Poblado neighborhood of Medel?n in April 
2009; Mauricio Cardona L?pez (?Yiyo?), who negotiated his surrender and extradition 
directly with the United States in July 2009; and Erick Vargas (?Sebastian?), Maximiliano 
                                                
6
 ?? brazo pol?tico de la banda los Paisas, componente rural de la oficina de Envigado.? Quoted in 
Agencia de Prensa IPC, ?Nueva Captura en la Corporaci?n Democracia.? 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 275 
Bonila Orozco (?Valenciano?), and Jader Botero (?Gancho,? or ?Hook?), al of whom 
were stil at large in mid-2009 (and one or more of whom were rumored in July to be 
secretly negotiating with the United States as wel). In Caicedo La Siera, the main fight 
was betwen trafickers asociated with ?Mem?n? L?pez?s organization and those 
asociated with ?Don Mario? Rend?n?s network, which by 2008 was starting to win 
control of the teritory in upper Caicedo La Siera formerly occupied by the Metro Bloc 
(and by M-6&7 before that).
7
 
Medel?n Mayor Alonso Salazar Jaramilo, in comments to the pres following 
?Job? L?pez?s asasination in July 2008, said that the ex-para had ?never realy 
disasociated himself from criminal activities ? it?s a reality that can?t be hidden ? 
we?re not talking about a hero of peace.?
8
 It was the first time Salazar ? either as mayor 
or as a top official within his predecesor?s administration ? had ever admited in public 
that a demobilized paramilitary leader was anything other than demobilized: previously 
he had always insisted that those acusing the ex-paras offer conclusive evidence, 
something few had been wiling to provide given the danger inherent in doing so. Now he 
was publicly vindicating those acusers, and the signal to the underworld was 
unmistakable: Not only had the national government given up on the reinsertion proces 
? it had been President Uribe?s decision to extradite the narco paras to the United States 
two months earlier ? but now the city government had given up on the proces, too: the 
                                                
7
 Information in this paragraph was extracted from dozens of articles published in the Colombian 
periodicals Semana, El Tiempo, El Colombiano, and El Espectador betwen April and July 209. Because 
some of those articles cite incorect or contradictory facts, I have tried to coroborate each piece of 
information in this paragraph from at least two independent sources. 
8
 ??l realmente nunca se desvincul? de las actividades delictivas, es una realidad que no se puede ocultar, 
no se trata de un h?roe de la paz.? Quoted in Juan Diego Restrepo E., ??Por Qu? Crece la Violencia en 
Medel?n?,? Semana, 11 August 208. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 276 
ex-paras involved in crime could now be expected to be treated like criminals instead of 
prodigal sons or ?heroes of peace.? Any pacts that had been agred to ? whether the 
rumored secret pacts to control violence, or the known public pacts for the reinsertion 
proces ? were no longer in efect. Murilo had ben extradited in May 2008, and so his 
ilicit network was up for grabs nationwide; his deputy ?Job? L?pez had just been 
asasinated in July 2008, so his ilicit network was up for grabs in Medel?n; and L?pez?s 
deputy ?Mem?n? had been arested in May 2008, so his ilicit network was up for grabs in 
Caicedo La Siera. At al thre levels, of course violence was going to increase. 
7.1. Medel?n: Dark Forces and Black Eagles 
There had been rumors about mas graves on the outskirts of the city for years. As 
early as 2004 a demobilized paramilitary was testifying to prosecutors that a lot of the 
city?s forced disappearances over the previous few years had involved taking people from 
their homes to areas where other victims had been buried, and kiling and burying them at 
the grave site. In late 2008, the city began searching for those mas graves, and 
discovered some just outside of Comuna 13 in the West and others near Comuna 8 in the 
East. Others were stil being sought as of mid-2009.
9
 The mas graves were understood to 
contain the remains of people kiled by armed actors on al sides of the conflict going 
back 20 or 30 years.
10
 ?It?s like a cemetery up there,? one resident of La Siera told a 
                                                
9
 Juliana Eusse Guera, ?No Hay T?cnica para Hacer Exhumaci?n en Comuna 13,? [?There Is No 
Technique for Comuna 13 Exhumations,?] El Colombiano, 20 August 209. 
10
 Agencia de Prensa IPC, ?En Medel?n, Bosque Del Bario La Siera Es Un Cementerio,? [?In Medel?n, 
the Wods in Bario La Siera Are a Cemetery,?] published electronically by Instituto Popular de 
Capacitaci?n (IPC), htp:/ww.ipc.org.co/page/index.php?option=com_content&task=view 
&id=1362&Itemid=368 (acesed 27 April 209); Juan Diego Restrepo E., ?Desenterar la Verdad de las 
Laderas de Medel?n,? [?Unearthing the Truth on the Outskirts of Medell?n,?] Semana, 21 November 208. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 277 
research and advocacy organization, refering to the unpopulated area just across the 
border from the bario. ?Over there in the woods, they buried people they took from 
nearby barios because they hadn?t paid their vacuna or ? because they were believed to 
be guerilas. There?s man up there from La Siera who used to have a store and they 
kiled him because he refused to pay the muchachos.?
11
 
Fear never fuly went away in the city?s peripheral barios, even during the period 
of peace known as the Medel?n Miracle. During those years, the fear had been 
counterbalanced, and during the best of those times perhaps even suppresed, by hope. 
But in 2007 the balance started shifting again, almost imperceptibly, and by the end of 
2008, a rise in threats and extortion, persistent rumors of beatings and mysterious 
disappearances, and reminders of the past being dug up from the mas graves al started 
fraying the nerves of those who were paying atention (many chose not to pay atention 
and thereby kept the fear at bay). 
As early as 2006 people were starting to talk about ?dark forces,? violence taking 
place against people who somehow had gotten on somebody?s wrong side. 
Pasenger cariers who operate in the informal sector and offer their 
services to thousands of customers from diferent barios in Comuna 13 
have been the objects of beatings by what they cal ?dark forces? for at 
least seven months. ?As informal cariers they walk al over us and one 
thing they resort to is the use of ?dark forces.? They show up, intimidate 
our drivers, take them away, and deliver beatings and threats,? said one 
director who asked that his name be withheld. He admits that those cases 
                                                
11
 ?Eso al? ariba es como un cementerio. ? En ese bosque enteraban la gente que sacaban de los barios 
cercanos porque no pagaban vacuna o, como en el caso de los tres morenos, porque se cre?a que eran 
guerileros. Al? est? un se?or de La Siera que ten?a una tienda y lo mataron porque se negaba a pagarle a 
los muchachos.? Quoted in Agencia de Prensa IPC, ?En Medell?n, Bosque del Bario La Siera Es un 
Cementerio.? 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 278 
wil go unpunished because they can?t be reported. The fear is that their 
drivers wil get kiled.
12
 
During my field research, I encountered a lot of profesed ignorance over who those dark 
forces represented. People might have had a general idea that they were asociated with 
the demobilized paramilitaries or with gangsters who had never demobilized but who 
nonetheles had been relatively inactive in recent years. During the more violent times, at 
least people usualy knew ho the perpetrators were, because those perpetrators were 
fighting to control teritory ? under the asumption that micro-teritorial control, and in 
many cases social control within that micro-teritory, would facilitate aces to local 
ilicit markets ? and it?s very dificult to control teritory and the people and resources in 
it without revealing your identity. Now, however, the fights didn?t sem to be over 
teritorial control or social control. Now, at least in Medel?n, the fights semed to be 
simply over aces to ilicit markets ? drug houses, smuggling routes, extortion rackets, 
and so on ? and over control of only those people involved in those markets. In that 
situation, not only is it possible to conceal one?s identity, it also is beneficial to do so. 
And so, in the face of this unknown, a lot of people started geting anxious about what 
these incidents might portend: the balance of fear and hope was beginning to shift in 
favor of the unwelcome emotional state. 
                                                
12
 ?Transportadores de pasajeros que operan en la informalidad y que prestan sus servicios a miles de 
usuarios de diversos barios de la Comuna 13 son objeto de golpizas por parte de lo que elos laman 
?fuerzas oscuras,? desde hace por lo menos siete meses. ?Como transportadores informales nos atropelan 
bastante y uno de los recursos es la utilizaci?n de ?fuerzas oscuras.? Elos legan, amedrentan a nuestros 
conductores, se los levan y nos los entreguen golpeados y amenazados,? cuenta un directivo que pidi? la 
reserva de la fuente. Admite que esos casos quedan en la impunidad porque no se puede denunciar. El 
temor es que les maten a varios conductores.? Juan Diego Restrepo E., ?Con Palizas Se Impone el Control 
Social en Comunas de Medel?n,? [?Social Control Imposed with Beatings in Medel?n?s Comunas,?] 
published electronicaly by Instituto Popular de Capacitaci?n (IPC), http:/ww.ipc.org.co/page/index.php 
?option=com_content&task=view&id=80&Itemid=368 (accessed 28 August 209). 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 279 
Despite the profesed ignorance of the identities of the perpetrators of violence, 
by the end of 2008 it was prety clear to the city?s and the country?s legal authorities that 
the main dispute in Medel?n was betwen the demobilized paramilitaries formerly 
asociated with Murilo?s network ? the Envigado Ofice, the Cacique Nutibara Bloc, 
Corporaci?n Democracia, Los Paisas, and the rest ? and groups of narcotrafickers 
asociated with the networks partialy coordinated by ?Don Mario? Rend?n, who in other 
parts of the country was known to be leading or coordinating some of the most dangerous 
?bacrim? ? bandas criminales emergentes, or emerging crime gangs ? in the country. 
These were a new generation of armed actor, whose operations and tactics very much 
resembled those of the AUC but whose ideology had almost nothing to do with guerilas 
and almost everything to do with cocaine: controlling drug markets, traficking groups, 
and teritories possesing coca farms, procesing facilities, or smuggling routes. Many of 
these bacrim (including some not controlled by Rend?n) had started caling themselves 
?guilas Negras, or Black Eagles, in 2006, and by mid-2007 there were at least 22 such 
groups operating in 200 municipalities in 22 departments throughout the country, 
including Antioquia. Some might have been new groups formed specificaly as ?Black 
Eagles,? but most were existing traficking or narco para groups that ? in much the way 
existing, independent paramilitaries started caling themselves ?blocs? of the AUC in 
1997 ? simply started caling themselves ?Black Eagles? to make them sem like they 
had more of a national reach than they did. In reality, most operated localy and were 
threats mostly to locals ? but they were very serious threats to locals.
13
 One of Rend?n?s 
groups named itself in a way that hinted at its narco para roots: Autodefensas Gaitanistas 
                                                
13
 Semana, ??Qu? Son las ?guilas Negras?,? [?What are the Black Eagles?,?] 18 August 207. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 280 
de Colombia (AGC: Gaitanist Self-Defense Forces of Colombia).
14
 Its emergence was 
announced in pamphlets that appeared in October 2008 in one part of the country, and 
soon a handful of afiliated groups were appearing al over the country. 
?Don Mario? Rend?n was born in the same vilage as the Casta?o brothers, and he 
grew up knowing the family wel. He and his brother, Freddy Rend?n Herera 
(?El Aleman,? or ?The German?), founded a paramilitary group that later joined the AUC 
as the ?lmer C?rdenas Bloc, which operated in northern Colombia until its 
demobilization in August 2006. But ?Don Mario? Rend?n, who had been his brother?s 
deputy, was not among the more than 1,500 combatants who demobilized with Freddy.
15
 
Instead, betwen the demobilization of Cacique Nutibara in 2003 and that of ?lmer 
C?rdenas in 2006, Rend?n built up a major drug-traficking army and took advantage of 
Murilo?s absence to expand his control of the production zones and traficking routes in 
the northern and western parts of the country. In 2008, he was the most wanted among 
the four most-wanted narcotrafickers in Colombia, and the increase in homicides 
Medel?n experienced during the last few years of the study period was widely atributed 
mainly to the presence of his network in the city (although there were others as wel), as 
he fought for aces to the narcotraficking resources left behind by ?Don Berna? Murilo. 
But ?Don Mario? Rend?n and a couple dozen of his men were captured on 
15 April 2009. The government?s hunt for him was so intense that at the time of his 
capture he had been reduced to hiding in a tiny space ? barely large enough for a mat to 
                                                
14
 The name refers to Jorge Eli?cer Gait?n Ayala, the populist presidential candidate whose asasination in 
1948 sparked a wave of violence that turned into La Violencia, a civil war that lasted more than a decade. 
Gait?n partisans take ofense at the AGC?s use of his name. 
15
 VerdadAbierta.com, ?Las Auc.? 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 281 
slep on ? next to a tre that had been surrounded by logs; the pictures of his hiding 
place brought to mind the hole in the ground where Saddam Hussein had been captured 
in Iraq in December 2003. Rend?n?s capture led some analysts to consider how far the 
government had come in its ability to capture major drug trafickers ? forcing them into 
les hierarchical and more difuse arangements that nearly guaranted an increase in 
tensions among the nodes of the network ? and how far the trafickers had falen since 
the heady days of the Cali and Medel?n cartels: 
What happens today is that the ?useful life? of the drug barons is shorter 
and shorter. The fragmentation of the busines had caused internecine 
fights to be to the death. Some of those who are considered the top chiefs 
of the mafias have died at the hands of their own men.
16
 
Others, more and more every year, were geting captured by the Colombian security 
forces. Soon after Rend?n?s capture, the newswekly Semana published a chart outlining 
the number of years diferent trafickers had spent as head of their respective 
organizations. The Cali Cartel?s leaders had lasted 15-17 years and the Medel?n Cartel?s, 
13-15 years; the next-generation North Valey Cartel?s leaders lasted just four or five 
years; and the ones who emerged after ?Don Berna? Murilo?s arest, including ?Don 
Mario? Rend?n, lasted just a year and half to two years.
17
 Moreover, the narcotrafickers 
of the current generation are not parastatal entities whose interests are aligned with those 
of the state, despite the eforts of some such groups to claim otherwise. (AGC had 
publicly claimed its objective was to protect citizens from guerilas, but nobody believed 
them: AGC was making too many drug deals with the guerilas? own trafickers.) Today, 
                                                
16
 ?Lo que ocure ahora es que la ?vida ?til? de los capos es cada vez menor. La atomizaci?n del negocio ha 
hecho que las luchas intestinas sean a muerte. Algunos de los considerados mayores jefes de la mafia 
murieron a manos de sus propios hombres ?.? Semana, ?El Teror de los Malos,? [?The Teror of the 
Vilains,?] 18 April 209. 
17
 Semana, ?La ?Vida ?til? de los Capos,? [?The ?Useful Life? of the Capos,?] 21 April 209. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 282 
the state both officialy opposes the ex-paras and the bacrim (as it had the AUC) and 
more often than not opposes them in practice as wel (as was definitely not the case with 
the AUC). There are exceptions, of course: the belief is widespread that many officials at 
al levels of government have been corrupted by narcotraficking money or fooled by 
narco paras claiming to be counterinsurgents. And the recent rise in violence has many 
concerned about a return to the bad days. But the bottom line is that the trend in the past 
25 years, and especialy in the past decade, has been mostly positive with respect to the 
government?s capacity and wilingnes to counter nonstate armed actors. 
There is a very high risk, however, that the real progres that has been achieved 
since the quiet reforms of the Pastrana era began showing public results during the Uribe 
era, could be overshadowed by the chalenges that remain. Some state actors might panic 
in the face of rising violence and either: quietly and ilegaly aly themselves with one 
side in the fight as a way of trying to control the violence, as many military and police 
units and individuals had done with gangs and paramilitaries up through the early 2000s 
(and as some are widely suspected of continuing to do today); or very publicly and very 
unwisely overeact to the threat in such a way that it leads to widespread abuses against 
the civilian population (against both ?suspects? and innocent bystanders), as has 
happened repeatedly throughout Colombia?s history. Neither approach, the experience in 
Medel?n has shown, is likely to achieve lasting peace: the first would legitimize the use 
of violence by nonstate armed actors and thereby prolong their ?useful life,? and the 
second would delegitimize or, worse, ilegitimize state actors as enforcers and promoters 
of social and political order and thereby forestal or reverse their consolidation of control 
and ability to govern. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 283 
During the Miracle period, the lull in violence gave state actors at the city, 
department, and national levels an opportunity to prove themselves capable of breaking 
the paterns of failure, corruption, and abuse to which many of their predecesors had 
become acustomed. Where there has been succes in this regard, the result has been a 
virtuous cycle in which succes has bred support, which has created the conditions for 
further succes. The temptation on the part of many state actors to fal back into the old 
paterns wil become increasingly acute in the next few years if the present trends in 
violence continue. 
Over the wekend of 27-31 March 2009, the Inter-American Development Bank 
(IADB or, in Spanish, BID: Banco Interamericano de Desarollo) held its 50
th
 
anniversary board meting in Medel?n. It was probably the most important international 
conference to take place in Medel?n in the city?s history, at least its recent history. No 
expense was spared, no efort neglected, to make certain the meting would leave all 
atendees with a positive impresion of the city?s progres. While some international 
media had been publishing articles about the Medel?n Miracle for a few years, this 
wekend promised to have the highest concentration of international media in the city 
since homicides began faling in 2003. The opportunity for good pres ? and the 
international investment that surely would follow ? had never been beter. The 
organizers were not disappointed by the outcome. Only the international leftist pres ? 
there to cover both the IADB conference and the international antiestablishment counter-
conference that took place the same wekend ? gave the city bad pres, and although in 
some respects its coverage was more honest (if rather exaggerated) about the chalenges 
that remained, its international readership was generaly limited to the choir it had already 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 284 
been preaching to. Most of the mainstream coverage generaly reported the oficial 
version of events, atributing the city?s peace to a succesful peace proces followed by a 
series of good-governance reforms and investments in social development and major 
infrastructure in poor communities.
18
 
Thre days after the meting was over and the international pres was gone, 
Medel?n suffered a spre of violence reminiscent of the days of Cacique Nutibara, even 
the days of Pablo Escobar. Betwen Friday, 3 April, and Tuesday, 7 April, 31 people 
were kiled in confrontations and executions among gangs and mafia figures. Four 
months later, homicide figures for the first seven months of the year were released: 1,043 
people were murdered betwen 1 January 2009 and 31 July 2009, almost exactly the 
number of people who had been kiled in al of 2008. The homicide rate had returned 
almost to 2003 levels.
19
 
The Medel?n Miracle was over. 
7.2. Caicedo La Siera: Rising Fear amid Hope for the Future 
7.2.1. The Fear of Strangers Returns 
It has long been the case in Caicedo La Siera, as in other neighborhoods on 
Medel?n?s periphery, that people would get anxious when they saw someone strange 
                                                
18
 For a typical example, see Chris Kraul, ?Medelin Has Undergone Renaisance: City Has Cleaned Up 
and Homicides Down 90 Percent,? The Boston Globe, published electronicaly by Boston.com, 
http:/ww.boston.com/news/world/latinamerica/articles/2009/03/29/medelin_has_undergone_renaisance 
(acesed 29 March 209). The official story of Medel?n?s suces was published in a boklet handed out 
at the conference: Gerard Martin, et al., Medel?n: Transformaci?n de Una Ciudad. 
19
 Secretar?a de Gobierno Municipal de Medell?n, Subsecretar?a de Orden Civil, Unidad de Convivencia 
Ciudadana. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 285 
entering their community, because strangers often presaged an outbreak of violence. One 
community leader told a story about a woman she knew ho was in the proces of seling 
her car to a pharmacist in the bario, when thre strange men showed up and said they 
wanted to buy it instead. She didn?t recognize them, so she told them she wanted to cal 
her husband to ask for his guidance. Instead, she caled one of the ?muchachos? 
(presumably an ex-para leader), who came imediately to check out what was going on. 
When the strangers saw the muchacho ariving, two of them took off, while the third 
stayed behind to talk. It turned out they were from an outside gang that was trying to 
infiltrate the neighborhood. ?So, you know, the comunity ? you live [here] a lot of 
years, you more or les know people, you know when someone?s a stranger. ? So the 
community would say that there?s someone, like, strange here, and hel would break 
loose. You have to be careful with the people who come in.?
20
 She told another story 
about a strange man who was sen walking up Sugarloaf Hil (Cero Pan de Az?car) with 
a young girl from the bario: 
As I said, when the community ses someone unknown ? you know, 
you?re a man and you?re taking a litle girl by the hand up some path, so, 
where you going with that girl? So the community saw that and said to a 
few kids, ?That guy went up around that side, go check out what?s going 
on.? So they went and the man had already gone some distance, and just as 
the kids arived he was about to rape the girl. So what did they do? They 
contacted the [community police] station at Vilatina and handed the man 
over to the police, because the community also helps out, because if the 
community doesn?t help out we?re not going to be able to do anything. 
We?d be like, any old person coming into whatever bario doing whatever 
they wanted.
21
 
                                                
20
 ?Entonces como la comunidad ? viva muchos a?os, usted mas o menos conoce las personas, usted 
conoce cuando alguien es extra?o. ? Entonces la comunidad misma dec?a, hay alguien como extra?o. 
Entonces ya se regaba la bola, hay que poner cuidado quien esta viniendo.? ?Flavia,? Interview No. 15. 
21
 ?Hace como dos a?os hubo un se?or que se meti? con una ni?a y lo vieron cero ariba de Pan de Az?car. 
Como yo te dec?a, cuando la comunidad ve a alguien que no es conocido ? pues si vos sos un se?or y 
llevas de la mano una ni?a y te estas entrando por un camino, entonces ?pa? donde vas con esa ni?a? 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 286 
As my research asistant and I had discovered when word on the stret got quickly to 
Ernesto, during my interview ith Cristiana, that a couple of strangers were hanging 
around and he arived to find out what we were doing in the bario (se ? 6.2), the 
communities in Caicedo La Siera were ever-vigilant. 
That vigilance, and the tensions that were giving rise to it, was growing as 
violence in the sector was intensifying. In February 2009, just a few eks before my 
interviews with Cristiana, Ernesto, and the others, the Organization for Peace and Social 
Development (Corpades: Corporaci?n para la Paz y el Desarollo Social) warned city 
officials that ?the situation in the barios La Siera, Las Mirlas, Las Estancias, and Vila 
Turbay [has gotten] very serious now that armed violence has left a death toll of thre 
murders and five injuries in les than one month.?
22
 Then in early April, a few days after 
the IADB conference had ended, came the city?s most violent wekend since the 
demobilization of the Cacique Nutibara Bloc. That?s around the time when people started 
teling me that it probably was not safe for me to continue my field research in Caicedo 
La Siera: strangers had always been considered a threat and, in the past, would often be 
shot on sight, without the niceties that Ernesto had offered: asking first. Then in May 
2009, ?Don Mario? Rend?n was captured in northern Antioquia close to the Caribbean 
coast. By this point in Medel?n?s history, everybody knew full wel what happens when 
                                                                                                                                            
Entonces la comunidad vio eso y le dijo a unos j?venes, ese se?or se entro por eso lado, mire a ver que 
pasa. Entonces se fueron y el se?or ya levaba un trayecto, cuando ya legaron preciso, y iba a violar la 
ni?a. Entonces, ?que hicieron elos? Se comunicaron con la estaci?n de Vilatina y le entregaron al se?or a 
la polic?a, porque la comunidad tambi?n ayuda, porque si la comunidad no ayuda no podemos hacer nada, 
Estar?amos cualquiera entrando a cualquier bario a hacer lo que quisieran.? Ibid. 
22
 ?? la situaci?n de los barios La Siera, Las Mirlas, Las Estancias y Vila Turbay es demasiado grave ya 
que la violencia armada deja en menos de un mes una cifra fatal de tres muertos y cinco heridos.? Juan 
Carlos Monroy G., ?Alcald?a niega denuncias sobre comuna 8? [?Mayor?s Ofice Denies Alegations about 
Comuna 8?], El Colombiano, 20 February 209. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 287 
the head of a major traficking organization is captured: his subordinates and adversaries 
start fighting for control over his asets, and violence soon rises. Don Mario had won 
some asets in upper Caicedo La Siera, and the expectation was that the violence would 
now get worse. That?s when people told me that, from that point forward, I would be 
endangering my interview subjects by being sen with them in the bario: in the past, the 
local muchachos would threaten or kil residents for interacting with outsiders, under the 
asumption that they were acting as informants for infiltrators trying to take over 
teritory, drug markets, or local extortion rackets. 
I had interviewed not quite as many people as I had planned, and did so during a 
period when it was safe for them to talk to me. With the interviews I had managed to do 
by that point, I was already hearing a reasonably consistent story, which I have since 
been able to corroborate with other sources. That story is the one I have told here. But 
with the security situation now having reached a turning point, and with warnings that 
doing further interviews in those barios would put people in danger, I decided that what I 
had would have to suffice. Studying legitimacy amid complex urban violence involves 
trying to measure a phenomenon that is unobservable in a place that is inacesible. I had 
already developed a framework for approximating the unobservable (se Appendix C); 
and now that I had managed briefly to get relatively fre aces to the bario during its 
most violent period in six years, I was confident that I had captured a coherent 
approximation of its history and the complex situation that was emerging. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 288 
7.2.2. Selective Violence Returns 
Soon after the first demobilizations, the city encouraged the demobilized paras to 
met with JAC presidents and the leaders of diferent community organizations. Ernesto, 
a community leader in Vila Liliam who also headed a community organization, 
remembered that, at some point, those metings started becoming coercive, as some of 
the ex-paras started presuring the organizations that had city contracts to let them in on 
the deal. 
At some point in one of those metings with the social organizations, JAC 
leaders, [and] community leaders, the [leaders of the ex-paras said]: 
?Anyone who?s here it?s because you?re with us. Anyone who?s not, it?s 
because they don?t want to be with us. And anyone who doesn?t want to 
be with us is against us, and if they?re against us we?l throw them in the 
water.? Me, I?l never forget those words. Strong-arm politics.
23
 
Then the vacunas returned: 
[The gangsters] are stil here in smal cels. They?re not taking any actions 
as such, nothing abrupt, but they go out in smal groups and they?re 
coming up with a new way of making themselves felt in the community. 
They go around seling tickets every wek, just like vacunas. They show 
up at al the stores with a litle ticket and [the owners] have to buy them. 
So, there, they?re coming back with ilegal activity.
24
 
Another resident described a similar problem: ?They have people asault us in the strets, 
rob us in our houses, steal our motorcycles, and when that happens, then they come by 
                                                
23
 ?En alg?n momento en una reuni?n as?, a las organizaciones sociales, a los l?deres de junta, a los l?deres 
comunitarios, personajes desde esa direcci?n: ?Quienes est?n aqu? es porque est?n con nosotros. Quienes no 
est?n, es porque no quieren estar con nosotros. Y quien no quiere estar con nosotros est? en contra de 
nosotros, y si esta en contra de nosotros lo tiramos al agua.? A mi no se me olvidan esa palabras. Una 
pol?tica de mostrar poder.? ?Ernesto,? Interview No. 14. 
24
 ?Ellos siguen estando por ah? en peque?as c?lulas. No tienen aci?n como tal, tan abrupta, sino que salen 
en peque?os grupitos, y se gesta un nuevo proceso de hacerse sentir ante la comunidad. Elos andan 
vendiendo boletas semanal, al estilo de vacunas, a toda tienda le legan con una boletica y la tienen que 
comprar. Entonces ah? est?n contestando con una acci?n ilegal.? Ibid. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 289 
and tel us, ?Se how dangerous it is? It?s best that you pay us.??
25
 The drug sales also 
returned: The Government Secretary of Medel?n was quoted in the pres as saying that, 
?In Comuna 8, more than anywhere else in the city, we have a problem with the gangs 
fighting over retail drug sales,? and that the city was learning about more and more drug 
markets there.
26
 ?There are clashes,? Ernesto said, ?betwen gangs and those things. So 
you?re starting to se one death, another death, in other words, isolated deaths, and that?s 
generating betwen them a new phenomenon that?s not going away.?
27
 
People were hesitant to admit that they knew ho was responsible for these acts. 
Or rather, some of the people I interviewed said that other people in the sector didn?t 
know or didn?t want to admit that they knew ho was responsible, even though they 
themselves did. ?Nobody wants to say who gives the orders,? a former resident who 
currently works on development projects in Caicedo La Siera told me.
28
 Nobody wanted 
to say who was behind the selective kilings or who was fighting in upper Caicedo La 
Siera either, he added. A community leader added that the uncertainty was creating a lot 
of anxiety in the sector: 
But currently there?s a litle anxiety, not just here in the bario or in the 
comuna, but in Medel?n, about what?s happening. So the comunity asks 
me, ?But what?s going on?? They?ve said that to me. ?We must be going 
                                                
25
 ?Ellos tienen a quienes nos asalten en las cales, nos roben en nuestras casas, nos hurten las motos, y 
cuando eso pasa, entonces vienen y nos dicen: ?ven c?mo hay de inseguridad? Lo mejor es que nos paguen 
a nosotros.? Quoted in Juan Diego Restrepo E., ?Con Palizas Se Impone el Control Social en Comunas de 
Medel?n.? 
26
 ?En la Comuna 8, m?s que en toda la ciudad, tenemos un problema con los combos que se disputan la 
venta de droga al detal. Hemos detectado muchas casas de vicio.? Quoted in Mary Luz Avenda?o, ?En la 
Mira de Delincuentes.? 
27
 ?Hay choques ? entre bandas y esas cosas. Entonces empieza a ver un muerto, otro muerto, es decir 
muertes aisladas, y eso genera entre ellos un nuevo fen?meno que no se desaparece.? ?Ernesto,? Interview 
No. 14. 
28
 Diego R?os, Interview No. 4. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 290 
back to the time before, when we couldn?t walk from one block to another 
because when we?d least expect it there?d be bullets.? Or: ?I?m not going 
to be able to send my kids to school alone. I?m going to have to take them 
there myself, because anything could happen to them.? So there?s an 
anxiety, but the anxiety is because of what I said before, that there are 
some groups or individuals, because we don?t realy know who they are, 
people who want to do some damage to society, to the city. But [it?s] for 
the reason I told you earlier as wel: the desire for power.
29
 
Of course the desire for power was part of it, but what they were fighting over 
was aces to the sector?s ilicit markets. And several people I interviewed, even after 
saying that nobody wanted to say who was responsible for the selective kilings, 
nevertheles repeated the widespread speculation (repeated as wel in media acounts) 
about who it involved. The rising violence was a consequence of the fight betwen ?Don 
Mario? Rend?n?s network and ?Don Berna? Murilo?s network, the later?s batle being 
fought by proxy by ?Job? L?pez and ?Mem?n? L?pez before their asasination and 
capture (respectively).
30
 But by late 2008, with Murilo, ?Job,? and ?Mem?n? al out of the 
picture, it was a fight betwen Rend?n and even lower-level subordinates from the 
disintegrating Envigado Ofice. ?Since December, Medel?n?s been having problems with 
?Don Mario?,? a community leader in La Siera told me. ?They?re kiling the demobilized 
paras.?
31
 (In fact it was rumored that, after ?Job? L?pez?s death, one of his aides switched 
sides and was now helping ?Don Mario? Rend?n take control of the area?s drug trade.) 
                                                
29
 ?Pero actualmente hay un poquito de zozobra, no solamente aqu? en bario, ni en la comuna, sino en 
Medel?n, por lo ocurido. Entonces a mi la comunidad se pregunta, ??Pero que est? pasando?? A m? me lo 
han dicho. ??Ser? que vamos a volver a la ?poca de atr?s, donde no pod?amos caminar de una cuadra a otra, 
porque cuando menos pens?bamos, bala?? O: ?Mis hijos no voy a poder mandarlos a la escuela solos, 
porque los tengo que levar porque les puede pasar alguna cosa.? Entonces hay una zozobra, pero la zozobra 
es por lo que yo le dec?a anteriormente, que hay unos grupos o personas, porque nosotros no sabemos en 
realidad quienes son, que quieren hacerle un poquito de da?o a la sociedad, a la ciudad. Pero por lo que yo 
te dec?a tambi?n: las ganas de poder.? ?Flavia,? Interview No. 15. 
30
 ?Danilo,? Interview No. 10. 
31
 ?Medel?n desde diciembre viene con problemas con Don Mario. Est?n matando a los desmovilizados.? 
?Boris,? Interview No. 8. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 291 
Now, since May 2009, with Rend?n out of the picture as wel, the full local story is 
simply untold: either it was the case that locals realy didn?t have any idea who could be 
behind the most recent wave of kilings ? al the big names from the previous chapters 
(plus many others whose inclusion would have needlesly complicated this already 
complicated story) were al out of the picture as of mid-2009, leaving today?s fight 
mostly to a bunch of mid-level and low-level unknowns ? or it was the case that locals 
realy were afraid to admit that they knew ho was behind the kilings because it had 
become so dangerous again that saying anything could be deadly. 
7.2.3. The Hopes of the Community 
Even given the fights over power, the selective kilings, the drug traficking, and 
the anxiety that by mid-2009 was growing wek by wek, if one compares the present 
with what was happening in Caicedo La Siera ten years ago, one would have to conclude 
that most residents were living with les worry today than before ? although that can 
change quickly.
32
 ?It?s like a drug market around here. Wel, they don?t sel it here, they 
sel it out there, but stil,? Cristiana said, refering to the drug dealers who lived in the 
bario but who, she believed, didn?t operate there. Stil, she said, ?I?m very much at 
peace, and I don?t think anything?s going to happen. There have been some things, but as 
you can se it?s just themselves against themselves.?
33
 (In fact, one police source told a 
pres agency that only 63 percent of the victims of homicides in 2009 were members of 
                                                
32
 R?os, Interview No. 4. 
33
 ?Por aqu? es como una plaza, que no la venden por aqu? sino por al?, pero ?. Yo tengo mucha 
tranquilidad y no creo que vaya a pasar nada. Ha habido cosas pero como uno ve que es de los mismos con 
los mismos.? ?Cristiana,? Interview No. 9. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 292 
drug gangs: it was not just ?themselves against themselves.?
34
) She admits it?s scary 
when somebody gets kiled, but she goes on with her life. ?I don?t pay atention to rumors 
about what?s going on with this or that.?
35
 It sems likely, however, that she would begin 
paying atention again if those kilings reached some unknown ?critical mas? in her 
bario or if the violence were to touch her family directly. 
Young people today, however, sem les interested in the path that leads to more 
violence, a local school oficial told me.
36
 A community leader also said that young 
people don?t have such a sentimental atraction to the gangs anymore. ?Before, they?d get 
involved with them ore, but now it?s changed.?
37
 Another suggested that the 
community in general, having experience a period of peace, has a strong desire not to 
regres to more violent times: ?When there was so much violence, the community was hit 
realy hard with a lot of resentment, because people who had nothing to do with it pased 
away.? 
So now when al of that ended the community started going out walking 
around a lot more, geting themselves involved in the story about how 
these kids now wanted to repair the damage they?d done. So in these seven 
years there?s been a transformation in which the community?s progresed 
a litle economicaly, in education, [and in the fact] that there have been 
asociations formed, organizations that want to get their hands dirty in 
community work.
38
 
                                                
34
 Agence France-Prese, ?Venganzas de ?Narcoparamilitares? Disparan Homicidios en Medel?n,? 
MetroLatinoUSA, 13 July 209. 
35
 ?No hacen caso de los rumores de que el que sigue es tal o cual.? ?Cristiana,? Interview No. 9. 
36
 ?Alicia,? Interview No. 7. 
37
 ?Antes se met?an m?s con elos, pero ahora ha cambiado.? ?Boris,? Interview No. 8. 
38
 ?Cuando hubo tanta violencia la comunidad qued? muy golpeada con mucho resentimiento, porque fal? 
gente que nada tenia que ver. Entonces cuando ya acabo todo esto la comunidad empez? a salir mucho m?s 
a caminar, a meterse en el cuento de que ya estos j?venes quisieron reparar el da?o que hicieron. Entonces 
en estos siete a?os ha habido la transformaci?n de que la comunidad ha progresado un poquito 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 293 
If one mesage came through clearly in al of the interviews, it was that the 
community longed for a constructive state presence, not the low-level resources provided 
through the JACs or the high-intensity police raids of the past, but sustained atention and 
protection by state actors who understand the importance of respecting the people who 
lived there. They had been ruled by leftist nonstate armed actors, right-wing nonstate 
armed actors, and non-political nonstate armed actors, and the experience of living in 
statelets at war had left them convinced that it was not a sustainable model. They did not 
necesarily trust state actors, but there were signs that, if nothing else, they were wiling 
to give the state a chance: 
The other isue would be, with regard to the comunity directly, with 
regard to the state, governance, the authorities, there has to be ? realy, 
the state has to show a real capacity to live [in the community,] 
approaching al the spaces. I don?t se how it?s so eficient to have this 
quantity of police patrolling [just] the main road. There should be a closer 
relationship ? betwen the community and the authorities. We?ve sen 
how hard that is, too, because people stil se the authorities as something 
that represes them. So from there, you?d have to start cultivating other 
kinds of relationships with the people who are here.
39
 
In contrast to how a lot of state authorities were viewed, the increased presence in 
Caicedo La Siera of the community police, units of the Imediate Atention Command 
(CAI: Comandos de Atenci?n Inmediata
40
), semed to be universaly welcomed, at least 
among those who were not involved in organized crime. In fact, everyone I spoke with 
                                                                                                                                            
econ?micamente, en educaci?n, que ha habido conformaciones de asociaciones, corporaciones que quieren 
meterle un poquito de mano al trabajo comunitario.? ?Flavia,? Interview No. 15. 
39
 ?Lo otro ser?a, eso en cuanto a la comunidad directamente, en cuanto al estado, la gobernabilidad, a las 
autoridades, tiene que haber, realmente, tiene el estado que mostrar una capacidad plena de vivir, 
abordando todos los espacios. Yo no lo veo tan eficiente de que est?n rondando cantidades de polic?as por 
una v?a principal. Deber?a haber mas acercamiento ? entre la comunidad y la autoridad. Tambi?n eso lo 
estamos viendo dif?cil, porque la gente sigue viendo la autoridad como la que reprime. Entonces desde ah? 
habr?a que empezar a gestar otro tipo de relaci?n con la gente que est? presente.? ?Ernesto,? Interview No. 
14. 
40
 Sometimes caled Imediate Atention Centers (Centros de Atenci?n Inmediata). 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 294 
agred that there were too few of them. ?That?s also a weaknes with the police, because 
they walk around the main roads but nowhere else,? one resident said. ?The rest [of the 
strets and aleys] are left very much on their own.?
41
 That could lead to a fragmentation 
of control betwen state and nonstate actors, if the nonstate actors were to decide that 
controlling the side strets and back aleys would be in their interest; so far they have not, 
but that could change quickly. Flavia said the community needed a lot more support for 
security in general. Earlier she had expresed real fear about the witneses who had 
testified against her boss, ?Mem?n? L?pez (se ? 6.2.1); clearly there were tensions 
betwen them and the people she worked with, and it is likely that she realy was in 
danger. And even though they were the ones who got kiled, not she, her point is taken 
that, with beter security for everyone in the barios, those sorts of disputes would have 
been much les likely to lead to violence, and there would be much les anxiety about a 
return to bad times.
42
 
Boris, who as a community leader in La Siera was probably the least hopeful 
about the future, nevertheles believed that there were thre factors that were keeping the 
situation from geting worse at the moment: ?the growth in education, the CAI, and the 
fact that the community is stil working.?
43
 In other words, education was helping young 
people envision a life beyond gangs and gang wars; the CAI were giving people a sense 
of security and somebody other than the ex-paras or other ?muchachos? to consult to get 
help with neighborhood conflicts; and a lot of the community organizations (though not 
                                                
41
 ?Tambi?n hay una debilidad de la polic?a, porque ela camina por las v?as principales nada m?s, y el resto 
se mantiene muy solo.? ?Ernesto,? Interview No. 14. 
42
 ?Flavia,? Interview No. 15. 
43
 ?? el crecimiento de la educaci?n, el CAI y que la comunidad sigue trabajando.? ?Boris,? Interview 
No. 8. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 295 
al) were providing esential services without leting themselves get dominated by the ex-
paras? own quasi-ilicit structures. What might prevent those structures from completely 
taking over the JACs and the city?s contracts in the sector, Boris thought and others 
agred, were the fact that the mayor?s office was phasing out the practice of giving cash 
directly to the JACs for city contracts, and that people were starting to view the new ten-
year Comprehensive Urban Plans (PUI: Planes Urban?sticos Integrales) as a framework 
through which to envision what their barios and their families? lives might look like in 
the future. The PUIs were funding infrastructure projects that would provide local, 
manual-labor jobs at least through the planning period. Those were the only kinds of jobs 
that many people in those barios were qualified to do, and so having that possibility 
could give them a broader horizon with which to plan their lives and thereby minimize 
the anxiety about the future that could be exploited by ilicit actors offering quick 
solutions.
44
 
In fact, unemployment was the only factor that everybody I talked to cited as the 
weakest link in everyone?s hopes for sustainable peace. ?I think that one of the main 
problems involving the conflict is the need [for] opportunities, for development, [so they 
can] live wel,? Ernesto said. 
Unfortunately, the entire population of these places is realy vulnerable. 
They have realy shameful economic levels. Our people?s culture isn?t 
?Give me such-and-such? but ?Give me a job, help me get a job.? So it?s 
realy hard, because young people, out of the desperation of hardship, go 
out in search of whatever opportunity or new income.
45
 
                                                
44
 ?Ernesto,? Interview No. 14; ?Boris,? Interview No. 8; ?Danilo,? Interview No. 10. 
45
 ?Yo pienso que una de las principales problem?ticas que conleva al conflicto es la necesidad de tener 
oportunidades, para el desarolo para vivir bien. Desafortunadamente toda la poblaci?n de estos espacios 
es muy vulnerable. Tiene unos niveles econ?mico muy lamentables ?. La cultura de nuestra gente no es 
reg?leme tal cosa, sino denme empleo, ay?denme a tener empleo. Entonces eso es complicad?simo, porque 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 296 
Without legal sources of income, he was suggesting, people would do what they needed 
to do so they could fed their families. ?We al know that the state can?t just have a 
military presence,? Danilo said, adding that while people recognized that the state had 
provided beter opportunities for education and other needs, it stil needed to help create 
job opportunities that could contribute more directly to the community and family 
development.
46
 Again, the atitude in the community was that the state would find fertile 
ground in which to lay down roots should it choose to do so in a long-term, constructive 
manner: protect them so they could go about their daily lives; give them opportunities 
and a framework through which they might make long-term plans for their lives; make 
life safe and predictable, residents and leaders were saying, and the community would 
have every reason to oppose nonstate armed actors who threatened to disrupt that 
stability. 
I ended some of the heavier interviews with two questions about the future. One 
of those questions was: What gives you hope that the peace your community has been 
experiencing might be sustainable? Boris had earlier cited education, the CAI, 
community work, and the PUI, but in answering this question now, he expresed much 
more pesimism. The words hope, expect, and wait for al translate into Spanish as 
esperar, and so while my question asked about hope, his response used the same word in 
the sense of expectation: ?One is ? expecting that they wil come for one and kil one,? 
he said, using the local manner of refering to oneself in the third person to expres 
uncomfortable thoughts. He was talking about the fear among many community leaders 
                                                                                                                                            
los j?venes del desespero de la necesidad salen a la b?squeda de cualquier oportunidad o ingreso nuevo.? 
?Ernesto,? Interview No. 14. 
46
 ?Sabemos que el Estado no s?lo tiene que hacer presencia militar.? ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 297 
that they might be kiled by the ex-paras or others, a fate that many of their colleagues 
throughout the city had suffered in recent months. Some believed it was just a mater of 
time. The community leaders had earlier tried to facilitate the reinsertion proces in their 
own barios, but they just ended up putting themselves and their families at risk for their 
eforts, and they did not have adequate protection. ?We?re just realy fucking tired of it. 
At first we thought we had the solution and that we could mediate, but now that one has 
been burned it?s beter to just leave it alone, because it?s a [crime] syndicate. One is just 
wasting time with them, because a distinction has to be made [to community members] 
that I am the facilitator and not [one of them].? With al that has happened, he said, al he 
can do is ?cross one?s fingers and se what happens.?
47
 Others were cautiously 
optimistic, recognizing that, having lived through a period of peace during which a new 
generation of young people have had the opportunity to envision a diferent future, 
people in the community would make a much stronger efort to prevent a return to the 
violence of the past, and would be very wiling to work closely with whomever could 
help keep the peace. 
At the end of our interview ith Ernesto, the community leader who had found 
out there were strangers interviewing Cristiana and came to investigate, I tried asking him 
the second question that I was asking people about the future: What would happen if 
some group other than the state were to appear in Caicedo La Siera and start ofering 
protection, stability, and opportunities again? Something got lost in my own translation, 
because he agred that it would be great to get more support from international 
                                                
47
 ?Uno est? ? esperando que vengan por uno y lo maten. ? Ya estamos mamados. En un principio 
cre?amos que ten?amos la soluci?n y pod?amos mediar, pero ahora uno se quema, es mejor dejarlo as? 
porque est? sindicado. Se boletea uno con elos, porque se tiene que diferenciar que estoy de facilitador y 
no dentro de [elos]. ? Cruzar las manos y dejar que suceda.? ?Boris,? Interview No. 8. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 298 
nongovernmental organizations (NGOs) and investments from abroad. My research 
asistant knew hat I was geting at, however, and rephrased the question for me: ?And if 
it weren?t a group of international corporations or NGOs, if it weren?t so legal, who came 
and offered young people a way of making money [for example], do you think the 
community could legitimize this group that arives, because they?re solving the 
community?s problems??
48
 After a brief pause, he answered with equal parts sadnes, 
resignation, and defiance: 
It could happen. You know, the need is so great that, if there are resources, 
nobody?s going to turn down the economic benefit. That?s the problem, 
that there are young people without jobs. We?ve heard it, that they?re 
drawing in young people and they?re going to pay them 500,000 pesos 
[about $250]. You hear that and right there the enthusiasm starts. I 
personaly think that, if that piece of news reaches some high school, al 
the kids wil drop out of school. Those things could happen. God, nobody 
wants that to cause new problems with order, but it?s very feasible that 
that could happen and that it could be acepted ?.
49
 
What is interesting is that, while everyone cited unemployment as the main 
potential source of instability, their own words and their own history suggested that 
unemployment was realy just one factor. Unemployment certainly was a problem in 
itself: it was causing hunger and anxiety and arguably could have been increasing the 
risks of intrafamilial violence (although I have not been able to find reliable evidence to 
make this later judgment with confidence either way). It was the isue that was most 
                                                
48
 ?Y si no fuera un grupo de corporaci?n internacional o ONGs, que no fuera tan legal, que viniera y 
ofreciera a los j?venes una forma de ganar dinero, ?usted cre que la comunidad puede legitimar ese grupo 
que lega, porque est? solucionando un problema de la comunidad?? In ?Ernesto,? Interview No. 14. 
49
 ?Puede pasar. Es que la necesidad es tan grande que si hay recursos, nadie se va oponer a un beneficio 
econ?mico. Ese es el problema, que hay muchos j?venes desempleados. Nosotros hemos escuchado, es que 
les est?n recogiendo los muchachos y le van a pagar de a 50,00 [pesos]. Uno escucha eso y ah? mismo 
empieza el entusiasmo. Yo personalmente pienso, que si una noticia de esas lega a un colegio, se retiran 
todos lo pelaos de colegio. Pueden pasar esas cosas. Dios, no quiera que sea generar nuevas problem?ticas 
de convivencia, pero eso es muy factible que pase y que sea aceptado ?.? ?Ernesto,? Interview No. 14. 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 299 
salient in their daily lives, so it makes sense that they would atribute any potential for the 
rise in violence to that particular chalenge. But its actual role with respect to the factors 
that had contributed to a rise in violence in the past had always been more instrumental 
than fundamental: it contributed to the more generalized problem of unpredictability that 
has caused them to resent (and ilegitimize) whomever was in charge at the time. Without 
reliable employment, one cannot know where one?s next meal is coming from or how one 
is going to pay the bils. When, on top of that, there has been a deficiency in physical 
security and public order as wel, then their lives have been filed with so much 
uncertainty that they have been wiling to acept help from whatever entity credibly 
promised a return to order: and that is what has led to violence asociated with contests 
for teritorial control and subsequently for the right to rule. In short, unemployment and 
insecurity have been ?bads? in themselves; but they have also been factors of 
unpredictability, which has ben instrumental to ilegitimacy, which has activated violent 
disputes over teritorial control. 
Physical security and social order provide the context for a predictable life, and 
the groups that have managed to control statelets in Caicedo La Siera for any length of 
time have been those who were able to provide the context for a predictable life. Once 
that changed, once people no longer knew hat to expect on a daily basis ? were no 
longer confident that going to the store or walking their children to school would not 
result in death, injury, or theft ? then the loss of predictability was soon followed by 
violence leading to the loss of teritorial control. Groups with relatively few resources 
have had to provide more than mere predictability: they have had to atend to the 
community?s values ? learn them, promote them, enforce them ? and to do so in a way 
Lamb, Robert D. | Chapter 7. A Window of Opportunity, Closing Quickly: 207-2009 300 
that was sen as fair or equitable or acesible. Once that changed, once the group in 
control could no longer use community values to win voluntary support (and since it did 
not have the resources to barter for that support), its loss of legitimacy, and then its 
ilegitimation, was soon followed by violence leading to its loss of teritorial control. 
If the state had and were wiling to invest the resources into addresing 
unemployment and poverty in Caicedo La Siera, and in Medel?n?s historicaly violent 
peripheral barios more generaly, it would have ben in a position easily to buy the 
loyalty of the communities in Caicedo La Siera. Although there were long-term plans in 
place to make such investments in the future, adequate resources were not imediately 
available as of mid-2009 (and, in fact, were probably insufficient beyond then as wel). 
But in mid-2009 security forces were ariving in those barios in numbers that were 
alarming to some human rights activists, whose views of the security forces were of 
people who commit or permit atrocities: they were refering to the recent increase in 
troops as the ?militarization? of the peripheral barios. However fair or unfair their 
characterization of the situation, the point is taken that, whatever else the security forces 
would do there, if the state?s goal were to keep violence at sustainably low levels, they 
would need to have a presence that would be capable not only of targeting criminals and 
other ilicit actors but capable as wel of protecting residents and providing them a 
context for a predictable life. But not just capable of protecting them: actualy protecting 
them, respecting them, and giving them opportunities to plan out their days, their weks, 
and ? if, against history, a constructive state presence could be sustained ? their 
futures. 
Lamb, Robert D. | Part I. Analysis and Conclusions 301 
Part I. Analysis and Conclusions 
? there has never been an example in Colombia of a para-statal security 
group that has not ultimately operated with wanton disregard for human 
rights or been corrupted by local economic interests. 
 ? Myles Frechete, U.S. Ambassador to Colombia (1994)
1
 
 
The epigraph above was from a cable writen by oficials at the U.S. Embasy in 
Colombia who were concerned about the potential destabilizing efects of the Colombian 
government?s Convivir proposal, which was to alow for the creation of ?special 
vigilance and private security services? in areas where state security forces were too 
weak to protect people from guerilas and criminals (se Chapter 5). The cable just as 
easily could have been refering to the Coosercom program, which alowed demobilized 
guerilas to provide security services (se ? 4.1.2), or to the Colombian state?s reliance 
on paramilitary groups to maintain public order amid civil war (se Chapters 5-6). In 
Colombia?s recent history, where there has been a legal framework for regulating the 
behavior of such groups, as in the case of Convivir and Coosercom, the state has never 
provided the resources or oversight needed to adequately enforce the regulations and laws 
or ensure that the groups would actualy try to acomplish the government program?s 
objectives. Where there has been no such legal framework because the relationships have 
been ilegal, as in the case of de facto, off-duty, or official-secret support for the 
                                                
1
 American Embasy Bogot?, ?Botero Human Rights Leter to A/S Shatuck,? 9 December 194, cable to 
United States Department of State, published electronicaly by The National Security Archive, 
http:/ww.gwu.edu/~nsarchiv/NSAEB/NSAEB217/doc01.pdf (acesed 30 June 2009). 
Lamb, Robert D. | Part I. Analysis and Conclusions 302 
paramilitaries, the state has rarely had the political wil or the resources to prevent those 
relationships or to mitigate the harms and abuses that inevitably would result from them. 
In al cases, even where some short-term objective has actualy been achieved (and it 
often has not even acomplished that much), it has come at the expense of the long-term 
stability and the legitimacy of state actors in the areas under the control of the nonstate 
armed actors ? not to mention at the expense of the security of unarmed local 
populations. 
Likewise, the state?s management of its various demobilization, disarmament, and 
reinsertion (DR) programs can be described at best as unfulfiled promises and at worst 
as unmitigated disasters. The observation, quoted earlier, that ?Colombians have acquired 
the status of serial demobilizers,? is worth repeating: 
The proces has become ritualized: cease-fires are declared and 
negotiators asemble in jungle clearings. After violations caried out by 
renegade elements on both sides, insurgents eventualy mas in designated 
areas. Amid proclamations before an asembled international pres that 
Colombia has, at last, turned the corner in its long history of armed 
violence, weapons are surrendered and peace declared. So far ? [this 
history] has merely transitioned, rather than terminated, violence ? for 
the same reasons: the state lacks the power and legitimacy to enforce the 
agrements and the resources to integrate demobilized fighters into the 
legitimate economy. As a result, enough bad actors remain to cary the 
violence forward.
2
 
The problem is that the state has been negotiating its demobilization frameworks from a 
position of weaknes (or coruption); alowing the disarmaments to involve mainly 
broken and unloved weapons; providing far too few resources for what is perhaps the 
most important step, the reintegration of the demobilized and disarmed combatants; and 
atending inadequately to the demands of victims, who have far outnumbered combatants 
                                                
2
 Douglas Porch and Mar?a Jos? Rasmusen, ?Demobilization of Paramilitaries in Colombia.? 
Lamb, Robert D. | Part I. Analysis and Conclusions 303 
(and many combatants had themselves been victims who took up arms in self-defense or 
revenge). 
Beneath both of Colombia?s main traditional approaches for dealing with the 
problem of violence ? arming nonstate actors in an efort to achieve victory, or 
disarming nonstate actors in an efort to achieve peace ? lies an ignorance about how 
these eforts might afect either the communities in which violence emerges or the 
dynamics of legitimacy betwen and among community, state, and nonstate actors. 
Ignoring either in the past has proven fatal to the long-term succes of these and other 
policies. In Part I of this thesis, I hope to demonstrate that, instead of ignoring them, 
explicitly acounting for them in policy might improve the chances of long-term succes. 
Chapter 8 describes the paterns of complex violence as they evolved in Medel?n 
betwen 1984 and 2009. It is intended to answer the first of this study?s thre research 
questions: what were the paterns of violence in Medel?n? Chapter 9 analyzes the 
dynamics of the legitimacy or the ilegitimacy of nonstate armed actors, community 
actors, and state actors in Medel?n. That chapter is intended to answer the second two 
research questions: what explains the paterns of violence in Medel?n, and what role did 
legitimacy or ilegitimacy play? Finaly, Chapter 10 lays out the study?s findings and 
disusses their implications in terms of some general propositions that might be tested in 
the future for their relevance to theory and policy. 
 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 304 
Chapter 8. Complex Violence in Medel?n 
Medel?n had been chosen as a case study of the relationship betwen legitimacy 
and violence for two main reasons: First, it was a perfect example of the kind of place 
where many diferent kinds of policy problems involving violence were active: terorism 
(car bombings, drive-by shootings, masacres), insurgency (guerilas, urban militias), 
paramilitaries (vigilantes, death squads, social cleansing), transnational crime (drug and 
arms traficking), organized crime, human rights abuses, gang warfare, humanitarian 
emergencies, and so on. Second, it had experienced a dramatic decline in most forms of 
violence in just a few short years ? a phenomenon so rare that some were caling it the 
?Medel?n Miracle? ? making it both an important case to study (what explains the 
Medel?n Miracle?) and a practical case to study (an outsider could live there safely). It 
turned out, however, to be an even more interesting place to study than expected: during 
my ten months of field research, it became clear that violence was rising again. The study 
therefore became a question of what has caused the paterns of violence to change there 
in the 25 years that violence has been the city?s most salient policy problem. 
Before one can explain any changes in the paterns of violence in Medel?n, 
however, one needs to begin by characterizing those paterns, and that is what this 
chapter endeavors to do: describe Medel?n?s complex violence. 
Violence, for the purposes of this study, is any behavior that causes one or more 
human beings death, bodily injury, physical pain, psychological trauma, suffering, or 
fear. Violence is manifested in two main ways: physicaly and psychologicaly. Physical 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 305 
violence is any behavior that involves physicaly touching the victim ? whether directly 
(beating, striking, shoving, etc.) or indirectly (throwing, launching, initiating, etc., as with 
a weapon, fire, energy, sound waves, etc.) ? in a way that causes the victim pain, injury, 
death, trauma, suffering, or fear. Psychological violence is any behavior that does not 
involve physicaly touching the victim but that nevertheles causes the victim to 
experience trauma, suffering, or fear. In other words, both manifestations of violence can 
result in psychological reactions ? trauma, suffering, and fear ? but only physical 
violence can result in pain, injury, or death as wel. 
In addition to these two manifestations of violence, al violence takes place in a 
particular context that can afect both the severity of the victim?s psychological response 
and, often, the brutality of the violent act itself. That context has to do with the 
relationship betwen victim and perpetrator, the motivations underlying the act of 
violence, or both. This implies thre levels of analysis: the individual level, the 
interpersonal level, and the collective level. 
Taking place at the individual level is self-inflicted violence, which involves only 
one person, who is both perpetrator and victim. Its most common physical manifestations 
are self-mutilation, suicide, and suicide atempts. Self-inflicted psychological violence 
often derives from mental ilnes and can include drug abuse and addiction, participation 
in extremely risky behaviors such as erotic asphyxiation and anonymous promiscuity, 
eating disorders such as anorexia, or self-neglect that leads to infectious disease or 
unsustainable weight loss or weight gain. 
Violence at the interpersonal level involves one or two perpetrators acting against 
one or more victims with whom they either are intimately related (as family members, 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 306 
life partners, close friends, etc.) or are pasing acquaintances or strangers. Intrafamilial 
violence is interpersonal violence that takes place inside the home or among close 
relatives or acquaintances (including spousal abuse, child abuse, elder abuse, pedophilia, 
date rape, etc.), while communal violence involves one or two perpetrators acting against 
one or more victims whom they may or may not know, and it may take place in 
someone?s home, on the stret, at a commercial property, or in some other public area 
(this includes most categories of common, rather than organized, crime: robbery, asault, 
rape, etc.). Note that this usage of the term communal difers from that used by some 
conflict researchers who use communal conflict or intercommunal conflict in a way that is 
similar to how I use collective violence. To avoid confusion, I wil usualy use the terms 
interpersonal-communal or intrafamilial-communal violence, but even without the 
modifiers the reader should keep my intended meaning in mind throughout this work. 
Violence at the collective level involves one or more perpetrators, who are 
members of an identifiable group, acting against one or more victims for social, 
economic, or political reasons. Collective violence is social violence if the group is a 
social (ethnic, linguistic, religious, etc.) group, formed for reasons of afection, common 
interest, common history, or identity (such as youth gangs, stret gangs, civic 
asociations, ethnic mobs, etc.) and if the act of violence is commited to enforce respect, 
expres pride in the in-group, expres disdain for an out-group (such as a competing gang 
or a diferent racial or ethnic group), to defend or acquire teritory for the purpose of 
social control, or to punish members for violating in-group rules, among other reasons. 
Collective economic violence involves any group formed for the primary purpose of 
making money, whether from the sale of goods or services (legal or ilegal) or by theft or 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 307 
fraud, which includes most organized crime groups and other criminal conspiracies: 
mafias, drug trafickers, money launderers, asasins for hire, crime rings, racketers, and 
otherwise legitimate busines managers using violence to coerce or enforce a cartel or 
monopoly arangement or to take or protect asets (as in the forced disappearance of 
trade-union organizers, or the forced displacement of peasants and the expropriation of 
their land). Finaly, collective political violence involves acts commited by one or more 
members of a group that makes or purports to be making decisions about political goods 
(e.g. they are doing the job they think a government should be doing), which may include 
agents of states and governments (police, military, and other security forces) whether 
acting in their official capacity or not, or may include nonstate or ilegal organizations 
such as paramilitaries, vigilante groups, self-defense groups, militias, or guerilas. Acts 
of political violence can include war, asasination, forced displacement, imprisonment, 
and the forced disappearance of political opponents and human-rights defenders, among 
other acts. Any given act of collective violence can count as more than one type, such as 
forcing guerila sympathizers off their land for political reasons but expropriating their 
land for economic reasons. 
This framework ? two ?manifestations? in six ?contexts? at thre levels ? is my 
modification of a framework developed by the World Health Organization for its study of 
violence as a public-health problem.
1
 The modified framework is discussed in detail in 
Appendix B; the reader is encouraged to consult that discussion before proceding. By 
breaking down and describing the manifestations and contexts of violence, the framework 
described there and used here is designed to make a study of complex violence tractable. 
                                                
1
 Etiene G. Krug, et al., World Report on Violence and Health. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 308 
Complex violence, as discussed in the introduction of Appendix B, involves multiform 
armed actors, in networked relationships, with shifting loyalties, and with diverse 
motives; complex urban violence is complex violence that takes place in cities. 
This framework was used to guide data collection on complex violence in 
Medel?n, although it was not possible to obtain data for most of the categories of 
violence. For that and other reasons, this study focuses its analytic atention primarily on 
interpersonal-communal violence and on the thre forms of collective violence, with most 
atention to their physical manifestations, since quantitative data are available, but some 
atention to their psychological manifestations as wel, to the degre that qualitative data 
could be derived or infered. Future work should expand this analysis to include al 
categories and both manifestations. Even with these limitations, however, the use of this 
framework made possible a richer description of the paterns of violence in Medel?n than 
had previously been available from other sources. 
8.1. Patterns of Violence 
This study was not a study of the ?decline? in violence in Medel?n in the 2000s, 
nor a study of the ?rise and fal? of violence since 1984. Rather, it was a study of the 
changes in the overal paters of violence in Medel?n over five periods of time since 
1984. Paterns of violence are the changes in either the manifestation of violence (the 
diferent physical and psychological forms) or the context of violence (the relationship 
betwen victim and perpetrator) over time, either quantitatively (e.g. a decline in 
homicide rates) or qualitatively (e.g. a shift to displacement instead). A quantitative 
change is the number or severity of incidents of violence or of the results of violence, 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 309 
such as the amount of fear people report, within the populations under study. A 
qualitative change is a shift in the form that violent acts take within that population, such 
as the shift in emphasis (during the 2003-2007 period when homicides declined) away 
from physical violence and towards psychological violence. 
In an efort to understand both the national context and the local details of the 
violence taking place in Medel?n, I tried to obtain data (both quantitative and qualitative) 
for al twelve categories of violence described in the framework not only for Medel?n 
itself but for Colombia as a whole and for a smal sector within Medel?n caled Caicedo 
La Siera. Quantitative data, however, were not available for al twelve contexts (self-
inflicted, intrafamilial, communal, social, political, and economic, in both their physical 
and psychological manifestations) at al thre levels of analysis (Caicedo La Siera, 
Medel?n, and Colombia) across al 25 years of the study period (1984-2009), as would 
have been ideal. Nonetheles, this chapter is the first to publicly review in one place al of 
the quantitative data that actualy are available for the city of Medel?n, that were 
reasonably reliable, and that were made available by their private owners or could be 
found in the public domain. In some cases where quantitative data did not exist (or could 
not be obtained), qualitative data were found or infered. These qualitative data did not 
alow for observations about relative magnitudes of violence across categories, but they 
did in some cases make possible general observations about yearly trends within each 
category of violence (i.e. whether it rose, fel, peaked, or hit a minimum in any given 
year) and about shifts betwen and among diferent categories over time. 
Time-series data within each study period were desirable because with time series 
one can describe trends in each form of violence rather than setle for static descriptions. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 310 
For example, rather than saying that more than 30,000 people were murdered in Medel?n 
during the 1992-1998 period (an average of about 240 murders per 100,000 inhabitants 
per year) or that about 620 people comited suicide (an average of about 90 people per 
year) during the period, the time series enables one to observe that the homicide rate 
declined by more than half during the seven-year period, while the suicide rate fluctuated 
widely. For those instances when time-series data were not available within time periods, 
some useful inter-period comparisons could stil be made using, for example, data from 
the censuses of 1983, 1995, and 2005. Sometimes the paterns had to be detected 
indirectly. For example, it was clear that, with the demobilization of the paramilitary 
blocs in Medel?n in the early to mid-2000s, political violence in the city was largely 
replaced by economic violence, as new reports and government investigations 
demonstrated that many of the former narco paras continued their involvement in 
organized crime. Combined, the quantitative and qualitative data that are described in the 
sections that follow, while stil unsatisfying as a comprehensive description, nevertheles 
provide a richer understanding of the overal paterns of violence in Medel?n than had 
been available from any other source. 
In many cases, it is not clear from the data whether a particular crime was an 
instance of interpersonal violence or collective violence. Rape is frequently used as a 
weapon of war (i.e. collective-political violence), but data for rapes even where available 
rarely identify the perpetrator or his motives, so it is not certain what portion are 
interpersonal rapes and what are collective rapes; finding that out requires that one look 
beyond raw number to the stories behind them. Homicide is an even beter example of 
this chalenge: of the more than 40,000 homicides in Medel?n during the 1990s, the data 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 311 
do not show, for example, how many were domestic incidents by jealous spouses 
(interpersonal-intrafamilial violence), how many were robberies gone wrong 
(interpersonal-communal violence), or how many were casualties of wars among stret 
crews (communal-social violence). This complicates eforts to analyze paterns ? and 
indeed demonstrates why it is necesary to study not only forms of violence beyond 
homicides but the context in which al of those forms of violence were taking place. 
8.2. Homicide Data 
Nonetheles, since aggregate data on homicides are the most readily available and 
widely discussed data on violence in Medel?n, it makes sense to begin with a discussion 
of paterns in homicides before reviewing the rest of the data. The quantitative data come 
from five main sources: the Secretar?a de Gobierno de Medel?n (the office of the 
Government Secretariat), the administrative and record-keeping arm of the city of 
Medel?n; the Fiscal?a General de la Naci?n, the national public prosecutor?s office; the 
Departamento Administrativo Nacional de Estad?stica (DANE), the Colombian census 
bureau; the Polic?a Nacional, the national police force, which has subunits in the 
country?s departments and cities; and the Instituto Nacional de Medicina Legal y 
Ciencias Forenses, the National Institute of Forensic Medicine and Sciences (most 
commonly caled simply Medicina Legal), which among other things operates the 
country?s morgues. The Secretar?a de Gobierno and the Fiscal?a build their data sets from 
the same source, public denunciations and prosecutions of homicide crimes, and have 
figures for Medel?n for al years in the study period (both for homicides and homicides 
per 100,000 inhabitants); neither agency releases data disaggregated by bario. DANE 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 312 
derives its data from multiple governmental sources (primarily Medicina Legal) and it 
has national-level data through 2006 and city-level data betwen 1990 and 2006. The 
Polic?a Nacional?s dataset is not of homicides but of homicide reports, data from which 
are available electronicaly from January 1999 through May 2009; it does not make city-
level data available at the bario level, but does release data by comuna, or ward (the four 
barios that make up Caicedo La Siera are part of Comuna 8). Medicina Legal?s data are 
based on forensic analyses of corpses that enter the country?s morgues, and data for 
Medel?n are available betwen 1988 and the present, disaggregated by date, time of day, 
and stret addres of death, making it possible to derive bario-level data for almost the 
entire study period.
2
 
The police reports tend to show the highest level of homicides relative to the other 
measures (presumably because more homicides are reported than bodies are found or 
murders prosecuted), but al of these homicide measures have correlation coeficients 
                                                
2
 In searching for data, I prioritized obtaining city-level over national-level data. My attempts to get 
national-level data from the Fiscal?a, DANE, Medicina Legal, and the Polic?a Nacional had not suceded 
by the time this thesis was submited. Bario-level data from Medicina Legal were derived by Maria 
Victoria Llorente of the Centro de Estudios sobre Desarolo Econ?mico (CEDE, the Center for Economic 
Development Studies) of the Universidad de los Andes in 206 and are used here with her permision (I 
derived the bario-level data for 207 and 208). 
Table 8-1. Correlation Coefficients among Five Measures of Homicides 
 Homicide 
Rate 
City 
Records 
Forensic 
Reports 
Police 
Reports 
Census 
Data 
Homicide Rate 1.000 ? ? ? ? 
City Records 0.983 1.000 ? ? ? 
Forensic Reports 0.988 0.992 1.000 ? ? 
Police Reports 0.998 0.998 0.995 1.000 ? 
Census Data 0.994 0.998 0.993 0.998 1.000 
 
Source: Author?s calculations based on data from Secretar?a de Gobierno (homicide 
rate and city records), Instituto Nacional de Medicina Legal y Ciencias Forenses 
(forensic reports), Polic?a Nacional (police reports), and Departamento Administrativo 
Nacional de Estad?stica (DANE, census data). 
 
 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 313 
exceding 99 percent, except the homicide rate, whose correlations are only slightly 
lower (se Table 8-1, p. 312), and the chart of the figures for Medel?n show clear 
agrement in the annual trends in data (se Figure 8-1). The data show clear peaks 
in homicides in 1991 and 2002, two years in which important urban wars were at turning 
points, and a valey in 1998, when the national paramilitary federation decided to go to 
war against guerilas and leftist militias in Medel?n, leading to the unsteady increase that 
reached its peak four years later. The figures are a bit contradictory across sources in the 
2005-2007 period, but it is clear that after 2005 something was changing in the city (?Don 
Figure 8-1. Five Measures of Homicides in Medel?n, 1984-2008 
 
 
homicide rate (per 
100,000) 
city records 
 
forensic 
reports 
census data 
 
police reports 
Sources: Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden 
Civil, Unidad de Convivencia Ciudadana (homicide rate and city records); Instituto 
Nacional de Medicina Legal y Ciencias Forenses (forensic reports); Departamento 
Administrativo Nacional de Estad?stica (DANE, census data); and Polic?a 
Metropolitana del Vale de Aburr? de la Polic?a Nacional (police reports). 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 314 
Berna? Murilo, the country?s most important organized-crime boss, who dominated 
Medel?n?s underworld as wel, was arested that year), and that beginning in 2007 
homicides were undeniably moving upward again, signaling at least a partial end to the 
Medel?n Miracle. 
To get a beter sense of what events might have driven these trends it is 
instructive to look at the monthly homicide data that are available (se Figure 8-2). 
Whereas the annual data show 1991 as the peak year for homicides in the city, the month 
with the highest number of murders was actualy December 1992, a few months after 
Figure 8-2. Homicides in Medel?n, by Month, 1987-2009 
 
Sources: Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden Civil, 
Unidad de Convivencia Ciudadana (January 1987 to December 205); Polic?a 
Metropolitana del Vale de Aburr? de la Polic?a Nacional, adjusted to 75 percent of their 
reported value (January 206 to January 209). Because data from the Secretar?a de 
Gobierno average 75 percent of the value of the data from the Polic?a Nacional during 
the years in which data are available from both (1999 to 2005), the data from 206 to 
2009 have been adjusted for greater visual acuracy. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 315 
Pablo Escobar, then the nation?s top drug boss, escaped from prison and waged war 
against the state from his home base in Medel?n, and a month after the national 
government declared a state of emergency due to the increase in violence both from drug 
gangs and leftist guerilas, including those operating in Medel?n. In 2002 the most 
violent months were June and August, with 360 and 359 homicides reported, 
respectively, a period in which the last of the guerilas and leftist militias were on the 
verge of being defeated. With the monthly data, it is much les obvious what events were 
more closely asociated with the valeys, since there were no dramatic downward spikes 
indicating some qualitative change, only fluctuations. The data suggest that the 1998 to 
2002 period ? the war betwen paramilitaries and guerilas ? began in October 1997, 
but there is no historical reason for that particular month to stand out as a low point in 
murders, aside from the trivial observation that a few months later the war began. 
(Likewise in September 2000: homicides suddenly but temporarily declined to just below 
that of October 1997, a brief period of relative peace amid war.) The same can be said for 
the most recent period: homicides were clearly faling until around 2005, when they 
fluctuated before beginning to rise again in 2007: the low point was May 2007, but again 
there is no reason that this particular month should be the most peaceful of the period. A 
possible explanation is that, because there were dozens of micro-conflicts taking place 
throughout the city throughout the period, and each micro-conflict had alternating periods 
of intensity and relative peace, there probably were some months when, simply at 
random, a number of micro-conflicts were going through periods of relative peace at 
once. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 316 
 A comparison of homicide trends in Medel?n with those at diferent levels of 
analysis demonstrates that possibility and the more general observation that 
disaggregated data paint more acurate pictures of lower-level dynamics (se Figure 8-3). 
Before 1990, annual homicide figures were rising in Colombia nationwide and in 
Medel?n, but they were rising as wel in Caicedo La Siera and in the two most violent of 
Figure 8-3. Comparison of Homicides acros Levels of Analysis 
 
 
Four barios of 
Caicedo La Siera: 
 
Las Estancias 
Vila Liliam 
 
La Siera 
Vila Turbay 
Sources: Instituto Nacional de Medicina Legal y Ciencias Forenses (four barios, 
Caicedo La Siera, and Medel?n); Departamento Administrativo Nacional de 
Estad?stica (DANE; Colombia). 
Colombia 
Medel?n 
 
Caicedo 
La Siera 
Four 
Barios 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 317 
the four barios that make up Caicedo La Siera, Las Estancias and Vila Liliam, which 
are both more populous and closer to the city center than the other two barios. Likewise, 
al four levels of analysis (national, city, sector, and bario) show clear downward trends 
in homicides beginning in 2002 as wel. The period betwen 1990 and 2002, however, 
shows more disagrement across levels: Medel?n?s homicide rate declined steadily 
beginning in 1991 until its ascent began in 1998, whereas Colombia?s declined only until 
1995 when it fluctuated for a few years before beginning its relentles ascent in 1997. 
While Medel?n and Colombia both experienced two major spikes in violence during the 
study period, Caicedo La Siera experienced four, in 1990, 1994, 1999, and 2002, 
representing the militia-gang wars, the militia-militia wars, the militia-paramilitary war, 
and the paramilitary-paramilitary war, respectively (se Chapter 5 for a les simplistic 
characterization). Consequently, the overal paterns in the top two levels of Figure 8-3 
look very diferent from those in the bottom two levels. This largely reflects the fact that 
homicides in Caicedo La Siera never made up more than 2 percent of the homicides in 
the city as a whole, so its dynamics stand out, whereas Medel?n?s share of Colombia?s 
homicides was much higher, averaging about 12 percent. But Medel?n?s homicides 
started out in 1990 with a 21 percent share of Colombia?s, faling to an average of about 
14 percent betwen 1992 and 1998, to 10 or 11 percent betwen 1998 and 2002, and 
down to 4 percent beginning in 2004 (the share of homicides is uncertain beginning in 
2007), which partly explains the divergence in the visual patern betwen Medel?n and 
Colombia over time. While Medel?n and Colombia both experienced spikes in homicides 
in 2002, for example, Medel?n had 42 percent fewer homicides in 2002 than it had in 
1991, whereas Colombia actualy had a few thousand more in 2002 than in 1991, 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 318 
suggesting that while Medel?n?s homicide problem was bad, it was worse in other parts 
of the country by that point in the city?s history ? and within a few years, it would 
become one of the least deadly of Colombia?s major cities. 
Before moving on to the other forms of violence in Medel?n, some historical 
context is in order, to demonstrate why 1984 was chosen as the beginning of the study 
period (se Figure 8-4, p. 319). That year, the homicide rate in Medel?n was lower than it 
had been during the late 1960s and early 1970s, when smugglers of cars, domestic 
appliances, gems, and other non-narcotic goods had begun making more money 
smuggling marijuana and cocaine (and therefore fighting to control the trade). But it was 
higher in 1984 than in the late 1970s, when the Medel?n Cartel had already consolidated 
control over the city?s ilicit economy and some of its leaders were trying to achieve 
legitimacy as businesmen and political leaders. While the upward trend in homicides 
began in 1976, it was betwen 1984 and 1985 ? the year the cartel?s leaders went into 
hiding and launched a war against the state ? that the rapid and relentles climb in 
deadly violence realy began in Medel?n. The frameworks and terms used throughout 
this work are a social scientist?s atempt to impose order upon chaos. The reality of 
collective and interpersonal-communal violence in Medel?n is too mesy to simply 
categorize diferent groups acording to some academicaly aceptable standard. The 
groups are not static, they are not designed; they emerge from individual decisions and 
path-dependent facts about the constrained social world their members inhabit. They 
might be a group of people one day, and crime ring for a month, then just a bunch of 
potheads again; one or two of them ight take a contract to asasinate someone, then 
use the money for a party with his boys; they might have an agrement with an outside 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 319 
group that provides them oney or weapons in exchange for inteligence, but then turn 
around and betray that group as soon as they look weak enough that they can get away 
with it. Then the charismatic leader could overdose or get a girlfriend, and everyone just 
drifts apart, maybe joining other gangs, maybe even fighting each other in the future. It?s 
happened that way a thousand times. 
8.3. Self-Inflicted Violence 
Reliable data on self-inflicted violence in Medel?n and at the national and bario 
(neighborhood) levels is dificult to come by beyond the aggregate number of suicides 
per year, especialy before 2000, but even after that date it is dificult to find even 
qualitative data about trends in self-inflicted psychological violence, such as levels of 
Figure 8-4. Homicide Rate in Medell?n, 1967-2008 
 
Source: Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden Civil, 
Unidad de Convivencia Ciudadana. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 320 
substance abuse. For example, annual data for the number of people commiting suicide 
in Medel?n during the period 1992-2000 from two diference sources (DANE and 
Medicina Legal) show some similarities (e.g. 1992, 1994, and 1999 difer by fewer than 5 
cases), but they difer significantly in the middle of the period: betwen 1994 and 1995, 
suicides fel from 91 to 84 acording to DANE but rose from 96 to 125 acording to 
Medicina Legal.
3
 The correlation betwen them for the entire period is just 39 percent, 
but if 2000 is excluded (when Medicina Legal?s figure is les than half of DANE?s, 63 to 
137, suggesting a major discrepancy in measurement), it is a much more respectable 85 
percent; stil, because data are not available from both sources for most years in the study  
                                                
3
 DANE data from DANE; Medicina Legal data from Rub?n Dar?o Manrique Hern?ndez, ?Caracter?sticas 
del Suicidio al Final del Siglo X en Medell?n,? [?Suicide Characteristics at the End of the 20th Century in 
Medel?n,?] Revista CES Medicina 14, no. 1 (200): 53. 
Figure 8-5. Suicides in Medel?n, 190-2006 
 
 forensic reports 
census data 
 male suicides 
female suicides 
 
Sources: Instituto Nacional de Medicina Legal y Ciencias Forenses (forensic reports); 
Departamento Administrativo Nacional de Estad?stica (DANE, census data, male 
suicides, and female suicides). 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 321 
 period, not much can be concluded from these data. 
Nevertheles, some general observations can be made about the overal trends in 
physical self-inflicted violence. First, acording to both of the above data sources and 
other available acounts, suicides fluctuated during the 1990s, with significant declines in 
1993 and 1997 but with dramatic increases in 1998 and 1999.
4
 Second, in the early 
2000s, suicides either fluctuated or fel slightly, remaining near the same high level they 
had risen to in 1999.
5
 Finaly, the increase in suicides in the late 1990s is atributable 
almost entirely to an increase in suicide among males; suicide among females fluctuated 
                                                
4
 Ibid. 
5
 Carlos Palacio-Acosta, et al., ?Characteristics of People Comiting Suicide in Medel?n, Colombia,? 
Revista de Salud P?blica 7, no. 3 (205): 243. 
Figure 8-6. Trends in Suicides versus Homicides 
 
 homicides  suicides  
Sources: Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden Civil, 
Unidad de Convivencia Ciudadana (homicides, 1986-2008; figures for 1984-1985 
estimated by author from rate data and population); Departamento Administrativo 
Nacional de Estad?stica (DANE: suicides). 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 322 
during the 1990s, miroring that of males, but stabilized beginning in 1998, remaining flat 
while male suicides skyrocketed (se Figure 8-5, p. 320). 
An interesting observation, given the data that are available and asuming for the 
moment that they are reliable, is what sems to have been inverse relationship betwen 
suicides and homicides: during the periods in which homicides were rising, suicides 
showed an overal negative trend, whereas those when homicides were faling, suicides 
had a positive trend. The correlation coeficient betwen suicides and homicides ranges 
from -0.64 for the Colombian census bureau?s homicide figures to -0.56 for the Secretar?a 
de Gobierno?s homicide figures. Data are likely to be more reliable in the future, as the 
city began a wide-ranging project during the administration of Sergio Fajardo to collect 
data on hundreds of indicators for the quality of life. One component of that system is the 
Medel?n C?mo Vamos (MCV: How We?re Doing in Medel?n) project, and in MCV?s 
analysis of suicides during the 2004-2007 period, when the homicide rate was 
plummeting, it found suicides rising from 4.2 people per 100,000 residents in 2004, to 5.4 
in 2005, and 6.3 in 2006, faling again in 2007 to 5.5, an overal increase of 30.9 percent 
during a period when the homicide rate fel 19.9 percent.
6
 While these figures are 
inconsistent with DANE?s figures, which found a slight drop in suicides betwen 2005 
and 2006, the overal patern of suicides moving contrary to homicides nevertheles 
generaly holds. The question of what explains this contrary motion is left to future 
research; an initial search for an explanation came up dry, and so this study has had to 
focus its explanatory eforts on interpersonal-communal and collective forms of violence. 
                                                
6
 C?mo Vamos en Seguridad: Informe de Indicadores Objetivos?Seguridad, 2004-2007 [How We?re 
Doing on Security: Objective Indicators Report?Security, 2004-2007] (Medel?n, Colombia: Medel?n 
C?mo Vamos, 208), htp:/ww.medelincomovamos.org. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 323 
While self-inflicted physical violence is reasonably wel documented, 
psychological variants are not. Two additional areas for potential future research, 
therefore, merit brief mention: trends in substance abuse, (data might be garnered from 
staf at treatment centers), and trends in risky behavior (such as the ?kamikaze kids,? 
Medel?n?s equivalent of suicide bombers: young males who joined hit squads fully 
expecting to die during an asasination atempt). Both areas would provide potentialy 
useful information about paterns of self-inflicted psychological violence. 
8.4. Interpersonal Violence 
To adequately describe paterns of interpersonal violence ? violence involving 
one or two perpetrators (who are not part of a larger group) acting against one or more 
victims whom they may or may not know ? would require reliable data on one form of 
violence that is notoriously undereported, intrafamilial, and on another that is generaly 
wel tracked, communal. I addres these in turn. 
8.4.1. Intrafamilial Violence 
Very litle reliable data are available regarding paterns of violence taking place 
inside the home or among close relatives or acquaintances, including spousal abuse, child 
abuse, elder abuse, pedophilia, date rape, etc. The data that are available are often 
contradictory, represent only a snapshot in time (i.e. no time series), or provide 
inadequate information about the magnitude of the problem. For example, in the 
department of Antioquia it was found in 1995 that about 95 percent of victims of abuse 
within the home were women or girls, and in 1998 that about 93 percent of victims of 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 324 
spousal abuse were wives, but the study citing these figures provided neither magnitudes 
nor trends.
7
 A public-health survey in Medel?n in 2003 and 2004, however, found that 
men and women were about equaly likely to be either a victim or a perpetrator of 
psychological violence or of physical violence that causes no serious injury, although 
women were about twice as likely as men to be victims when the physical violence 
caused an injury. The conclusion of that study was therefore not that some men were 
violent against some women and others weren?t, but that some couples were violent to 
each other while others weren?t.
8
 (The author of that study, initialy puzzled by these 
results, said he then consulted experts on the mater in other countries and it turned out 
that these counterintuitive results were, in fact, consistent with findings on intrafamilial 
violence in many parts of the world.
9
) The two studies ? the one finding mostly female 
victims, the other finding a rough equality ? were done at diferent time periods and at 
diferent levels of analysis, but those diferences alone are unlikely to acount for such a 
large discrepancy in the findings; more likely they used diferent methods or 
asumptions, which makes it very dificult to say anything productive about the main 
paterns of intrafamilial violence. For that reason, this study did not include interpersonal-
intrafamilial violence as part of the dependent variable violence in the explanation of 
paterns. The remainder of this section, therefore, focuses mainly on describing, rather 
than explaining, what data could be found. 
                                                
7
 Rafael Rinc?n P., Antioquia, Fin de Milenio: ?Terminar? la Crisis del Derecho Humanitario?, 249. 
8
 Luis Fernando Duque, La Violencia en el Vale de Abur?: Caminos Para la Superaci?n, 52. 
9
 Luis Fernando Duque Ram?rez, personal comunication (Interview No. 19), 13 May 209. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 325 
The public-health survey on violence in Medel?n and the Aburr? Valey found 
that 25.7 percent of people in Medel?n had been victims and 24.4 percent had been 
perpetrators of psychological violence with their partner during the previous year, while 
9.0 percent were victim and 7.2 percent were perpetrator of physical violence without a 
weapon, and 1.8 percent were victim and 1.4 percent were perpetrator of physical 
violence that caused injury. The lifelong figures were 15.3 percent victims and 13.1 
percent perpetrators of unarmed physical violence, 4.1 percent victims and also 4.1 
percent perpetrators of injurious physical violence, and 31.0 percent victims and 29.3 
percent perpetrators of psychological violence at some point in their lives.
10
 
(Psychological violence is not the term used in this survey, but rather verbal abuse, which 
is likely a much weaker standard than I use: most instances of verbal abuse probably 
would not cause the kind of trauma or harm that would make it count as true 
psychological violence. I leave the exercise of teasing out the violence within the broader 
category of verbal abuse to future research.) 
The percentages of children reported as having been abused by a parent are as 
follows: 6.1 percent in the past year and 8.4 percent in their lives were victims of 
unarmed physical violence, 1.3 percent in the past year and 2.5 percent in their lives were 
victims of injurious physical violence, and 6.1 percent in the past year and 7.2 percent in 
their lives were victims of psychological violence. Of adolescents aged 12 to 15 surveyed 
about mistreatment by one or both parents at some point in their lives, 63.0 percent 
reported that a parent had hit them with some object, 19.4 percent reported having been 
injured by a parent, 5.4 percent kicked or biten, 1.8 percent atacked with a firearm or 
                                                
10
 Luis Fernando Duque, La Violencia en el Vale de Abur?: Su Magnitud y Programa Para Reducirla, 
201. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 326 
knife, 1.3 percent burned, and 1.1 percent had required medical atention after having 
been atacked by a parent. In Comuna 8, betwen 8.4 and 9.4 percent of this age group 
reported having been a victim of violence by one or both parents at some point during 
their lives.
11
 
One of the dificulties of finding good data in Medel?n is that betwen 81.3 and 
98.3 percent of al incidents of physical violence not involving a weapon, betwen 97.5 
and 98.5 percent of al incidents of psychological violence, and betwen 88.8 and 
92.4 percent of al incidents of sexual violence (either atempted or succesful) were 
never reported to the authorities when the perpetrator was the victim?s intimate partner 
(e.g. spouse), parent, child, sibling, or other family member; only about five to eight 
percent of intrafamilial violence involving a weapon were reported to authorities.
12
 In the 
past, it has been dangerous to report it. There was a time when staf at a local high school 
in Caicedo La Siera would report such abuses to the authorities, an administrator told 
me, but the mothers would deny that they had been abused or that the children had been 
abused, and then acuse the school of siding with the ?enemy? ? and in a war zone, that 
acusation can be dangerous. ?You have to be very careful.?
13
 One staf member at that 
school did tel me, however, that ?there is a lot of intrafamilial violence in these barios? 
and that she had once met with city authorities to encourage them provide beter 
protections for those who reported it. She said she knew of cases where a man of the 
house would work during the wek, but after drinking until drunk on the wekends would 
                                                
11
 Ibid., 205. 
12
 Luis Fernando Duque, La Violencia en el Vale de Abur?: Caminos Para la Superaci?n, 51. 
13
 ?Alicia,? Interview No. 7. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 327 
try to rape his daughter; somebody would cal the police to report it ?but they [the police] 
never protect confidentiality ? so it?s beter not to get involved.?
14
 
A staf member at the city?s Ofice of the Family told me, however, that during 
the last ten years not only has there been a growing awarenes of the problem and of the 
rights of victims, but also the system for reporting intrafamilial violence has improved 
substantialy. As a consequence, she said, reports of intrafamilial violence have risen over 
the past decade (although she could not provide figures to support this claim).
15
 Of 
course, rising reports amid a changing reporting system cannot be considered an indicator 
of any underlying trend. So the problem remains: approximate magnitudes of 
intrafamilial violence can be identified, but how those magnitudes have changed over 
time cannot be discerned with the data I was able to obtain. 
Risk factors, however, have been identified. One study found that the risk factors 
for commiting intrafamilial physical violence that left no serious injury included 
unemployment and the perpetrator?s belief that there was a lot of violence in the 
community; risk factors for intrafamilial violence leaving injuries were the perpetrator?s 
approval of the use of violence to defend one?s comunity (including social cleansing), a 
history of having been abused, a cynical atitude about corruption in public officials, and 
frustration with his or her own personal and profesional achievements.
16
 
                                                
14
 ?Cristiana,? Interview No. 9. 
15
 Soraya Betancur, personal comunication (Interview No. 20), 18 May 209. 
16
 Luis Fernando Duque, La Violencia en el Vale de Abur?: Caminos Para la Superaci?n, 51. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 328 
8.4.2. Communal Violence 
Interpersonal-communal violence involves one or two perpetrators acting against 
one or more victims whom they may or may not know, and it may take place in 
someone?s home, on the stret, at a commercial property, or in some other public area. 
This form of violence includes most categories of common crime: robbery, asault, rape, 
etc. (Organized crime would be a form of collective violence, discussed in the next 
section.) Besides data for homicides, I was not able to get good crime data before 1999, 
and of the data that were made available since then, it is not entirely clear what trends are 
implied by them beyond the few key observations made here. First I discuss the 
observations that can be made about the homicide data, and then those regarding other 
crime data. Homicide data, as noted earlier (?? 8.1-8.2), includes not only those murders 
commited in the context of common crime or personal disputes (interpersonal violence) 
but also those commited by hit squads, militias, paramilitaries, social-cleansing groups, 
drug gangs, and other groups (collective violence). Very litle good data are available to 
tease out the relative magnitudes of communal versus collective homicides, let alone how 
those magnitudes have changed over time. 
But enough hints exist to make some reasonable general observations. One study, 
by Giraldo Ram?rez, of the period betwen 2003 and 2005 ? only one of the five periods 
I studied ? found that ?the homicide rate shows a directly proportional relationship with 
the intensity of the armed conflict,?
17
 but he did not specify what that correlation implied: 
Who was doing the kiling? Was the decline in homicides due entire to a decline in 
                                                
17
 Jorge Giraldo Ram?rez, ?Conflicto Armado Urbano y Violencia Homicida: El Caso de Medel?n,? 
[?Urban Armed Conflict and Homicidal Violence: The Case of Medel?n,?] Urvio: Revista 
Latinoamericana de Seguridad Ciudadana No. 5 (208): 99. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 329 
collective violence during that period, or did communal violence also fal as the rule of 
law as restablished once the collective violence was under control? Jurgen Brauer, 
Alejandro G?mez-Sorzano, and Sankar Sethuraman studied homicides in Colombia 
betwen 1946 and 1999 and were able to separate out the ?permanent? homicide rate 
from the ?cyclical? murders, the later being asociated with periods of unrest in the 
country. For the post-1991 period they discovered an increase in the permanent (i.e. non-
cyclical) component of homicides, which they explained by suggesting that ?the 
increasing dollar value of the drug trafic may have pushed up the slope of the permanent 
murder series ?, perhaps reflecting entangled political and economic interests.?
18
 This is 
relevant to the Medel?n case because of the very close correlation betwen homicides at 
the city and national levels during the 1990s ? the correlation coeficient is 0.85
19
 ? 
plus the fact that Medel?n was the epicenter of the drug trade in the country. But do 
Brauer et al?s findings suggest that murders asociated with common crime in Colombia 
(i.e. economicaly motivated interpersonal-communal violence) were beginning to 
displace collective-political violence, or that collective-economic violence (i.e. the 
organized drug trade) was beginning to displace colective-political violence? In other 
words, what share of the murders was communal and what was collective, and how did 
that ratio change over time? 
Beyond these general studies, several hints can be found within some of the data 
that are available. (And in some that are not available: one police-inteligence source told 
                                                
18
 Jurgen Brauer, Alejandro G?mez-Sorzano, and Sankar Sethuraman, ?Decomposing Violence: Political 
Murder in Colombia, 1946-1999,? European Journal of Political Economy 20, no. 2 (204): 447. 
19
 Author?s calculation based on homicide data for Colombia and Medel?n betwen 191 and 199 from 
the Colombian census bureau, Departamento Administrativo Nacional de Estad?stica (DANE: Department 
of National Statistics Administration). 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 330 
a pres agency that only 63 percent of the victims of homicides in 2009 were members of 
drug gangs, suggesting that a significant minority of murders were not directly related to 
the organized drug trade: at least a third were interpersonal-communal murders.
20
) One 
hint comes from a Colombia-wide study by Sa?l Franco Agudelo, who found higher rates 
of violent deaths in regions of the country with higher indicators of impunity.
21
 Another, 
looking at Medel?n specificaly, comes from a study of violence that spanned thre of the 
five periods I looked at. Marleny Cardona et al studied homicides in Medel?n betwen 
1990 and 2002, when 85.8 percent of al murders took place in barios clasified as 
?middle,? ?middle-low,? ?low,? or ?low-low? on the city?s socioeconomic scale; most 
such barios are on the city?s periphery, where most of the armed conflicts took place. In 
89.7 percent of al murders in the city, there was only one victim involved in the 
homicide event in; in 93.0 percent of those cases, nobody else was even injured. In other 
words, most of the homicides semed to be neither military casualties nor civilian 
casualties (?collateral damage?) in urban warfare, nor instances of indiscriminate 
violence due to terorism. This possibly suggests that what was happening was not 
primarily organized violence but rather a more difuse phenomenon, smal-scale atacks, 
primarily one-on-one, not necesarily centraly directed. In only 15.4 percent of cases 
could a motive for the murder be identified, but the most common motives that could be 
identified were revenge (7.0 percent of al cases) and armed robbery (4.5 percent); the 
motives underlying the rest of the cases (84.6 percent in total) were unknown, which 
                                                
20
 Agence France-Prese, ?Venganzas de ?Narcoparamilitares? Disparan Homicidios en Medel?n.? 
21
 Sa?l Franco Agudelo, El Quinto, No Matar: Contextos Explicativos de la Violencia en Colombia, cited in 
Jaime Arturo G?mez Corea, et al., ?Estado del Conocimiento Sobre la Violencia Urbana en Antioquia en 
la D?cada de los Noventa,? 175. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 331 
makes it dificult to identify the most important causes of homicides more generaly. 
(Unfortunately, the Cardona et al study also did not try to tease out trends over time. How 
did the ratios of single-victim versus multiple-victim homicides, or of interpersonal 
versus collective motives, difer betwen periods? The authors claimed that the 
?characteristics of homicides in Medel?n have remained unchanged since the 1980s,? but 
did not specify what, exactly, that meant: Did the relative proportions of motives for 
murder not change, or just the characteristics of the victims, types of weapons used, 
etc.?)
22
 
While by no means conclusive, these hints are strongly suggestive that much of 
the violence in Medel?n was asociated not only with a war among competing factions 
but also with a breakdown in the rule of law, a dynamic in which political, social, or 
religious authorities were incapable of preventing violence, punishing violence, or 
instiling norms against violence during the study period. This is consistent, for example, 
with the work of Francisco E. Thoumi, who, in his analysis of the necesary, sufficient, 
and contributing conditions involved in the emergence of narcotraficking, argued that 
Colombia, unlike other places with similar economic, demographic, and political 
characteristics, has a culture and a social order that instils few significant self-generated 
behavioral controls: ?the rules of law are weak, social capital is poor, and solidarity and 
trust are lacking? ? al elements, he argued, that would be necesary for such a violent 
industry to emerge.
23
 Those elements are present in Medel?n, distiled into their very 
                                                
22
 Marleny Cardona, et al., ?Homicidios en Medel?n, Colombia, Entre 1990 y 2002: Actores, M?viles y 
Circunstancias.? 
23
 Francisco E. Thoumi, ?Introduction?; se also Francisco E. Thoumi, ?The Colombian Competitive 
Advantage in Ilegal Drugs: The Role of Policies and Institutional Changes.? 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 332 
esence, as Duque observed: ?In the culture of ilegality, the feature most asociated with 
severe violence is that norm of Antioquian culture: ?Make money honorably, m? boy, but 
if that?s not possible, make money.?
24
 In his public-health studies, Duque found that 
among the strongest risk factors for an individual?s engaging in most forms of violence in 
Medel?n was a wilingnes to break certain legal, moral, or political norms, including 
those against the use of violence, and that in Medel?n a very high percentage of the 
population were so wiling: 75.2 percent approved of the use of ?extreme? violence (that 
is, murder) to defend one?s family or to get some economic or political benefit, 
44.9 percent approved of the use of extreme violence to defend one?s community, and 
51.7 percent approved of earning money by ilegal means. In East Central Medel?n, the 
zone where Comuna 8 and therefore Caicedo La Siera is, 15.6 percent approved of 
murder for familial self-defense or political or economic benefit, 17.1 percent approved 
of murder for community self-defense, and 17.6 percent approved of earning money 
ilegaly (although these figures probably do not reliably reflect the views of those in 
Caicedo La Siera, since the East Central zone includes the downtown district, which has 
diferent dynamics from the city?s residential zones and tends therefore to bias data that 
are aggregated at the zone level).
25
 When asked about the conditions under which the use 
of violence was legitimate, 36 percent of respondents in Medel?n said that violence is 
                                                
24
 ?En la cultura de la ilegalidad, el rasgo m?s asociado a la violencia severa es esta norma de la cultura 
antioque?a: ?consiga plata, mijo, honradamente, pero si no es posible, consiga plata.?? Luis Fernando 
Duque, La Violencia en el Vale de Abur?: Caminos Para la Superaci?n, 50. This same observation is 
made, and a variant of this adage is quoted, in Alonso Salazar J., No Nacimos Pa? Semila, 196. 
25
 Luis Fernando Duque, La Violencia en el Vale de Abur?: Caminos Para la Superaci?n, 90. It is od 
that the authors of that study decided to ask, in a single question, about the use of violence to defend one?s 
family ?or? to get some economic or political benefit: it would have ben far more informative to separate 
out those two isues, as more people would likely be wiling to kil in familial self-defense than to kil 
simply to make money. This is even oder considering that the question about comunity self-defense did 
not have the ?or? clause atached. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 333 
legitimately used for educational purposes (as either an ?eye for an eye? response to 
aggresion or as corporal punishment for children), 29 percent considered violence 
legitimate to protect one?s family or society (including kiling home invaders, kiling in 
self-defense, torturing to extract information, and kiling as capital punishment), and 
23 percent considered violence and intimidation to be legitimate ways to deter atacks 
from others or to expres anger toward others. The internal barier to violence, in other 
words, was already low in Medel?n, so wherever or whenever external controls ceased to 
function, one would expect violence to rise there. 
In short, what semed to be the case in Medel?n, acording to these studies and to 
substantial anecdotal evidence, was that during the study period there was a base level of 
interpersonal-communal homicides (the ?permanent? murder rate of the Brauer et al 
study of Colombia), which is sort of the background noise of crime in any place in the 
world and which was probably higher in Medel?n than elsewhere due to the structure of 
norms regarding legality and ilegality. But during periods when conflicts flared up in the 
city (or to be more precise, in micro-teritories within the city), the ?cyclical? murders, 
the temporary increases in homicides, had two components: collective, deriving directly 
from the conflicts; and interpersonal-communal, deriving indirectly from the conflicts. 
The direct component derived from military or terorist actions, as two or more groups 
vied for control (over teritory or resources) and engaged in collective violence against 
each other and each other?s supporters. The indirect component derived from both the 
private exploitation of the larger conflict (as civilians falsely denounced their private 
adversaries as agents of the enemy, as a way to have them kiled or displaced by one of 
the fighting groups) and from a further breakdown in the rule of law, as the dominant 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 334 
 power shifted resources and atention to defensive and offensive operations and away 
from maintaining general order in their micro-teritories. That inatention to order created 
an atmosphere of impunity (or perhaps further impunity) for common crime, enabling 
and exacerbating existing cultural tendencies to solve private conflicts through violence. 
Thus was the rise of collective homicide acompanied by a rise in interpersonal-
communal homicide as wel. 
Given these studies, the data, and anecdotal evidence, we can say with reasonable 
certainty that interpersonal-communal murders generaly rose and fel with collective 
murders over al five study periods, but what about forms of violence other than 
Figure 8-7. Trends in Comon-Crime Reports versus Homicides 
 
 homicides 
kidnaping reports 
batery reports 
 robery reports 
extortion reports 
land-piracy reports 
 
Sources: Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden Civil, 
Unidad de Convivencia Ciudadana (homicides, 1986-2008; figures for 1984-1985 
estimated by author from rate data and population);  Polic?a Metropolitana del Vale de 
Abur? (kidnaping, batery, robery, extortion, and land-piracy reports, 199-2008). 
Kidnaping reports are hard to se on this chart, but they peaked at 58 in 200 and fel 
until 2003, then plummeted to single digits. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 335 
homicide? It is clear that during the first study period, 1984-1992, al forms of violence 
rose together; I have found no source claiming otherwise and many sources confirming 
that observation (se Chapter 3). Evidence is mixed for the second study period, 1992-
1998, when homicides were generaly faling, but given the extremely rapid increase in 
al forms of violence during the first period, it sems only logical that they had to have 
falen, or at least to have fluctuated, during the second period; litle more can be said 
about it beyond that. For the third study period we have crime data to show that, while 
homicides were rising, other forms of violence were generaly faling ? or at least 
reports of other forms of violence were faling, which indicates either that the underlying 
violence was actualy faling or that the violence was fluctuating or increasing but a 
climate of fear, a lack of trust in authorities, or some other factor increasingly prevented 
people from reporting those crimes. In the fourth study period, as the homicide rate began 
to plummet, crime reports either rose slightly or their declines slowed in 2003, but after 
that they generaly continued to fal through 2007. Almost al forms of violence, 
however, began to rise again in 2008, the beginning of the fifth study period. In short, in 
thre of the five, and probably in four of the five, study periods, non-fatal forms of 
violence, in both their physical and psychological manifestations, tended to rise and fal 
with homicides, or, to be more precise, the wave of homicides that peaked in 1991 was 
acompanied by a wide range of non-fatal forms of violence, while the wave that peaked 
in 2002 was primarily a homicide phenomenon (se Figure 8-7, p. 334). 
The Medel?n C?mo Vamos surveys for 2006, 2007, and 2008 found that betwen 
9 and 11 percent of residents had been the victim of some kind of crime in the previous 
12 months; 64 percent of those were stret robberies or atacks, although it is not 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 336 
specified whether these were violent crimes or, for example, pickpocketing. 
(Pickpocketing is extremely common: almost everyone, it sems, has had a cel phone or 
walet snatched in Medel?n, including both my wife and me. But it is not a form of 
violence, and so its presence in crime data makes that data useles as an indicator of 
violence.) The trend from 2006 to 2008 was a step decline in reporting of the crime to 
authorities: In the 2006 survey, 52 percent of respondents who said they had been the 
victim of a crime in the previous year said that they had reported that crime to the police; 
in 2007, that figure fel to 43 percent, and in 2008 it was down to 36 percent. In 2006 and 
2007, the most common reason given for not having reported the crime was either not 
having evidence or not knowing whom to report it to; in 2008, the main reason given was 
not trusting the authorities. Given that 2007 was the year homicides and other forms of 
crime began to rise in the city, these findings suggest that we should be cautious with the 
conclusions we draw from any of the crime data based on voluntary reporting, as any 
decline in crime reports might be was due more to a declining trust in authorities than to 
an underlying decline in crime. Stil, this caution is offered based on a time series of only 
thre years, during a period when it was not entirely clear which direction homicide and 
crime rates were realy heading.
26
 
Looking in a litle more detail one finds that the two main indicators of physical 
interpersonal-communal violence (batery reports and kidnapping reports) are strongly 
positively correlated with four diferent measures of homicides, at least during the 
periods when the crime data were available to measure the correlations (1999 to 2008). 
                                                
26
 Encuesta de Percepci?n Ciudadana 208 [Public Perception Survey 208] (Medel?n, Colombia: 
Medel?n C?mo Vamos, 208), htp:/ww.medelincomovamos.org. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 337 
Since the crime data on batery and kidnapping are based on voluntary reporting of 
incidents, one would expect that any external factor that afected wilingnes to report ? 
fear of retaliation, trust in authorities, etc. ? would have afected all of the crime-report 
data, including homicides, and so one would expect those data to be highly corelated 
with each other. For that reason, it is important to cross-check crime-report data against 
not only homicide reports but also against homicide data from sources that do not depend 
on voluntary reporting (or do not depend entirely on voluntary reporting; se ? 8.1 for a 
discussion of the sources of homicide data). Doing that, one stil finds strong corelations: 
batery reports (physical atacks that leave injuries) have correlation coeficients with 
four diferent measures of homicides ranging from 0.85 to 0.90, and kidnapping reports 
(a form of physical violence, since it involves physical contact) have correlations with the 
homicide measures of betwen 0.81 and 0.87.
27
 This provides evidence in favor of the 
observation that several diferent physical forms of violence generaly moved together 
(se Table 8-2). 
                                                
27
 Se source note in Table 8-2. 
Table 8-2. Crime Report Corelations with Diferent Homicide Measures 
 Homicide: 
Crime: 
City 
Records 
Forensic 
Reports 
Police 
Reports 
Census 
Data 
Batery 0.8 0.89 0.90 0.85 
Kidnaping 0.85 0.85 0.87 0.81 
Extortion 0.74 0.76 0.76 0.67 
Land Piracy 0.65 0.67 0.6 0.5 
Robery ?0.52 ?0.55 ?0.53 ?0.63 
 
Source: Author?s calculations based on data from Secretar?a de Gobierno (city records 
of homicides, 1986-2008), Instituto Nacional de Medicina Legal y Ciencias Forenses 
(forensic reports of homicides, 198-2008), Departamento Administrativo Nacional de 
Estad?stica (DANE, census data on homicides, 190-2006), and Polic?a Nacional 
(police reports of homicides, battery, kidnapping, extortion, land piracy, and robery, 
1999-2008). 
 
 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 338 
Reliable indicators of psychological interpersonal-communal violence are harder 
to come by. In terms of the absolute and relative magnitudes of physical versus 
psychological forms of interpersonal-communal violence, Duque?s surveys from 2003 
and 2004 found that 24.7 percent of survey respondents in Medel?n had been the victim 
of a threat (interpersonal-communal psychological violence) and 15.1 percent had been 
the victim of a physical atack without a weapon (interpersonal-communal physical 
violence) during the previous year. But this says litle about how those magnitudes 
changed over time.
28
 Two of the thre available crime-data indicators for psychological 
violence have positive correlations with homicides, although somewhat les strong than 
the physical forms have: extortion (mostly the notorious vacuna system, or protection 
rackets) has correlations ranging from 0.67 to 0.76, while land piracy (usualy, armed 
robbery of delivery vehicles) ranges from 0.55 to 0.67. The exception is robbery, which 
has a negative correlation with al of the homicide data, ranging from ?0.52 to ?0.63.
29
 
But none of these psychological indicators can be considered indicators for interpersonal-
communal violence: vacunas and kidnappings tend to be perpetrated primarily by 
organized criminals and so would more appropriately be considered instances of 
collective rather than communal violence, while the robbery data include pickpocketings 
along with armed and otherwise coercive robberies, and so that figure, dominated by 
nonviolent crimes, cannot be considered an indicator of violence. 
In short, the crime data and other evidence suggest that physical communal 
violence, psychological collective violence, and both collective and communal homicides 
                                                
28
 Luis Fernando Duque, La Violencia en el Vale de Abur?: Su Magnitud y Programa Para Reducirla, 
188. 
29
 Se source note in Table 8-2. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 339 
al rose and fel together, at least during the last thre study periods, and probably at least 
during the first study period as wel. It is probably the case that psychological communal 
violence (i.e. interpersonal threats) also rose and fel with these other forms, but the 
evidence is not conclusive. What al of this implies is that the spikes in homicides 
experienced in the city were generaly acompanied by spikes in other forms of physical 
and psychological violence in both their communal and collective forms: while the peaks 
and valeys in these waves of violence do not match up perfectly, one may nevertheles 
speak coherently of a ?rise in violence? and a ?decline in violence? without fear of 
contradicting too much of the available evidence. 
The most significant observation that emerges from this analysis is that most of 
the violence throughout the study period was interpersonal-communal, even during the 
peaks when the highest-profile forms of violence were collective-political (e.g. police 
masacres of young people in peripheral barios, asasinations of Patriotic Union party 
members, wars betwen militias and paras), collective-social (e.g. stret-gang turf wars, 
?social cleansing? death squads), and collective-economic (e.g. organized crime, drug 
wars). The spikes in collective violence created a context in which the rule of law as too 
weak to control interpersonal-communal violence, and those who engaged in the later 
often used that broader context to their own advantage, either by recruiting or deceiving 
collective-violence organizations to cary out personal vendetas, or simply by taking 
advantage of the environment of impunity. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 340 
8.5. Collective Violence 
The terminology to describe the types of violent groups that have emerged in 
Medel?n is as varied and as dificult to define as the groups are themselves: some people 
use banda, combo, gallada, and pandila more or les interchangeably, meaning street 
gang; others use both combo and oficina to refer to a mafia cel or other smal organized-
crime structure; stil others use bandas and combos interchangeably, to refer either to 
stret gangs or mafia cels; most claim a combo is a gang that is les organized than a 
banda, while some claim just the opposite. The terminology, however, maters les than 
the observation that the ilicit landscape in Medel?n was complex and not easily 
described, despite the false order that academics and historians have atempted to impose 
upon it. 
In this study, the false order I atempted to impose upon it was as follows: I used 
the terms gang or crew to describe a nonstate group of people who use violence or 
commit crimes. A crew or stret gang is a group of people who asociate primarily for 
social (e.g. identity) reasons and are les formaly structured and have les strict 
membership requirements than gangs; most people who use the term combo mean it as a 
synonym for crew. A gang tends to have a more formal structure (which does not 
necesarily imply it was hierarchical) and more stringent membership requirements, and 
its members tend to asociate for both social and economic reasons or for primarily 
economic reasons. It might be asociated with organized crime (organized-crime gang), 
narcotics traficking (drug gang), or asasination-for-hire (asasin gang or hit squad), or 
it might be afiliated with nobody else (criminal gang). Most people who use the term 
banda mean it as a synonym for gang. I use the term militia to describe a nonstate group 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 341 
either with political aspirations or whose stated goal is to maintain order and protect 
communities; the term is most often asociated with leftist politics in Medel?n, but there 
have been right-wing paramilitary militias as wel, not to mention non-political 
community self-defense or civil-defense groups (variously caled autodefensas, 
encapuchados, capuchos, or vigilantes). I use the term paramilitary or para to describe 
those militias ? sometimes caled bloques or blocs ? that were afiliated with the 
national paramilitary asociation, the United Self-Defense Forces of Colombia (AUC: 
Autodefensas Unidas de Colombia), or one of its afiliates. 
This categorization, as loose as it is, does not do justice to the fluid nature of ilicit 
activity and collective violence in Medel?n over the past 25 years. Some gangs and crews 
evolved into militias in the late 1980s (se Chapter 3). Most militias devolved into gangs 
in the early 1990s (se Chapter 4). The paras at the end of the 1990s came in two flavors: 
those with purely political aspirations (political paras) and those with links to organized 
crime and narcotraficking (narco paras, criminal paras, or simply gangs) (se Chapter 5). 
After the paramilitary demobilizations of the early 2000s (se Chapter 6), there were, 
realisticaly, no more truly political paras in Medel?n, only criminal gangs formerly 
asociated with paras, some with a vague, and at best secondary, antisubversive ideology 
that was clearly dominated by economic interests (se Chapter 7). Indeed, many groups 
that had been formed for primarily political reasons (e.g. leftist militias, right-wing 
paramilitaries) or primarily social reasons (e.g. self-defense, vigilante, or social-cleansing 
groups) also acted in ways that were indistinguishable from the behavior of criminal 
gangs and organized crime bosses. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 342 
Stil, we can make some general observations about trends in collective violence 
in Medel?n. Of the thre types of collective violence, social violence was asociated most 
strongly with crews, stret gangs, and self-defense groups (whose goals revolved 
primarily around identity or social order); political violence, most strongly with militias 
and paras (whose goals were teritorial control and political power); and economic 
violence, most strongly with criminal gangs, organized criminals, oficinas, hit squads, 
and narco paras (whose main goal was to make money: even if they participated in social 
or political violence, they were mainly hiring themselves out to the highest bidders). 
These asociations were far stronger in the later study periods than they were early on, 
however: the militias of the 1980s and early 1990s engaged in al thre forms of violence 
(trending toward more economic violence in the end), the paras of the late 1990s and 
early 2000s were les focused on social order and social cleansing than on political and 
economic goals, and the post-para criminal gangs of today hardly care about politics or 
social order, just money. 
Teasing out the relative proportions of groups involved in social versus political 
versus economic violence is not easy, but some hints exist regarding the aggregate 
number of groups engaged in collective violence. Good numbers are hard to come by 
before the mid-1990s, but by 1995 it was estimated that there were 156 such groups 
operating in the city, a number that surely was approximate (it came from a gangster-
sponsored informal census) and that may have been either higher or lower than the 
number ten years earlier.
30
 (There were probably more gangs in the mid-1980s than the 
mid-1990s, since the late 1980s is known to have ben a period of consolidation; see 
                                                
30
 ?Esos territorios se convirtieron en zonas vedadas para el resto de la comunidad. Traspasar sus fronteras 
pod?a costar la vida.? Semana, ?Paz de verdad,? [?Real peace,?] 8 May 200. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 343 
Chapter 3.) By the end of the 1990s, a diferent source estimated, there were about 8,000 
people, most betwen the ages of 15 and 25, who were members of somewhere betwen 
180 and 220 diferent gangs involved in ?criminal activities such as asasination-for-
hire, kidnapping, extortion, bank robbery, vehicle theft, etc., whether on their own 
initiative or under contract to criminal, guerila, or self-defense organizations.?
31
 The 
number of gangs, acording to stil another source, was about 200 in 2003, 150 in 2004, 
and 100 in 2005 while the total number of ilicit armed actors in Medel?n went from 
about 7,000 in 2003 to about 4,000 in 2005.
32
 
Because these figures come from diferent sources whose methods are not 
transparent, it is impossible to know the degre to which they are comparable across time 
periods. However, these figures are not necesarily inconsistent with the qualitative 
history told in Part I. Given the qualitative and quantitative evidence available, therefore, 
it would not be too much of a stretch to cautiously suggest that, very roughly speaking, 
there were probably about 200 gangs asociated with collective violence in Medel?n 
around 1990, about 150 around 1995, about 200 around 2000, and about 100 around 
2005, figures that roughly corespond to the trends in homicides. (Other estimates of 
groups asociated with diferent forms of collective violence are not consistent with these 
figures; se the next thre sections.) Even if there is a rough correlation betwen 
homicide rates and number of gangs or gang members, what cannot be known is whether 
the gangs were a generator of homicides or the homicides were a generator of gangs. 
                                                
31
 ?? actividades criminales como el sicariato, el secuestro, la extorsi?n, el asalto a bancos, el robo de 
veh?culos, etc., bien por su iniciativa o bien porque son contratados por organizaciones criminales, guerila 
y autodefensas.? Luis Fernando Duque, Antecedentes y Evoluci?n del Programa de Convivencia 
Ciudadana de Medel?n, 20. 
32
 Jorge Giraldo Ram?rez, ?Conflicto Armado Urbano y Violencia Homicida: El Caso de Medel?n.? 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 344 
Perhaps homicides and gang formation were elements of a fedback loop driven by 
security dilemas, in which people formed gangs in self-defense out of fear of rising 
homicides, but the defensive posture was mistaken for offensive intention, triggering 
conflict and further homicides, which increased fear, which triggered further gang 
formation, and so on. But this is only speculation to be considered for future research. 
In the public-health study discussed in the previous section, Duque suggested that 
some of the risk factors for violence can be explained by the possibility that many crimes 
commited with a weapon were probably perpetrated by members of gangs, militias, or 
paramilitaries. If that is the case, then the presence of a weapon in the commision of a 
crime might be a rough indicator of collective violence, and Duque?s survey might 
therefore provide some clues regarding the approximate magnitude of collective violence 
in Medel?n. Asked about their lifelong experiences with violence, 42.9 percent of 
residents surveyed reported themselves as having been physicaly atacked with a weapon 
(although the type of weapon or the nature of the atack is not specified) and 38.0 percent 
as having been threatened with a weapon at some point in their lives. On the perpetrator?s 
side of such incidents, 6.3 percent of survey respondents said that they had threatened 
somebody with a weapon, and 3.6 percent had physicaly atacked somebody with a 
weapon. In other words, a litle more than 40 percent of the population of Medel?n have 
been victims, and a litle les than five percent have been perpetrators, of physical 
collective violence, while a litle les than 40 percent have been victims and a litle more 
than five percent have been perpetrators of psychological collective violence ? or, to 
keep it simple, about 40 percent of Medel?n residents have been victims and about five 
percent have been perpetrators of collective violence at some point in their lives. (By the 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 345 
same logic, Duque?s survey would suggest the rough magnitude of the problem of 
interpersonal-communal violence: 15.1 percent had been victims and 17.0 percent had 
been perpetrators of one or more physical atacks at some point during the year before the 
survey [physical manifestation], while 38.4 percent had been victims and 3.9 percent had 
been perpetrators of one or more ?severe? threats without a weapon at some point in their 
lives [psychological manifestation].)
33
 
The next thre sections briefly review the general trends in social, political, and 
economic violence in the city. 
8.5.1. Social Violence 
Collective-social violence is perpetrated by one or more members of a social, 
ethnic, linguistic, religious, or other such group formed for reasons of afection, common 
interest, common history, or identity, including youth gangs, stret gangs, civic 
asociations, ethnic mobs, and so on. Such violence is commited to enforce respect, 
expres pride in the in-group, expres disdain for an out-group (such as a competing gang 
or a diferent racial or ethnic group), to defend or acquire teritory for the purpose of 
social control, or to punish members for violating in-group rules, among other reasons. 
Litle data are available to quantify the problem of social violence. The only data sets I 
was able to acquire were the People?s Training Institute (IPC: Instituto Popular de 
Capacitaci?n) a research and advocacy organization whose definitions and methods were 
not at al transparent (?homicides due to social intolerance,? 1988-1997), and by 
Colombia?s medical examiner, the National Institute of Forensic Medicine and Sciences 
                                                
33
 Luis Fernando Duque, La Violencia en el Vale de Abur?: Su Magnitud y Programa Para Reducirla, 
188. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 346 
(Instituto Nacional de Medicina Legal y Ciencias Forenses) whose standards were also 
not transparent (?homicides with a social-cleansing motive,? 1998-2004). Both provided 
data for al of the department of Antioquia, so there were no disaggregated figures 
available for Medel?n. Nevertheles, taken together, those figures had a correlation 
coeficient of 0.40 with homicides in Medel?n, and a look at the chart of the two figures 
does show some common movement (se Figure 8-8, p. 347).
 34
 
The exception is for the year 1997, when homicides asociated with social 
cleansing spiked from 73 in 1996 to 204 in 1997 (removing that outlier, the correlation 
coeficient with homicides is 0.69
35
). It is hard to know what to make of the outlier, 
especialy given the magnitude reported by Medicina Legal the following year. Either 
something happened somewhere in Antioquia in 1997 that pushed the social-cleansing 
figure to astronomical heights, or there was some kind of measurement bias in the IPC 
study. In Colombia, many advocacy organizations that collect data on violence against 
the populations they represent have a tendency to report that the year leading up to the 
publication of any given report, and especialy their annual reports, had been the ?worst? 
year in memory: the figures are almost always ?bad,? the trend is almost always ?worse,? 
and the methods that lead them to come to these conclusions are almost always opaque. 
IPC is usualy much more reliable than most of its peers, however, so I am hesitant to 
characterize the outlier as an anomaly of its data or the organization as having a deficit of 
integrity. The most likely explanation is that, as the Metro Bloc prepared to take over 
                                                
34
 Author?s calculations based on Base de Datos del Instituto Popular de Capacitaci?n (IPC), published in 
Guera, Paz y Derechos Humanos en Antioquia [War, Peace, and Human Rights in Antioquia] (Medel?n, 
Colombia: IPC, 198), for 198-1997, and Medicina Legal for 198-2004. 
35
 Ibid. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 347 
Medel?n in 1997, it entered surrounding vilages and some of the city?s barios and 
engaged in social-cleansing campaigns as a way of ingratiating itself to the communities 
fed up with the abuses of the militias (se ? 5.1.1). This almost certainly acounts for 
much of that spike, but 1997 sems a litle bit too early, and an almost-tripling of social 
cleansings sems a litle bit too high, for that to be the full explanation, so until beter 
data are acquired, this spike wil have to remain at least partly unexplained. 
Nevertheles, what can be observed from this data are otherwise fairly consistent 
with anecdotal evidence. Social cleansing rose during the first study period (1984-1991) 
as militias targeted the ?social undesirables? who were posing a threat to Medel?n?s 
Figure 8-8. Social Cleansing in Antioquia versus Homicides in Medel?n 
 
 homicides  social cleansing  
Sources: Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden Civil, 
Unidad de Convivencia Ciudadana (homicides, 1986-2008; figures for 1984-1985 
estimated by author from rate data and population);  Base de Datos del Instituto 
Popular de Capacitaci?n (IPC), published in Guera, Paz y Derechos Humanos en 
Antioquia [War, Peace, and Human Rights in Antioquia] (Medel?n, Colombia: IPC, 
1998) (homicides due to social intolerance, 1988-1997);  Instituto Nacional de 
Medicina Legal y Ciencias Forenses (homicides with a social-cleansing motive, 198-
2004). 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 348 
peripheral communities, then fel at the beginning of the second study period (1991-
1998) as the militias consolidated their control, but then fluctuated and rose as the 
militias became corrupted and started overeaching in their social-cleansing campaigns 
around mid-decade, and as members of the community, and later the paras, responded 
with social-cleansing campaigns of their own ? with some militants now defined as 
socialy undesirable. During the third study period (1998-2002), the distinctions among 
social cleansing, political violence, and economic violence became increasingly fuzzy, 
which might acount for the rise in social cleansing beginning in 1999 (e.g. paras 
targeting socialy undesirable, drug-dealing militants; se Chapter 5). As the fourth study 
period (2002-2007) got underway, al forms of violence began to fal, and it looks like 
social violence might have done so as wel, even though there are no data available after 
2004. I have not found any evidence that social cleansing has begun to rise during the 
wave of violence that began in 2007, but absence of evidence is not evidence of absence. 
Another form of social violence is perpetrated by stret crews, but unfortunately 
there are even les data for that than for social cleansing. One study found that, during the 
fourth study period, the number of stret crews (combos) fel from 6,300 in 2003, to 
5,900 in 2004, to 4,000 in 2005, a 37 percent decline in thre years, which tracks very 
wel with the decline of the homicide rate.
36
 
8.5.2. Political Violence 
Collective-political violence involves acts commited by one or more members of 
a group that makes or purports to be making decisions about political goods; for example, 
                                                
36
 Jorge Giraldo Ram?rez, ?Conflicto Armado Urbano y Violencia Homicida: El Caso de Medel?n.? 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 349 
they are doing the job they think a government should be doing. Such groups might 
include agents of states and governments (police, military, and other security forces) 
whether acting in their oficial capacity or not, or might include nonstate or ilegal 
organizations such as paramilitaries, vigilante groups, self-defense groups, militias, or 
guerilas. Acts of political violence can include war, asasination, forced displacement, 
imprisonment, and the forced disappearance of political opponents and human-rights 
defenders, among other acts. While economic violence has been the most pervasive form 
of violence in Medel?n (se next section), political violence is what has been studied the 
most. 
Yet much of the writing is polemical in nature, and most of the data are simply 
untrustworthy, gathered and published by advocacy organizations whose every incentive 
is to find ways to claim that the problem they track is geting worse every single year. For 
anyone trying to argue that some aspect of life in Colombia has improved or is 
improving, this population is the toughest audience imaginable, because they do not trust 
that things have not been manipulated only to seem to be improving: One might say: 
?Homicides have falen.? And the response would be: ?Disappearances are up.? One 
might insist on the facts: ?Actualy, reports of disappearances are also down.? And the 
response would be: ?But they?re higher in Colombia than in other countries. ?Okay, but 
masacres have falen.? ?No, they?ve risen.? ?But only because the definition of 
massacre has been expanded from five people kiled at a time, to four, to thre, as the 
actual number of masacres has falen.?
37
 ?But human rights violations against trade 
unionists have increased.? ?Actualy, atacks against them have falen.? ?But threats 
                                                
37
 Andr?s Balesteros, et al., ?The Work of Amnesty International and Human Rights Watch: Evidence 
From Colombia,? working paper, Documentos de CERAC 003639 (Bogot?: Centro de Recursos para el 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 350 
                                                                                                                                            
An?lisis de Conflictos, 1 February 207), htp:/ideas.repec.org/p/col/00150/03639.html (acesed 2 
August 209). 
Figure 8-9. Trends in Political Violence versus Homicides 
 
 extrajudicial executions 
disappearances (IPC) 
disappearances (CINEP) 
 total deaths in combat 
civilian combat deaths 
no. of cobat events 
 
Sources: Secretar?a de Gobierno Municipal de Medel?n, Subsecretar?a de Orden 
Civil, Unidad de Convivencia Ciudadana (homicides, 1986-2008; figures for 1984-
1985 estimated by author from rate data and population); Banco de Datos de 
Derechos Humanos y Violencia Pol?tica, CINEP (Diciembre 204) (extrajudicial 
executions and disappearances, 198-2003); Base de Datos de Instituto Popular de 
Capacitaci?n (IPC) (disapearances, 1986-1998);  Centro de Recursos para el 
An?lisis de Conflictos (CERAC) Base de Datos de Conflicto (CDB-CERAC V.8.) 
(combat events, civilian deaths in combat, and total deaths in combat). 
Combat Events 
and Combat 
Fatalities 
in Medell?n 
Homicides in Medel?n 
Extrajudicial 
Executions 
and Forced 
Disappearances 
in Antioquia 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 351 
against them have skyrocketed.? ?What counts as a threat?? ?Wel, atacks against trade 
unionists are stil higher in Colombia than elsewhere.? And so on. I?ve had many 
informal conversations like this with otherwise reasonable people in Colombia. 
Stil, the times series that are available, and that had been collected with 
reasonably consistent and transparent definitions and methods, tel a story that is 
consistent with the story developed in Part I of this study. Groups perpetrated political 
violence by atacking each other or by atacking each other?s supporters or innocent 
civilians. Figure 8-9 (p. 350) compares the time series for homicides in Medel?n (top 
panel) with some indicators for violence against each other (botom panel) and some 
indicators for violence against civilians (middle panel). 
The indicators for political violence against civilians ? forced disappearances 
(two diferent indicators, with a correlation coeficient of 0.79 betwen them) and 
extrajudicial executions ? are for Antioquia as a whole, because none were available for 
just the city of Medel?n. What stands out most in these figures is the relentles rise in 
extrajudicial executions (selective kilings of civilians outside of the formal justice 
system) beginning in the mid-1990s, as wel as the les dramatic overal increase in the 
two measures of forced disappearances. Some of these were perpetrated by death squads 
(the 1993 spike was probably atributable to the Pepes; se ? 2.1.3), state security forces, 
and leftist militias or guerilas. But the dramatic increase in executions and 
disappearances was due mainly to the activity of the paramilitaries, who nevertheles, in 
late 2001, after their long patern of human rights abuses had earned them the official ire 
of the United States, made an active decision to lower their profile and switch from a 
paramilitary to a ?parapolitical? strategy (se ? 6.1.1); hence the major decline in 2002. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 352 
The indicators for political violence betwen and among the diferent armed 
actors themselves ? combat involving guerilas (and leftist militias), state security 
forces, or paramilitaries ? are for batles that took place within Medel?n?s city limits: 
combat events, total combat fatalities, and the ?collateral damage? of such combat (the 
proportion of combat fatalities in which the victims were civilians). What stands out here 
is how wel the paterns of combat match the paterns of homicides more generaly, with 
clear peaks in 1990 and 2002. Since the number of total combat deaths is only a tiny 
proportion of total homicides (3.3 percent in 1990; 2.5 percent in 2002), this is very 
strong evidence that the interpersonal-communal murders that made up the majority of 
the homicides in the city during study period were enabled by a breakdown in the rule of 
law brought about by what sems to be mainly collective-political violence ? although it 
should be noted, as it is in the next section, that what has semed on the surface to be 
political violence since around 2000 has in fact been largely economic in nature due to 
the changes in Colombia?s political landscape during the 1990s. 
8.5.3. Economic Violence 
Collective-economic violence involves any group formed for the primary purpose 
of making money, whether from the sale of goods or services (legal or ilegal) or by theft 
or fraud. This includes most organized crime groups and other criminal conspiracies: 
mafias, drug trafickers, money launderers, asasins for hire, crime rings, oficinas, 
racketers, and otherwise legitimate busines managers using violence to coerce or 
enforce a cartel or monopoly arangement or to take or protect asets. The forced 
disappearance of trade-union organizers, and the forced displacement of peasants and the 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 353 
subsequent expropriation of their land, are two examples of economic violence. On the 
surface they are political acts undertaken in the context of counterinsurgency, but in 
many cases the counterinsurgent justifications have not withstood scrutiny, especialy as 
the perceived threat of Communism fel with the Soviet Union beginning in the early 
1990s: union organizers have been targeted not because of their presumed or suspected 
subversive sympathies but because unions are threats to corporate profits; peasants in 
guerila teritory have likewise been ejected from their lands not because of their 
presumed or suspected support for guerilas but because those doing the ejected wanted 
their land. 
Collective-economic violence has been the most visible form of violence in 
Medel?n throughout the entire period under study: the first period was characterized by 
economic violence from narcotraficking gangs and hit squads asociated with the 
Medel?n Cartel; the second period, from militias that had been corrupted by the 
opportunities provided by narcotraficking and the sale of security services; the third, 
from the Ofice of Envigado and the narco paras of the Cacique Nutibara Bloc; the 
fourth, from the narco paras who had never demobilized and from a minority of the paras 
who demobilized but then returned to narcotraficking; and the fifth, from the latest 
generation of narcotrafickers, including those once asociated with the narco paras. This 
has also been the form of violence most commonly cited as reason for concern. One 
recent survey found that 37 percent of residents considered the greatest source of 
insecurity in their barios to be drug traficking and 25 percent said the greatest threat 
were stret muggings, both asociated with collective-economic violence; 17 percent 
cited the presence of ?pandilas? or stret crews, but unfortunately the survey did not 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 354 
characterize what type of gangs these are, so it cannot be concluded whether they were 
engaged in primarily social or primarily economic violence.
38
 
Ilicit nonstate actors today are primarily engaged in economic violence, which 
suggests that they?re not interested in teritorial control. Again, anti-union violence in the 
past had been a species of political violence because it took place in the context of an 
insurgency whose goal was to impose Communism over Colombia, but it was also 
economic violence then, because wealthy corporations used Communism as an excuse to 
get rid of chalengers to their profits. In other words, anti-guerila violence, of which the 
perpetrators considered anti-union violence to be a species, had primarily political 
objectives and secondarily economic objectives. Today, anti-guerila violence is mostly 
over control of markets. stil both economic and political, but the emphasis has changed 
such that it is primarily a species of economic violence: during the 1990s, the guerilas 
became primarily economic actors, and during the 2000s the main anti-guerilas, the 
paras, either demobilized or became primarily economic actors as wel. 
The most significant long-term trend observed over the 25-year study period was 
a gradual ?corruption? of collective-social and collective-political violence by collective-
economic violence: the implicit ratio of economic violence to political or social violence 
has increased over time, although due to the paucity of reliable data, this ratio has not 
been able to be quantified. This has come about as violent organizations dedicated to 
ilicit profits (primarily narcotraficking mafias and hit squads) infiltrated and corrupted 
violent organizations dedicated to social order (e.g. community self-defense groups), 
political change (e.g. anti-guerila paramilitaries), or both (e.g. urban militias). The ilicit 
                                                
38
 Encuesta de Percepci?n Ciudadana 208. 
Lamb, Robert D. | Chapter 8. Complex Violence in Medel?n 355 
organizations dedicated to profit were non-ideological: they would work with whatever 
guerila or paramilitary group was in control of whatever teritory or micro-teritory, as 
long as it could help them ake money. As Brauer et al argued: 
One might therefore argue that the character of Colombia?s ?political? 
violence changed during the 1990s, i.e., that although it was cloaked in 
terms of revolutionary and counterevolutionary language, the observed 
violence is linked to the economics of the drug trafic. ? Revolutionaries 
and counterevolutionaries become bandits who defend their respective 
teritories and interests with murder. If this is correct, the war in Colombia 
in the 1990s was esentialy an economic war over aces to and 
exploitation of natural resources, not unlike those we observed in Africa in 
the 1990s (e.g., Siera Leone, Liberia, Angola).
39
 
To conclude from this, however, that Medel?n?s violence ? or that Colombia?s violence 
? has been generated only from conflicts over aces to ilicit resources is to ignore a 
substantial body of evidence regarding the role that community actors, and not just state 
and nonstate armed actors, have played in the violence. Community members were not 
only victims or collaborators in violence, but also perpetrators, enablers, denouncers, 
informants, pacifiers, supporters, financiers, and recruits. Understanding the roles played 
by nonstate actors and state actors, but not by community actors, therefore mises a 
significant part of the story, a story that Part I told and that Chapter 9 wil explain. 
 
                                                
39
 Jurgen Brauer, Alejandro G?mez-Sorzano, and Sankar Sethuraman, ?Decomposing Violence: Political 
Murder in Colombia, 1946-1999,? note 10. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 356 
Chapter 9. Legitimacy and Ilegitimacy in Medel?n 
Legitimacy is worthines of support or a right to loyalty. To claim that something 
is legitimate is to give a moral or normative reason (?it is right?) to obey, support, acept, 
imitate, comply with, or refrain from opposing it within some bounded range of activity 
or experience. To say that one should ofer such support, or that something is worthy of 
such support, is diferent from saying that one merely does offer such support: support 
can be externaly motivated as wel ? it can be coerced or purchased, for example ? but 
loyalty and self-motivated support are what make, for example, a social relationship or a 
governance structure legitimate, stable, and sustainable. 
Ilegitimacy is not merely the absence of legitimacy but a worthines of 
opposition: to say that something is ilegitimate is to give a moral or normative reason to 
ignore, disobey, reject, or oppose it, actively or pasively. An obligation of disobedience 
or a duty to oppose would both be based on an asumption or an argument that that which 
is to be disobeyed or opposed is ilegitimate. Legitimacy and ilegitimacy do not 
necesarily reside on a continuum; they are related but distinct phenomena: one a 
worthines to support, the other a worthines to oppose. The absence of one does not 
imply the presence of the other. For that reason, delegitimation, the proces in which a 
worthines of support is lost, should be considered a separate, although perhaps prior, 
proces from what I cal ilegitimation, the proces in which a worthines of opposition is 
confered. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 357 
A proper analysis of legitimacy should begin by identifying or defining the 
confere and the referee. The confere is the entity upon which or upon whom legitimacy 
is or is not to be confered. Legitimacy may be confered upon a role (within a social, 
political, economic, or cultural structure or relationship), a policy (se next section for 
definition), a distribution (of wealth, power, prestige, status, etc. across a defined set of 
individuals or groups), or a structure (which entails roles, policies, and distributions). 
Identifying the confere answers the question: legitimacy of what? 
The referee is the person who is judging the degre to which the confere is or is 
not legitimate. Referes include both outsiders and insiders. Outsiders are people who 
neither are members of nor are afected by the role, policy, distribution, or structure in 
question, such as authors of academic papers about legitimacy in other places. Insiders 
are people who are part of the structure or relationship in question, are afected by the 
actions of the entity occupying the role, or are afected by the policy or distribution; 
insiders can be members of high-status or low-status groups. Identifying the refere 
answers the question: legitimacy according to whom? 
Life is lived at multiple levels simultaneously, and legitimacy, being a human 
phenomenon, is a multi-level phenomenon. A proper analysis of legitimacy, therefore, 
should identify indicators across multiple level of analysis. Identifying such indicators 
answers the question: legitimacy by what criteria? In other words: what criteria does the 
referee use to judge the degre to which the confere is or is not legitimate? In diferent 
studies of legitimacy, various criteria have been used to identify either some set of causal 
indicators (also caled composite or constitutive indicators), which are understood to 
collectively constitute a measure (index, scale) of legitimacy; or some efect indicator 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 358 
(also caled a proxy or substitutive indicator) that is understood to measure some 
phenomenon that comes about as a consequence of the presence of legitimacy and can 
therefore stand in as a measure of legitimacy.
1
 
This study asked thre questions: What were the patterns of violence in Medel?n? 
What explains the changes in the patterns of violence in Medel?n? And, What role did 
legitimacy and ilegitimacy play in those changes? The first question was answered in 
Chapter 8. The second two are related and are discussed in the present chapter. The third 
question was meant to go beyond those works that study conflict as something that takes 
place mainly betwen and among armed actors, whether agents of the state or of nonstate 
entities; such works ignore the role played by members of the communities where the 
violence takes place. Community members are not only victims or collaborators in 
violence, but also perpetrators, enablers, denouncers, informants, pacifiers, supporters, 
financiers, and recruits. Understanding the roles played by nonstate actors and state 
actors, but not by community actors, therefore mises a significant part of the story. 
To get an acurate picture of the role of legitimacy, I took two approaches. First, I 
modified and expanded a framework I had previously developed for analyzing 
legitimacy. The intention was to overcome the greatest weaknes of legitimacy research: 
measurement validity. Most authors who atempt to measure legitimacy acknowledge that 
they cannot be certain that it is legitimacy and not something else that they are 
measuring. Those who asert that legitimacy resides primarily in individual belief 
acknowledge the dificulties of recal and other biases inherent in measuring opinions. 
                                                
1
 Keneth A. Bolen and Richard Lenox, ?Conventional Wisdom on Measurement: A Structural Equation 
Perspective?; Keneth A. Bolen, ?Latent Variables in Psychology and the Social Sciences?; Bruce Giley, 
?The Meaning and Measure of State Legitimacy.? 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 359 
Those who asert that legitimacy resides primarily in group behaviors acknowledge the 
dificulty of determining whether certain behaviors derive from belief rather than 
coercion. And those who asert that legitimacy resides in the objective characteristics of 
the structure under study acknowledge that their outsider judgment of the system?s 
legitimacy may wel difer from that of insiders. Yet these authors draw the data or 
observations that underlie their studies of legitimacy from only one or, at best, two of 
these levels of analysis (micro, meso, or macro). The framework I designed is intended to 
be used to look for evidence at al thre levels and, by doing so, to provide a higher 
degre of certainty about what is being measured: if individuals say they believe some 
structure to be legitimate, and groups act as if they believe that structure to be legitimate, 
and that structure has characteristics that suggest it operates legitimately, then it is very 
dificult (albeit not impossible) to argue that legitimacy is not at work in that structure; 
but if one of those levels does not agre with the others, that suggests that something 
other than legitimacy is at play (coercion, for example). Furthermore, this framework 
does not measure only proxies for legitimacy, nor does it measure only causal indicators: 
rather, it measures both a proxy variable and six causal indicators. This framework is 
discussed in detail in Appendix C (?? C.3-C.7); the reader is strongly encouraged to 
review the discussion in that appendix to be sure the terms are understood precisely as 
used in the analysis that follows. 
Second, I tried to obtain or derive data for as many of these variables as possible 
across al thre levels of analysis (individual, group, and system) by exploring published 
works and opinion polls and interviewing experts and residents about violence, 
legitimacy, governance, and teritorial control in Medel?n and a sector in East Central 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 360 
Medel?n caled Caicedo La Siera. This was an efort to understand both the 
microdynamics and the multi-level dynamics of legitimacy. Using the framework 
described in Appendix C to organize the analysis, I tried to find evidence related to the 
legitimacy or the ilegitimacy of community actors, nonstate actors, and state actors. I 
was not able to extract data for al seven indicators for al thre types of actor at al three 
levels of analysis for al 25 years of the study period. However, this study was the first 
even to atempt such a systematic analysis of the dynamics of legitimacy amid the 
complex violence of Medel?n. As such, its findings provide a richer explanation of the 
causal mechanisms among teritorial control, governance, legitimacy, and violence in 
Medel?n than had previously been found (or sought), and therefore provides a sound 
foundation upon which future research on the dynamics of complex urban violence may 
be based. 
The study of extreme cases ? cases with extremely high or extremely low values 
on the study variables ? can help to iluminate more general relationships betwen and 
among the phenomena under study.
2
 Medel?n is an extreme case of high violence, high 
both in its level of violence (it had the highest murder rate in the Western Hemisphere in 
1991 and again in 2002) and in its level of complexity (in terms of the variety of actors, 
their shifting motivations, and the instability of their aliances). But Medel?n is also an 
extreme case of instrumental social relations: social capital and trust among strangers are 
extremely low, and the tendency for expediency or exploitation to be used to achieve 
short-term results at the expense of long-term relationships is extremely high; it?s the 
kind of place for which Imanuel Kant must have imagined his hypothetical Kingdom of 
                                                
2
 Robert K. Yin, Case Study Research: Design and Methods, 47; Stephen Van Evera, Guide to Methods for 
Students of Political Science, 24. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 361 
Ends as a corrective.
3
 As such, it can be argued that, if a role for legitimacy can be found 
in an explanation for violence and stability in a place such as Medel?n, then legitimacy 
can be said to at least partly explain something about violence and stability more 
generaly. Does legitimacy ? or its opposite, ilegitimacy ? add anything to our 
understanding of violence in Medel?n that cannot be adequately explained by incentives? 
If so, what does that imply for theorists of legitimacy or, more importantly, for policy 
makers charged with reducing violence in complex environments? 
The analysis proceded as follows. First, I identified a set of proxy (efect) 
indicators, at thre levels, for legitimacy and ilegitimacy (and in some cases for 
neutrality). This step was useful as a quick first cut at an analysis because of the smal 
number of variables that needed to be evaluated. 
Second, I identified a set of causal indicators, also at thre levels, and also for 
legitimacy and ilegitimacy (and neutrality wherever possible). This step supplemented 
the analysis of the proxy variables, and was necesary because proxies can take on the 
same value whether they are caused by legitimacy or by some other phenomenon, such as 
coercion or fear, making it impossible to determine which of those phenomena actualy 
generated the proxy. Identifying causal indicators ? measures of phenomena that are 
known to contribute to the emergence of legitimacy ? helps determine whether 
legitimacy was involved: if the sorts of things that are known to cause legitimacy are 
present, and if the value of those causal variables are congruent with the value of the 
proxy variables identified earlier, then that gives a prety good reason to believe that 
                                                
3
 Immanuel Kant, Groundwork of the Metaphysic of Morals, trans. H. J. Paton (New York: Harper & Row, 
1967 [1785]; reprint 1958). 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 362 
legitimacy was involved. (Se ? C.7 for a discussion of how proxy and causal variables 
can be measured.) 
Third, I determined whether each indicator was congruent across al thre levels 
of analysis. This step helped to tease out the degre to which legitimacy was involved 
relative to other potential explanations: a higher degre of congruence suggests higher 
legitimacy and offers higher confidence in the judgment about legitimacy?s involvement. 
Fourth, I determined whether the refere indicators were congruent betwen high-
status and low-status insiders. This step helped to acount for the possibility that support 
came simply as a result of expedience (i.e. people benefit) rather than as a result of 
congruence betwen confere atributes and refere values; higher congruence betwen 
high-status and low-status groups suggests higher legitimacy. 
Finaly, I reversed the direction of the analysis so that the conferes (nonstate 
armed actors, ? 9.3.1) became the referes, and the referes (community actors, ? 9.3.2) 
become the conferes. This step, reflecting the observation that legitimacy is a two-way 
stret, helped to acount for the possibility that low-status groups? support derived from 
adaptive preferences: if low-status residents (as referes) consider the power of nonstate 
armed actors to be legitimate, but the nonstate armed actors (as referes) do not consider 
the power of the low-status residents to be legitimate, then this lack of congruence 
suggests that the low-status residents had a reason to adapt their preferences (thereby 
indicating lower legitimacy), whereas congruence suggests they did not have a reason to 
adapt. (A final section in the analysis, ? 9.3.3, briefly considers the role of state actors in 
the dynamics of legitimacy, but does not dwel on that role since it was so limited until 
late in the study period.) 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 363 
One can claim with confidence that legitimacy was involved if: the things that 
legitimacy causes are present; the things that cause legitimacy are present; both sets of 
things are present and congruent at thre levels of analysis; both sets of things are present 
and congruent across two levels of status; and both sets of things are present and 
congruent in two directions. If this is the case, then one can stil argue that legitimacy had 
nothing to do with the support or stability that is observed, but the burden of proof at that 
point wil have shifted to the person making that claim. And if one can find this in 
Medel?n, an extreme case of instrumental social relations, then one can stil argue that 
there does not exist a role for legitimacy in support or stability more generaly, but again 
the burden of proof wil be on the person making that argument. 
9.1. Proxy Indicators 
One can imagine any number of public-attributes proxy indicators for the 
legitimacy of the nonstate armed actors who controlled micro-teritories or statelets in 
Medel?n, but the most direct proxy is probably the extent of internal represion or 
coercion: If the nonstate armed actors believed that they had the support of the residents 
of the areas they controlled, then they would have no reason for represion or coercion. If 
they believed they had litle or no internal support (low legitimacy or neutrality), their 
behavior would tend to be coercive. And if they believed there was a chance of internal 
rebelion (ilegitimacy), their behavior would tend to be represive. Those groups to 
whom local residents voluntarily contributed money, supplies, and volunters, therefore, 
would be considered more legitimate than those who raised resources through the vacuna 
system (extortion rackets), who in turn would be considered les ilegitimate than those 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 364 
who regularly kiled or displaced internal adversaries or sealed off the borders to 
outsiders (out of fear that insiders who wished to rebel would sek help from outsiders 
who wished to control the area). 
Closely related, of course, would be a group-behavior proxy that measures acts of 
support by residents, such as voluntary participation in programs sponsored by the armed 
actors or voluntary contributions to the armed actors? budget. The degre to which these 
behaviors were voluntary would be partly indicated by the amount of punishment visited 
upon those who did not participate or contribute (a public-atribute measure), but it is the 
group behavior itself that is the indicator of legitimacy. Non-participation, a reluctance to 
contribute, and protests against some aspect of the armed actors? leadership would 
indicate neutrality or low legitimacy, while acts of sabotage or theft against the armed 
actors? resources and atacks against the armed actors themselves (or their supporters) 
would indicate ilegitimacy. 
Finaly, an individual-belief proxy might measure the way residents expresed 
support, opposition, or indiference (verbaly or in writing). Expresing pride in a family 
member or a friend who had become a member of the armed actor?s organization, 
expresing satisfaction with their leadership or with the overal direction that their bario 
was taking, or expresing a generaly favorable impresion of the armed actors 
themselves would al count as proxy indicators for a belief in their legitimacy, while low 
values on these expresions would indicate low legitimacy or neutrality. Disowning 
family and friends who joined the armed actors, and expresing disgust with the way 
things were going or disdain for the armed actors themselves would be proxy indicators 
for ilegitimacy. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 365 
9.2. Causal Indicators 
Because proxy indicators are efects indicators and can therefore have multiple 
causes (legitimacy being only one potential cause in this case), using proxies alone does 
not enable one to identify which cause was active in any given instance. Participation, for 
example, is a proxy for legitimacy since one of the things that legitimacy does is 
encourage participation. But other things can encourage participation, too ? some 
reward for participating, some punishment for not participating ? and so it would be 
helpful to identify indicators for other phenomena that cause legitimacy but that do not 
cause those other things. 
These indicators used in this analysis were based on criteria derived from a 
review of the literatures on legitimacy across several academic and policy disciplines, 
including political science, sociology, psychology, philosophy, organizational studies, 
and military doctrine (se ? C.6 for a discussion). This review revealed what writers have 
said over the centuries about what sorts of things make people more likely to offer or 
withdraw voluntarily support or loyalty: law, tradition, leadership, efectivenes, consent, 
norms, and so on. These criteria are sometimes caled the ?sources? or ?factors? of 
legitimacy, and the problem with many such lists and typologies is that they have 
provided neither a ?thick? enough nor a ?thin? enough acount of the sources of 
legitimacy: not thick enough because they cannot be used as a field guide to any 
particular population?s reasoning about legitimacy, and not thin enough because they 
cannot be used as a general guide to human reasoning about legitimacy.
4
 For the 
framework used in this study (se Appendix C), I opted to develop a thinner acount to 
                                                
4
 Michael Walzer, Thick and Thin; Bruce Gilley, ?Thick Or Thin? An Empirical Intervention.? 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 366 
demonstrate the very human reasoning that underlies judgments and behaviors about 
what counts as right or wrong, as worthy or unworthy to support, as a duty to comply or 
to oppose. By doing so, I was able to capture and incorporate into a single, simple 
framework the broad range of sentiments and motives underlying the sources identified in 
the existing literatures. This framework provides a more useful starting point from which 
the particular, complex, mesy details of the dynamics of legitimacy in the real world can 
later be uncovered. If nothing else, it provides a baseline for the types of questions that 
should be asked for any study of legitimacy: no fewer questions than are suggested by the 
framework, but no limits to the questions that can be asked beyond that. 
The six criteria that I ended up identifying as basic, ?thin? causal indicators for 
legitimacy are: transparent, credible, justifiable, equitable, acesible, and respectful. 
Transparent and credible describe the most basic criteria that motivate people to support 
a confere, because together they make life predictable: people want to know, for 
example, what the rules are and know that they can be enforced non-arbitrarily, even if 
they don?t agre with the rules, and even if they don?t benefit from the rules. These two 
?predictability? criteria do not actualy confer legitimacy; they are more like contributory 
or background conditions, necesary but not sufficient, that make it possible for people to 
live their daily lives and plan out personal projects within the given constraints. To put it 
another way: predictability wil not generate legitimacy (and voluntary support), but its 
significant absence wil almost certainly generate ilegitimacy (and active opposition). 
The justifiable criterion tends to be the central component of legitimacy, as it captures the 
values people hold most dear ? their judgments about what is right (in acord with 
valuable norms or rules), good (in acord with valuable outcomes), proper (in acord 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 367 
with valuable proceses), and admirable (representative of their values) ? and therefore 
worthy of their support or loyalty. The equitable criterion reflects ideas about fairnes: 
people want to be asured that inequalities are justified and that, if they have les of 
something that someone else has, it is for a good reason. The accesible criterion captures 
much of the literature on consent as the basis for legitimacy, but goes beyond what many 
authors consider to be a strictly democratic basis; regardles of the specific system of 
consent or public reason, what people want is some asurance that they have a voice, 
some say in how the things that afect them operate. Finaly, the respectful criterion 
captures the literature on human dignity and pride: consistently disrespectful treatment, 
even if everything else is justified, equitable, and acesible, tends to create tension with 
people?s desire and ability to be loyal or offer support. 
These six criteria together represent a rather thin conception of legitimacy: 
everybody can agre in principle that, for example, the rules regulating political and 
social relations should be transparent, credible, justifiable, equitable, acesible, and 
respectful. Actualy measuring legitimacy, however, is a mater of measuring what these 
adjectives mean in a real-world context. What they mean wil difer depending on 
whether the confere is a role, a policy, a distribution, or a structure. Table C-2 in 
Appendix C lists each of these criteria and suggests how the basic questions might be 
formulated based on the type of confere (it does not list structure, because structure 
encompases roles, role-holders, policies, and distributions); for conceptual 
completenes, it additionaly includes a generic proxy indicator. This framework and 
these criteria guided the identification of causal indicators, which are discussed presently. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 368 
9.2.1. Transparent 
Did the nonstate armed actors in control of teritory let internal residents know the 
terms of engagement: how they were going to govern; what the rules, the reasoning 
behind the rules, and the consequences of breaking the rules were; and so on? Eforts to 
communicate (honestly) with the community and to explain their reasoning, in language 
the community can understand (whether writen or oral), would count as a high level of 
transparency and would indicate stronger legitimacy; weak eforts to communicate and 
explain, or doing so incomprehensibly, would be lower transparency and indicate weaker 
legitimacy; and deception would indicate opacity and no legitimacy or ilegitimacy. 
These are public-attributes indicators. A group-behavior indicator of legitimacy by 
transparency might be residents? compliance with the armed actors? rules and commands 
(since one cannot comply unles one knows what it is that one is complying with), 
whereas an indicator for low legitimacy or even ilegitimacy might be conscientious 
objection. An individual-belief indicator would simply be the residents? ability to 
acurately state the rules, the reasoning, or the governance objectives of the armed actors, 
or a statement that they believe they know what the rules and so on are. 
9.2.2. Credible 
Were the nonstate armed actors capable of carying out their promises and 
obligations? Were their rules enforceable and enforced? Did they govern in a way that 
enabled residents to cary out daily activities and engage in long-term planning? A 
history of succes in implementing their policies (i.e. outcomes acord with their stated 
objectives) and efectivenes in creating a predictable living environment would be signs 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 369 
of credibility and public-attributes indicators of legitimacy by credibility, whereas a 
history of failure would indicate the opposite. Evidence that residents actualy do go 
about their daily activities or engage in some degre of long-term planning might work as 
group-behavior indicators of credibility (legitimacy), whereas evidence that they live ?for 
the moment? or that they habitualy disobey orders and break rules might indicate an 
absence of credibility (neutrality or ilegitimacy). Individual-belief indicators for 
credibility might include the fact that people talk about the future (especialy if they 
expres hope for the future) or expres confidence (or even fear) that the armed actors can 
do what they say they wil do (legitimacy), while indicators for a lack of credibility might 
be reports by individuals that they do not trust the armed actors, have no hope for the 
future, or believe that you can get away with breaking the rules (neutrality) or that the 
armed actors habitualy lie or that they wil probably be out of power soon (ilegitimacy). 
9.2.3. Justifiable 
Did the nonstate armed actors enforce, promote, embody, or otherwise behave in 
acordance with the community?s sense of the right, the good, the virtuous, and the 
admirable? Or did they transgres the behavioral or symbolic norms of the community? 
To measure this, it helps to know a litle something about the range of things that 
the community in question actualy values, because the best public-attributes indicators 
wil be those that measure the degre of congruence betwen those values and the 
structure, behavior, or identity of the armed actors. What traits or virtues are most 
important to the most influential community actors, or which do they consider most 
admirable: loyalty, courage, equanimity, bombast, charisma, integrity, empathy, 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 370 
diligence, creativity, generosity, humility, self-confidence, prudence, boldnes, etc.? Do 
they value material succes above al else? Do they root for the underdog or for the 
proven winner? Do they tend to be more authoritarian or libertarian in their internal social 
relations? More ?conservative? or more ?liberal? (in the local interpretation of those 
terms)? To what degre is pluralism valued? What roles do religion and tradition play, 
and what roles do people think religion and tradition should play? Do people idolize the 
same historical figures and role models and merely interpret what they represent in 
diferent ways, or do diferent subgroups idolize diferent figures entirely? How strong 
are nationalist or tribalist sentiments, in-group favoritism, out-group prejudices, or 
bigotry in general? How easily is disgust expresed? What sorts of carers or spouses do 
parents aspire for their children to have? What are their views of equality (se equitable, 
below)? To what degre is disent permited? To what degre is fredom of expresion 
valued? What are the most important manners and folkways and how strictly do 
community actors insist upon their observance? Do they recognize other ways of life as 
being valuable (if perhaps localy inappropriate) or do they judge them as simply being 
wrong? With this information, and with an efort on the part of the analyst to keep his or 
her own values out of the analysis, one can make a judgment about the degre to which 
the nonstate armed actors acorded with residents? values and beliefs (legitimacy) or 
antagonized their most deeply held convictions (ilegitimacy). 
The best group-behavior indicators for the justifiable criterion might measure the 
presence of or participation in demonstrations of support (legitimacy), protests (low or no 
legitimacy), or outright rebelions (ilegitimacy) in which the language of morality, 
religion, values, or virtue played a prominent role. Individual-belief indicators might 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 371 
measure atitudes toward those group behaviors or, in the absence of such behaviors, 
might analyze the language people used to refer to the nonstate armed actors and what 
they did or represented. For example, what epithets did they use to expres disapproval 
(e.g. in the United States, caling a politician ?French? is considered an insult), and did 
they use words and phrases like moral, imoral, disgusting, admirable, hero, good for 
us, bad for us, inspiring, holy, righteous, evil, and so on to describe the nonstate armed 
actors or their actions or policies? 
9.2.4. Equitable 
Did the distribution of influence, respect, and wel-being among the residents 
living in the nonstate armed actors? teritory acord with the residents? views of merit? 
Things can be unequaly distributed without there being any adverse influence on the 
level of legitimacy, as long as that inequality was not unjustified acording to the 
standards of the community. A public-attribute indicator should not measure simple 
inequality, as that would mis the point of inequity: what maters is not the degre of 
inequality but the degre of locally unjustified inequality, which is to say, inequity. Stil, 
equity is dificult to measure directly, so it is probably necesary to combine public-
atribute measures of equality ? wealth is usualy the most acesible measure, but one 
can also consider the degre of social or economic mobility ? with group-behavior 
indicators, such as demonstrations and protests that used the language of identity, justice, 
privilege, or pride, or with individual-belief indicators, such as statements about the 
degre to which people believed they either had or could get what they deserved, perhaps 
derived from an analysis of how low-status individuals explained their own status. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 372 
9.2.5. Accesible 
Did residents have (or believe they had) influence over the nonstate armed actors? 
policies and policy outcomes? Evidence that policy proceses alowed for some degre of 
public participation, even if it was as simple as alowing residents to ask for help in 
resolving a conflict or hosting community metings at which people could air grievances 
or request asistance, might work as public-attributes indicators of legitimacy by 
acesibility. The best group-behavior indicator is simply the level of actual 
participation: public demonstrations of support (legitimacy), protests (low legitimacy or 
neutrality), or acts of sabotage or rebelion (ilegitimacy). But others can be imagined: 
Did people regularly approach the armed actors seking help? In electoral systems, did 
people vote? Individual-belief indicators might measure the degre to which people 
reported feling empowered, disempowered, or marginalized. 
9.2.6. Respectful 
Were residents? identities, values, opinions, membership in the community, and 
input into policies treated as being valuable? The habitual humiliation or denigration of 
residents or some subgroup of residents is a clear public-attribute causal indicator of 
ilegitimacy, whereas treating people acording to the community?s manners indicates 
legitimacy. The way people acted toward each other in the stret is a group-behavior 
indicator of the degre to which people thought they are treated with respect: if residents 
had taken to making threatening demands for respect in the stret, that suggests they 
believed they were not geting enough respect from either higher-status community 
members or from the nonstate armed actors who controlled their teritory (neutrality or 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 373 
ilegitimacy), while a culture in which people were generaly civil to one another on the 
strets might indicate that they did not fel they were habitualy disrespected. Finaly, 
individual-belief indicators that people felt respected might include statements about the 
degre to which they felt they were treated with dignity or recognized as valued members 
of the community, or the degre to which they believed the nonstate armed actors had 
created an environment that was elevating and civil. 
9.3. Analysis 
To answer the question about the role that legitimacy or ilegitimacy played in the 
changes in the paterns of violence in Medel?n, I took the steps described in the previous 
sections and tried to obtain or derive data for as many of the framework?s variables as 
possible across al thre levels of analysis (individual, group, and system) by exploring 
published works and opinion pols and interviewing experts and residents about violence, 
legitimacy, governance, and teritorial control in Medel?n and the sector Caicedo La 
Siera. I tried to find evidence related to the legitimacy or the ilegitimacy primarily of 
nonstate armed actors and secondarily of community and state actors. I was not able to 
find quantitative or qualitative data for al of the framework?s indicators for al three 
types of actor for al five study periods, but keeping in mind the framework?s questions as 
I read the literature, interviewed subjects, and analyzed what data I could find gave the 
overal analysis a sound basis upon which to build judgments about the microdynamics 
and multi-level dynamics that are reported in the sections that follow. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 374 
9.3.1. Legitimacy of Nonstate Armed Actors 
A few years ago I was at a restaurant in Washington, D.C., with a group of 
people. I was discussing my research with a friend of a friend, a Ph.D. student in political 
science who late in the conversation told me he considered the field of Policy Studies in 
general, and the subject of my research in particular, to be a joke: Policy Studies because 
it produces analysis and not theory, and my research because, as he put it, ?there?s no 
such thing as the legitimacy of anything other than states.? He was wrong. He would 
have been wrong had he said it during the 1990s, when nonstate actors were barely 
geting any pres, but he was definitely wrong when he said it in the mid-2000s, when 
even a cursory reading of power dynamics in Afghanistan and Pakistan would have 
demonstrated that tribal actors rather than state actors were the leaders who commanded 
loyalty in the Pashtun areas where Osama bin Laden (and other nonstate actors about 
whom the United States was suddenly concerned) was now believed to be hiding. The 
same observation can be made regarding rural Colombia, where the dynamics of power 
and loyalty have long centered more around local strongmen than around state 
representatives. Sanho Tre, a drug-policy analyst, once told me that whenever he travels 
far from Colombia?s city centers to its more remote regions, such as the borders with 
Venezuela or Ecuador, he likes to pose a particular question to locals: What makes you 
?Colombian?? That is, in what aspect of your identity do fel that you are part of this 
country caled ?Colombia?? Almost invariably, he said, he gets the same response: 
soccer. ?They?re Colombian because they root for Colombian soccer teams. That?s a very 
thin basis for legitimacy.?
5
 State actors, in other words, are not always the most salient 
                                                
5
 Sanho Tre, personal comunication, 14 April 208. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 375 
players in Colombian politics. The next chapter argues that policy makers today have no 
choice but to pay close atention to the behavior, power dynamics, and, yes, legitimacy of 
nonstate actors; for the present purposes, I consider the point proven. 
To what degre were nonstate actors in Medel?n worthy of the support of the 
people who lived in the micro-teritories they controlled? The power that nonstate armed 
actors in control of teritory have, by the definition of control (se Appendix A), is the 
power to prevent governance and therefore the power to define who governs and how 
they wil govern. The question, then, is whether those in Medel?n had exercised that 
power in a way that could be considered worthy of supporting. Proxy indicators for this 
revolve around the extent to which the nonstate armed actors resorted to internal 
represion and coercion against residents to keep themselves in power: the habitual 
kiling and displacement of internal adversaries and excesive border enforcement (i.e. 
the automatic kiling of strangers and harsh punishment against insiders sen with 
outsiders) would tend to indicate that they were not considered worthy of voluntary 
support, that is, that they were ilegitimate; excesive coercion, especialy vacunas, would 
tend to indicate neutrality or low legitimacy; and the absence of anything above minimal 
levels of both would tend to indicate higher legitimacy, as would evidence of residents? 
voluntary contributions of money, supplies, and volunters. Which groups had these 
features, and when? For the causal indicators, one can ask: Which groups acted in a way 
that was transparent, credible, justifiable, equitable, acesible, and respectful? Which 
groups did residents believe ? and which groups did residents act like they believed ? 
were transparent, credible, justifiable, equitable, acesible, and respectful? 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 376 
To answer these questions, I drew on the cases presented in Part I to review the 
strategies and behaviors employed by the various crews, gangs, militias, and 
paramilitaries that have sought to win teritorial control or local aces to global ilicit 
markets and the reactions of community actors to those strategies and behaviors. 
The poor hilside barios on Medel?n?s periphery had started out as nearly 
?ungoverned? but relatively stable areas. There was violence and crime, but they were at 
a level that most residents could tolerate and addres within existing social structures. 
Evidence suggests that their tolerance was due to the strength of community?s social 
institutions ? there were usualy a church, some schools, reasonably strong families, and 
nosey neighbors ? and those social institutions had the ability to maintain a respectable 
degre of social control, despite the scant presence of state institutions (such as police) or 
other armed actors wiling and able to maintain order. In short, control was maintained by 
the community. 
But as the population of Medel?n?s hilside slums grew ith people displaced 
from the country?s conflict in rural areas, the tenuous social order that existed among the 
populations that lived there was chalenged beyond its capacity to continue to instil 
effective social controls: the influx of outsiders into those communities, including 
outsiders with diferent sets of values and mores, simply put too much presure on their 
social institutions. Many young people without jobs or social recognition formed stret 
crews and gangs, fought each other, often atacked members of their own communities, 
and in general both emerged from and contributed to the breakdown in social order in the 
city?s periphery. Violence skyrocketed, as did behaviors deemed offensive or deviant 
acording the socialy conservative standards of many longstanding members of those 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 377 
communities. By the end of the 1980s, one might say that social control lost ground to 
teritorial control under the law of the jungle: the criminals, crews, and gangs who 
emerged amid the overcrowding fought over micro-teritories and local aces to global 
ilicit markets, and they clearly had the capability to prevent at least some forms of 
outside governance and self-governance. But as rulers of the bario they were completely 
ilegitimate in the eyes of the bario communities: Young people who lived where those 
gangs controlled became known as the ?No Future? generation; visitors were feared and 
sometimes kiled upon arival; and deviant behavior (acording to the socialy 
conservative norms of most residents) was unopposed and so proliferated. The peripheral 
barios came to be characterized by an extreme of unpredictability, instability, and 
insecurity ? not to mention growing violence ? and the residents considered the way 
the gangs controlled them to be very worthy of opposition: there was a terified demand 
for a return to social control. 
When the M-19 guerilas set up ?peace camps? in some of those barios, they 
were welcomed by many as the only force both wiling and able to confront the growing 
insecurity and social deviance (se ? 3.1.1). These peace camps offer a fairly typical 
example of a strategy of legitimation. The M-19 ?sponsored community sancochos 
[parties at which a traditional stew is made], retreats, or other activities to win the 
community?s afection, they organized commisions to solicit cash donations from the 
busines community ? [and these] donations were made without major reservation, since 
the M-19?s camps had offered security and recreation to these forgotten sectors.? The 
stret crews and gangs, who had found themselves opposed by most residents, lost their 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 378 
control over teritory (and many lost their lives) to the peace campers, who had won the 
support of some residents (most residents remained neutral).
 
But when the M-19 left the barios that they had briefly controlled and governed, 
some of its members who had stayed behind formed new groups that operated more like 
the gangs they had been organized originaly to counter, and ?the donations that at first 
were given wilingly became obligatory payments or vacunas.? The punishments for not 
paying for the new gangs? protection ?services? became increasingly represive. One 
gang that had been founded by former M-19 peace campers, Los Nachos, once punished 
a small-busines owner who had not paid his vacuna by seting one of his buses on fire ? 
with thre tenagers who worked for him stil inside.
6
 Without guidance from leaders 
with a political program, and with growing aces to opportunities to make easy money, 
the former peace campers abandoned their legitimation strategy, and so abandoned their 
legitimacy. 
The urban militias that formed during the late 1980s and early 1990s (to combat 
both the earlier type of stret crew and the later, more capable gangs whom the peace 
camps had trained) emerged in those barios by an explicit strategy of legitimation, as the 
M-19 had before them. These early militias were succesful because they created statelets 
(se ? A.1) ? they acted as ?a state inside the state? ? and governed in a manner 
consistent with their legitimacy-building strategies. They provided security to 
communities suffering from a severe breakdown in social order; they achieved a great 
                                                
6
 ?.. cuando los campamentos realizaban un sancocho comunitario, una retreta y otra actividad para 
ganarse el afecto de la comunidad, organizaban comisiones para solicitar la colaboraci?n en especie a los 
negocios del sector. .. El aporte del comercio se daba entonces sin mayores reparos pues los campamentos 
del M-19 hab?an ofrecido seguridad y recreaci?n a estos sectores olvidados. Cuando los jefes del M-19 
salieron de El Popular, los aportes que en principio eran de buena voluntad se convirtieron en cuota 
obligatoria o vacuna ..? Gilberto Medina Franco, ?Una Historia de las Milicias de Medel?n (Historia Sin 
Fin),? 24. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 379 
deal of social support as a result of this security and the social work that folowed; and 
they were explicitly dedicated to encouraging and promoting the communities? own 
moral values within the poor barios they controlled.
7
 
The militia that formed in upper Caicedo La Siera, the November 6 and 7 Militia 
(M-6&7), was a fairly typical case in point. M-6&7 started out with a good deal of 
support from most community actors based on a relationship of legitimacy. I base this 
conclusion on an analysis using the framework described in this chapter. With respect to 
the public-atributes causal indicators, M-6&7 made its rules known (transparent), 
demonstrated its capability to enforce them (credible), became advocates for the 
communities? own moral values (justifiable), alowed people to approach them to resolve 
conflicts and disagrements (acesible), and by most acounts treated most community 
members with dignity (respectful), at least at first. (The bario?s low-status members ? 
the so-caled disposable people ? certainly were treated very unequaly, but I found no 
reliable information to support any claims about how inequitably they or other members 
of the community felt the militants had acted.) 
Most group-behavior and individual-belief causal indicators pointed to M-6&7?s 
relatively high legitimacy as wel (relative to other nonstate armed actors who had 
controlled that and other teritories on Medel?n?s periphery): 
? Transparent. Residents (both high-status and low-status) semed to know the 
rules, and high-status residents usualy (and low-status residents most of the 
time) generaly complied with them. 
                                                
7
 Francisco Guti?rez San?n and Ana Mar?a Jaramilo, ?Crime, (Counter-)Insurgency and the Privatization 
of Security.? 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 380 
? Credible. At al status levels, people would expres confidence in the militias? 
ability to maintain stability and enforce the rules, and they demonstrated that 
confidence by actualy going about their daily activities within the constraints 
of those rules. 
? Justifiable. Most residents, even those who disagred with the militias? 
politics, would say they generaly approved of the way the militias ruled, not 
only with respect to security but also with respect to the way they addresed 
what most community members considered dangerous or deviant behavior 
(such as drug sales and prostitution); many young people talked about their 
dreams of joining the militias, an indicator of their admiration for them; and a 
lot of people (high- and low-status alike) participated in militia-sponsored 
community activities such as festivals, although I did not find much more 
evidence about participation than that. 
? Equitable. I found some weak evidence regarding the degre to which the 
militias were considered equitable, mainly in the form of anecdotes about how 
even mothers whose children were socialy cleansed or violently punished for 
breaking curfews supposedly agred that their children deserved their 
punishment; and low-status residents did not generaly protest their status, but 
I could not find much verbal evidence to determine whether that was due to 
fear of punishment or to a belief that they could be beter of if only they 
would try harder (i.e. that their status was not the fault of the militias). 
? Acesible. Residents of al statuses clearly believed the militias were 
acesible, because they constantly approached militants with requests to 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 381 
solve problems or provide some benefit, such as a job or housing. (I could not 
find reliable individual-belief evidence for this criterion.) 
? Respectful. In the stret people were generaly civil to each other, indicating 
by their behavior that the militias had created a generaly respectful 
environment. (I could not find reliable individual-belief evidence for this 
criterion either.) 
Meanwhile, there continued to be many barios that the militias never took control 
over. Those barios remained in the hands of crews, crews confederated with larger 
gangs, or those larger gangs themselves, usualy with funding from narcotraficking or 
asasination contracts. A lot of those gangs were inconsiderate or abusive to their 
neighbors, able to maintain control of their barios mainly by virtue of their economic 
and military power (i.e. they were able to buy off, displace, or kil any significant source 
of opposition). But many others took a leson from the golden era of the militias (pre-
1991) ? that legitimacy lowers the costs of teritorial control ? and ?were vigorously 
promoting communitarian causes to win the support of the population?: 
They learnt self-discipline and started to impose some basic regulations on 
criminal activity (you shal not steal in your own bario, etc.). Other 
groups ? opposed to both the gangs and the militias ? appeared, but they 
too semed to have learned that to maintain teritorial control they had to 
offer security, some kind of self-discipline, and a constructive, 
communitarian set of activities.
8
 
This dynamic could be sen in lower Caicedo La Siera, where no militia had ever 
emerged to take control, either because they never tried to take control of some blocks or 
because where they tried they failed because the gangs that were there were too wel 
armed. The most significant of these wel armed gangs to emerge was La Ca?ada, whose 
                                                
8
 Ibid. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 382 
strength derived mainly from the economic resources it was able to bring to the fight. La 
Ca?ada was not founded with any political or social program; rather, it grew based on its 
ability to win economic resources from the ilicit economy, primarily through contracts 
from, and drug sales for, the Medel?n Cartel. La Ca?ada did enjoy some degre of 
support from the community, which derived from the fact that they did protect favored 
residents against rival gangs and the militants from above, and took some other steps to 
keep the community happy. ?If the people from the community are happy, you can 
manage things beter,? one resident said, since it gives people a reason to act as 
informants; plus, in a community where a lot of people have already suffered, he added, 
further harming them would risk turning them into informants for the gang?s rivals.
9
 
La Ca?ada?s control, therefore, derived primarily from its military strength (it 
could win acquiescence through coercion) and its economic strength (it could win support 
through barter), and only secondarily from any relationships of legitimacy: from these 
public-atributes proxies we can conclude that it had low legitimacy at best. I could not 
find a good group-behavior proxy, and in terms of individual beliefs, I could find only 
evidence that a lot of young people, at least, admired the gangsters. So the proxy 
indicators across levels did not match up wel, suggesting that what legitimacy they had 
was weak or that what support they had was coerced or purchased. 
The causal indicators painted an even more mixed picture. La Ca?ada did a 
reasonably good job of making its rules and expectations known, and people generaly 
complied (transparent). The gang certainly was capable of enforcement, and people (both 
high-status and low-status residents) generaly went about their busines within the 
                                                
9
 ?Si la gente de la comunidad est? contenta usted maneja mejor las cosas.? ?Danilo,? Interview No. 10. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 383 
constraints those rules imposed (credible). But La Ca?ada did not explicitly work to 
embody or promote community values (for example, they permited drug use, which most 
in the conservative communities could not approve of), even if they did engage in some 
social cleansing of the type that at least the high-status residents semed to support; I 
found only weak anecdotal evidence regarding the views and behaviors of low-status 
residents for the justifiable criterion. I could not find good evidence for any of the 
equitable indicators, but several residents mentioned that they thought the gangsters were 
respectful and acesible (e.g. they could be asked to resolve conflicts in the community). 
By comparison to the legitimacy of M-6&7 in Upper Caicedo La Siera, therefore, 
La Ca?ada had relatively low legitimacy, or possibly neutrality, in Lower Caicedo La 
Siera. Rather than maintaining its teritorial control primarily through a strategy of 
legitimation (actively making itself worthy of the community?s support), it semed 
implicitly to be engaging primarily in a strategy of ilegitimacy-avoidance (working 
merely to avoid triggering a moral obligation of opposition), by contrast to M-6&7, 
whose strategy was explicitly one of legitimacy-building. 
The fact that both M-6&7 and La Ca?ada could maintain control for relatively 
extended periods in their respective teritories demonstrates that, at least in the short 
term, having legitimacy acording to the community is a useful, but neither a necesary 
nor a sufficient, condition to winning or maintaining teritorial control. Beyond Caicedo 
La Siera, the nonstate armed actors who were able to maintain control over their own 
micro-teritories or statelets for any extended period were those who had aces to 
needed resources (through local resource extraction or from external backers) and who 
stayed within certain bounds of behavior with respect to their community (ilegitimacy-
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 384 
avoidance). The later, more capable gangs who succesfully controlled statelets (as 
opposed to the earlier smal-time gangs who had merely controlled micro-teritories) 
were flush with cash from narcotraficking and sicariato (hired asasination), while the 
militias who succesfully controlled statelets (even those who enjoyed modest support 
from rural guerilas) had to depend mostly on local sources of funding. 
This analysis suggests that, at least in Medel?n, legitimacy has lowered the costs 
of teritorial control while ilegitimacy has raised the costs of teritorial control. One 
would therefore expect a resource-poor organization to succed only if it engaged in a 
strategy of legitimation, whereas a resource-rich organization could succed with merely 
a strategy of ilegitimacy-avoidance; that is exactly what was observed with respect to the 
relatively resource-poor M-6&7 (legitimacy-building) and the relatively resource-rich La 
Ca?ada (ilegitimacy-avoidance). 
This conclusion is strengthened by observing how the militias fel (se Chapter 4) 
and how the paramilitaries rose (se Chapter 5). Both in Caicedo La Siera and Medel?n 
as a whole, the relatively resource-poor militants failed in their legitimation strategies and 
ilegitimized themselves by failing to maintain predicable living environments or by 
transgresing what was moraly aceptable to the communities (hypocriticaly alowing 
drug use again, for example). Two things raised the costs of the militias? teritorial 
control and led ultimately to their inability to hold on. First, during the 1990s they faced 
growing presures from state and nonstate armed actors; as long as they had internal 
support, they could use what resources they had for external defense. Second, however, 
their own behavior within their statelets sparked a backlash against their rule, and so they 
had to use their resources for internal defense as wel. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 385 
On just about every indicator, the militias first delegitimized and then 
ilegitimized themselves. Consider one proxy for legitimacy: internal represion, in this 
case in the form of social cleansing. At first welcomed as the only available mechanism 
for maintaining a defensible social order, these social-cleansing campaigns ? the kiling 
of people who engaged in deviant behaviors ? meant that, over time, more and more 
people in the communities were suffering the loss of a family member or a friend to 
social cleansing. Moreover, the militias kept expanding the types of behavior that they 
judged as deviant: from dangerous behaviors (e.g. drug seling and treason) to merely 
disapproved-of behaviors (e.g. homosexuality and poor parenting). In both public 
atributes (represion) and individual beliefs (expresions of disgust with their behavior), 
proxies for the militias? legitimacy were showing rapid declines through the mid-1990s. 
Other indicators suggested their ilegitimation as wel. With the lure of easy 
money from narcotrafickers, and the radicalization and fragmentation brought about by 
the militias? growing pains, the militias? own behavior was ?characterized by the resort to 
summary executions, arbitrary rule, the community?s ignorance about civil and military 
decisions, [and] the resort to murder as a tool for resolving conflicts within the group.?
10
 
They were becoming unpredictable, and their conflict-resolution methods were sen as 
increasingly unjustifiable. ?Even worse, the negotiation proces [that led to the Santa 
Elena Acords] weakened the broad social support for the militias. As soon as they 
abandoned their military activities, they forfeited the power that alowed them both to 
                                                
10
 ?Este periodo, entre 191-1994, se caracteriza por el recurso a las ejecuciones sumarias, la arbitrariedad, 
el desconocimiento de la comunidad para la toma de decisiones militares y c?vicas, el recurso al homicidio 
como instrumento de resoluci?n de conflictos al interior del grupo.? Arleison Arcos Rivas, ?Ciudadan?a 
Armada,? 108. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 386 
control the gangs and to promote their cherished moral order,?
11
 a devastating loss of 
credibility that acelerated their delegitimation and ultimately resulted in ilegitimacy. 
Their loss of support (neutralization) was reflected in part in their poor showing in 
subsequent elections: they simply ?did not recognize the enormous gap betwen their 
imagined constituency, the manual workers, with their complex political culture, and 
their own rather obscure, revolutionary insider frame of reference.?
12
 Their ultimate 
ilegitimation was reflected in the communities? wilingnes, beginning in the late 1990s, 
to give the paramilitaries a chance. 
In short, the main indicators of the militias? ilegitimation were: an increased 
reliance on represion (a proxy indicator); declining levels of transparency, credibility, 
and acesibility (causal indicators); and their failure to continue ?to promote their 
cherished moral order? and their disconnect from community perspectives (in both cases 
reflecting the justifiable causal indicator). 
The paramilitaries? Metro Bloc displaced these militias, beginning in the late 
1990s, by using a combination of military force, deal-making, and a weak version of the 
same legitimation strategies that the militias themselves had used a decade earlier to 
displace gangs from some of the same barios (se ? 5.1.1): they offered security against 
abusive strongmen, removed low-status social groups from the community, provided 
some benefits to supporters and neutral community members, outlawed the sale and use 
of drugs, and threatened death or displacement to anyone who offered opposition instead 
of compliance. Where they were particularly efective were in barios where armed actors 
                                                
11
 Francisco Guti?rez San?n and Ana Mar?a Jaramilo, ?Crime, (Counter-)Insurgency and the Privatization 
of Security.? 
12
 Ibid. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 387 
already operated with the support of the community and could simply be coopted into 
Metro, as was the case with M-6&7 under the leadership of Jason and ?dison, who had 
efectively staged a coup against the founding militants of M-6&7 who had turned 
abusive (se ? 5.2). When Metro approached them, ?dison was already acting as de facto 
mayor of upper Caicedo La Siera, was wel liked and respected in the community, and 
had a reputation for usualy being respectful to residents and fair in his judgments; the 
name change did not afect those relationships. These are al strong causal indicators of 
legitimacy: equitable and respectful treatment, transparent and justifiable rules, acesible 
leaders, and a credible manner of rule. 
More generaly, however, whatever support the rest of the paras enjoyed during 
the late 1990s and at the turn of the milennium derived mainly from their efectivenes 
in defeating the hated militias and driving them out of the city, rather than from any 
warm relations with community actors, who knew, despite the paras? public protestations 
to the contrary, that the units that made up the Cacique Nutibara Bloc were simply the 
same drug gangs who had already been working with the Envigado Ofice. To the degre 
legitimacy played a role, it was their legitimacy acording to state actors outside of the 
barios they controled, rather than acording to comunity actors inside. Diego R?os, 
who had grown up there, told the story of a city investigator who came to Comuna 8 to 
respond to a citizen?s complaint during this period: when the investigator met the woman 
who had caled, the first thing he asked her was whether she had brought the isue up 
with ?Job,? ?Don Berna? Murilo?s top aide in that sector. ?Imagine that,? R?os said: ?A 
city detective asking a citizen if she?d taken her complaint to the criminal! Who?s realy 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 388 
in charge here??
13
 The state actors knew ho they were working with (transparent), and 
they worked with them because they semed to share their antisubversive political 
outlook (justifiable) and knew that they were efective (credible). (For more on state 
actors? historical support and legitimation of paramilitaries, se the introduction to 
Chapter 5.) Community actors generaly appreciated Cacique Nutibara?s credibility but 
did not necesarily offer the higher degre of support that some Metro block units might 
have enjoyed. 
Even after the paramilitaries demobilized, they did not return to their communities 
as common residents. Demobilized paras continued working in the barios as community 
leaders under the same military-like structures they had operated through during their 
armed control over the barios, but they were publicly coy about the degre to which they 
continued to be involved in ilegal activities such as drug traficking, extortion rackets, 
and forced disappearances (se Chapters 6 and 7). One resident of West Central Medel?n 
complained that, at least under militia rule, people knew ho was dangerous: 
In this bario, with the paramilitaries, the law of silence reigns. You could 
say we were beter off with the guerila, because they acted openly. You 
could get kiled in a shooting, but at least you knew here the bullets were 
coming from. With the paramilitaries everything happens in secret. We do 
not know exactly who is responsible for the disappearances.
14
 
Similar sentiments were expresed in East Central Medel?n: that a lot of mid-level 
paramilitary commanders were active, but ?nobody wants to say who?s in charge? or 
admit to knowing who was responsible for the selective kilings.
15
 Since the most basic 
                                                
13
 Diego R?os, Interview No. 4. 
14
 Quoted in Ralph Rozema, ?Urban DDR-Proceses.? 
15
 ?Nadie quiere decir qui?n manda.? Diego R?os, Interview No. 4. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 389 
requirement for a confere?s legitimacy is that the referes know who the confere 
actualy is, it would be dificult to argue that the paras? rule was legitimate, either 
acording to the residents who lived there or acording to the public atributes of their 
rule. Their secret control over the places they lived was, almost by definition, 
ilegitimate. 
To summarize the analysis: The indicators for the legitimacy of the peace-camp 
guerilas, of the early-period militias, and, to a leser degre, of certain paramilitary units 
asociated with the Metro Bloc, al semed to point in the same general direction and 
were generaly congruent with one another across levels of analysis and status groups, 
suggesting moderate to relatively high levels of legitimacy (although not very high in 
absolute terms). By contrast, the indicators for the legitimacy of most stret crews and 
gangs, the militias during their corrupted late period, and the paramilitaries asociated 
with Cacique Nutibara were much les consistent across levels and groups, suggesting 
relatively low levels of legitimacy or more likely neutrality, which implies that 
phenomena other than legitimacy were most likely behind whatever support or teritorial 
control they enjoyed. 
Of course, these indicators were not static. Most people, most of the time, 
remained neutral toward al of the armed actors, considering them neither legitimate nor 
ilegitimate but merely a fact of life, someone to cooperate with when expedient or to 
oppose when necesary. But active and moraly motivated voluntary support, and active 
and moraly motivated voluntary opposition, stil did wax and wane in diferent places at 
diferent times as diferent armed actors emerged, took control, governed wel or poorly, 
were chalenged, and ultimately fel. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 390 
The indicators for ilegitimacy were highest for those groups who faced credible 
chalenges to their control and then could not hold on much longer. This observation 
enables us to abstract from the specific situations in the micro-teritories and statelets to a 
general observation about the role that ilegitimacy played in the city?s complex paterns 
of violence: When those in control of teritory governed in a manner characterized by a 
lack of transparency and credibility ? when they acted unpredictably and their manner 
of governance made life intolerably unpredictable for residents ? they made it very 
likely that the people they governed would decide that they should be opposed. This 
ilegitimation raised the cost of teritorial control, making feasible a contest for control by 
rivals and thereby triggering collective-violent conflicts (acompanied by interpersonal-
communal violence), which complicated eforts to govern in a manner that could make 
life predictable, which further raised the costs of teritorial control, which further 
alienated residents, and so on, as the cycle repeated until the teritory was lost to a 
diferent armed actor. In short, ilegitimacy has been an ?intervening? phenomenon in the 
city?s cycles of violence. 
9.3.2. Legitimacy of Community Actors 
The analysis of the previous section summarized the dynamics of the legitimacy 
and ilegitimacy of Medel?n?s nonstate armed actors, both acording to community 
actors? beliefs and behaviors and acording to the armed actors? public atributes. If we 
were to reverse that analysis, making the armed actors the referes of the legitimacy of 
the community actors, then we would discover the degre to which these dynamics were 
symmetrical (a two-way stret). That is the purpose of this section. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 391 
It might not be readily apparent what it would mean for a community or for 
community actors to be legitimate under the framework developed for this study. It is 
more familiar to speak of communities as the referes (those who confer legitimacy upon, 
for example, a government) and not as the conferes (those upon whom legitimacy is 
confered). So it is worth going through the framework step by step to demonstrate. As it 
was stated earlier, to claim that something is legitimate is to give a moral or normative 
reason to obey, support, acept, imitate, comply with, or refrain from opposing it within 
some bounded range of activity or experience. What was the ?bounded range of activity 
or experience? here? What did they demand (obedience, support, imitation, etc.) and of 
whom did they demand it? And who were ?they?? 
To begin with, ?they? were the people who lived in the peripheral barios where 
most of Medel?n?s violence took place. We could consider them collectively (i.e. 
residents of barios designated as ?middle low? or below on the city?s socioeconomic 
scale); we could disaggregate them by zone, comuna, bario, or Community Action 
Board (JAC) jurisdiction in the oficial city-planning documents; or we could 
disaggregate them by their own local spatial identification, that is, by how locals identify 
where they themselves live, for example by neighborhood (a few strets or city blocks) or 
by gang teritory (areas controlled by a nonstate armed actor). For the present analysis, I 
focused more or les on gang teritory; in Caicedo La Siera, for example, the main 
division was betwen los de arriba (?those from above,? who lived in M-6&7 or Metro 
Bloc teritory) and los de abajo (?those from below,? who lived in the teritory of the 
networks of La Ca?ada and the Cacique Nutibara Bloc). As conferes, the communities 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 392 
in question could be considered social structures, with roles, rules (however implicit), and 
distributions of goods and status. 
What did they demand, and of whom (i.e. to which referes) did they demand it? 
Given that this was a study of violence and that most of Medel?n?s violence was 
asociated with contests over teritorial control (and the resultant breakdowns in social 
order and the rule of law; se Chapter 8), the relevant referes were primarily the nonstate 
armed actors who controled their teritories and secondarily the state actors whom many 
residents believed should be in control of their teritories. Given that this was a study of 
legitimacy (?worthines of support?), the relevant ?support? that the community 
demanded of them was, at minimum, non-interference with their daily lives and personal 
projects, which, due to the low quality of life in these areas, generaly revolved around 
eforts to get adequate food, water, shelter, and a source of income. Beyond that 
minimum, what they more commonly demanded was protection against violence and an 
environment in which they could secure those esential services themselves; at a 
somewhat higher level they demanded actual asistance to secure those services and to 
defend the social order in acordance with their own values. 
In other words, the communities claimed that they had a right to be governed and 
that those who had managed to take control of their teritory by force of arms had a 
corresponding duty to govern, and not merely to control, them. (For an extended 
discussion of the diference betwen control and governance, se Appendix A.) Was this 
demand to be governed a legitimate demand? Were the people making this demand 
making it legitimately? This section addreses these questions, but they are only part of 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 393 
the story; the rest of the story involves the question of what power the community actors 
actualy had. 
Studying the legitimacy of the power of community actors turns on its head the 
traditional way of thinking about political authority, in which power is held to be 
legitimate to the degre it acords with the values, preferences, or interests of ?the 
governed? and, once legitimized, those in power have certain rights that ?the governed? 
have a duty to respect. Theories of democracy are theories about rule by the people, in 
whom power is said ultimately to be held. Yet the ?power of the people? is rarely, if ever, 
questioned in the theoretical literature on democratic legitimacy: the power of the demos 
is asumed to be legitimate, whereas the power of leaders, oligarchs, aristocrats, dictators, 
kings, appointes, elected officials, and so on is asumed to be the power that needs to be 
legitimized. Because it is an asumption, theorists of legitimacy do not always bother 
putting much efort into demonstrating that citizens have a right to the power they have 
and that political leaders therefore have a duty not to undermine that power. 
In the real world, however, leaders, oligarchs, aristocrats, dictators, kings, 
appointes, elected officials, and so on do not necesarily share the asumptions of 
Western, liberal, democratic theorists about the legitimacy of the power of the people: to 
them, their own power is legitimate ? even dictators and their closest followers convince 
themselves of their own legitimacy ? whereas the legitimacy of other centers of power is 
something to be demonstrated. Those of us who study legitimacy, democracy, or politics, 
therefore, should demonstrate, rather than asume, that the power of the people is a 
legitimate power; otherwise, we cannot hope to influence such leaders. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 394 
What power do ?the people? have? (Joseph S. Nye?s definition of power as the 
ability to influence outcomes should suffice here.
16
) The epigraph by Nicol? Machiaveli 
at the beginning of this chapter demonstrates two potential sources of power that 
communities might have to influence outcomes: the implicit threat of taking up arms 
against unloved leaders, and the availability of ?foreigners to asist them? when they so 
decide to take up arms.
17
 In Medel?n, communities did organize self-defense and 
vigilante groups to remove armed actors from their barios on many occasions, and any 
number of outside armed groups (?foreigners?) made themselves available to asist them. 
However, not al of the communities actualy had the capability to overthrow localy 
unloved leaders or the capability to aly themselves with powerful outsiders. 
But al people and al communities do have an even more basic power, and those 
in Medel?n were no diferent: the power to define for themselves the way of life that is 
worth taking up arms to defend. People can be coerced to mouth their support for 
someone else?s prefered way of life (and to repres their own). But the power to silently 
define for oneself the way of life that is worth taking up arms to defend is a power than 
cannot be taken away. 
To what degre did community actors have a right to that power? To what degre 
did community actors exercise that power legitimately? I evaluate these questions based 
on proxy and causal indicator for legitimacy acording to public atributes of the 
communities in question, acording to the behavior and beliefs of the various high-status 
                                                
16
 Joseph S. Nye, Soft Power: The Means to Succes in World Politics, 1st ed. (New York: Public Afairs, 
2004), 1. 
17
 Nicol? Machiaveli, The Prince, trans. Luigi Rici and Eric Reginald Pearce Vincent (New York: 
Mentor/Penguin, 1935 [1513]), 108. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 395 
and low-status nonstate armed actors (insider referes) who until the mid-2000s 
controlled the teritories where the community actors lived, and acording to the behavior 
and beliefs of the high-status and low-status state actors (a second set of insider referes) 
who tenuously controlled that teritory beginning in the 2000s. 
Some proxy indicators for community legitimacy can be derived by analogy to the 
proxies used in the analysis of nonstate armed actors? legitimacy and ilegitimacy (se 
previous section). In that analysis, the proxy for ilegitimacy was a high degre of internal 
represion, which was asumed to indicate that the armed actors who had the power to 
rule did not believe that they had efectively made their case to residents that they also 
had the right to rule or, at least, that their rule was worthy of support: represion by 
conferes is an expresion of fear of, or contempt for, referes. Is there something similar 
that might indicate that community actors in Medel?n did not believe that they had 
efectively made their case to other actors in Medel?n that they had not only the power to 
define their own way of life but also the right to define their own way of life or that their 
way of life was worthy of support? Or to put it more simply: to what degre had 
community actors made their case that their way of life was worth defending? 
When armed actors believe that they have efectively made their case for a right 
to rule, then they trust that residents of their statelets wil not rebel; likewise, when 
community actors believe they have efectively made their case for a right to define their 
own way of life, then they trust that others wil not try to undermine that right. Mistrust 
or fear of strangers, then, might be a reasonable public-atribute proxy indicator for 
ilegitimacy, while social capital might be a (weak) proxy for legitimacy. And, in fact, a 
prominent feature of Medel?n?s social landscape was low social capital and a profound 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 396 
lack of trust in others, especialy of strangers; during periods of high violence this lack of 
trust has often translated into fear of strangers or outsiders, especialy in the peripheral 
barios. (These features are sometimes masked by another important feature of the social 
landscape: their strong sense of hospitality toward the strangers with whom they 
interacted, which is not by any means a mutualy exclusive feature; it is entirely possible 
to be kind and gracious toward someone whom you fear or do not trust.) 
The other two levels of proxy indicator (group behavior and individual belief) 
measure the insider referes? judgments. Four categories of insider refere are relevant 
here: high-status nonstate armed actors (e.g. the founders or leaders of militias, gangs, 
and paramilitaries who set policy or give commands), low-status nonstate armed actors 
(e.g. their lowest-ranking members and asorted hangers-on), high-status state actors (e.g. 
those who set policy or give commands), and low-status state actors (e.g. those who 
implement policy, cary out commands, or otherwise provide support to high-status state 
actors). If these referes voluntarily supported the community actors? right to define their 
own way of life, then they would have participated in or otherwise contributed to the 
community?s defense of that way of life (a group-behavior proxy) or would have 
expresed their support for the community?s self-determination of that way of life (an 
individual-belief proxy). Non-participation, a reluctance to contribute, or noncommital 
comments about community rights would indicate low legitimacy or neutrality toward the 
community, while the theft of community resources, atacks against community actors, 
and expresions of disdain for the communities would indicate a judgment that the 
communities? way of life was ilegitimate. By these indicators, the evidence regarding the 
community actors? legitimacy is mixed. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 397 
It was generaly the case that the nonstate armed actors who most considered the 
community they controled legitimate were those who had grown up in those same 
communities. That is because in most cases they had formed those groups specificaly to 
defend their families, their neighbors, and their strets in the first place. It should be 
noted, however, that this legitimacy did not always run both ways: a stret crew formed 
by young people to defend their community might believe their community had a 
legitimate right to define its own way of life (i.e. the nonstate actors were the referes 
who confered legitimacy upon community actors), but members of the community did 
not always return the favor by considering the stret crew?s exercise of control over their 
bario to be legitimate (i.e. the community actors were the referes who did not confer 
legitimacy upon nonstate actors). Al else equal, this asymmetry suggests a weaknes in 
the overal dynamics of legitimacy in those communities. Legitimacy did run both ways 
in some cases, however. In places controlled by groups such as M-6&7 and the Metro 
Bloc (se previous section), the group in power legitimized, defended, and spoke wel of 
the community?s rights, and the communities returned the favor by more or les 
legitimizing their rule. 
A general observation can be made as wel, then, that the referes who considered 
and treated the communities under their control as legitimate were the same groups who 
actualy made an efort to govern them and not just to control them: the peace campers, 
the early-period militias, and some of the Metro Bloc units that had emerged from the 
communities themselves (rather than those units that had been imposed on the 
communities by outsiders), such as Metro?s La Siera unit (se ? 5.2). 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 398 
Beyond those groups, most of the nonstate armed actors, most of the time, 
controlled their teritories in a way that suggested neutrality at best or contempt at worst 
for the community actors? right to define their own way of life. Many, though not al, of 
these were outsiders who came in to control teritory, or insiders who alied themselves 
with outsiders who wanted to control that teritory. (The most important exception was 
the peace campers, the guerila outsiders who entered communities through an explicit, 
and succesful, strategy of legitimation.) The indicators that in the last section acted as 
public-atributes proxies for these armed actors? ilegitimacy (acording to community 
actors) operate in the present section as group-behavior proxies for the community actors? 
ilegitimacy (acording to the armed actors): the fact that the armed actors demanded 
vacunas and engaged in extensive internal represion (i.e. public-atribute indicators of 
the armed actors? ilegitimacy acording to the comunity) suggests something about 
those armed actors? views of the legitimacy of the communities they abused, namely that 
they considered the communities? right to define their own way of life to be not worth 
defending ? that is, to be ilegitimate (i.e. group-behavior indicators of community 
actors? ilegitimacy acording to the armed actors). 
The trend over time suggests that nonstate armed actors increasingly recognized 
the value of publicly afirming the legitimacy of community actors, and so they 
increasingly paid lip service to community rights and publicized their own eforts to 
govern. Their deeds, however, generaly continued to be fairly abusive. 
The evidence for community legitimacy acording to state actors was equaly 
mixed. Some low-status state actors, such as some police officers not part of the 
community-police units, would expres contempt for members of the community, while 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 399 
others would simply follow orders to raid the community, having no opinion about their 
way of life whatsoever. But other low-status state actors, such as the social workers, 
community police, and public-school teachers who were regularly sent to work in those 
communities were some of city?s strongest advocates for those communities? rights, 
indicating high legitimacy. Similarly, some high-status state actors, especialy during the 
first half of the study period, ordered raids and masacres or expresed contempt for 
residents of the peripheral barios, characterizing them as animals or lumping al 
community members together with the gangsters among them. 
The trend over time, however, was toward stronger legitimacy as measured by 
high-status state actors? public statements and activities. For example, top city officials 
increasingly recognized the communities? right to self-determination, to the point where 
Mayor Alonso Salazar in the early 2000s instituted a participatory budgeting proces that 
enabled the communities to develop and the city to fund their own plans for development. 
This suggests an overal increase in the legitimacy of community actors acording to state 
actors ? a stabilizing trend that policy makers should take note of. (The legitimacy of 
citizens acording to their political leaders is a topic taken up again in Appendix A.) 
Does this overal analysis suggest that the communities of the peripheral barios 
were mostly ilegitimate during most of the study period ? that they did not have the 
right, even though they technicaly had the power, to define the way of life that they 
considered worth defending, or that they exercised that power in a way that could not be 
considered worthy of the state and nonstate actors? support until later in the study period? 
If the proxy indicators discussed in the previous paragraphs were the only 
indicators evaluated, then the answer would have to be that the communities did begin 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 400 
the study period as largely ilegitimate in these respects but over time became les so, and 
may even have begun to enjoy some neutrality or low legitimacy acording to most of the 
nonstate and state actors who wanted to control their teritories or the resources available 
in their barios. 
There is some danger in this interpretation, however. Recal that the problem with 
relying on proxy indicators to evaluate a latent phenomenon such as legitimacy is that 
proxies cannot help determine whether that latent phenomenon, and not some other 
phenomenon, is what had generated the value on the proxy. In this case, community 
actors? mistrust of strangers might not have derived from a fear that their own 
ilegitimacy might cause armed actors to want to harm them; it might have derived 
instead from a general cultural tendency of mistrust. Likewise, the harsh treatment the 
nonstate armed actors showed the communities might not have derived from a view that 
the community?s exercise of its power to define its way of life was not worth supporting; 
it might have derived instead from simple gred or clas hatred. And the recent good 
treatment by high-status state actors might not have derived from a strengthening belief 
in the community?s right to self-determination; it might have derived instead from a 
simple desire for votes. 
Therefore, causal indicators must be sought to supplement the analysis. 
Any outsider evaluation of the transparency and credibility of what community 
actors said they wanted is going to have to grapple with a number of local cultural 
tendencies. While people in Medel?n are extremely gracious and hospitable, even they 
themselves admit that, as a general cultural tendency (i.e. usualy, but not al people and 
not al the time), they tend to treat promises and commitments very casualy. I myself 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 401 
was offered several paid asignments that never materialized, and no fewer than six 
potential research asistants who said they would work for me never started the job; in 
most of those cases, we had agred upon specific starting dates and salaries, and they had 
semed to indicate enthusiastic interest in the topic (and the salary). There is a tendency 
toward conflict-avoidance as wel. Not knowing an answer is not always considered a 
good reason for not giving an answer, if giving an answer wil help the speaker avoid 
witnesing the listener?s disappointment; offers of help to friends and strangers sem 
often to be made just to enjoy hearing expresions of gratitude, regardles of the 
speaker?s intent to follow through when needed. There are culturaly appropriate methods 
that locals use to get others to follow through on their commitments, but several 
independent sources told me (or demonstrated to me) that it involves acting superior to 
someone or making someone fel very uncomfortable. Moreover, social relations tend to 
be highly instrumental, and so loyalties and aliances, beyond a very tight circle of family 
or friends, tend to be highly unstable. 
Work does nevertheles get acomplished in Medel?n, and people do stil work 
together on community projects, so I would not want to conclude that most of the 
community actors under study suffered a complete deficit of transparency or credibility. 
But there was clearly a deficit, and that deficit might have been one factor in the 
dificulty the community had geting support from nonstate armed actors and the state: 
being unreliable and unpredictable would tend to weaken one?s potential for relationship-
building. 
Did those referes of the communities? legitimacy ? the state and nonstate actors 
? think and act as if the communities suffered these deficits? Surprisingly, not always. It 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 402 
is probably fair to say that, for most of the study period, most of the city?s gangs, 
paramilitary units, and city workers were simply too uninterested in engaging the 
communities for it to mater; to them, the communities were neither legitimate nor 
ilegitimate, merely neutral. But some important state actors, most notably the JAC 
leaders (who were from and were elected by the comunities they represented), and 
some important nonstate actors, especialy those who emerged from the communities 
whose teritories they controlled (such as M-6&7, Metro Bloc, and, to a much leser 
degre, La Ca?ada), already knew the communities wel enough that the transparency and 
credibility deficits were not much of an isue: they already knew hat most community 
members envisioned for their lives, they already knew how to engage and mobilize 
community members to work toward achieving that vision when needed, and they often 
did so succesfully. In other words, they treated the communities, in word and deed, as if 
the communities were transparent, credible, and acesible. 
Moreover, some of these referes ? including many of the early-period militias, 
Metro Bloc units, JAC leaders, later elected officials, and community police officers ? 
actualy shared or promoted the communities? own visions, or tried with some succes (in 
the case of the peace campers and early-period militias) to influence it: they treated the 
communities, in word and deed, as if their views were justifiable or acesible. 
The analysis of these referes suggests a congruence betwen individual beliefs 
and group behaviors with respect to the transparent, credible, justifiable, and accesible 
criteria. This congruence betwen two of the thre levels of analysis ? individual belief, 
group behavior, and public atribute, with the later in this case being the incongruent 
level ? offers weak evidence in favor of legitimacy: significant, perhaps not decisive, 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 403 
but certainly not dispositive either. On balance, it is probably fair to say that this set of 
referes considered the communities legitimate and treated them acordingly. 
By now it should be clear that a patern is emerging. Some groups merely 
controlled teritory: they had the capability to prevent others from governing (the 
definition of control) but did not themselves try to govern, or if they did try, they 
governed very litle or very poorly. Other groups controlled teritory but also tried to 
govern, even if they did not always do it wel or with great succes. Clearly this is a 
continuum and not a dichotomy: some governed more, others les, some governed wel, 
others les wel, and so on. I do not want to go too far beyond my evidence base and 
claim too much about the general relationship betwen legitimacy and governance; I 
simply want to make an observation that in Medel?n the groups who tried to govern also 
tended to support community actors and their way of life (as shown so far in this section), 
and the community actors they supported tended to support them as wel (as shown in the 
previous section). 
The causal indicators for some of the criteria for legitimacy on balance support 
this observation. Regarding the group-behavior and individual-belief indicators: the 
groups who tried to govern (cal them the ?governors?), in comparison to those who tried 
merely to control (cal them the ?controllers?), semed more likely to talk and act as if 
they believed the communities they were trying to govern had transparent, credible, 
justifiable, and accesible views about how they wanted to live their lives, and that the 
communities were generaly respectful toward them (I discuss the equitable criterion 
below). Regarding the public-atributes indicators: in the places that these ?governors? 
controlled, even if the communities? views about how they wanted to live their lives were 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 404 
not (from a public-atributes perspective) necesarily transparent, at least their demands 
were reasonably credible (in the sense that they could be complied with: ?leave us 
alone,? ?don?t govern us arbitrarily,? ?protect us against violence,? ?help us atain 
esential services,? etc.); their views were also reasonably justifiable (in the sense that the 
underlying values were usualy not inconsistent with the governors? own values) and 
reasonably accesible (in the sense that they were generaly open to the governors? 
leadership on questions of community development); and for the most part the 
community actors were respectful toward the governors, despite the always-lingering 
trust isues discussed earlier. Inequities, on the other hand, were quite rampant: There 
was always a lot of talk about human rights and equality in the communities under study 
(despite some clear gender biases), but in practice most people stil held very strong in-
group biases that amounted to very unequal and, compared to their stated beliefs, very 
inequitable treatment. 
At the end of the study period, an important question can be asked: Were the 
agents of the state acting more as ?controllers? or as ?governors? of the peripheral barios 
that had long been the main generators (and victims) of violence in the city? On balance, 
it sems fair to conclude, most state actors semed sincerely to be making an efort to be 
governors. After the paramilitaries demobilized, the city took advantage of the peace that 
followed to improve development (e.g. libraries, public transportation), security (e.g. 
community police and occasional military patrols), and participation (e.g. participatory 
budgeting) in those barios. Those eforts semed to reflect an underlying belief in the 
legitimacy of those communities as members of the larger city community and worthy of 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 405 
the city?s support ? a refreshing contrast to the general atitudes and behaviors of the 
early study period. 
This is exactly what it means for a community to have legitimacy acording to 
state actors (or to nonstate actors for that mater). Given that legitimacy is a worthines of 
support or a right to loyalty, we can ask: To what degre do those in power consider 
diferent communities within the teritories in which they wield power to be worth 
supporting? To what degre do they have a right to demand protection, governance, and 
inclusion? A government that neglects or abuses an entire clas of people or an entire 
sector of a city (such as a shantytown) is a government that does not consider those 
people to be legitimate members of the political or social community: ?Those people 
don?t have rights, and as a government oficial I therefore have no duties to them.? 
To what degre, then, do those in power protect the rights of those not in power? 
To what degre do the agents of the state or the statelet do things for common people that 
they do not personaly want to do but believe they have a duty to do? To the degre those 
agents so protect and respect community actors, provide them with a predictable 
environment to cary out daily activities and long-term projects, and treat them as they 
treat members of their own in-group ? to the degre they do these things, they are 
thereby increasing the overal legitimacy of their society. 
A visitor went to Medel?n during the Miracle period (se ? 6.1) and observed the 
impresive changes that had been taking place in the city?s peripheral barios. He had 
lived in Rio de Janeiro, Brazil, another city with a marginalized and violent periphery 
(the slums are caled favelas there), and he had thought a lot about what could possibly be 
done to deal with such a complex problem in such a complex environment. ?I don?t have 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 406 
any original, clever answer to the problem of Rio?s favelas,? he wrote. ?Frankly, I just 
don?t know enough about the complex sociological and economic isues involved.? But 
after visiting Medel?n and seing the beginnings of a transformation in the periphery?s 
relationship with the rest of the city, he came to understand one important point: ?no 
amount of economic development or job training or infrastructure building or education 
is going to have any efect until the residents of Rio?s favelas are sen as legitimate 
citizens of society, as the valuable contributors they actualy are.? 
The Brazilian government and al the NGOs in the world can pour as 
many resources into these communities as they want, but they need to 
learn one important leson from Medel?n: it is only when you promote a 
genuine mesage of inclusion both in public and behind closed doors, a 
mesage backed up by concrete action, that you are able to inspire the 
energy and wil from al sectors of society that you need to conquer such 
an intractable problem.
18
 
Medel?n?s ?intractable? problems had not yet been conquered by the end of this study, 
and in fact much of its progres was being eroded as the most recent wave of armed 
actors reinitiated conflicts over aces to local ilicit markets and over local aces to 
global ilicit markets, a variation on a familiar theme. But the dynamics of legitimacy 
betwen the nonstate armed actors who had controled teritory in the past and the 
community actors who faced the choice of cooperating with, supporting, being neutral 
toward, or opposing them ? those dynamics offer a leson to the state actors who today 
are trying to establish their own control and legitimacy in the periphery. 
                                                
18
 ?Medel?n, Colombia vs. Rio de Janeiro, Brazil: Lesons to be Learned,? Tiago in Colombia, 
http:/ww.tiagoforte.com/2009/07/medelin-colombia-vs-rio-de-janeiro.html (acesed 29 July 209). 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 407 
9.3.3. Legitimacy of State Actors 
There is, of course, such a thing as the legitimacy of the state, as the friend-of-a-
friend political scientist had pointed out to me in the introduction to this section (? 9.3). 
But I hope that by this point I have demonstrated some of the more interesting non-
traditional dynamics of legitimacy as wel. The reader who has suffered through the 
extended discussions of those dynamics in the previous two subsections probably has by 
now a clear sense of what my analysis of the legitimacy of state actors might entail: a 
step-by-step review of the proxy and causal variables at thre levels of analysis, with 
pairwise, bidirectional comparisons of the dynamics betwen diferent sets of state actors 
and diferent sets of referes, and so forth. For the interested reader, I leave that analysis 
as an exercise. Here, rather than applying the full framework to the legitimacy of 
Medel?n?s multifarious state actors, I make some general observations that by now are 
probably fairly obvious but worth emphasizing nonetheles. 
In my interviews with experts who have studied Medel?n?s violence, the question 
of legitimacy appeared frequently, but the term itself semed often to be used ironicaly, 
whether intentionaly so or not. Diego R?os told me that most people in Medel?n, 
wanting to avoid problems, have simply stayed neutral with respect to whatever political 
movement or armed conflict happened to be present in their bario that year; they didn?t 
explore too deeply what those movements and conflicts were about, and they tried not to 
get involved if they could avoid it.
19
 Sometimes those residents have found it necesary 
to engage with the armed actors who controlled their barios, and C?sar Mendoza 
Gonz?lez used the phrase ?legitimacy of convenience? to describe the relationship 
                                                
19
 Diego R?os, Interview No. 4. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 408 
betwen the communities and those armed actors; the state was not present, he said, so 
residents would approach the gangsters or militants to solve their problems ? and their 
problems would get solved, but in exchange they had to subject themselves to extortion 
and control.
20
 
Juan Carlos Palou described the dynamics of legitimacy in Medel?n as 
Hobbesian: the main source of legitimacy for diferent armed actors has been the 
protection they ofered for life and property. When protection is the only public good 
provided by those in power, he said, life is impoverished, both materialy and moraly. In 
Medel?n, even that impoverished form of protection has been taken away when it has 
ceased to be convenient. As a result, communities? loyalties in Medel?n have tended to 
be very weak, shifting from guerilas, to militias, to paramilitaries, and now, perhaps, to 
the state, as each group has come into control. ?But what other option do people have?? 
Palou asked: people can subject themselves to the new power, they can collaborate with 
it, or they can resist it, and in Medel?n people have done al thre, depending on the 
circumstances. He observed that people have learned that they have to play betwen the 
legal and the ilegal: they might use the state for its public services (infrastructure, social 
work, etc.) but they wil turn to whatever group controls their bario to resolve their 
personal and community problems.
21
 
Many others, experts and common people alike, expresed a similar skepticism 
about legitimacy in Medel?n. In a place where social relations are so relentlesly 
instrumental, they implicitly argued, legitimacy cannot reliably explain much: Al 
                                                
20
 C?sar Mendoza Gonz?lez, Interview No. 6. 
21
 Juan Carlos Palou, Interview No. 3. 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 409 
relations betwen common people and nonstate armed actors have been relations of 
expediency; al support of armed groups has come from a desire for protection or a fear 
of reprisal; al opposition to those groups has come from self-defense, gred, or 
opportunism; and al support that those groups have offered to residents has come from 
deception, since pretending to care about the community simply made it easier for them 
to control teritory and thereby aces and exploit that teritory?s resources (drug markets, 
smuggling routes, brothels, rackets, etc.). The state ? not just Medel?n, but the 
Colombian state as a whole ? is likewise widely believed to be irelevant at best and 
ilegitimate at worst. A deep cynicism about social and political relations pervades 
Medel?n?s society. 
But I don?t think that cynicism is entirely justified. To be fair to the cynics, 
Colombia in general and Medel?n in particular can be a demoralizing place for someone 
who recognizes that social capital and legitimacy in the exercise of power tend to have 
salutary efects on the quality of life, stability, and long-term development. But 
legitimacy and ilegitimacy have, despite widespread claims to the contrary, played an 
important role in regulating relations betwen community actors and the nonstate or state 
actors who have tried to control or govern them. 
With respect to the nonstate armed actors, as the last two subsections and many 
sections in Part I should have demonstrated, there have been many diferent kinds of 
armed actor who have sought and won teritorial control in Medel?n?s periphery over the 
past 25 years, and in their relationships with the people who lived there, the levels of 
legitimacy and ilegitimacy varied widely. Most such relationships were probably 
characterized by neutrality, but many were characterized by ilegitimacy, some by weak 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 410 
legitimacy, and even a few by moderately strong legitimacy. Moreover, if one is 
concerned about explaining the violence there, ilegitimacy played a non-trivial role. It 
wasn?t al just fear, gred, and opportunism that led to violence, although those things 
certainly were involved in most instances of violent opposition to armed rule. Much of 
that opposition was strongly tinged with moral and social disapproval as wel, especialy 
against those gangs involved in the drug trade and against those who had once been 
defenders of the community?s moral and social order but later lost their way. Wherever 
there was active support or violent opposition, there were always mixed motives, but that 
does not necesarily imply that the subset of motives that derived from what people 
thought was right or wrong played no role: the presence of ilegitimacy ? moral 
judgments about worthines of opposition ? certainly semed to activate latent 
opposition. 
With respect to the state, the cynicism about legitimacy in Colombia is not 
entirely justified either. There is a strong tendency in Colombia ? not just in Medel?n ? 
to insist that the Colombian state is ilegitimate, an atitude that was systematicaly 
rejected by Eduardo Posada Carb? in 2003, when the period of relative stability had 
barely even begun. Posada Carb? offered evidence that, while many in Colombia have 
complained biterly about the state?s ilegitimacy, citizens? group behaviors, the state?s 
public atributes, and even some opinion polls have often pointed in the opposite 
direction, toward legitimacy. For example, he countered criticisms about the quality of 
the country?s democracy (or in my terms, complaints that the political system was not 
acesible to comon people) by citing the fact that not only have elections taken place 
regularly in Colombia for more than a century, but reforms to the system have steadily 
Lamb, Robert D. | Chapter 9. Legitimacy and Ilegitimacy in Medel?n 411 
extended the franchise to more and more people ? and despite citizens? claims that there 
is no democracy in Colombia, a lot of those people continue to vote election after 
election, and the losers of those elections have stepped down peacefully. He countered 
criticisms about the state?s credibility ? in particular its ability to deliver political goods 
to citizens ? by pointing out that a lot of state institutions actualy operate very wel and 
have often received positive evaluations in public opinion surveys; he cited the country?s 
world-clas public-transportation networks, the high quality (if not always the equity) of 
public education and health services, a strong tradition of policy planning, and steady 
improvements in the profesionalism of military and police forces, among others.
22
 
Certainly it has long been the case that the state has had litle if any presence in 
many of Medel?n?s peripheral barios. Despite that absence, many communities have 
longed for a constructive state presence, and there is a long history of informal 
community leaders approaching the state with requests, for example, for a stronger police 
presence, or for a JAC and a budget line. (Se Chapter 7, especialy ? 7.2.3, for a 
discussion of what community actors want from the state.) That in itself could be 
considered a proxy for legitimacy: were the state to show up and do the things that the 
militias, for example, were doing, it was clear that it would be welcomed and supported. 
Few people wanted a nonstate actor to govern them, but in the absence of the state, they 
took what they could get, and if what they could get was prety good at governing, they 
were wiling to lend a degre of legitimacy to that efort. Were the state to put down roots 
in their communities instead, it would find fertile ground. 
 
                                                
22
 Eduardo Posada Carb?, ?Ilegitimidad? del Estado en Colombia. 
Lamb, Robert D. | Chapter 10. Implications 412 
Chapter 10. Implications 
10.1. Sumary of Findings 
The preceding chapters have demonstrated that the relationship betwen 
legitimacy and violence in Medel?n has been shaped by isues such as the manner of 
governance (whether or how the rule of law as maintained or the social order 
promoted), aces to resources (relative to rivals? aces), and the relative costs of 
teritorial control. The findings described at length in Chapters 8 and 9 may be 
summarized as follows: 
? Violence fel once someone capable of keeping order was in charge. That is, 
most decreases in collective and interpersonal-communal violence, in both 
micro-teritories and the city as a whole, were explained by increases in 
teritorial control. 
? Violence rose when inefective, unpredictable governors atracted violent 
opposition. More precisely, most increases in collective violence were 
explained by a proces of ?ilegitimation,? in which an unpredictable living 
environment sparked internal opposition to local rulers and raised the costs of 
teritorial control, which increased rulers? vulnerability to rivals. 
? Most of the city?s violence was due to common crime rather than organized 
crime (i.e. more interpersonal-communal violence than collective violence). 
Lamb, Robert D. | Chapter 10. Implications 413 
? But the violence due to common crime was enabled by the violence due to 
organized crime: most increases in interpersonal-communal violence were 
explained by the breakdowns in social order and the rule of law brought about 
by collective violence asociated with teritorial control (e.g. gang wars and 
mafia wars). 
? Most of those engaged in organized violence today are in it for the money: the 
true believers of the armed political and social movements of the past have 
been marginalized or corrupted over time. As a consequence of this gradual 
shift in motives from a bias for the political and social toward a bias for the 
economic, the complexity of the city?s violence has diminished over time. 
? Legitimacy has lowered the costs of teritorial control: relatively resource-
poor nonstate armed actors have won and held teritory against rivals mainly 
by strategies of legitimation (winning the support of residents). 
? Ilegitimacy has raised the costs of teritorial control: relatively resource-rich 
nonstate armed actors have been able to win and hold teritory against rivals 
using force, coercion, and barter but otherwise engaging in strategies of 
ilegitimacy-avoidance (maintaining a tolerable and predictable daily living 
environment). 
? The most important factors driving ilegitimacy have involved an intolerable 
unpredictability in the daily living environment; the main features of 
unpredictability have been a deficit of transparency (acurate, corect, and 
comprehensive publicity of the rules, rights, duties, and identities of those 
Lamb, Robert D. | Chapter 10. Implications 414 
who are in control) and a deficit of credibility (capable and non-arbitrary 
enforcement or fulfilment of rules, rights, and duties). 
These findings, more than any others, acount for most of the dynamics observed 
in dozens of micro-teritories and statelets within Medel?n (including in Caicedo La 
Siera) across al of the time periods studied and (during the 2003-2007 ?Miracle? period) 
in the city as a whole. 
10.2. Implications for Policy 
These findings provide a glimpse of what it might take for policy makers to 
succed in sustainably reducing violence in places such as Medel?n, and here I interpret 
these findings as policy recommendations. The temptation is always to alow the scope of 
one?s recommendations to exced the scope of one?s findings; I have endeavored to resist 
that temptation and so offer only a limited set of recommendations that can be derived 
directly from the main observations presented above. 
10.2.1. Protect and Respect Residents of Peripheral Barrios 
Violence in the city?s peripheral barios increased dramaticaly during the first 
half of 2009. The murders semed mainly to be instances of collective-economic violence 
(mainly drug trafickers against drug trafickers, plus innocent bystanders), which is 
consistent with the findings that collective violence has become increasingly motivated 
by economic rather than political or social concerns. But the murders sem also to be 
related to disputes over control of resources rather than direct control of teritory: that is, 
they sem to be fighting to control local drugs and rackets markets or local aces to 
Lamb, Robert D. | Chapter 10. Implications 415 
global ilicit markets ? wholesale drug markets, smuggling routes, etc. ? but not to 
control and govern entire neighborhoods where those resources are present. If true, and 
several sources have suggested it is (se Chapter 7), then this is somewhat diferent from 
what has been found in the past, when teritorial control was often considered necesary 
to control the economic resources it contained. 
This observation, however, does not directly chalenge the main findings of this 
study. If we asume (as I think we rightly can) that state actors are the ones who currently 
(albeit tenuously) control the teritories where most of the violence is taking place, but 
that their control is beginning to break down, then we can se, based on the city?s history, 
what wil probably happen if that breakdown gets to a point where residents start feling 
that life has become intolerably unpredictable: the more the current wave of violence 
afects their daily routines, those residents wil ilegitimize and begin to oppose the state 
actors who claim to be in control, and that opposition wil dramaticaly and quickly 
increase the costs of controlling those barios, making the state actors vulnerable to 
atacks by nonstate armed actors who do not want to control the teritory but who do not 
want the state to control it either. In other words, if the state ilegitimizes itself to its 
citizens in those barios, it wil end up facing opposition from both its own citizens and 
those nonstate armed actors (thereby increasing collective violence), which wil make it 
harder to maintain order (thereby increasing interpersonal-communal violence). If that 
happens, Medel?n wil be back to where it was before 2002. This outcome is not 
inevitable, because I don?t think the city has gotten to the point of ilegitimation, even if it 
is geting very close to unpredictability. 
Lamb, Robert D. | Chapter 10. Implications 416 
To prevent this outcome, the state should consider itself resource-poor relative to 
the potential resources available to potential rivals (drug trafickers): given the magnitude 
of the problem of violence historicaly, the state today realy does not have the resources 
it needs to fully control the peripheral barios and protect them against armed trafickers. 
Acording to the findings of this study, resource-poor actors have only ever succeded in 
Medel?n when they have engaged in strategies of legitimation: winning the voluntary 
support of the communities. That would be the ideal strategy for city and state actors. 
(Se ?? 9.2, 10.3, and C.6 for observations about what legitimacy-building strategies 
might entail, including criteria for measuring progres.) 
Given the city?s history, however, it is very unlikely that enough state actors ? in 
city hal, the mayor?s office, the baracks, or the front lines ? would work in solidarity 
toward a goal of legitimation long enough to make any such strategy work. Therefore, I 
focus here on a somewhat higher-risk recommendation that asumes the ilicit actors who 
are the main perpetrators of today?s collective-economic violence wil continue to be 
engaged in a contest for control over markets rather than over teritory (at least in the 
short term) and wil continue to target the state?s control over teritory only to the degre 
the state?s weaknes gives them fre aces to those ilicit resources. If they do not 
contest the state?s control of teritory directly, then the state should be able to hold 
teritory in the short term with a strategy of ilegitimacy-avoidance. 
As the violence in the city has begun to return to its formerly high levels, city and 
national security forces have been sent in to the peripheral barios to keep order. To 
follow an ilegitimacy-avoidance strategy, the imediate priorities of the oficials who 
manage or have oversight of those security forces should be: 
Lamb, Robert D. | Chapter 10. Implications 417 
1. to ensure that residents of those barios know al relevant laws, their rights and 
obligations under the law, and the rights and obligations (e.g. the rules of 
engagement) of the security forces patrolling or raiding their neighborhoods; 
2. to strictly enforce rules against corruption within the security forces, and to 
prosecute to the fullest extent of the law any violations of those rules; these 
first two recommendations are intended to promote transparency, which is one 
of the two components of predictability; 
3. to give the security forces the personnel, resources, and training they need to 
protect residents? rights and enforce the law hile engaging in operations 
against nonstate armed actors; this recommendation is intended to promote 
credibility, the other component of predictability; and 
4. fully empower, support, and expand the local Imediate Atention Command 
units (CAI: Comandos de Atenci?n Inmediata) to act as intermediaries 
betwen residents and security forces to ensure that residents do not turn to 
nonstate armed actors to resolve their grievances as they have in the past; this 
recommendation is intended to improve the state?s credibility relative to that 
of the nonstate armed actors. 
These recommendations, if implemented, should help to maintain a predictable 
and tolerable quality of life and avoid any ilegitimizing actions that would raise the costs 
of control, make the state more vulnerable to rivals, and begin a new cycle of violence. It 
should be noted, however, that, if the nonstate armed actors who are the target of security 
forces? operations decide to directly contest control of the peripheral barios? teritories 
and not just their resources, then the state should recognize that its own resources wil be 
Lamb, Robert D. | Chapter 10. Implications 418 
inadequate for the task and that it wil need to enlist the active and voluntary support of 
the community. 
10.2.2. Protect Busineses from Protection Rackets (?Vacunas?) 
Along with drug sales and asasination contracts, protection rackets have 
historicaly been among the most important sources of income for nonstate armed actors 
who have controlled micro-teritories and statelets within Medel?n. Not only has it 
funded their control, and thereby facilitated their other ilicit activities, it also has created 
a hostile and unpredictable environment for local busineses and has restrained the city?s 
potential for much-needed economic development. Making it a priority to protect local 
busineses from extortion not only would contribute to the creation of a predictable 
operating environment and thereby contribute to an ilegitimacy-avoidance strategy but 
would also send a signal to busines and residents alike that the state is the armed actor 
most worthy of their support and thereby contribute to any strategy of legitimation that 
would be needed if (or when) any nonstate actors decide to contest control of the 
periphery again. 
10.2.3. Prepare for ?Invasions? by Internally Displaced Persons 
The history of Medel?n ? as in many Latin American, African, and Asian cities 
? shows that violent conflict and economic depresion in the countryside leads 
inevitably to ilegal urbanization, as the rural poor arive in cities they cannot aford to 
live in and therefore engage in squating, ?bario piracy,? and ?invasions.? The patern is: 
they arive, they squat, they lack resources and so they pirate them (electricity, water, 
Lamb, Robert D. | Chapter 10. Implications 419 
etc.), they lack security and so they provide it themselves (via gangs or self-defense 
groups), they are unrecognized by the city and so they are left to fend for themselves 
economicaly, and finaly when their number is too high to ignore or the violence in their 
setlements begins to leak out into ?legal society,? the city finaly recognizes them, gives 
them title to their lands, provides some services ? but stil largely leaves them to fend 
for themselves, until, again, the negative externalities become too severe to ignore. 
This has happened repeatedly in Medel?n and has long been among the main 
drivers of social instability in the city?s periphery (se Chapters 2-3). It continues to 
happen today (for example, in setlements just outside the legal boundary of the city?s 
northeastern zone), and it wil happen again in the future. So instead of doing ad hoc and 
post hoc planning, the city?s planners should premptively ?legalize? setlements before 
they are setled, by finding the areas of the city most likely to be setled by internaly 
displaced persons in the future and planning for their rapid development when the time 
comes. This policy would be a buffer against future disruptions of social order and would 
prempt any nonstate eforts to control teritory that the city has not already claimed for 
itself. 
10.2.4. Kep Data-Collection Programs Funded and Transparent 
As Chapter 8 demonstrated, basic data are not available for many variables that 
might be important to understanding the problem of violence in Medel?n. This gap in 
data has been addresed recently in city policy: Medel?n C?mo Vamos and other data-
collection eforts begun under the administration of Sergio Fajardo have been 
enormously helpful for city officials doing long-term planning for development, anti-
Lamb, Robert D. | Chapter 10. Implications 420 
violence, and other eforts. These projects should be given a reliable funding source so 
that they may continue indefinitely. Further, regulations requiring transparency of data-
collection methods should be strengthened, and al collected data should be made frely 
available online, disaggregated at the lowest level consistent with appropriate privacy 
protections. These steps wil enable outside researchers to act as ?force multipliers? in the 
analysis of the city?s most presing problems, including violence. 
10.3. Implications for Strategy 
This study?s findings are suggestive not only of the policy recommendations 
above but also of several general propositions that merit further study in terms of their 
implications for strategies to implement anti-violence policies. These general 
propositions are as follows: 
? Any strategy of violence reduction should begin with a capability to prevent 
rival actors from governing; that is, it should begin with control, especialy 
teritorial or social control. 
? In any efort to atain or sustain teritorial control, resource-poor actors should 
engage in a strategy of legitimation while resource-rich actors should engage 
in, at minimum, a strategy of ilegitimacy-avoidance. 
? The esential element of an ilegitimacy-avoidance strategy is predictability 
(in both action and outcome), and two of the main contributors to 
predictability are transparency and credibility. As noted, transparency 
involves acurate, correct, and comprehensive publicity of the rules, rights, 
duties, and identities of those who are in control, and credibility involves 
Lamb, Robert D. | Chapter 10. Implications 421 
capable and non-arbitrary enforcement or fulfilment of rules, rights, and 
duties. 
? Tenuous teritorial control, if it is not contested, can be strengthened by 
greater resources or greater legitimacy. 
More broadly, this study suggests some considerations that should be acounted 
for when dealing with isues related to what is variously caled state failure, state 
weaknes, or state fragility. Colombia has long been considered a fragile state, with weak 
governance and weak legitimacy, and the dynamics in Medel?n?s periphery have long 
been considered a symptom of that fragility. In such areas, when some nonstate armed 
actor has ?taxed? local busineses, patrolled their neighborhoods, resolved disputes, lent 
money, fed poor families, helped their neighbors pirate electrical service from the public 
grid, let people in the neighborhood know what the rules are, and punished those who 
broke them, what those armed actors were doing in those areas was governing. They 
might not have been doing it wel or fairly, but they were governing. Were they the 
?legitimate? government of the areas they controlled? It depends: Did the people who 
lived there believe the militia or the gang was worthy of their support? Some did, and 
some did not. But what was certain to most of them ost of the time was that the state 
was not the legitimate government, because it did none of the things the militia or the 
gang did: the state did not govern them, and so it had no legitimate claim to their loyalty. 
This suggest a general observation. Fragile states are fragile for a reason; 
?ungoverned? areas are not under control of the state for a reason. That reason usualy 
has to do with political relationships betwen the social elites who happen to control the 
state institutions, and the other populations who happen to live in the state?s nominal 
Lamb, Robert D. | Chapter 10. Implications 422 
teritory, such as the residents of Medel?n?s peripheral barios. If certain populations do 
not identify with the state or its system of governance, it often is because the state ? 
whether from lack of capacity, ethnic enmity, or historical mistakes of cartography ? has 
never governed them, has failed to protect them, has not advanced their interests, has 
exploited their resources, or has outright harmed them. Consequently, local power 
structures ? family, community, tribe, clan, gang ? often end up with stronger claims to 
the locals? loyalty and support than does the state: from their perspective, the legitimate 
governing structure is whatever entity governs them in a way that they consider right or 
worthy. In fact, a social, political, or jurisdictional dispute betwen the nominal 
government of a state and some local, provincial, tribal, or autonomous government is 
often precisely the factor that enables certain ilicit actors to operate with impunity in 
such places, hiding betwen gaps in governance and legitimacy. 
In many places, of course, people who do not have any real relationship with the 
state want nothing more than for the state to protect them, to educate them, to pave their 
roads, to provide electric and water services, to help them start busineses. To say that the 
authority of the state has been eroded by globalization (se Appendix A) is not to say that 
states no longer represent the last, best hope of many people. Many indigenous people in 
southern Colombia have wanted to be left alone by the state to maintain their native way 
of life; when that way of life has been threatened by the encroachment of insurgents and 
narcotrafickers, they often have wanted the state to help protect them ? but to do so in a 
way that would enable them to maintain their autonomy, not in a way that would be 
exploitive. Likewise, many residents of the peripheral barios of Medel?n in the 1980s 
and 1990s protected themselves by joining gangs or forming self-defense groups: they 
Lamb, Robert D. | Chapter 10. Implications 423 
had been neglected by the state, and their only interaction with the state was during police 
raids, many of which turned into masacres. But many residents nevertheles held out 
hope that the police would build a station in their neighborhood and go on patrol to 
protect them, or that the city would help resolve conflicts betwen gangs. Sometimes 
their pleas worked; sometimes they had to depend on NGOs or the church to play that 
role. But they would have prefered a ?normal? life, recognized and protected as full 
members of the Colombian citizenry. 
In strategies aiming toward long-term teritorial control and stability, many policy 
makers increasingly recognize the centrality of legitimacy-building. What this study 
suggests ? and this is a proposition that merits further study ? is that legitimacy-
building entails governing: sometimes the prize goes to whomever shows up. If state 
actors want to control teritory, if resources are scarce, and if mere ilegitimacy-
avoidance is not likely to succed, then legitimacy-building wil most likely become the 
key strategy. This study suggests that legitimacy-building likely involves state actors? 
recognizing the legitimacy of the populations living in the teritory in question (se 
? 9.3.2) and, above al, actualy governing them. 
10.4. Implications for Theory 
Likewise, this study?s findings suggest several propositions that merit further 
study with respect to their relevance to theory: 
? The distinctions this study made among legitimacy, ilegitimacy, and 
neutrality, and among delegitimation, ilegitimation, and neutralization, merit 
theoretical elaboration. 
Lamb, Robert D. | Chapter 10. Implications 424 
? The negative correlation (within certain bounds) that this study found betwen 
complex urban violence and teritorial control partly supports existing conflict 
theory but needs further testing of asumptions. It tends to confirm the work 
of Kalyvas, but Kalyvas?s work addreses traditional civil wars, not complex 
urban violence. Therefore, further study is merited to determine the full sets of 
circumstances under which the relationship in question is operative. 
? The negative correlation found betwen teritorial control and ilegitimacy 
supports existing legitimacy theory (se Appendix C) but needs further testing 
in cases outside of Medel?n. 
? The positive correlation found betwen ilegitimacy and unpredictability 
represents a new proposition that merits further study. 
? The proces aspect of legitimacy merits further study (se ? C.5). 
? The distinction betwen long-term teritorial control and short-term teritorial 
control should be further specified. A preliminary specification could be as 
follows: for terms of teritorial control, the short-term ay be defined as the 
period during which either legitimacy or resources are fixed, and long-term 
may be defined as the period after which either could be adjusted. In the short-
term, teritory can be held either: (1) by an actor with the ability to raise 
resources but with low support (fixed legitimacy) or high opposition (fixed 
ilegitimacy); or (2) by an actor with fixed resources but the ability to raise 
legitimacy; but not (3) by an actor with fixed resources and fixed legitimacy 
or fixed ilegitimacy. To reach the long-term, the actor would have to make 
Lamb, Robert D. | Chapter 10. Implications 425 
some structural adjustment such that new resources become acesible or 
legitimacy becomes feasible. This proposition, however, merits further study. 
? Hybrid categories of violence should be considered for inclusion in 
frameworks for future work on complex violence. The violence framework 
used in Chapter 8 and discussed in Appendix B provides a finer-grained look 
at violence than is usualy provided in studies of violence. Yet there are 
important cases it does not capture wel because they fal betwen categories, 
for example betwen communal and collective. The most important of these is 
violence that is perpetrated by an individual who does not have a direct or 
two-way relationship with a larger group or movement but is nevertheles 
inspired by that movement to act violently: such violence is not collective 
(since the individual is not acting on behalf of a group), but it is not merely 
interpersonal either (since is inspired by collective and not merely individual 
motives). It might therefore be worth considering adding a hybrid category of 
?communal-collective? violence, or perhaps ?networked? violence, to acount 
for those organizations that operate on an ?open-source? model ? perhaps a 
violent version of Anonymous, the anti-Scientology movement started by a 
smal group of people who posted a video on YouTube encouraging certain 
kinds of actions (and discouraging violent actions) to disrupt that 
organization, with the ultimate aim of dismantling it
1
 ? or to acount for 
those individuals who take inspiration from some larger movement and 
                                                
1
 Se, for example, Marcus Baram, ?Scientology?s Anonymous Critics: Who Are They?? published 
electronically by ABCNews.com, http:/abcnews.go.com/US/story?id=4513883 (acesed 3 September 
2009). 
Lamb, Robert D. | Chapter 10. Implications 426 
become ?copy cats? of others in the movement without ever actualy 
communicating with them or taking orders from them. 
? A final proposition that merits further study is the likelihood that legitimacy is 
a path-dependent phenomenon; if it is, then its path-dependence should be 
acounted for in models that incorporate some measure of legitimacy. For 
example, in the adaptive-agent model of civil violence produced by Joshua 
Epstein, John D. Steinbruner, and Miles T. Parker, grievance, G, is a function 
of both hardship, H, and perceptions of legitimacy, L; and legitimacy is 
considered a constant (it doesn?t need to be, but the authors do so for the sake 
of simplicity).
2
 If G=H(1-L), as that model proposes, then L=1-G/H. In that 
case, one can predict that someone with few grievances and a lot of hardship 
must think the authorities have a lot of legitimacy, otherwise they would rebel. 
That makes sense. But with this equation, you can?t predict what they think 
about legitimacy if they have few grievances and litle hardship: this equation 
would suggest that they think the authorities have a lot of legitimacy, but 
unles you know what their grievances and hardship levels were before, you 
wouldn?t necesarily come to this conclusion: If, for example, they once had 
few grievances and a lot of hardship, but now they have the same grievances 
but litle hardship, why would we now think that they believe the authorities 
have les legitimacy, as this equation would suggest: shouldn?t they have 
more? Maybe so and maybe not; either way, it should be demonstrated rather 
                                                
2
 Joshua Epstein, John D. Steinbruner, and Miles T. Parker, ?Modeling Civil Violence: An Agent-Based 
Computational Aproach,? working paper, Center on Social and Economic Dynamics Working Papers 20 
(Washington: The Brookings Institution, January 201). 
Lamb, Robert D. | Chapter 10. Implications 427 
than asumed. By looking at legitimacy over time, the phenomenon might be 
beter understood. 
Lamb, Robert D. | Part II. Apendices 428 
Part II. Apendices 
 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 429 
Apendix A. Policy, Globalization, and Governance 
This study was undertaken, ultimately, to help policy makers and implementers 
beter understand what can and cannot be done to control complex urban violence. To 
that end, it is necesary not merely to understand certain complex phenomena, urban 
dynamics, and paterns of violence; it is equaly important to understand the context, 
chalenges, and limitations of policy making itself as an instrument of change. For that, 
certain questions have to be answered: What is policy? Who makes policy? Through 
what governance structures is policy made? How are those structures changing, and what 
are the causes and consequences of those changes? How does globalization afect policy 
making and policy makers? To what ends can states no longer efectively or legitimately 
addres their policies? Finaly, for a study of policy as an instrument for controlling 
violence, what kinds of policies should be studied, and how should they be measured? 
These questions are addresed in this chapter. The concepts defined and discussed here, 
and the analysis presented in the preceding chapters, are mutualy reinforcing: the 
Medel?n case provides a local context for how these phenomena operate in the real 
world, while the discussion of the phenomena provides a global context for how the 
events in Medel?n unfolded. 
As globalization erodes the monopoly that states were once expected to have on 
force, loyalty, and the provision of political goods and the regulation of social relations, 
policy makers on the payroll of a state (and the people who have to implement their 
policies) are finding their work to be a growing chalenge. Questions of governance and 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 430 
legitimacy, once comparatively straightforward, are blurring the boundaries among 
important policy isues, creating uncertainties regarding which institutions have the 
appropriate authorities and capabilities to addres those isues, and complicating the 
ability of those institutions even to formulate the questions their policies are meant to 
addres ? much les to formulate efective policies that have a chance of succeding in 
the real world. Some isues that once were in the purview of states (or their provincial or 
local governments) are today more efectively addresed bilateraly, or by international 
institutions, inter-governmental organizations (IGOs), international non-governmental 
organizations (INGOs), or non-governmental organizations (NGOs) that work at the 
community or national levels. Some isues simply are more efectively addresed by 
public-private partnerships, or through ?open-source? networks, by local communities, or 
even by private foundations and influential individuals than they are by states. 
The sovereignty of states has always been what is commonly termed a 
?convenient fiction?: states have always intervened in the afairs of other states, while 
proclaiming their respect for the norm of external sovereignty (non-interference); and 
many states have long been unable to atain or to maintain internal sovereignty (efective 
control) over their teritory or the people under their nominal jurisdiction. Yet over the 
past several decades, and acelerating during the 1990s, globalization has exacerbated the 
presures that have always chalenged the state system, making the fiction of sovereignty 
les and les convenient with each pasing year. 
Not only are a growing number of nonstate, substate, transstate, and stateles 
actors taking responsibility for governance functions once reserved exclusively for states, 
but a growing number of individuals and groups are shifting their primary identities and 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 431 
loyalties away from teritorialy defined states and toward local communities or tribes on 
the one hand and teritorialy dispersed social networks and social movements on the 
other ? thereby adding their number to those many communities around the world that 
have never identified with any state in the first place. The growing acesibility and 
afordability of global communication, transportation, and financial networks, which 
facilitate global commerce and make it easy for people to stay in touch with distant loved 
ones, also make it possible for ilicit actors to smuggle and trafic in human slaves and 
dangerous products, plan and implement terorist atacks, launder il-gotten finances, and 
co-opt or destabilize whatever formal governance structures threaten their activities. Most 
policy makers, meanwhile, are constrained to act within the teritorial boundaries of 
states, the functional boundaries of international organizations, and the moral boundaries 
of the constituents they purportedly represent. With few exceptions, ilicit actors are not 
so constrained. 
There is a sense among many policy makers and implementers that, within this 
dificult governance environment, legitimacy and ?good? governance ? perhaps in the 
context of human development, economic development, or iregular war, depending on 
the policy maker you?re talking to ? must somehow play a role in any sustainable 
solution to these problems. But legitimacy is a dificult concept: dificult to define, 
dificult to measure, and, in the real world, dificult to influence. And the combination of 
globalization and ilicit activity has shown ?bad? governance to be a sticky phenomenon 
in the environments where reform is said to be needed most. 
One of the purposes of this study was to clarify the relationships among 
legitimacy, governance, violence, and social or teritorial control in the context of a 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 432 
globalizing world, to help policy makers formulate relevant questions and efective 
answers. This appendix steps back from the analysis of the preceding chapters to discuss 
several concepts related to policy, governance, and control. Below, after defining key 
terms used throughout this work (? A.1), I discuss globalization and the chalenges it has 
created for policy making (? A.2), then define governance and discuss its growing variety 
and complexity amid the declining of the authority of the state (? A.3). 
A.1. Definitions 
A policy is a plan of action, or a statement about what one wil do under what 
conditions, how one wil do it, and, often, to what ends it wil be done. Policies can be 
formal (writen, legal, explicit, or created by institutions formed specificaly for policy 
making), or informal (verbal, customary, de facto, implicit). Rules, regulations, laws, 
strategies, statements of obligations and penalties, and the dictates of tribal or customary 
law are al examples of policies. So are declaratory policies, which are either declarations 
about what one wil do under what circumstances (i.e. the promulgation of policies) or 
policies about what one wil say under what circumstances (i.e. public afairs or public 
diplomacy). Rules and proceses for developing policies are themselves policies: policies 
about policy-making. Individuals, groups, organizations, governments, and other entities, 
whether they be state or nonstate actors, al are capable of developing, promulgating, 
implementing, and (it should be noted) mismanaging policies. A public policy is any 
policy that is developed and enforced collectively (as by a government), or that purports 
to confer a benefit or a right, or to impose a cost or an obligation, upon al members of a 
given group of people whether they were directly involved in its development or not. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 433 
The development, promulgation, implementation, and revision of policies al take 
place in specific social contexts; afect and are afected by those contexts; and often are 
the subject of intense conflict if not outright antagonism. Conflict is a natural feature of 
al human societies: diferent people and diferent groups have diferent abilities, needs, 
opinions, worldviews, values, interests, and priorities, and where these are not in 
acordance across diferent individuals and diferent groups, the result is conflict. The 
conflict may be over the distribution of wealth, power, or prestige; it may be over the 
content of policy, of memory and history, or of what counts as truth; it may be over 
goods that are divisible (an art collection, money) or goods that are indivisible (the 
economic system, the external security of the community); and it may be resolved 
violently or nonviolently. 
Conflict creates tension with human beings? innate desire for some degre of 
social order ? most of us want to be safe, most of us want to be more or les ?normal? 
acording to the standards of our reference groups, and most of us want life to be more or 
les predictable so that we can go about our day?s activities and cary out our life?s plans 
and projects without too many unpleasant or harmful surprises. (Pleasant surprises, of 
course, we generaly welcome, yet some people are unsetled even by pleasant surprises.) 
Social order emerges from interactions among individuals motivated to make life safe, 
predictable, and ?normal,? a proces through which ?conduct is defined as desirable or 
undesirable, approved or deviant, permited or prohibited.?
1
 Through the emergence of 
social order, some conflicts are prempted by defining the behavior that would give rise 
                                                
1
 ?Mediante el control social se definen las conductas deseables e indeseables, conformes o desviadas, 
l?citas o prohibidas ?? Hernando Le?n Londo?o Ber?o, Ricardo Le?n Molina L?pez, and Juan David 
Posada, ?Pol?tica Criminal y Violencia Juvenil,? 31. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 434 
to them as socialy unaceptable, and some conflicts are resolved by punishing behavior 
that violates those norms. Systems of social control evolve to shape and maintain this 
social order. Some social scientists distinguish active social control, through which 
norms are constructed and socialized by the institutions of society (family, school, 
church, media, recreation, etc.), from reactive social control, through which violations of 
those norms are defined ?as pathological, disapproved of, deviant, or criminal? and so are 
punished, either formaly by criminal justice institutions or informaly by social 
institutions and social presures (shaming, mocking, etc.).
2
 
Formal and informal systems of social control have evolved or been established in 
al societies, and while a great variety of social structures therefore exists, some general 
observations can be made about al of them: 
Social life is structured along the dimensions of time and space. Specific 
social activities take place at specific times, and time is divided into 
periods that are connected with the rhythms of social life ? the routines 
of the day, the month, and the year. Specific social activities are also 
organized at specific places; particular places, for instance, are designated 
for such activities as working, worshiping, eating, and sleping. Teritorial 
boundaries delineate these places and are defined by rules of property that 
determine the use and posesion of scarce goods. Additionaly, in any 
society there is a more or les regular division of labour. Yet another 
universal structural characteristic of human societies is the regulation of 
violence. Al violence is a potentialy disruptive force; at the same time, it 
is a means of coercion and coordination of activities. Human beings have 
formed political units ? within which the use of violence is strictly 
regulated and which, at the same time, are organized for the use of 
violence against outside groups.
3
 
                                                
2
 Ibid. 
3
 Encyclop?dia Britanica Online, s.v. ?social structure,? htp:/ww.britanica.com (accessed 2 February 
2009). 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 435 
Diferent societies and diferent social groups may resolve their conflicts diferently, may 
legitimize and use violence diferently, and may develop diferent systems of social 
control, but conflict, violence, and social control are universal features of humanity. 
Conflicts over two fundamental isues can arise in any group: who has control? 
and how do they exercise that control? In a sense these are meta-conflicts, conflicts over 
(among other isues) who gets to define how conflicts get resolved. The way these meta-
conflicts are resolved by any given human group fundamentaly defines the nature of that 
group?s social structure or political regime: how do they resolve conflicts over questions 
of control and governance? These questions are fundamental to the functioning of any 
human grouping, and they involve concepts that were central to this study: governance 
and control, states and statelets, and the relationships betwen and among state and 
nonstate actors at diferent levels of aggregation. 
Governance is the particular manifestation of collective behavior that involves 
policy-making, public-goods delivery, institution-building, and network management. 
Policy-making is geared toward regulating social, political, and economic relations. 
Institution-building and network management are geared toward the same goal ? to 
influence policy outcomes ? but operate under diferent asumptions: institution-
builders asume policy outcomes can be achieved through linear proceses managed by 
hierarchical organizations, while network managers asume policy outcomes are the 
result of often-complex interactions among heterogeneous actors, with diverse incentives 
and motivations, in the public, private, and voluntary sectors. Either can be the case, 
depending on the isue and the actors involved; in the real world, governance often 
involves both networks and hierarchies. A public good is technicaly anything whose 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 436 
production or delivery necesarily benefits everyone within a particular population 
because, once produced or delivered, it cannot be selectively withheld from any 
subpopulation unles it is withheld from the entire population (e.g. the only way to 
prevent one person on a stret from enjoying the benefits of a stret light at night would 
be to turn off the stret light entirely); a public good is sometimes said as wel to be 
something whose use or enjoyment by one person does not diminish its ability to be used 
or enjoyed by anyone else (e.g. if one person ?uses? a trafic signal, by looking at it to 
determine how safe it would be to proced through the intersection, that act does not 
prevent anyone else from similarly using it). 
Many things that are claimed to be public goods only approximate the various 
technical definitions that academics use, and so ?the delivery of public goods? often ends 
up, in practice, to mean ?the delivery of political goods? or, equivalently, ?the work done 
by governance institutions,? a standard list of which might include ?security, judicial, 
legal, regulatory, inteligence, economic, administrative, social, and political goods and 
public services.?
4
 It would be circular, of course, to define governance as ?the delivery of 
political goods? and define political goods in terms of those things that are delivered by 
governance institutions. But many academic concepts get translated by the public (and by 
policy makers) that way, and in any event the popular misinterpretation does not afect 
the definition used here, which focuses on the broader concept of public goods: 
Governance involves policy-making, public-goods delivery, institution-building, and 
network management. Governance institutions (security, judicial, legal, etc.) make rules, 
deliver public goods, and manage networks of influence. To say that some governance 
                                                
4
 Robert D. Lamb, Ungoverned Areas and Threats From Safe Havens, 17. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 437 
institutions also do other things under the guise of public-goods delivery is only to say 
that some governance institutions overeach; it does not mean that they are not 
governance institutions. 
Governance can be undertaken by state, nonstate, substate, and transstate actors 
alike, and it can be done wel or poorly; fairly or unjustly; democraticaly or dictatorialy; 
through linear or nonlinear proceses; through hierarchical organizations, markets, or 
networks; or through any combination of the above. In other words, governance should 
not be confused, as it too often is, with government, nor with ?good? or ?authoritative? 
governance, nor with any particular system of governance, such as state governance or 
democratic governance, nor with the more limited concept and practice of public 
administration. 
Given this definition, one can now say that to govern is to deliver public goods; to 
make, implement, and enforce policies; and to build and maintain the institutions through 
which, and to try to ster the networks through which, these activities take place. A 
government is any entity formaly (i.e. specificaly) constituted for the purpose of 
governing (e.g. local governments cary out certain governance functions in a given 
teritory; student governments govern students; etc.). A governor is any individual or 
group that actualy governs an area or a population; it need not be a government but can 
be an organization formed for some other purpose (such as a busines enterprise, as in old 
factory towns) that controls an area or a population and also caries out some governance 
functions. 
Control is the capability to prevent governance, as measured by ?the probability 
that a certain event or clas of events wil not occur within a defined area within a defined 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 438 
period of time.?
5
 It ?can be defined and measured empiricaly, using various indicators 
such as the level of, presence of, and aces enjoyed by political actors in a given place 
and time.?
6
 A gang controls a neighborhood if it can prevent the city (and other gangs) 
from entering and governing there (e.g. to provide police protection or social services); 
insurgents control a vilage if they can keep the state?s military forces out. Having the 
capability (or ?capacity?) to govern or to prevent governance does not necesarily imply 
that one actualy governs or prevents governance: control can be used to selectively alow 
governance as wel as to forbid it (e.g. the guerila militias of Medel?n?s peripheral 
barios invited state social workers and development profesionals into the barios they 
controlled, but chalenged the presence of state police forces). Teritorial control is the 
ability to prevent others from governing in a specific place (e.g. a gang can prevent the 
police from entering a bario). Functional control is the ability to prevent others from 
delivering some public goods and esential services but not others (e.g. a gang can 
prevent the police from entering a bario, but cannot prevent the people who live there 
from receiving public asistance deposited electronicaly into an acount acesible by 
debit card: the gang has control over security functions but not social-services functions). 
Institutional control is the ability to prevent others from using the controlled institution to 
govern (e.g. the state inteligence service has been ?captured? by individuals not 
necesarily loyal to the constitution, and the president and prime minister are unable to 
reliably make and implement inteligence policy as a result of their not having control 
                                                
5
 Jefrey Race, War Comes to Long An: Revolutionary Conflict in a Vietnamese Province (Berkeley: 
University of California Pres, 1972); quoted in Stathis N. Kalyvas, The Logic of Violence in Civil War, 
210. 
6
 Ibid. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 439 
over that institution). Temporal control is the ability to prevent others from governing a 
specific place, population, function, or institution during certain periods of time (e.g. the 
police control a neighborhood during the day, but a gang controls it at night). Finaly, an 
important aspect of social control, discussed earlier, is the ability to prevent others from 
governing a specific population or subpopulation (e.g. a religious leader claims the 
loyalty of a religious minority who obey his dictate not to comply with the state?s 
demands; or a union leader selectively delivers or withholds the votes of its members). 
One form of social control is the ability to prevent a population from governing itself 
(e.g., the governor imprisons, executes, or ?disappears? certain political activists; or 
criminal and gang activity in a neighborhood undermines the authority and capabilities of 
social institutions to control vice and deviant behavior). 
A policy event is a significant occurrence that takes place as a result of a policy, 
whether the event or its outcome had been intended or not, and regardles of the identity 
of the policy maker (the city, a militia, the chamber of commerce, etc.). Examples of 
policy events include military operations, development projects, hostage releases, 
legislation, gang wars, elections, the publication of an article or opinion poll, a tipping 
point or critical mas in an emergent phenomenon, or the development and promulgation 
of a policy. 
A control policy event is one in which an actor gains, atempts to gain, or 
announces the intent to gain power at the expense of some other actor; examples might 
include a military operation to eject a militia (even if it does not succed), or an 
asasination of a rival gang leader. The discussion above listed five kinds of control: 
teritorial, functional, institutional, temporal, and social. Al teritorial disputes are 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 440 
control events. Corruption scandals or the infiltration of ilicit actors into positions of 
power are policy events related to institutional control. Threats to community or church 
leaders are maters of social control; the forced displacement of a family afects both 
social and teritorial control. Personnel changes ? hiring more police or public school 
teachers, stepping up recruitment or conscription ? can change the balance of functional 
or temporal control. The circulation of a pamphlet seting a 10 p.m. curfew ? a common 
occurrence in Medel?n ? is an example of temporal control. Control is a prerequisite for 
governance (by definition: control is ability to prevent governance): if you do not control 
a place (or an institution, or a function, etc.), you cannot govern it. 
A governance policy event is one in which an actor uses, tries to use, or 
announces the intent to use power to deliver public goods or esential services to 
residents; to make and enforce rules in the community; to build institutions; or to 
influence the networks that influence policy outcomes. Examples might include the 
announcement of a development project, the establishment of an emergency response 
system, or an election. As the discussion above noted, governance has a number of facets 
(policy-making, institution-building, network management, and public goods delivery) 
and some definitions have also speled out lists of functions (security, law-enforcement, 
justice, economic regulation, esential services, disaster response, etc.). 
Control implies a concentration of power. Governance implies a dispersion of the 
benefits of power. (Se the discussion of ?controllers? versus ?governors? in ? 9.3.2.) 
Social control is a special case: when it is maintained by social institutions within the 
community it more closely resembles self-governance, but when it is maintained by an 
identifiable subgroup, especialy one that operates by preventing self-governance, it more 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 441 
closely resembles control. Many events, in fact, are likely to have the character of both. 
For example, instaling a police station in a particular bario is a clear signal of an intent 
to control it, but it also is there to facilitate the protection of residents, a governance 
function; on the other hand, if the police posted there operate in such a way that the safety 
of residents is routinely sacrificed or disregarded during searches for insurgents or gang 
leaders, then its presence would tend to be an indicator of control rather than governance. 
Protecting privileges is control; protecting residents is governance. 
A state, pragmaticaly speaking, is the government, bureaucracies, ofices, 
institutions, and so on that, collectively, are internationaly recognized as the de jure 
sovereign ruler of a particular teritory (e.g. as a member of the United Nations). 
Acording to customary international law, a state is an entity that posseses ?(a) a 
permanent population; (b) a defined teritory; (c) government; and (d) capacity to enter 
into relations with the other states.?
7
 In Colombia, for example, the state includes al 
documented citizens; the physical teritory in the northern Andean region of South 
America; and the national, departmental, and local governments, as wel as the courts, the 
military, the police, government social service agencies, public transportation agencies, 
and other government offices. Charles Tily provides a useful characterization of a state 
as being an entity whose agents 
characteristicaly cary on four diferent activities: (1) War making: 
Eliminating or neutralizing their own rivals outside the teritories in which 
they have clear and continuous priority as wielders of force. (2) State 
making: Eliminating or neutralizing their rivals inside those teritories. 
(3) Protection: Eliminating or neutralizing the enemies of their clients. 
                                                
7
 ?Montevideo Convention on the Rights and Duties of States.? 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 442 
(4) Extraction: Acquiring the means of carying out the first thre 
activities ? war making, state making, and protection.
8
 
Agents or actors (here I use the terms interchangeably) are people (whether 
working alone, in informal groups, or in formal groups such as agencies or organizations) 
who work for, represent, govern, or control some entity (group, organization, institution, 
corporation, etc.) within a social structure or a network. Examples of agents include gang 
leaders, governors, treasurers, detectives, arms traders, social workers, teachers, soldiers, 
bureaucrats, financers, etc. State actors, or state agents, are the agents of a state. Nonstate 
actors, or nonstate agents, are agents of nonstate entities such as NGOs, militias, 
companies, mafias, insurgent groups, etc. Substate actors, or substate agents, are the 
agents of local, provincial, tribal, autonomous, or other governments or governors at a 
scale below that of the state (e.g. they control or govern either a teritory that is smaler 
than, and fully within, a state?s teritory, such as a province or a statelet, or some sub-
population within the state); substate actors may be subsidiary to (and therefore part of) 
the state, or they may be nonstate actors separate from or competing with the state (e.g. 
the governors or agents of a statelet). Transstate actors operate in, exist in, or move 
among the teritories of multiple states; more commonly caled transnational actors and 
sometimes caled stateles actors, they may be individuals, groups, governments, 
governors, or any other kind of entity; they may operate on a contiguous teritory that 
spans the borders of two or more states (as the FARC have at Colombia?s border with 
Venezuela), or they may operate as geographicaly dispersed networks, cels, or 
movements (as the Cacique Nutibara Bloc operated in Medel?n). Ilicit actors are groups 
or individuals who ?use[] or incite[] armed violence (or who asist[] those who use or 
                                                
8
 Charles Tily, ?War Making and State Making as Organized Crime,? 181. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 443 
incite armed violence) for political or private gain in ways that threaten?
9
 other people; 
they can be nonstate, substate, transstate, or even state actors (such as ?rogue elements of 
the state, for example, a police force that has been ?captured? by organized crime?
10
). 
Examples of ilicit actors include actors commonly labeled as terorists, insurgents, 
trafickers, smugglers, pirates, money launderers, extortionists, kidnappers, and so on. 
A statelet is an entity that does not enjoy any recognition as a state but whose 
agents nonetheles cary out some form of war making (keeping the state out), state 
making (keeping rivals out), protection (of local alies and constituents), and extraction 
(taxes, smuggling, extortion, etc.) in a particular teritory with a particular population. A 
statelet is the de facto governor of a teritory that is under the nominal jurisdiction of one 
or more recognized states that, in fact, lack the capacity to actualy govern or control it. 
Examples include Somaliland, Transdneister, Gaza, and the tribal lands of the Pakistan-
Afghanistan border region. Neighborhoods or housing developments in the inner cities of 
some otherwise strong states, places wel known as ?no-go? zones for police, have 
operated esentialy as statelets. In Colombia, there have been many statelets, both in 
rural and urban areas, where the state has ceded control, either by design or by default, to 
a militia, paramilitary, or guerila organization, some of which have actualy governed as 
wel as controlled the area in question. The teritories in Caicedo La Siera controled by 
M-6&7/Metro Bloc and La Ca?ada/Cacique Nutibara Bloc were statelets. Statelets can be 
governed strongly or weakly, wel or poorly, and legitimately or ilegitimately, just like 
states. 
                                                
9
 Robert D. Lamb, Ungoverned Areas and Threats From Safe Havens, 15. 
10
 Ibid. at note 7. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 444 
Some authors and policy makers refer to statelets as ungoverned areas, but this is 
often inacurate. Any area that looks ungoverned from the outside is more commonly 
characterized in reality by one of thre situations: either the ?ungoverned? area is in fact a 
collection of very smal statelets whose agents have setled into stable arangements with 
their neighbors regarding who wil control what areas; or the ?ungoverned? area is in fact 
a contested area, a zone of conflict where two or more actors are actively vying for 
control, usualy violently; or the ?ungoverned? area is in fact governed by a state, but it is 
governed weakly, poorly, iresponsibly, or in a way outsiders simply do not like. 
Finaly, subsidiarity has to do with the relationship betwen higher and lower 
levels of governance, such as that betwen state and substate governments or betwen 
state and substate actors (as such, the term is closely related to federalism and 
decentralization). As a normative principle, subsidiarity prescribes a rebuttable 
presumption that any policy decision should be made at the lowest governance level that 
is competent to make and implement it (i.e. higher levels are subsidiary to lower levels). 
As a practical mater, it refers to the principles (whether explicitly negotiated or 
implicitly acepted) that guide agrements over which levels or actors of government wil 
have jurisdiction over which functions of government. The term originated in the 
Catholic church but is most commonly used today in the context of European Union (EU) 
treaties that limit the jurisdiction of EU rulemaking and recognize the jurisdictional rights 
of EU member states. Despite its restricted use today, the underlying concept can usefully 
be generalized and applied to the analysis of the dynamics of governance in any number 
of situations. Many fragile states, for example, are fragile precisely because the society in 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 445 
question has not established a commonly acepted set of subsidiarity principles to 
regulate relations betwen center and periphery. 
A.2. Policy and Globalization 
State-based policy-making today faces two basic chalenges: First, the problems 
that many state-generated policies are meant to addres ? such as the international 
narcotics trade that has been a driver of much of Colombia?s violence ? have sources 
and consequences that no longer (if they ever did) respect administrative boundaries 
betwen and within states. As a consequence, state policy makers are forced either to 
cooperate with policy makers from other jurisdictions and from nonstate organizations, or 
to concede their incompetence to make eficient or efective policy on some isues. 
Second, the people over whom state policy makers believe they have authority ? such as 
the residents of Medel?n?s peripheral barios ? do not necesarily share that belief to the 
degre they once did; many consider themselves instead to be subject to the rights and 
responsibilities defined by their religious communities, their tribal leaders, or 
international law, to take just thre examples, and not by the state, or not primarily by the 
state. Both chalenges arise as a consequence of globalization, and it has only been in the 
past few years that these chalenges have even been recognized as such. In this section I 
briefly criticize those who have ignored globalization and the nonstate actors it has 
empowered, then discuss the chalenges that globalization increasingly poses for policy 
makers. 
In 2001, a few months before the United States suffered 3,000 casualties in a 
coordinated atack by four separate teams of terorists, John Mearsheimer?s The Tragedy 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 446 
of Great Power Politics was published. In his discussion of the causes of war, 
Mearsheimer explained that states, or at least great powers, strive toward regional 
hegemony ?because hegemony is the ultimate form of security,? then followed that 
statement with the astonishing claim that ?there are no meaningful security threats to the 
dominant power in a unipolar system.?
11
 The term terorism did not appear in the book?s 
index. His neglect of nonstate threats folowed from the thre ?core? realist asumptions 
he cited: 
1. states are the principle actors in world politics, but realists ?focus mainly on 
great powers ? because these states dominate and shape international politics 
and they also cause the deadliest wars?; 
2. the structure of the international system is the main influence on the behavior 
of great powers, not the ?internal characteristics? of the states; and 
3. power calculations ?dominate states? thinking? as they ?compete for power 
among themselves.?
12
 
Mearsheimer was correct that history?s great power wars have been enormously 
destructive. But to argue that great powers have nothing to fear from any source other 
than other great powers was to walk with blinders through recent history. It is true that 
states are the primary actors on the world stage and great powers are the stars of the 
show. However, a theatrical production can be ruined not only by the mistakes of lead 
actors, but also by minor actors turning in a particularly bad performance, and even by 
disruptive audience members. Great powers often do have other great powers to fear, but 
                                                
11
 John J. Mearsheimer, The Tragedy of Great Power Politics (New York: Norton, 201), 345. 
12
 Ibid., 17. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 447 
not always, and not exclusively. Sometimes leser powers act with imunity in 
opposition to the national interests of great powers ? how were the OPEC states made to 
suffer during the oil crises of the 1970s, when high energy prices harmed great-power 
economies? ? and sometimes nonstate actors do the same. The point is taken that the 
9/11 atacks, while tragic and disturbing, kiled far fewer Americans than died in the 
Vietnam War, the Korean War, either of the World Wars, or certainly the American Civil 
War; historicaly, wars have indeed been more destructive than terorist atacks. And, yes, 
realism is a theory of state actions, not of the behavior of nonstate actors. But so much 
the worse for the theory if it fails to look at states not only as actors in international 
afairs but also as the acted-upon. And so much the worse for a theorist who lacks the 
imagination to anticipate the mas of casualties that could result from terorists? 
acquisition of weapons of mas destruction ? or from the target society?s overeaction to 
a smaller-scale terorist atack ? or from the smal arms that transnational criminals 
deliver to conflict zones worldwide. 
Hans Joachim Morgenthau proves a partial corrective to Mearsheimer?s parochial 
outlook. While acknowledging ?interest defined as power? as the ?perennial standard by 
which political action must be judged and directed,? and the nation-state as the ?ultimate 
point of reference of contemporary foreign policy,? Morgenthau insisted that ?the 
contemporary connection betwen interest and the nation state is a product of history, and 
is therefore bound to disappear in the course of history.?
13
 He asumed larger units would 
one day supplant the nation-state as the interest-bearing unit of analysis and insisted that 
nothing in realist theory precluded that outcome. But that just suggests that other units ? 
                                                
13
 Hans Joachim Morgenthau, Politics Among Nations: The Strugle for Power and Peace, Brief ed. (New 
York: McGraw-Hil, 193), 12. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 448 
larger, smaler, or diferent in form ? could easily do the same. In other words, in 
realism there is nothing sacrosanct about the nation-state, or more generaly the state. 
Mearsheimer, in stating that realism ?merely requires anarchy,? does briefly 
acknowledge that ?it does not mater what kind of political units make up the system. 
They could be states, city-states, cults, empires, tribes, gangs, feudal principalities, or 
whatever.?
14
 Fair enough, but then why ignore the cults, tribes, gangs, mafias, organized 
criminals, and other nonstate actors who today actualy are chalenging the authority of 
nation-states, and who have at various times in recent history managed quite succesfully 
to disrupt their functioning? Even Morgenthau, being a realist, was stil mostly concerned 
with interstate war ? his masterpiece, after al, was an analysis of politics among nations 
? so he, too, leaves out nonstate actors as significant, meaningful factors in world 
afairs. But as Robert Gilpin notes, ?Realism is a philosophical position and an analytic 
perspective; it is not necesarily a moral commitment to the nation-state.?
15
 The state, he 
says, continues to be the primary actor in international afairs, but ?realism should 
acknowledge the importance of such nonstate actors as multinational firms, international 
institutions, and nongovernmental organizations (NGOs) in the determination of 
international afairs.?
16
 Since Gilpin?s primary focus is economic globalization pre-9/11, 
he could perhaps be forgiven for failing to mention terorists, trafickers, and warlords. 
Today, with U.S. and other troops fighting insurgencies and terorists in Afghanistan and 
Iraq, with terorist atacks against civilians on the rise worldwide, and with ?criminal 
                                                
14
 John J. Mearsheimer, The Tragedy of Great Power Politics, 365. 
15
 Robert Gilpin, Global Political Economy: Understanding the International Economic Order (Princeton, 
N.J.: Princeton University Pres, 2001), 15. 
16
 Ibid., 17. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 449 
insurgencies? tearing apart some areas in places such as Colombia and Mexico, such a 
failure could no longer be forgivable. 
Bruce W. Jentleson, in a self-reflective look at the policy relevance of his own 
scholarly discipline in the wake of the 9/11 terorist atacks, found that political scientists 
in the previous thre to five years had failed to adequately addres important ?questions 
that addres terorism as a particular sociopolitical phenomenon and security threat ? 
Who are the terorists? How are they organized? What are their strategies? Their 
operational relationships to states? Their imediate political and regional contexts?? 
Major scholarly journals such as World Politics, International 
Organization, Security Studies, the American Political Science Review, 
and the American Journal of Political Science did not publish a single 
article in at least the past thre years that had questions such as those 
posed above as the primary focus. International Studies Quarterly had one 
article, although it concluded that transnational terorism was declining. 
International Security had only one article, on domestic preparednes for a 
terorist atack. ? For the 2001 annual conference of the American 
Political Science Asociation, examining five sections demed most 
potentialy relevant, and their 101 panels averaging about 3 papers each, 
only 2 papers had terorism as their primary focus. For the 2001 
conference of the International Studies Asociation, with many more 
potentialy relevant sections and panels, 20 papers were identified from 
the pool of about 2,000 papers.
17
 
His point in criticizing the discipline was not to advocate that it become the academic 
equivalent of a think tank or, indeed, the discipline of policy studies, but simply to 
criticize ?the broader disciplinary privileging of general theory with its many ??isms? 
(e.g., realism, liberalism, and constructivism) over middle-range theory with its more 
limited claims and greater self-consciousnes of conditionalities yet greater utility for 
                                                
17
 Bruce W. Jentleson, ?The Ned for Praxis: Bringing Policy Relevance Back in,? International Security 
26, no. 4 (2002): 169. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 450 
being applied to the ?how? questions of strategy? and to advocate merely for a shift in the 
discipline?s emphasis toward policy-relevant middle-range theory.
18
 
Mois?s Na?m, editor of Foreign Policy, has twice asked prominent international 
relations scholars to review the state of their discipline for his journal: Stephen M. Walt 
in 1998 and Jack Snyder in 2004.
19
 In both cases, one before 9/11 and one after, the 
asumption that states are the central actor in international relations remained 
unquestioned. ?Just as his colleague Walt had done six years earlier, Snyder too 
concluded that the realist model stil ofered the most reliable lenses to ases the 
direction of global politics,? Na?m wrote in an extended criticism of the field; even the 
alternatives to realism ? constructivism and liberalism ? continue to view the world 
through the ?lens? of the state.
20
 
If an academic discipline?s focus is on international security and if international 
security has traditionaly been merely a polite antonym for the risk of war betwen states, 
then one should not be surprised that such scholars failed to produce research suggestive 
of the modifications to security arangements that would be needed to prevent medium-
scale terorist atacks such as 9/11.
 
The Global War on Terorism (GWOT) that followed 
those atacks was, acording to those undertaking it, a war betwen a great power and 
nonstate actors (and, where applicable, their state sponsors): at least among policy makers 
and implementers, the consensus was shifting to an acknowledgement that international 
                                                
18
 Ibid. 
19
 Stephen M. Walt, ?International Relations: One World, Many Theories,? Foreign Policy (198): 29; Jack 
Snyder, ?One orld, Rival Theories,? Foreign Policy (204): 52. 
20
 Mois?s Na?m, Illicit: How Smuglers, Trafickers, and Copycats Are Hijacking the Global Economy, 1st 
Anchor Boks ed. (New York: Anchor Boks, 206), 269. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 451 
security was increasingly about more than merely interstate wars. Transnational crime 
and terorism were at least as relevant. 
That has been clear to some scholars for years. In the summer of 1968, a Palestine 
Liberation Organization (PLO) cel hijacked a commercial airline flight on its way to 
Israel and threatened to kil everyone on board if Israel refused to release Palestinian 
terorists from prison. With that act, wrote Bruce Hoffman, who was studying terorism 
long before 9/11, ?the nature and character of terorism demonstrably changed ? 
facilitated by the technological advances of the time that had transformed the speed and 
ease of international commercial air travel and vastly improved both the quality of 
television news footage and the promptnes with which that footage could be broadcast 
around the globe.?
21
 Were those hijackers threats to Israeli teritory, or to the integrity of 
Israeli state institutions, or to Israel?s very survival? No, they were not. But does that 
mean they were not threats to Israel?s national security, despite the traditional 
definitions? Surely, they were threats to the individual security of the hostages over 
whose lives the terorists had undeniable power. Militarily they certainly were no match 
against Israeli security forces. But that is exactly why they used the tactics they did: the 
Palestinians had no military or even economic power to speak of, so they resorted to the 
only power available to them: force on a smal scale as an instrument of extortion and 
public opinion on a large scale. At any rate, Israel certainly treats such tactics as national 
security threats. 
What makes such asymmetric tactics increasingly efective ? whether used by 
terorists or trafickers, for political objectives or for personal profit ? is globalization. 
                                                
21
 Bruce Hofman, Inside Terorism (New York: Columbia University Pres, 198), 68. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 452 
The term globalization has come to refer to nearly everything and so in many contexts it 
has come to mean almost nothing. Those who do define it usualy do so in limited ways, 
most frequently to refer to the interdependence of national economies or to the efects of 
instantaneous telecommunication on cultures and values. But globalization is not limited 
to these spheres; political, military, economic, social, cultural, migratory, technological, 
ecological, and other socialy relevant proceses al take place on a global scale, asisted 
by developments in communication and transportation technologies. These proceses 
may be considered diferent dimensions of globalization. It is vital to keep in mind this 
multidimensional character, ?because changes in the various dimensions of globalization 
do not co-vary?: the period 1914-1945, for example, was characterized by military 
globalization but economic de-globalization.
22
 
David Held and colleagues identified six dimensions, or ?domains,? of 
globalization ? political, military, economic, cultural, migratory, and ecological ? and 
defined the term as 
a proces (or set of proceses) which embodies a transformation in the 
spatial organization of social relations and transactions ? generating 
transcontinental or interegional flows and networks of activity, 
interaction, and the exercise of power. In this context, flows refer to the 
movements of physical artefacts, people, symbols, tokens and information 
across space and time, while networks refer to regularized or paterned 
interactions betwen [sic] independent agents, nodes of activity, or sites of 
power.
23
 
                                                
22
 Robert O. Keohane and Joseph S. Nye, ?Introduction,? in Governance in a Globalizing World, ed. Joseph 
S. Nye and John D. Donahue (Washington, DC: Brokings Institution Pres, 200), 6. 
23
 David Held, et al., Global Transformations: Politics, Economics and Culture (Stanford, Calif.: Stanford 
University Pres, 199), 16. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 453 
Les formaly, globalization they defined as the ?widening, deepening, and speeding up 
of global interconnectednes,? which they argued tends to magnify the efects of distant 
events such that even local developments can have global consequences.
24
 
Robert O. Keohane and Joseph S. Nye usefully distinguished globalization from 
globalism and interdependence. ?Globalism is a state of the world involving networks of 
interdependence at multicontinental distances. ? Globalization and deglobalization refer 
to the increase or decline of globalism.? Interdependence is a characteristic of a 
reciprocal relationship betwen or among entities; it is ?part of contemporary globalism 
but by itself is not globalism.? They identify four dimensions: economic (?the long-
distance flow of goods, services, and capital, and the information and perceptions that 
acompany market exchange?), military (?long-distance networks of interdependence in 
which force, and the threat or promise of force, are employed?), environmental (?the 
long-distance transport of materials in the atmosphere or oceans or of biological 
substances such as pathogens or genetic materials that afect human health and wel-
being?), and social and cultural (?movements of ideas, information, and images, and of 
people ? who of course cary ideas and information with them?).
25
 
Graham Alison similarly used a network metaphor, defining globalization as ?the 
creation or expansion of an identifiable network ? [that] connects points and people 
around the globe on some specified dimension.? Where specific connections betwen 
specific points cannot be identified, he noted, there is no globalization. Where reciprocal 
relationships can be identified, he added, they may be unequal: a rural farmer, for 
                                                
24
 Ibid., 14. 
25
 Robert O. Keohane and Joseph S. Nye, ?Introduction,? 2. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 454 
example, may benefit from the extension of the electricity grid to his farm, but ?no one 
imagines that the farmer?s influence upon the power generator is equivalent to the 
generator?s influence on him.?
26
 Inequalities of this sort are among the main features, for 
example, of the international security environment, particularly with respect to the 
distribution of power among states and nonstate actors and to diferent forms and 
qualities of governance at diferent levels of governance (local, regional, etc.). 
Orlando Acosta and Jorge Iv?n Gonz?lez went a step further and defined 
globalization in terms not just of interdependent networks but of interacting subsystems. 
In their perspective, globalization is a hierarchical, self-organizing, open, complex system 
brought about by ?multidimensional interactions among individual, communal, national, 
and regional agents through institutional codes that articulate international, transnational, 
and global dynamics.?
27
 As a complex adaptive system it is constantly changing, with no 
predetermined endpoint, no predictable outcomes, systemic features that cannot be 
extrapolated from the features of subsystems, and, occasionaly, some very nasty 
surprises. 
In sum, interdependence is a characteristic of a relationship involving reciprocal 
effects, globalism is a state of the world involving interdependent networks and 
interconnected subsystems at transregional distances, and globalization is a proces 
involving the creation or expansion of globalism ? that is, of an identifiable global 
                                                
26
 Graham Alison, ?The Impact of Globalization on National and International Security,? in Governance in 
a Globalizing World, ed. Joseph S. Nye and John D. Donahue (Washington, D.C.: Brokings Institution 
Pres, 200), 72. 
27
 ?? originado en las interaciones multidimensionales entre agentes individuales, comunidades, naciones 
y regiones a trav?s de c?digos institucionales que articulan las din?micas internacionales, transnacionales y 
globales.? Orlando Acosta and Jorge Iv?n Gonz?lez, ?Globalizaci?n: Una Aproximaci?n desde la 
Evoluci?n Biol?gica y los Sistemas Complejos Auto-Organizativos,? [?Globalization: A Perspective from 
Evolutionary Biology and Self-Organizating Complex Systems,?] An?lisis Pol?tico No. 61 (207): 101. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 455 
network connecting specific people, states, or nonstate actors in reciprocal relationships 
? in ways that cannot always be predicted. 
The transportation and communication technologies that have been the driving 
force of globalization are indiferent to the desirability of the exchanges they enable: 
terorists can travel as readily as scientists; arms can be traded as readily as tulips; and 
extremism can be communicated as readily as pluralism. One might usefully (if 
simplisticaly) speak, therefore, of a continuum betwen beneficial forms of globalism 
(?good? globalism) and dangerous forms (?bad? globalism). Moreover, exchanges of 
ideas, people, and economic goods betwen and among states take place at a rate that 
enables actions in one place to have substantial efects abroad. Unsustainable debt 
acumulation in Indonesia, South Korea, and Thailand led to a collapse of their 
currencies in 1997 and significant damage to the Asian regional economy, which 
triggered riots in Indonesia the following year and played a role in the civil violence in 
East Timor in 1999. It remains to be sen what the security implications of the global 
financial crisis that began in 2008 wil have, but that crisis demonstrated just how tightly 
woven the international financial system had become, as some extremely large gambles 
(such as ?credit-default swaps?) by a smal number of people interacted with some 
relatively smal gambles (such as high-risk mortgages) by a very large number of people 
to create waves of financial instability throughout the world, afecting even those who 
placed no such bets. 
Globalism and interdependence complicate policy making not only for economic 
policy but for security policy as wel. Terorism is a good example. For centuries 
terorism was a localized phenomenon ? the earliest known terorists used daggers 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 456 
against their victims ? but developments in technology that enabled legitimate 
busineses to travel and communicate more quickly and securely across long distances 
enabled terorists to do the same, and in the second half of the twentieth century terorist 
groups increasingly crossed national borders to atack their targets, trade weapons, 
recruit, train, and plan.
28
 The Soviet military?s withdrawal from Afghanistan in 1989 left 
a power vacuum that eventualy was filed by the Taliban, which atracted Al Qaeda and 
others from throughout the region seking a haven in which to plan atacks against distant 
targets; this globalization of terorism reached into U.S. teritory in 2001. 
The international trade in narcotics is another example, one that was key to the 
current study (se Part I). Latin American drug trafickers have been innovators in 
underground globalization for at least thre decades. The cocaine trade betwen the 
United States and the Andean region has existed for more than a hundred years and has 
been ilegal for almost that long. Until the late 1970s, however, cocaine was usualy 
smuggled into the United States by ?mules,? travelers who hid smal quantities of cocaine 
in their luggage and then sold the drugs to U.S. distributors. When American demand for 
cocaine rose in the late 1970s, the need to transport ever-larger quantities triggered the 
development of more complex networks of growers, producers, smugglers, and money 
launderers throughout the Western Hemisphere. As policy makers and counternarcotics 
officials in the United States responded with stepped-up law enforcement, demand-
reduction eforts, and counternarcotics aid beginning in the mid-1980s, these 
international networks became even thicker and more sophisticated. They were early 
adopters of cel phones and computer technology and set up innovative and complex 
                                                
28
 Bruce Hofman, Inside Terorism, 67. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 457 
busines arangements and transportation networks worldwide. By the early 1990s, South 
American trafickers were transporting tons of cocaine at a time by land, sea, and air and 
coordinating their activities with dozens, sometimes hundreds, of other groups: growers 
and insurers throughout the Andes; producers operating masive drug laboratories 
(protected by guerilas or paramilitaries) in the jungles of Colombia; transporters in 
Central America, Mexico, and the Caribbean; international arms dealers; and investors 
and organized crime syndicates in the United States, Europe, and Asia. The wealth and 
power of these international traficking groups corrupted the already weak institutions of 
governance in the Andean states. The drug lord Pablo Escobar was so wealthy at the 
height of his power that he offered to imediately pay off Colombia?s entire foreign debt 
in exchange for amnesty, an ofer the state rejected. 
During the late 1990s and early 2000s much of the debate over U.S. aid to 
Colombia centered on the degre to which U.S. asistance should (or could) strengthen 
those institutions, and on whether the military components of the aid package might 
weaken them further. U.S. policy makers faced presure on these questions from a 
number of domestic interest groups, but just as the problems the aid sought to addres 
were now globalized, the presure on policy makers came from global sources as wel. 
For example, international networks of human rights, environmental, and (perhaps 
ironicaly) anti-globalization activists worked with their counterparts within the United 
States to either oppose or restrict U.S. military aid to Colombia. Presure came also from 
foreigners who supported an expanded mision for U.S. aid, including the government of 
Colombia and multinational corporations whose operations there were frequently under 
atack by guerilas. The Colombian government had the ear of American policy makers 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 458 
during the Clinton and second Bush administrations both because it was a wiling partner 
in the war on drugs and, later, the war on terorism, and because the country was an 
important trading partner and a growing source for oil imports. This advance of bilateral 
trade reflected a broader trend in which countries throughout Latin America were 
increasingly, if at times reluctantly, opening their markets to trade more frely with the 
United States. As globalization tied the U.S. economy more strongly to those other 
economies, stability and security within those countries became a growing concern to 
U.S. policy makers. But while Colombia opposed international traficking and terorism 
and was wiling to cooperate in multilateral eforts to oppose them, it also was unable to 
exercise efective control over much of its teritory and suffered from pervasive 
corruption at al levels of government; as a consequence, the efectivenes of its 
cooperation was deeply compromised, and the costs ended up being far higher than they 
might otherwise have been. 
Latin American drug trafickers are not the only globalized ilicit actors. There are 
also arms trafickers in Africa, human trafickers in Asia, money launderers in the 
Caribbean, poppy and heroin smugglers in Afghanistan, intelectual-property pirates in 
China, actual pirates off the coast of Somalia, nuclear expertise trafickers, organ 
smugglers, art thieves, even international garbage brokers. While many transstate 
criminals stil specialize in particular ?products? (organs, art, girls, poppies), while other 
ilicit actors use weapons and tactics typical of their field (insurgents have military 
doctrine, while many crime rings use nothing more powerful than a shotgun, etc.), and 
while some organized crime is stil ?organized? in the sense of having hierarchical 
organizations with reasonably efective command and control systems, some evidence 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 459 
suggests that in general ilicit actors are increasingly indiferent to the products they 
trafic in and the tactics they employ, and they increasingly form themselves into non-
hierarchical organizations and networks. In other words, global ilicit activity is 
increasingly characterized by product indiference, difusion of tactics, and difusion of 
structures: 
? Product indiference. More and more, ilicit actors specialize in proceses 
rather than products; for example, they develop a smuggling route and open it 
up to shipments of any number of products or people. Human ?mules? have 
been found smuggling, in their stomachs, not only condoms filed with 
cocaine but also condoms filed with hundred-dollar bils. Any human 
traficking network that can smuggle sex slaves across borders can smuggle 
wanted criminals or known terorists across borders. A system used to 
smuggle a ton of cocaine can as easily smuggle a crate of automatic rifles. 
? Difusion of tactics. Stret gangs are no longer the switchblade and length-of-
chain stret fighters imagined in old films; now they have aces to military-
grade weapons and, most recently, formal military training in urban warfare. 
The Zetas, one of the most dangerous gangs of Mexico?s ?criminal 
insurgency,?
29
 are made up of former Mexican special-forces troops who 
realized they could make a lot more money working for drug trafickers than 
for the government. Reports have surfaced of American gangsters joining the 
U.S. armed forces, receiving basic training in firearms and urban warfare 
                                                
29
 John P. Sulivan and Adam Elkus, ?Plazas for Profit: Mexico?s Criminal Insurgency,? Small Wars 
Journal, published electronicaly by Smal Wars Foundation, htp:/smalwarsjournal.com/blog 
/209/04/plazas-for-profit-mexicos-crim/ (accesed 10 May 209). 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 460 
tactics, serving out their tour of duty in Iraq or Afghanistan, and returning 
home to their gangs at the expiration of their contracts. ?Such military training 
could ultimately result in more organized, sophisticated, and deadly gangs, as 
wel as an increase in deadly asaults on law enforcement officers.?
30
 
Likewise, insurgents and paramilitary organizations increasingly raise money 
by traficking in narcotics; after a while, many become litle more than drug 
traficking organizations themselves, even while continuing to proclaim a 
broader political program. This was sen with both the FARC and the AUC in 
Colombia (se Chapter 6). 
? Difusion of structure. The hierarchical organized crime structures of the past 
? kinship and cartel arangements, for example ? have increasingly given 
way to network structures and social movements, facilitated by cel phones, 
airplanes, cable news, and the Internet. The evolution of the drug trade in 
Colombia is a perfect example of this trend: Pablo Escobar persuaded the sons 
of Fabio Ochoa Restrepo to use their father?s smuggling routes (which had 
been developed to get domestic appliances stolen from the United States into 
Colombia) to smuggle cocaine from Colombia into the United States (an early 
example of product indiference). Their organizations formed a cartel 
arangement with other traficking organizations, with Escobar?s group 
playing the role of coordinator. After Escobar?s Medel?n Cartel was 
dismantled, his succesors organized the drug trade as more of a hub-and-
spokes network, with Diego Murilo?s ?office? at the center (se Chapters 5 
                                                
30
 National Gang Inteligence Center, Gang-Related Activity in the U.S. Armed Forces Increasing, 
Inteligence Asesment (12 January 207), htp:/ww.stripes.com/07/feb07/gangs/ncis_gangs.pdf, 3. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 461 
and 6). As Murilo?s network was targeted, and he himself was eventualy 
extradited to the United States, new network structures emerged with far more 
nodes, many of them now in Mexico. Al Qaeda has similarly evolved from an 
organization, to a network with sleper cels, to a globalized social movement. 
This is only a very smal sampling of the innovations in organization and in the 
global trade in ilicit goods and services made possible by the same technologies and 
social phenomena that have empowered and motivated multinational corporations, 
international activists, and humanitarian-aid workers.
31
 Some of the most important and 
dificult policy problems the world faces today have become globalized. 
This state of afairs is unlikely to change any time soon. Airplanes, telecom 
systems, and satelites cannot be uninvented, and people who want or need to 
communicate with distant friends and colleagues, send them things, visit them, or help 
them out, have no incentive to stop doing so and have every incentive to find cheaper 
ways to do so. Even with the financial crisis in 2009 there is litle evidence of 
deglobalization, with the exception of some cals for economic protectionism. In the 
absence of major catastrophes (e.g. unles the financial crisis were to turn the current 
global recesion into a global depresion), this growing interdependence is not likely to 
be reversed in the short run. Furthermore, demographic, economic, technological, and 
environmental trends are such that the dificulties involved in addresing the efects of 
globalization wil become increasingly complicated in the long run: 
An unprecedented 90 milion people are added to the planet each year, 
most of them in the poorest countries, which are least able to 
acommodate them. Poverty, disease, and hunger continue to blight the 
lives of hundreds of milions of people. Valid concerns persist about the 
                                                
31
 Mois?s Na?m, Illicit. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 462 
sweping global economic changes that could eliminate jobs and 
livelihoods, undermining whole communities; about rising economic 
disparities; about failing governments and worsening social conditions. 
It?s certainly possible that this generation?s legacy to the next wil be an 
Earth poisoned by industrial toxins, shorn of virgin forests, and commited 
to an altered climate. ? Over the next half century, human society wil 
undergo a profound demographic transformation, experience fundamental 
shifts in the global balance of economic and political power, and cope 
with nearly continuous technological change. These transformations are 
inevitable ? the forces that compel them are already in place ?.
32
 
Just as balistic misiles made it possible for one state to kil milions on the other 
side of the planet within hours, and satelites made it possible to locate targets, guide 
missiles, and photograph the results anywhere in the world, new developments in 
technology are likely to complicate maters even further. Examples include advanced 
research on biological pathogens that makes it increasingly possible to manipulate basic 
life proceses, and that risks acidental or intentional escape of new, virulent strains of 
those pathogens; and work on miniaturization of computer systems, the development of 
nano-technology, and other technological innovations that may make it possible, for 
example, to target individuals for asasination at great distances while offering to the 
asailant a high degre of plausible deniability.
33
 
In short, policy makers facing diplomatic, economic, security, and other 
chalenges are increasingly finding their hands tied with respect to what they can 
acomplish within the confines of ?foreign? or ?domestic? policy, a distinction that 
scarcely makes sense in a world in which local events can have global consequences and 
                                                
32
 Allen L. Hamond, Which World? Scenarios for the 21st Century (Washington: Island Pres, 198), 3. 
33
 John Steinbruner and Nancy Galagher, ?Constructive Transformation: An Alternative Vision of Global 
Security,? Daedalus 13, no. 3 (204): 83; John D. Steinbruner, Principles of Global Security 
(Washington: Brookings Institution Pres, 200). 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 463 
global phenomena afect even the most remote locations. Globalization, in other words, 
has greatly complicated governance in both concept and practice. 
A.3. Policy, Governance, and ?Ungoverned? Areas 
The ancient Grek words kybernan (?to ster a ship?) and kybernetes (?the 
stersman?) entered the English language by way of the French. Kybernetes evolved via 
Latin into cybern?tique, which in the 1830s meant the ?art of governing,? and from there 
entered English in 1948 as cybernetics (the study of automatic control and 
communication systems). Kybernan similarly evolved (Latin: gubernare, ?to ster, direct, 
or guide?) into gouverner (?to govern?); both noun variants gouvernement and 
gouvernance were used synonymously in the 13
th
 century to mean ?the art or manner of 
governing? but later they diverged, with gouvernance coming to refer to the ?burden of 
governing? and gouvernement to the institution. In 1297, the verb gouverner was 
translated to Middle English as the verb govern, and in the 14
th
 century, gouvernance 
pased into Middle English as governance, retaining the sense of its Old French 
predecesor, ?the art or manner of governing.? The Governance of England, the first 
constitutional treatise in English, was writen around 1470,
34
 when governance and 
government were stil used more or les synonymously; like their French counterparts, 
                                                
34
 Sir John Fortescue, The Governance of England, otherwise caled The Diference betwen an Absolute 
and a Limited Monarchy, ed. Charles Plumer (Oxford: Clarendon Pres, 185, orig. ca. 1470); cited in E. 
Maunde Thompson, ?Review,? The English Historical Review 1(3), no. 3 (186): 568. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 464 
however, the two terms later diverged, sometime during the 16
th
 century, to distinguish 
the art from the organization.
35
 
Current usage of governance derives indirectly from disparate ideas introduced in 
the field of economics during the 1930s but left unconnected and undeveloped for nearly 
four decades. The most important of these was Ronald Coase?s 1937 article, ?The Nature 
of the Firm,? in which he argued that organizational behavior ? for example, managers? 
decisions about whether to acquire another company and thereby grow larger, or simply 
to enter into a contract with it and thereby stay the same size ? could not be explained 
simply in terms of market prices, but rather had to be explained as wel in terms of 
marketing costs, contracting costs, and other transaction costs.
36
 Coase did not actualy 
use the word governance in print in 1937. But the idea that the structure and behavior of 
organizations and the nature of organizational decision-making needed to be derived 
rather than asumed resonated with the institutional economists of the 1970s, who cited 
Coase as the intelectual forefather of their work on corporate governance. Corporate 
governance came to be sen as a broader concept than corporate management, since 
the governance role is not concerned with running the busines of the 
company per se, but with giving overal direction to the enterprise, with 
overseing and controlling the executive actions of management and with 
satisfying legitimate expectations for acountability and regulation by the 
interests beyond the corporate boundaries.
37
 
                                                
35
 Corine Huynh-Quan-Suu and Service de Traduction de l?Union Europ?ene, ??tymologie du Terme 
?Gouvernance?,? Gouvernance dans l?EU, published electronicaly by Comision europ?ene, 
http:/ec.europa.eu/governance/docs/doc5_fr.pdf (acesed 20 April 209); Online Etymology Dictionary, 
s.v. ?govern? and ?cybernetics,? htp:/ww.etymonline.com (acessed 19 April 209). 
36
 R.H. Coase, ?The Nature of the Firm,? Economica 4, no. 16 (1937): 386. 
37
 R. Ian Tricker, International Corporate Governance (Englewod Clifs, N.J.: Prentice Hal, 1984), 6; 
quoted in R.A.W. Rhodes, ?The New Governance: Governing Without Government,? Political Studies 
XLIV (196): 652. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 465 
In their use of this term one can hear the echo of its ancient origins: governance as 
stering. This was folowed in the 1980s, during the mas migration of private-sector 
ideas and practices into the field of public management, when the idea took hold that 
what the public sector needed was ??les government? (or les rowing) but ?more 
governance? (or more stering).?
38
 Proponents of this school of thought argued that the 
offices of government should set the rules and the tone, but that the work of government 
would be more eficient and more just if it operated les like a hierarchy and more like a 
market; privatization, outsourcing, structural adjustment, private-sector management 
practices, and a loosening of regulations thus became the order of the day for many 
governments.
39
 
From the ?new institutional economics? that introduced corporate governance to 
the discipline that studies market mechanisms, and the ?new public management? that 
introduced market mechanisms to the discipline that manages governments, emerged a 
broader understanding of the concept that introduced ?good governance? to the 
disciplines of international development and conflict management. Whereas contracts and 
transaction costs were central to much of the corporate governance literature of the 
1970s, and the eficiencies of markets over hierarchies were central to much of the new 
public management literature during the 1980s, the good governance literature of the 
1990s focused on public goods and the institutions that deliver them. 
                                                
38
 Ibid; citing David Osborne and Ted Gaebler, Reinventing Government: How the Entrepreneurial Spirit Is 
Transforming the Public Sector (Reading, Mas.: Adison-Wesley Pub. Co, 192), 34. 
39
 Oliver Wiliamson, ?Corporate Governance,? The Yale Law Journal 93, no. 7 (1984): 1197; R.A.W. 
Rhodes, ?The New Governance: Governing Without Government?; Claire Launay, ?La Gobernanza: 
Estado, Cuidadan?a y Renovaci?n de Lo Pol?tico: Origen, Definici?n e Implicaciones del Concepto en 
Colombia,? [?Governance: State, Citizenship and Political Renewal: Origin, Definition, and Implications of 
the Concept in Colombia,?] Controversia no. 185 (205): 91. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 466 
The World Bank Group was the bridge betwen the economic-liberalization 
conceptualization of the new public management school and the later, broader 
conceptualizations that followed. In a 1989 report on Africa, the Bank argued that the 
underlying cause of underdevelopment on that continent was a ?crisis of governance.?
40
 
Thre years later, defining governance as ?the manner in which power is exercised in the 
management of a country?s economic and social resources for development,? it 
elaborated a vision of development that equated ?good? development with good 
governance. Although it could not say so explicitly, being limited by charter to 
addresing a country?s development and not its politics, the Bank nevertheles defined 
good governance in terms scarcely distinguishable from the standards of the economic 
and political institutions of Western liberal democracies.
41
 
Other authors, unrestrained by a development charter, quickly recognized the 
breadth of the concept inherent in the Bank?s work and set about elaborating it 
acordingly. ?Governance is the word that describes the tension-filed interaction 
betwen citizens and their rulers, and the various manners in which diverse kinds of 
governments enable their constituents to achieve satisfaction and material prosperity, or 
to thwart those and related aspirations,? wrote Robert I. Rotberg and Deborah L. West in 
a study based on Rotberg?s definition of governance as ?the delivery of political goods to 
citizens: the beter the quality of that delivery and the greater the quantity of the political 
                                                
40
 World Bank, Sub-Saharan Africa: From Crisis to Sustainable Growth (Washington: The World Bank, 
1989). 
41
 World Bank, Governance and Development (Washington: The World Bank, 192); Adrian Leftwich, 
?Governance, Democracy and Development in the Third World,? Third World Quarterly 14, no. 3 (193): 
605; Claire Launay, ?La Gobernanza: Estado, Cuidadan?a Yrenovaci?n de Lo Pol?tico: Origen, Definici?n 
e Implicaciones del Concepto en Colombia.? 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 467 
goods being delivered, the higher the level of governance.?
42
 Following Rotberg, Marie 
Besan?on defined governance as ?the management, supply and delivery of some or most 
of these [political] goods,? which in her conceptualization included, among other things, 
security, the rule of law, political and civil fredoms, health care, education, 
infrastructure, and a functioning money and banking system, with security being the most 
important: ?Good governance results when nation-states provide a high order of certain 
political goods ? when the nation-states perform efectively and wel on behalf of their 
inhabitants.?
43
 For B. Guy Peters, governance involved building ?institutions designed to 
exercise collective control and influence over the societies and economies for which 
[governments] have been given responsibility.?
44
 Daniel Kaufmann, Aart Kray, and 
Pablo Zoido-Lobat?n defined the term even more broadly, as 
the traditions and institutions by which authority in a country is exercised. 
This includes (1) the proces by which governments are selected, 
monitored and replaced, (2) the capacity of the government to efectively 
formulate and implement sound policies, and (3) the respect of citizens 
and the state for the institutions that govern economic and social 
interactions among them.
45
 
A few years into the new milennium, so much work was being done on ?good 
governance? as the solution to development that the World Bank was able to identify 
                                                
42
 Robert I. Rotberg and Deborah L. West, The God Governance Problem: Doing Something About it, 
WPF Reports, 39 (Cambridge, Mas.: World Peace Foundation, 204), htp:/ww.orldpeacefoundation 
.org/WPF39.GovernanceRanking.pdf, 1. 
43
 Marie Besan?on, Good Governance Rankings: The Art of Measurement, WPF Reports, 36 (Cambridge, 
Mas.: World Peace Foundation, 203), 1; Robert I. Rotberg, ?The Failure and Colapse of Nation-States: 
Breakdown, Prevention, and Repair,? in When States Fail: Causes and Consequences, ed. Robert I. 
Rotberg (Princeton, N.J.: Princeton University Pres, 204), 2. 
44
 B. Guy Peters, The Future of Governing, 2nd ed. (Lawrence: University Pres of Kansas, 201), 1. 
45
 Daniel Kaufman, Aart Kray, and Pablo Zoido-Lobat?n, ?Governance Maters,? working paper, World 
Bank Policy Research Working Papers 2196 (Washington: World Bank, October 199), 
http:/ww.orldbank.org/wbi/governance/pubs/govmaters.html, 1. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 468 
more than a hundred relevant datasets, asesments, and ongoing studies sponsored by the 
Bank and other researchers, and Besan?on was able to review 47 projects measuring 
governance and related isues in democracy, coruption, and risk.
46
 
The growth of the good governance literature paraleled that of the ?state failure? 
literature ? many of the same thinkers worked on both topics ? and this led inexorably 
to policy makers? concern over ?ungoverned? teritories in the wake of the terrorist 
atacks of 11 September 2001 (9/11). In Rotberg?s introductory chapter to his second 
edited volume on state failure, he described a ?strong state? (in contrast to ?weak,? 
?failed,? and ?collapsed? states) in the same terms as his work on good governance, even 
explicitly linking the concept of state strength to empirical studies of governance 
quality
47
: 
Strong states unquestionably control their teritories and deliver a full 
range and a high quality of political goods to their citizens. They perform 
wel acording to indicators like GDP [gross domestic product] per capita, 
the UNDP [United Nations Development Programe] Human 
Development Index, Transparency International?s Corruption Perception 
Index, and Fredom House?s Fredom of the World Report. Strong states 
offer high levels of security from political and criminal violence, ensure 
political fredom and civil liberties, and create environments conducive to 
the growth of economic opportunity. The rule of law prevails. Judges are 
independent. Road networks are wel maintained. Telephones work. Snail 
mail and e-mail both arive quickly. Schools, universities, and students 
flourish. Hospitals and clinics serve patients efectively. And so on. 
Overal, strong states are places of enviable peace and order.
48
 
                                                
46
 World Bank, ?Consolidated Matrix of Both Internal and External Governance Data Sets and 
Instruments,? Excel spreadsheet, published electronicaly by The World Bank Group, 
www.worldbank.org/wbi/governance (acesed 20 April 209); Marie Besan?on, Good Governance 
Rankings. 
47
 Robert I. Rotberg, ed., When States Fail: Causes and Consequences (Princeton, N.J.: Princeton 
University Pres, 204); Robert I. Rotberg and Deborah L. West, The God Governance Problem: Doing 
Something About it. 
48
 Robert I. Rotberg, ?The Failure and Colapse of Nation-States: Breakdown, Prevention, and Repair,? 4. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 469 
The obvious asumption of the state failure literature that the nation-state was the 
appropriate unit of analysis pervaded the literature on governance as wel. Besan?on, 
citing Rotberg?s work on failed states, explains why: ?In this era, nation-states are 
responsible for the task of governing and providing goods to those who reside within 
their borders. Many of these nation-states have corrupt leaders who drain the country?s 
treasures and provide litle or no security, education, infrastructure, or any other public 
good to their constituents.?
49
 
Yet states are not the only coordinators of collective action and are not the only 
actors corrupted by power. Keohane and Nye argued that governance is a broader concept 
than is suggested by the literature; to them, governance encompases 
the proceses and institutions, both formal and informal, that guide and 
restrain the collective activities of a group. Government is the subset that 
acts with authority and creates formal obligations. Governance need not 
necesarily be conducted exclusively by governments and the international 
organizations to which they delegate authority. Private firms, asociations 
of firms, nongovernmental organizations (NGOs), and asociations of 
NGOs al engage in it, often in asociation with governmental bodies ? 
sometimes without governmental authority.
50
 
Governance takes place at multiple levels (subnational, national, and supranational), and 
sometimes betwen levels, and it is exercised through multiple instruments (laws, norms, 
markets, and architecture): ?one can slow trafic through a neighborhood by enforcing 
speed limits [(laws)], posting ?children at play? signs [(norms)], charging for aces 
[(markets)], or building speed bumps in the road [(architecture)].?
51
 States can operate at 
any of these levels or use any of these instruments, but they need not: nonstate actors are 
                                                
49
 Marie Besan?on, Good Governance Rankings, 2. 
50
 Robert O. Keohane and Joseph S. Nye, ?Introduction,? 12. 
51
 Ibid; citing Lawrence Lesig, Code and Other Laws of Cyberspace (New York: Basic Boks, 199), 8, 
207. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 470 
increasingly capable and active. The central governments of many states do not exercise 
de facto control over the entire teritories under their de jure jurisdictions. Yet while 
Keohane and Nye correctly observed that governance takes place within and among 
private, governmental, and third-sector actors, even they failed to acknowledge that 
governance is at times undertaken by, for example, guerilas who have wrested control 
of an area from the central government (or are atempting to).
52
 Colombia, for example, 
did not exercise control over some regions of its teritory for decades, occasionaly even 
relinquishing its own sovereignty over certain teritories during peace negotiations; in 
some towns they controlled, guerilas maintained local order and provided social 
services, until ejected by paramilitary forces, some of whom then played that same role 
(se Chapter 6). 
It was concern over precisely such teritories ? especialy in the wake of the 9/11 
terorist atacks, which had been planned from a ?lawles? and ?ungoverned? 
Afghanistan ? that led the U.S. Department of Defense (DoD) to launch a thre-year 
study of ?ungoverned? areas that transnational terorist, insurgent, or criminal networks 
could use as safe havens. Its initial conceptualization of the isue was almost entirely in 
terms of state failure and state sponsorship of terorism. As defined during the 2006 
Quadrennial Defense Review, the mision of what came to be known as the Ungoverned 
Areas Project was to figure out how to contribute to ?building partnership capacity? by 
enabling ?strategic partnerships to extend governance to under- and ungoverned areas?; 
                                                
52
 Nelson Kasfir, ?Dilemas of Popular Suport in Guerila War: The National Resistance Army in 
Uganda, 1981-86? (paper presented at the Laboratory in Comparative Ethnic Proceses (LiCEP) 6, 
Dartmouth Colege, University of California-Los Angeles, November 202). 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 471 
the ?partners? in question were asumed to be state actors, especialy those receiving U.S. 
foreign asistance.
53
 
The terminology raised some eyebrows outside of the department, because the 
term ungoverned historicaly has been used by the powerful as a prelude and justification 
for atempts to repres ?ungoverned pasion? (the ?undisciplined? or ?unbridled? 
sexuality of women); to ?civilize? (or, more recently, ?democratize?) an ?ungoverned? 
(?uncivilized?) people; or to take over some ?ungoverned? (?uncolonized? or 
?unclaimed?) teritory to extract its resources ? in short, as a prelude to oppresion, 
imposition, or exploitation.
54
 
Administration officials did not ease such concerns by the way they used 
?lawles? and ?ungoverned? (or, worse, ?ungovernable?) in public in the early days of 
the Global War on Terorism, as at times they semed to suggest they were considering 
some kind of military approach to the problem, even if it was just by increasing military 
asistance to states with such areas. In an interview ith The New York Times les than 
four months after 9/11, then-Deputy Secretary of Defense Paul Wolfowitz was asked to 
clarify comments he had made ?about regions that you consider ungovernable or 
ungoverned ? sort of lawles regions where it?s easy for terorist cels and terorist 
group to bred.? While stipulating ?very clearly, we?re not talking about future 
                                                
53
 United States Department of Defense, Building Partnership Capacity: QDR Execution Roadmap 
(Washington: 206), Tasks 3.3.4 and 3.3.5; United States Department of Defense, Quadrenial Defense 
Review Report. 
54
 Early instances of the use of ?ungoverned? in this maner include: The New York Times, ?Sentimental 
Philanthropy,? 29 January 1852, 2; The New York Times, ?Washington ? the Aproaching Meting of the 
Virginia Legislature ? the Charges of Coruption Against Kansas Members of Congres ? an 
Unorganized Teritory  the Georgia Case in Washington ? Mr. Hear and Mr. Sawyer of South 
Carolina,? 7 February 1870, 1; Herbert Spencer, ?Behavior and Ceremony,? The New York Times, 27 
January 1878. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 472 
operations,? his response did suggest he believed that non-state forms of governance as 
such were the problem: 
At the same time that one has to wory about governments that actively 
support terorism, what this whole crisis has revealed to us is that places 
with no government at al become very dangerous as havens for terorists 
as wel. ? I haven?t thought enough about what I want to mean by that 
term [ungoverned] ? I think what it realy means is at some level it?s 
governed by not what we can naturaly think of as governments ? it 
might be more benign if they were truly ungoverned, but they are in fact 
governed by tribal groups or anarchic groups or ethnic militias. ? [There] 
are probably going to be many cases where the best thing we can do is to 
provide the training and equipment and possibly some of the backup 
support from us to permit countries that realy want to deal with terorists, 
[that] have the wil but lack the means, lack the ability, to take on the 
problem.
55
 
The conventional wisdom, which these comments reflected at the time, asumed 
that the problem of ?under- or ungoverned areas? was a problem of state weaknes or 
state fragility, an unwilingnes or inability on the part of some states to govern al their 
teritory, and that therefore the appropriate policy response was to encourage and enable 
those states to fil those ?gaps? in governance.
56
 A simplistic take on this conventional 
view held that this could be acomplished primarily by increasing military, inteligence, 
justice, and law-enforcement capacity, but that take scarcely withstood scrutiny: the first 
managers of the Ungoverned Areas Project imediately recognized that states with 
?ungoverned? areas would need asistance with ?governance capacity? defined more 
broadly: not just to improve counterterorism capacity but to strengthen administrative 
                                                
55
 Paul Wolfowitz, interview of 7 January 202 with James Dao and Eric Schmit of The New York Times, 
M2 Pres IRE, 14 January 2002, published electronicaly by M2 Comunications Ltd., 
http:/ww.m2.com (acesed 15 April 2008); James Dao and Eric Schmit, ?A Nation Chalenged: 
Strategy: U.S. Ses Batles in Lawles Areas After Afghan War,? The New York Times, 8 January 202. 
56
 United States Department of Defense, Building Partnership Capacity: QDR Execution Roadmap, 12. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 473 
capacity, economic management, political institutions, and civil society as wel.
57
 
Likewise, the most systematic study of the problem at the time recognized the relevance 
of nonsecurity institutions, demographics, and other ?soft? factors, in addition to military 
and security factors, in such ?ungoverned teritories.?
58
 
The asumptions underlying the conventional wisdom at the beginning of the 
project ? that inadequate ?state penetration? was the problem, and that building state 
capacity was the solution ? were chalenged during the course of the project itself. As 
staf studied the problem and solicited input from experts and experienced military 
personnel (I was the project?s co-manager during this stage), it quickly became clear that 
what actualy makes some places suitable as sanctuaries for ilicit transnational activity is 
not their lack of governance but their manner of governance: many states govern in ways 
that are dangerous to both insiders and outsiders, and many places not controled by a 
state are not dangerous to either. 
The first of the project?s thre main findings was simply this: ?Few places in the 
world are literaly ?ungoverned?: where the central government is weak or mising, 
another governing body ? usualy a provincial, local, tribal, or autonomous government, 
but sometimes an informal or nongovernmental organization ? tends to emerge (or 
already exists) to maintain order and deliver needed services.? It went on: 
                                                
57
 The first action oficers to manage the Ungoverned Areas Project were Leslie Hunter, a civil servant, and 
Colin Kahl, a consultant; I joined the project in 206 and co-managed it with Major Sandra Reyna of the 
U.S. Army. The project was directed at diferent times by Bary Pavel, Thomas ahnken, Amanda Dory, 
Kathlen Hicks, Alisa Stack O?Conor, Eric Her, Katherine Johnson, and Laura Coper. Contributions to 
the project came as wel from staff in more than forty offices throughout the U.S. government, including 
the Defense Department, the State Department, USAID, the Inteligence Comunity, and the National 
Security Council; from participants in a series of workshops hosted by The RAND Corp; and from 
consultations with private-sector experts. 
58
 Angel Rabasa, et al., Ungoverned Teritories. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 474 
The pertinent question with respect to ungoverned areas is not about the 
degre of governance but about the manner of governance: Who is, and 
who is not, governing an area, and what are the consequences of the 
particular way they govern? If a semi-autonomous tribal government is 
wiling and able to govern its teritory in a way that is inhospitable to 
transnational ilicit activity, then that is likely to be satisfactory ?.
59
 
The project got a strong push toward this conclusion at an August 2007 conference at the 
Naval Postgraduate School in Monterey, California, where contributors to a forthcoming 
edited volume on the isue implicitly criticized the asumptions underlying most of the 
literature on governance, al of the literature on state failure, and the conventional 
wisdom on ?ungoverned? areas that had reified state capacity-building during an era 
when the authority of the state had been diluted, in the organizers? analysis, by: the 
authority of the market (as a result of the spread of neoliberal ideology), the authority of 
international organizations (as a result of the spread of human rights ideology), the 
authority of civil society (as a result of the spread of democratic ideology), and the ability 
of common people to communicate and travel globaly and cheaply (as a result of the 
spread of technology).
60
 
The second finding of the Ungoverned Areas Project followed from the first. If 
the problem was not who governs but how they govern, then the solution should aim not 
to improve specific types of government (i.e. states) but to achieve specific results with 
respect to the quality of governance, whether the ?partner? is a state or not: 
In many cases, provincial, local, tribal, or autonomous governments ? 
and in some cases, other countries, corporations, or organizations ? are 
simply beter positioned than the central government to addres the local 
conditions that enable ilicit actors to operate there. It often wil be more 
eficient and efective to influence and enable those entities rather than ? 
                                                
59
 Robert D. Lamb, Ungoverned Areas and Threats From Safe Havens, 18. (emphasis in original). 
60
 Anne L. Clunan and Harold Trinkunas, Ungoverned Spaces? 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 475 
or, preferably, in addition to and with the asistance of ? the host state in 
the short term. For diplomatic, legal, and practical reasons, the host state 
cannot be ignored or bypased, but nor should it be permited to impede 
progres against safe havens when other entities are positioned to help. An 
appropriate balance is needed.
61
 
In other words, policy on ?ungoverned? areas should be based, not on the narower 
conceptualizations of governance reflected in the state failure literature, but on the 
broader conceptualizations found in the best of the ?good governance? literatures ? at 
least as broad as that of Keohane and Nye (if not of the ?cybernetic? approaches, to be 
discussed shortly). By focusing on state failure and conventional notions of governance, 
those involved in developing policies and strategies to confront nontraditional security 
threats involving nonstate actors had been blinding themselves to the fact such threats can 
emerge in strong and otherwise wel governed states as wel as in weak, failing, or failed 
states. The Ungoverned Areas Project had litle to say regarding ?ungoverned? areas in 
strong states, beyond recommending further study on the dynamics of governance that 
make urban, maritime, and ?virtual? safe havens possible in weak and strong states 
alike.
62
 (The current study folows directly on the recommendation to study urban 
dynamics.) 
The third and final finding was based on an observation, expresed by many 
consulted during the course of the project, that today?s nonstate actors semed to 
understand the new dynamics of governance beter than state actors, and that nonstate 
ilicit actors had been quicker to adapt to globalization and to states? eforts to counter 
them than states had been to adapt to the changing nature of transnational crime, iregular 
                                                
61
 Robert D. Lamb, Ungoverned Areas and Threats From Safe Havens, 5. 
62
 Ibid., 24 and 49. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 476 
war, and ?ungoverned? areas. ?Our ability to develop coherent policy for safe havens is 
constrained by policy proceses that depend on separate institutions to addres isues that, 
by their nature, cannot efectively be dealt with separately.? The chalenge was 
summarized as follows: ?criminals, insurgents, terorists, and warlords increasingly 
borrow each other?s tactics, buy each other?s services, and exploit each other?s misions. 
The [United States Government], by contrast, has mostly separate doctrines for each type 
of adversary or type of conflict, with limited overlap.?
63
 The problem with having 
diferent agencies, diferent units, and diferent personnel developing and implementing 
policy for diferent types of security problem or ilicit activity is that the nature of these 
isues have changed, becoming increasingly fluid across diferent disciplinary boundaries 
as new innovations alow (se the introduction to Appendix B). While it is clear that the 
United States and other states have recognized this problem and have taken some steps to 
overcome it through various interagency coordination proceses, those steps have been 
tentative and contested. 
The findings of the Ungoverned Areas Project represented a strong critique of the 
way governance had been conceptualized and the way its misconceptions had 
misinformed policy, strategy, and doctrine for a particular set of security problems (and, 
indeed, continue to do so). Yet in some ways it did not go far enough in its critique of the 
way governance had been conceptualized: in the spectrum of conventionality in the 
?ungoverned? literature, the Ungoverned Areas Project lay somewhere betwen the best 
of the conventional wisdom as articulated by Angel Rabasa and his colleagues at RAND, 
which it had build upon, and the more daring alternatives to state authority considered in 
                                                
63
 Ibid., 47. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 477 
the Monterey conference and edited volume by Anne L. Clunan and Harold Trinkunas, 
which it anticipated.
64
 
For a wide range of problems, beyond security and crime, alternative forms of 
governance have arisen in recent decades precisely in response to the presures of 
globalization and the waves of ideologies (fre markets, human rights, democracy) and 
technologies (global communications, global finance, global transportation) that have 
eaten away at the authority and capacity of the state. Liesbet Hooghe and Gary Marks 
convincingly argue that governance, if it is to have any hope of succeding in today?s 
world, must operate at multiple levels. ?Large (i.e., teritorialy extensive) jurisdictions 
have the virtue of exploiting economies of scale in the provision of public goods, 
internalizing policy externalities, alowing for more eficient taxation, facilitating more 
eficient redistribution, and enlarging the teritorial scope of security and market 
exchange. Large jurisdictions are bad when they impose a single policy on diverse 
ecological systems or teritorialy heterogeneous populations?: 
Centralized government is not wel suited to acommodate diversity. 
Ecological conditions may vary from area to area. Controlling smog in a 
low-lying flat area surrounded by hils (such as Los Angeles) poses a very 
diferent policy problem than smog control in a high plateau such as 
Denver. Preferences of citizens may also vary sharply acros regions 
within a state, and if one takes such heterogeneity into acount, the 
optimal level of authority may be lower than economies of scale dictate. In 
short, multi-level governance alows decision makers to adjust the scale of 
governance to reflect heterogeneity.
65
 
Through their analysis of five more or les separate literatures on the difusion of state 
authority, they distinguish betwen two general types of multi-level governance: general-
                                                
64
 Angel Rabasa, et al., Ungoverned Teritories; Ane L. Clunan and Harold Trinkunas, Ungoverned 
Spaces? 
65
 Liesbet Hoghe and Gary Marks, ?Unraveling the Central State, But How? Types of Multi-Level 
Governance,? American Political Science Review 97, no. 2 (203): 23. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 478 
purpose jurisdictions (Type I) and task-specific jurisdictions (Type I). Type I multi-level 
governance, in their conception, is the familiar federal structure in which ?every citizen is 
located in a Russian Doll set of nested jurisdictions, where there is one and only one 
relevant jurisdiction at any particular teritorial scale.? Such structures govern 
communities rather than specific policies and tend to be highly hierarchical by design. 
Type I multi-level governance is a set of more fluid structures ?fragmented into 
functionaly specific pieces ? say, providing a particular local service, solving a 
particular common resource problem, selecting a particular software standard, monitoring 
the water quality of a particular river, and adjudicating international trade disputes.? The 
variety of arangements is vast, including everything from local school districts (to 
govern just the school system), to multi-province districts such as the Delaware River and 
Bay Authority (to operate bridges and feries), to international environmental treaty 
bodies (to monitor and enforce, or ?name and shame?), to the World Trade Organization, 
to any number and variety of partnerships betwen and among the public, private, and 
voluntary sectors.
66
 
Largely overlooked in policy circles until only very recently has been the ?other? 
branch in the linguistic family tre of governance: kybernetes (the stersman), 
cybern?tique (the art of governing), and cybernetics (the study of automatic control and 
communication systems). Conspiracy theorists ? of which there semed to be no 
shortage in Medel?n ? tend to atribute undesirable outcomes to the intentional 
behaviors of powerful, if secret, government or corporate agents: by asking, ?Who 
benefits?? they believe they can identify the perpetrator, to whom they apparently 
                                                
66
 Ibid. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 479 
atribute vast, and almost always fully unwaranted, amounts of competence (se ? 6.1.3). 
Against the logic of conspiracy ? an extreme version of what John Steinbruner caled 
the analytic paradigm
67
 ? one may pose the logics of adaptation, of cybernetics, or of 
complex systems, in which outcomes are path-dependent, dificult to predict, constantly 
changing, and ?emergent? from interactions among networks of non-centralized agents. 
Governance, in these approaches, is ?directed at the creation of paterns of interaction in 
which political and traditional hierarchical governing and social self-organization are 
complementary, in which responsibility and acountability for interventions is spread 
over public and private actors.?
68
 Or in the words of R.A.W. Rhodes, ?Governance is 
about managing networks?: 
Interorganizational linkages are a defining characteristic of service 
delivery and I use the term network to describe the several interdependent 
actors involved in delivering services. These networks are made up of 
organizations which need to exchange resources (for example, money, 
information, expertise) to achieve their objectives, to maximize their 
influence over outcomes, and to avoid becoming dependent on other 
players in the game.
69
 
As globalization makes the world more complex, governance structures have 
evolved to met the chalenges it has generated. Innovations in governance have come 
about among organizations of al kinds: ilicit actors, multinational corporations, 
international organizations, public-private partnerships, social movements, volunter 
networks, and even states have al experimented with new structures and proceses of 
governance ? to varying degres of succes. It can be easy to fal into the simplistic 
                                                
67
 John D. Steinbruner, The Cybernetic Theory of Decision: New Dimensions of Political Analysis 
(Princeton, N.J.: Princeton University Pres, 202), 25. 
68
 Jan Koiman, Modern Governance: New Government-Society Interactions (Newbury Park, Calif.: Sage, 
1993), 252; quoted in R.A.W. Rhodes, ?The New Governance: Governing Without Government,? 657. 
69
 R.A.W. Rhodes, ?The New Governance: Governing Without Government,? 658. 
Lamb, Robert D. | Appendix A. Policy, Globalization, and Governance 480 
conclusion that ilicit actors have adapted wel while states have not, but clearly there 
have been many ilicit actors whose internal governance has been disastrous for their 
organization or movement, while many states have shown creativity and flexibility by, 
for example, adopting private-sector management practices for some tasks, creating 
formal mechanisms for interagency coordination within existing hierarchical structures, 
and encouraging (or sometimes merely not discouraging) the formation of informal 
networks such as ?communities of interest? on particular isues. Governance, then, as 
James N. Rosenau has writen, is a broader phenomenon than government, encompasing 
formal political institutions as wel as social institutions, social networks, and 
organizations in the private and voluntary sectors as wel as the public sector: 
? government suggests activities that are backed by formal authority, by 
police powers to insure the implementation of duly constituted policies, 
whereas governance refers to activities backed by shared goals that may or 
may not derive from legal and formaly prescribed responsibilities and that 
do not necesarily rely on police powers to overcome defiance and atain 
compliance. ? Put more emphaticaly, governance is a system of rule that 
works only if it is acepted by the majority (or, at least, by the most 
powerful of those it afects), whereas governments can function even in 
the face of widespread opposition to their policies.
70
 
For policy makers hoping to influence the complex dynamics of teritorial and social 
control, governance, violence, and legitimacy that take place in cities such as Medel?n, it 
is important to be able to recognize the distinction betwen governance and control, and 
especialy the diference betwen mere control (the ability to prevent governance) and 
?bad? governance (inefective policies, poor management, doing just enough to be 
acepted only ?by the most powerful of those it afects,? etc.). 
 
                                                
70
 James N. Rosenau and Ernst Oto Czempiel, eds., Governance Without Government: Order and Change 
in World Politics (New York: Cambridge University Pres, 192), 4. 
Lamb, Robert D. |  Appendix B. Complex Violence 481 
Apendix B. Complex Violence 
Many of the world?s most dificult security problems can be made to sem ore 
intractable simply by listing the categories of problems or armed actors involved. In 
Afghanistan, for example, there is a terorism problem (al Qaeda), an insurgency problem 
(the Taliban), an unconventional warfare problem (parastatal militias), a transnational 
crime problem (arms and opium poppies), an ?ungoverned? areas problem (the 
?Pashtunistan? safe haven), a human rights problem (the violent oppresion of women 
and religious minorities), a humanitarian problem (internaly displaced persons), and 
most generaly a ?post-conflict? reconstruction problem. Pakistan, Sudan, Somalia, and 
the Democratic Republic of Congo have a similarly large number of problem categories, 
as do any number of cities or vilages throughout the world in places such as Nigeria, 
Indonesia, and, perhaps to a leser extent, now Mexico. Each of these problem categories 
has traditionaly been addresed by its own policy discipline (by analogy to academic 
disciplines), and the complexity of such places can be roughly measured by how many 
policy disciplines can reasonably claim jurisdiction over a significant subset of its 
problems. 
Those problems can be made to sem even more intractable stil by recognizing 
that certain individuals and groups who operate there do not fit neatly or stably within 
any one of these problem categories. An arms dealer, for example, might take opium 
poppies as payment, sel them to insurgents, then launder the money through a charity 
that is a subcontractor for a development project. Or a guerila unit made up of fighters 
Lamb, Robert D. |  Appendix B. Complex Violence 482 
recruited from a refugee camp might avoid defeat by defecting to the competing 
paramilitary group, and then to fund future operations they might sel security services to 
poppy farmers and opium producers in the war zone, then eventualy take over the 
production proces directly. Or a farmer might acept some money from an insurgent 
group to shel a neighboring vilage one night, and only that night. Are these people 
insurgents, narcotrafickers, transnational criminals, unconventional forces, or just 
common people? They can be any or al of the above at any given time. 
Complex situations such as these, involving multiple problem categories and 
multiform actors operating from ambiguous motives, present a real puzzle to policy 
makers who must decide which of their existing capabilities, agencies, and authorities are 
best suited to addresing them: Is it mainly a defense problem or mainly a development 
problem, and if it?s both which should take priority? Is it mainly a military problem or 
mainly a law-enforcement problem? Which doctrines, planning frameworks, 
?paradigms,? ?best practices,? or, most generaly, policy disciplines should be brought to 
bear upon the problem: reconstruction and stabilization, democracy and governance, 
conflict mitigation and management, counterterorism, counternarcotics and transnational 
crime, counterinsurgency, unconventional warfare, urban warfare, humanitarian relief ? 
which one, or which combination, and with what priority? 
As was discussed in Appendix A, trends in globalization, urbanization, iregular 
warfare, and other phenomena have created a world in which these problem categories 
and policy disciplines cannot hope to adequately capture or addres the complexities 
present in certain areas of the real world today. These problems require interdisciplinary 
solutions, and both in the field and back at headquarters, policy makers and those 
Lamb, Robert D. |  Appendix B. Complex Violence 483 
involved in implementation have been improvising such solutions, often with useful and 
innovative results. 
But social scientists and policy analysts can contribute to the search for sensible, 
eficient, and lasting solutions to today?s complex problems by studying phenomena that 
cut across or take place in the spaces betwen our existing categories of knowledge and 
practice. 
Complex violence is one such phenomenon, cutting across and sometimes hiding 
betwen the gaps of our existing policy disciplines. Complex violence involves multiform 
armed actors, in networked relationships, with shifting loyalties, and with diverse or 
ambiguous motives; complex urban violence is, to keep the definition simple, complex 
violence that takes place in densely populated areas. 
The central problem of each of the security-policy disciplines ? stability 
operations, transnational crime, counterterorism, etc. ? is the use of violence by 
diferent actors; the central problem of complex environments involving multiple 
security-policy disciplines is complex violence. By focusing on the common problem 
within these complex environments ? complex violence ? the analysis of the problem 
and the search for a solution becomes more tractable than would trying to understand al 
the diferent motivations and methods and loyalties of al the diferent violent actors. (By 
reference to the complex-systems literature, I am tempted to say that certain forms of 
violence might be analogous to an attractor, the study of which makes it possible to 
understand important dynamics within a complex system without having to know much 
about al of the system?s components. Given the data requirements for studying complex 
systems, however, I am not wiling to claim anything beyond the analogy.) 
Lamb, Robert D. |  Appendix B. Complex Violence 484 
As a thought experiment, consider the diference betwen narcotraficking today 
and narcotraficking in some hypothetical world in which no violence were asociated 
with narcotraficking: if the proceds of the poppy trade in Afghanistan went to building 
mansions and buying expensive cars instead of also funding insurgency operations and 
terorist atacks; or if, upon the arest of a drug kingpin in Colombia, questions about 
succesion were resolved by nonviolent procedures within the traficking organization 
instead of by mafia wars; or if, in Los Angeles, gangs who distribute drugs were to divide 
turf through negotiation and ongoing consultation. Narcotraficking in the real world is a 
danger; narcotraficking in this hypothetical world is a nuisance, not entirely unharmful 
and not exactly legal, but at least no more violent than, say, pineapple-juice traficking: 
The cultivation, procesing, transport, and distribution of cocaine difers from the 
cultivation, procesing, transport, and distribution of pineapple juice only in that the 
cocaine cycle is ilegal and involves violence, whereas the pineapple-juice cycle is legal 
and does not, generaly, involve violence (at least not as much as it once did in Latin 
America). 
As another thought experiment, consider the diference betwen a group of people 
using indiscriminate violence against innocents as a way to coerce an occupying power to 
leave the area, versus a group of people publicizing the occupiers? abuses as a way to win 
domestic and international support for eforts to get them to leave. The first is a terorist 
group; the second is a political movement. As a final thought experiment, consider a 
place where social order has broken down and people resolve their conflicts by 
threatening or kiling their adversaries, versus a place where communal, tribal, or legal 
Lamb, Robert D. |  Appendix B. Complex Violence 485 
proceses exist to resolve conflicts and people actualy use those proceses when 
problems arise. 
The point of these thought experiments is not that finding a way to get 
narcotrafickers to use les violence would be a worthy intermediate policy goal since it 
would mitigate the harms of the ilegal activity (even if stopping short of ending it). Nor 
is it that finding a way to get terorists to participate nonviolently in political proceses, 
or finding a way to establish conflict resolution procedures where none currently exist, 
would be worthy goals as wel. These certainly would be worthy goals, and in fact 
achieving them in the real world would be impresive acomplishments. But there is a 
larger point. 
The larger point of these thought experiments is that the central problem, in al of 
our prexisting categories of security problems, is the use of violence by some actor to 
expres some emotion or to achieve some end; isolate violence from any one of these 
problems and the problem reduces to an injustice or a nuisance. This is not meant to 
trivialize injustices or social nuisances. But the introduction of violence complicates any 
situation and exacerbates any conflict, whereas its elimination simplifies both. 
These observations sem so obvious that they would not be worth making ? if 
not for the fact that making them explicit can help reduce what sems to be an intractable 
set of problems to what sems to be a potentialy tractable problem: We can view certain 
policy environments as having a such complex mix of (our own prexisting) problem 
categories that we can never hope to adequately map them out for the purposes of 
designing comprehensive and rational strategies to deal with them, or we can view such 
policy environments as problems of complex violence, stil dificult to understand and 
Lamb, Robert D. |  Appendix B. Complex Violence 486 
addres, but analyticaly more tractable. It is the later approach that has been taken for 
this study. 
The story of violence in Medel?n cuts across and hides betwen so many diferent 
categories and levels of analysis that studying it requires either an intradisciplinary 
analysis of some subset of the problem or an interdisciplinary synthesis by a team of 
experts.
1
 For this study, I tried to take a somewhat diferent approach. In Part I, a rather 
complicated story was told of armed actors, innocent bystanders, shifting loyalties, 
surprise betrayals, evolving policies, and policy failures. But the main character 
throughout the story was violence itself, or rather complex urban violence, from its 
conception and birth, its juvenile outbursts, its gradual evolutions and multiple 
dimensions, its sudden transformations and crises of identity, to where it is today, stil 
carying its distinctive character, but now (to extract diminishing returns from the 
metaphor) in mature form, without the baggage and idealism of its youth, and perhaps 
showing some early signs of aging. By making violence the main character, I hoped to 
demote the very complicated set of problem categories and armed actors in the city, from 
a cast of divas fighting for the spotlight, to a chorus of supporting actors, there mainly to 
help with the exposition and development of the main story line. 
In Chapter 8, I hoped to demonstrate that the multifarious security problems of 
Medel?n during the past 25 years could be analyticaly reduced to a problem of complex 
violence, and that the paterns of that violence could be described in a way that hints at 
                                                
1
 A god example of an interdisciplinary synthesis is Pablo Emilio Angarita Ca?as, H?ctor Galo, and 
Blanca In?s Jim?nez Zuluaga, Din?micas de Guera y Construci?n de Paz: Estudio Interdisciplinario del 
Conflicto Armado en la Comuna 13 de Medel?n. 
Lamb, Robert D. |  Appendix B. Complex Violence 487 
potential paths for intervention. In this appendix, I discuss the conceptualization and 
framework that informed Chapter 8?s description of the city?s paterns of violence. 
Thre pathologies taint the literature on violence in Medel?n, not only the 
academic literature but the popular pres and policy analyses as wel. First, too many 
studies and policy makers look simply at homicide rates and asume (without 
demonstrating) that they stand as an adequate proxy for ?violence? when even a casual 
review of the data on violence suggests it is a much more involved phenomenon than can 
be captured by only one indicator. In other words, in trying to explain violence, their 
dependent variable is homicides. Second, many recognize that Medel?n suffers not from 
violence but from ?violences? and their work describes the many types of violence in 
detail; but in trying to explain these violences (plural), their dependent variable ends up 
simply being violence (singular), lumping together al the diferent types and explaining 
them in aggregate as being ?the? result of culture, narcotraficking, gred, etc. A third 
pathology in the study of and reporting on violence in Medel?n is a tendency to ignore 
or, worse, explain away trends and paterns over time. This particular pathology afects 
mostly those organizations, social workers, and researchers who work with victims of 
violence. Typical is the claim by some people who work with trade unionists, for 
example, that violence, human rights violations, and violations of international 
humanitarian law against trade unionists has only one trend: worse. That claim is made 
even in years when homicides, disappearances, atacks, instances of torture, and so on 
have al declined in the previous several years; in the face of such declines, some people 
wil point to the one type of violence that has increased (threats) and to the fact that even 
if things were beter this year than last, they are worse here than elsewhere. 
Lamb, Robert D. |  Appendix B. Complex Violence 488 
Here, I studied violence across two dimensions over time. The first dimension is 
the relationship betwen perpetrator and victim, which I cal the context of violence, and 
the second is the manifestation of the act of violence itself. The framework developed for 
the World Health Organization?s (WHO) study of health and violence takes a similar 
approach, and, with some significant modifications informed partly by the work of 
Caroline O.N. Moser and Cathy McIlwaine and party by a desire to focus on violence 
resulting from action rather than neglect, it was largely the WHO framework that was 
used as a model for this study.
2
 The next section discusses the manifestations of violence 
(? B.1) and the one after discusses context (? B.2). The data that were collected about 
violence in Medel?n were presented in Chapter 8, which described the overal paterns of 
violence in Medel?n and, where available, Caicedo La Siera. 
Some definitions are in order. Violence, for the purposes of this study, is any 
behavior that causes one or more human beings death, bodily injury, physical pain, 
psychological trauma, suffering, or fear. Violence can manifest itself physicaly or 
psychologicaly, intentionaly or unintentionaly, and, when intentional, for instrumental 
or expresive purposes. It can be self-inflicted, perpetrated by one or two individuals 
either within the home or out in the community, or perpetrated by members of groups for 
social, economic, or political reasons. Death is the loss of human life. Bodily injury is 
nontrivial damage to a limb, organ, sense, or system resulting in a temporary disability 
(i.e. the injury can heal) or a permanent disability (it cannot heal). Physical pain is a 
                                                
2
 The definitions and categories used in this Chapter were partly informed by: Etiene G. Krug, et al., 
World Report on Violence and Health; Caroline O.N. Moser and Cathy McIlwaine, ?Latin American Urban 
Violence as a Development Concern: Towards a Framework for Violence Reduction?; Caroline O.N. 
Moser, ?Urban Violence and Insecurity: An Introductory Roadmap?; Cathy McIlwaine and Caroline O.N. 
Moser, ?Violence and Social Capital in Urban Por Comunities: Perspectives From Colombia and 
Guatemala?; Panos D. Bardis, ?Violence: Theory and Quantification.? 
Lamb, Robert D. |  Appendix B. Complex Violence 489 
nontrivial, unpleasant, and distresing sensation in any part of the body. Psychological 
trauma is a nontrivial, unpleasant, and distresing emotion that results in a temporary or 
permanent behavioral or personality disorder. Sufering is an unpleasant and distresing 
emotion arising in response to nontrivial pain, a nontrivial loss, or the anticipation of 
either. Fear is an unpleasant and distresing emotion arising in response to nontrivial 
danger, such as the anticipation of death, injury, or pain. 
Under these definitions, violence can be commited against nonhuman animals. 
Similarly, some might consider vandalism ? defacing, damaging, or destroying 
inanimate objects ? to be a form of violence against property. For the purposes of this 
study, however, both vandalism and violence against nonhuman animals wil be ignored, 
unles that behavior is intended as a threat, in which case it wil be considered an instance 
of psychological violence. Unintended violence wil likewise be ignored, as the purpose 
of this study is to understand how policy might be used not to prevent acidents but to 
change violent behavior. Finaly, the definition of violence proposed here stands a middle 
ground betwen those who would define violence in strictly physical terms (by excluding 
psychological violence) and those who would define it in broadly difuse terms (by 
including nonhuman ?systemic? perpetrators, as discussed in the next paragraph). 
To the thre categories of perpetrator proposed in this definition (self, individual, 
group), some writers would add system, using the term structural violence, incidental 
violence, or variants to describe the way social, political, and economic relations in 
society can result in fear, suffering, trauma, pain, injury, or death for disadvantaged 
members of society, usualy over a long period and out of the sight of the privileged. 
Similarly, some writers would expand the definition of violence beyond discret acts or 
Lamb, Robert D. |  Appendix B. Complex Violence 490 
behaviors to include ongoing relationships, for example, the ongoing exploitation of the 
weak or the marginalized by powerful individuals and groups. While structural violence 
and exploitive relationships certainly result in the same sorts of harms caused by acts of 
violence, they are far more difuse forms of violence, in which specific, individual, 
human perpetrators cannot easily be identified.
3
 They are diferent enough from the 
discrete behaviors perpetrated by self, individual, and group that they would more 
usefully be analyzed and addresed under other frameworks, such as human development 
or human security, designed to addres precisely those broader concerns.
4
 
B.1. Manifestations of Violence 
Violence is manifested in two main ways: physical and psychological. Physical 
violence is any behavior that involves physicaly touching the victim ? whether directly 
(beating, striking, shoving, etc.) or indirectly (throwing, launching, initiating, etc., as with 
a weapon, fire, energy, sound waves, etc.) ? in a way that causes the victim pain, injury, 
death, trauma, suffering, or fear. Psychological violence is any behavior that does not 
involve physicaly touching the victim but that nevertheles causes the victim to 
experience trauma, suffering, or fear. In other words, both manifestations of violence can 
result in psychological reactions ? trauma, suffering, and fear ? but only physical 
violence can result in pain, injury, or death as wel: behavior that does not result in pain, 
                                                
3
 Se, for example, Mary R. Jackman, ?License to Kil: Violence and Legitimacy in Expropriative Social 
Relations,? in The Psychology of Legitimacy: Emerging Perspectives on Ideology, Justice, and Intergroup 
Relations, ed. John T. Jost and Brenda Major (New York: Cambridge University Pres, 201), 437. 
4
 See, for example: United Nations Development Programe, Human Development Report 200: Human 
Rights and Human Development (New York: Oxford University Pres, 200); Comision on Human 
Security, Human Security Now (New York: Comision on Human Security, 1 May 203), 
http:/ww.humansecurity-chs.org. 
Lamb, Robert D. |  Appendix B. Complex Violence 491 
injury, or death but that does result in trauma, suffering, or fear counts as psychological 
violence if it does not involve physical contact betwen the victim and the perpetrator (or 
an object used by the perpetrator); it counts as physical violence if it does involve such 
physical contact. Obviously, then, suicides, homicides, masacres, and beatings wil 
always be considered examples of physical violence; rape is physical violence as well, 
even in those cases when it does not result in physical pain or injury, since it involves 
direct and very intimate physical contact. By contrast, any threat of physical violence 
wil always be considered a form of psychological violence (unles and until the threat is 
caried out, at which point it becomes physical). Likewise, some forms of extortion, such 
as protection rackets, that threaten nontrivial losses are forms of psychological violence 
as wel (but most forms of blackmail, for example, would not be considered a form of 
violence at al, because what is threatened would cause only a trivial form of suffering: 
embarasment). A single act, such as torturing a man in front of his wife, can count as 
both physical violence (against the man) and psychological violence (against the woman). 
Some writers exclude psychological forms of violence from their work, arguing 
that it is les direct, les measurable, or les damaging to the victim than physical 
violence. This, however, is shortsighted for two reasons. First, the two manifestations of 
violence are closely linked: psychological violence often results from the threat of 
physical violence. An armed robbery (psychological violence) might not technicaly 
count as a ?violent crime? in legal circles, but the chances are very high that it could, if 
just one thing ?goes wrong?; the fear of its going wrong is very real, and often very 
traumatic, to the victim. Forced prostitution (psychological violence) might be very 
diferent from forced rape ? the client does not need to use overwhelming force to get 
Lamb, Robert D. |  Appendix B. Complex Violence 492 
her to do what she wants, since she is pretending to want it or enjoy it ? but it does cary 
an implicit or explicit threat of physical violence (from her pimp) for noncompliance, and 
as such forced prostitution is no diferent from (and in fact, may be defined as) coerced 
rape: have sex with him, and pretend to like it, or face the devastating consequences. 
(Coercion is discussed below.) Likewise, merely witnesing an act of physical violence 
? a gang forces a man to watch them rape his wife, a mother receives a finger in the mail 
from her son?s kidnappers, a soldier is the sole survivor of a batle his platoon lost ? can 
create lasting harm to the witnes, in the form, for example, of post-traumatic stress 
disorder (PTSD) or chronic major depresion: the loss of one?s personality is not trivial. 
Second, psychological trauma is no les destructive to an individual than physical 
harm, and in many cases it can be even more devastating. One may physicaly recover 
from a shooting and go on to live a happy life; or the shooting might have been done 
under circumstances that were so traumatic that the victim?s family and friends say that 
?he was never quite himself again?: he recovered from the injury but lost a part of his 
identity or his humanity in the proces. Someone may be kidnapped without ever having 
been physicaly touched; but they suffer the very nontrivial loss of their fredom. A 
farming family might receive a series of writen death threats, followed by the nighttime 
kiling of their cows: they fle and suffer the nontrivial losses of their land, their home, 
their livelihood, and perhaps, depending on where they end up, their culture or their 
identity. The fact that kidnapping and forced displacement (more acurately caled 
coerced displacement) are forms of psychological violence does not make them any les 
devastating than if the victims had been beaten until their ribs were broken. 
Lamb, Robert D. |  Appendix B. Complex Violence 493 
Both physical and psychological forms of violence may be used for either 
instrumental or expresive purposes. Expresive violence is violence (physical or 
psychological) that is caried out for the psychological fulfilment of the perpetrator: for 
?fun,? for ?sport,? for revenge, or out of hatred, frustration, or boredom. Many cases of 
spousal abuse result from a (usualy) man?s inability to deal with life?s frustrations in a 
constructive manner, and he takes those frustrations out on his wife (usualy coming up 
with some elaborate post hoc rationalization having to do with her behavior). Some serial 
kilers were known earlier in their lives to have tortured animals for fun before moving 
on to human beings. Sadists are people who get enjoyment or sexual pleasure out of 
causing pain in others. Some gangs get started by tenagers with a lot of time on their 
hands and no social responsibilities; imagining a conflict with the kids a block over 
makes life a lot les boring, and fights break out as a result. Revenge kilings are the 
driving force behind some of the world?s most intractable conflicts, many of which are 
started over some relatively trivial event: 
The war betwen the Handa clan and the Ombal clan began many years 
ago; how many, Daniel didn?t say, and perhaps didn?t know. ? Among 
Highland clans, each kiling demands a revenge kiling, so that a war goes 
on and on, unles political considerations cause it to be setled, or unles 
one clan is wiped out or fles. When I asked Daniel how the war that 
claimed his uncle?s life began, he answered, ?The original cause of the 
wars betwen the Handa and Ombal clans was a pig that ruined a 
garden.?
5
 
Chapter 5 discusses the gang wars of Caicedo La Siera; despite the ostensibly political 
nature of the conflict some years after, the gang wars actualy originated in an argument 
betwen children, which generated cycles of resentments and revenge. In short, 
                                                
5
 Jared Diamond, ?Vengeance Is Ours: What Can Tribal Societies Tel Us About Our Ned to Get Even?? 
The New Yorker (208). 
Lamb, Robert D. |  Appendix B. Complex Violence 494 
expresive violence (whether physical or psychological) is done to no particular ends: 
nothing is asked of the victim, and nothing is given to the perpetrator but the satisfaction 
of having commited the violent act. 
Instrumental violence, by contrast, is violence done to achieve an end: to defend 
against or deter an atack, to take someone else?s property, to compel compliance with a 
demand, etc. Instrumental violence may take a physical or a psychological form, and the 
diference betwen instrumental-physical violence and instrumental-psychological 
violence is the diference betwen force and coercion, as Scheling eloquently described 
in Arms and Influence: it is the ?diference betwen taking what you want and making 
someone give it to you, betwen fending of asault and making someone afraid to asault 
you, betwen holding what people are trying to take and making them afraid to take it, 
betwen losing what someone can forcibly take and giving it up to avoid risk or 
damage.?
6
 In Scheling?s view, suffering, the psychological response to pain or to the 
threat of pain, is the bridge betwen force and coercion: 
To inflict suffering gains nothing and saves nothing directly; it can only 
make people behave to avoid it. The only purpose, unles sport or revenge, 
must be to influence somebody?s behavior, to coerce his decision or 
choice. To be coercive, violence has to be anticipated. And it has to be 
avoidable by acommodation. ? Coercion requires finding a bargain, 
aranging for him to be beter of doing what we want ? worse off not 
doing what we want ? when he takes the threatened penalty into 
acount.
7
 
Force (instrumental-physical violence) uses power or pain to induce behavior, while 
coercion (instrumental-psychological violence) uses fear or suffering to induce voluntary 
behavior. Coercion achieves compliance by credibly expresing an intent to use physical 
                                                
6
 Thomas C. Scheling, Arms and Influence (New Haven: Yale University Pres, 196), 2. 
7
 Ibid. 
Lamb, Robert D. |  Appendix B. Complex Violence 495 
violence (or further violence) or the loss of something of value as punishment for 
noncompliance, while also credibly asuring the victim that that outcome wil be avoided 
by complying. A single event can count as force against one victim and coercion against 
another. If a militia enters a vilage, kils the men, rapes the women, castrates the boys, 
sels the girls into slavery, then burns the vilage to the ground, but spares al of the 
neighboring vilages to give them the opportunity to surrender or suffer the same fate, 
then the militia is using force (physical violence) against one set of victims as a means of 
coercion (psychological violence) against another set of victims. If a drug dealer is kiled 
by a vigilante group, whose members then mutilate his corpse and display it in public 
with a sign around his neck teling other drug dealers to leave the neighborhood (a not-
uncommon occurrence in Medel?n?s history), then the specific way the physical violence 
was caried out against the victim amounted to an intentional act of psychological 
violence against other neighborhood drug dealers. Likewise, physical violence can be 
used as a strategy of psychological violence against one and the same victim. In a strong-
arm robbery, the perpetrator punches or knocks the victim to the ground, then stops and 
demands money: the victim?s compliance derives from the (psychological) fear of further 
(physical) violence. 
B.2. Context of Violence 
Al violence takes place in a particular context that can afect both the severity of 
the victim?s psychological response and, often, the brutality of the violent act itself. That 
context has to do with the relationship betwen victim and perpetrator, the motivations 
Lamb, Robert D. |  Appendix B. Complex Violence 496 
underlying the act of violence, or both. This implies thre levels of analysis: the 
individual level, the interpersonal level, and the colective level. 
Taking place at the individual level is self-inflicted violence, which involves only 
one person who is both perpetrator and victim. Its most common physical manifestations 
are self-mutilation, suicide, and suicide atempts. Self-inflicted psychological violence 
often derives from mental ilnes and can include drug abuse and addiction, participation 
in extremely risky behaviors such as erotic asphyxiation and anonymous promiscuity, 
eating disorders such as anorexia, or self-neglect that leads to infectious disease or 
unsustainable weight loss or weight gain. 
Violence at the interpersonal level involves one or two perpetrators acting against 
one or more victims with whom they either are intimately related (as family members, 
life partners, close friends, etc.) or are pasing acquaintances or strangers. Intrafamilial 
violence is interpersonal violence that takes place inside the home or among close 
relatives or acquaintances (including spousal abuse, child abuse, elder abuse, pedophilia, 
date rape, etc.), while communal violence involves one or two perpetrators acting against 
one or more victims whom they may or may not know, and it may take place in 
someone?s home, on the stret, at a commercial property, or in some other public area 
(this includes most categories of common, rather than organized, crime: robbery, asault, 
rape, etc.). Note that this usage of the term communal difers from that used by some 
conflict researchers who use communal conflict or intercommunal conflict in a way that is 
similar to how I use collective violence. To avoid confusion, I wil usualy use the terms 
interpersonal-communal or intrafamilial-communal violence, but even without the 
modifiers the reader should keep the intended meaning in mind throughout this work. 
Lamb, Robert D. |  Appendix B. Complex Violence 497 
Violence at the collective level involves one or more perpetrators, who are 
members of an identifiable group, acting against one or more victims for social, 
economic, or political reasons. For the purposes of this definition, a group is considered 
to include thre or more people who interact on a more or les regular basis for social, 
economic, or political reasons; the choice of thre people rather than two people is 
esentialy arbitrary, but it does fit with common usage (two people make a ?couple? or a 
?pair,? not a ?group?) and does have some precedents (for example, Medel?n?s morgue 
defines a masacre as the kiling of thre or more people during a single event). 
Table B-1. Contexts and Manifestations of Violence, with Examples 
Context Manifestation 
Level Type Physical Violence Psychological Violence 
Individual  Self-inflicted ?suicide 
?self-mutilation 
?substance abuse 
?self-destructive behavior 
Interpersonal Intrafamilial ?spouse, child, or elder 
abuse 
?pedophilia, date rape 
?child or elder neglect 
?stret children (proxy) 
 Comunal ?assault and battery 
?rape 
?robery/armed robery 
?fear of crime (proxy) 
Colective Social ?turf wars 
?hate crimes 
?social cleansing 
?mutilation of corpses as public 
intimidation 
?stret gangs, vigilantes 
 Economic ?kidnapping 
?mafia wars 
?anti-union violence 
?organized crime 
?extortion 
?coerced prostitution 
 Political ?war/combat 
?masacres 
?displacement 
?disappearance 
?terorism 
?imprisonment 
?state security forces, 
guerillas, paramilitaries 
?deterence and compellence 
policies 
?terorism as public 
intimidation 
The examples in this table are ilustrative only. Some (identified as ?proxy?) are not acts of 
violence per se, but consequences of violence, included here as potential indicators for the 
category. Others (such as ?street gangs?) are meant simply to identify the type of perpetrator 
with which that category of violence is most strongly identified and whose presence alone is 
intimidating (and so a potential indicator). 
Source: World Health Organization, with author?s modifications and examples 
Lamb, Robert D. |  Appendix B. Complex Violence 498 
Collective violence is social violence if the group is a social group, formed for 
reasons of afection, common interest, or identity (such as youth gangs, stret gangs, 
civic asociations, etc.) and if the act of violence is commited to enforce respect, expres 
pride in the in-group, expres disdain for an out-group (such as a competing gang or a 
diferent racial or ethnic group), to defend or acquire teritory for the purpose of social 
control, or to punish members for violating in-group rules, among other reasons. 
Collective economic violence involves any group formed for the primary purpose 
of making money, whether from the sale of goods or services (legal or ilegal) or by theft 
or fraud, which includes most organized crime groups and other criminal conspiracies: 
mafias, drug trafickers, money launderers, asasins for hire, crime rings, racketers, and 
otherwise legitimate busines managers using violence to coerce or enforce a cartel 
arangement or to take or protect asets (as in the forced disappearance of union 
organizers). 
Finaly, collective political violence involves acts commited by one or more 
members of a group that makes or purports to be making decisions about political goods 
(e.g. they are doing the job they think a government should be doing), which may include 
agents of states and governments (police, military, and other security forces) whether 
acting in their official capacity or not, or may include nonstate or ilegal organizations 
such as paramilitaries, vigilante groups, self-defense groups, militias, or guerilas. Acts 
of political violence can include war, asasination, forced displacement, imprisonment, 
and the forced disappearance of political opponents and human-rights defenders, among 
other acts. 
Lamb, Robert D. |  Appendix B. Complex Violence 499 
This framework is summarized in Table B-1, which describes the context of 
violence in terms of both the level at which the perpetrator operates (individual, 
interpersonal, or collective) and the specific type of relationship the perpetrator has with 
the victim; and describes the manifestation of violence in both its physical and its 
psychological forms, with examples, proxies, and indicators that might be useful as 
measures. 
How is it useful to categorize violence in the way described in this framework? 
Two ways: First, it provides the vocabulary to describe any given act of violence: A guy 
stops a bus at the end of its route and tels the driver to pay his boys a large amount of 
money to wash and guard his bus overnight. The driver refuses, saying the amount of 
money is too high for a cleaning, and he?s never had problems with vandalism before so 
he doesn?t need a guard. He goes home for the day. He returns the next morning to find 
his bus a chared wreck, completely burned inside and out. The bus driver is not a victim 
of physical violence, because he himself was never touched: it was an inanimate object 
that was damaged. It was, however, psychological violence, because the driver suffered a 
nontrivial loss (of his only source of family income). Second, analyzing violence through 
these categories helps to generate a reasonable initial asesment both of the general 
structure of the problem ? for example, if most violence is collective (that is, organized) 
rather than communal (a more difuse problem), that suggests the underlying problem is 
not primarily a breakdown in social order but a reflected of some larger conflict ? and of 
the general motivations driving the violence (e.g. some forms of political violence could 
represent real grievances against the government, whereas purely economic violence is 
litle more than an expresion of gred). Robin M. Wiliams gives an example as wel of 
Lamb, Robert D. |  Appendix B. Complex Violence 500 
the diferences in psychological motivations behind interpersonal versus collective 
violence: 
those persons who commit violent crimes difer greatly in personality 
characteristics from persons who serve in wartime military forces. 
[Interpersonal] violence tends to isue from persons who have dificulty 
establishing satisfactory group ties and enduring interpersonal 
relationships, who frequently clash with established authority and group 
norms, and whose violence is often self-defeating and self-punishing. In 
contrast, ?good soldiers? are highly responsive to demands for social 
conformity, readily adjust to giving and receiving orders, have strong 
afiliative capacities, and function reliably as loyal members of groups. ? 
Of course, there wil be intermediate cases ? for example, asasinations 
or acts of sabotage caried out by isolated individuals who nevertheles 
se themselves as participants in mas movements.
8
 
The utility of the conceptualization and framework presented here are perhaps 
easiest to se in a real-world context, and the reader is refered to Chapter 8, which 
applied this framework to a review of the quantitative and qualitative data that were 
available about violence in Medel?n. 
                                                
8
 Robin M. Wiliams, Jr., ?Legitimate and Ilegitimate Uses of Violence: A Review of Ideas and Evidence,? 
27 and n. 5 (his citation omited). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 501 
Apendix C. Legitimacy and Ilegitimacy 
The question may be discussed thus: a prince who fears his own people 
more than foreigners ought to build fortreses, but he who has greater 
fear of foreigners than of his own people ought to do without them. ? 
Therefore, the best fortres is to be found in the love of the people, for 
although you may have fortreses they wil not save you if you are hated 
by the people. When once the people have taken arms against you, there 
wil never be lacking foreigners to assist them. 
 Nicol? Machiaveli (1513)
1
 
 
Legitimacy is a fundamental concept in several fields of human knowledge; it 
plays an important role in the political, sociological, anthropological, psychological, 
philosophical, busines management, and organizational studies literatures and, 
increasingly, in policy and doctrine for diplomacy, development, warfare, and other 
endeavors. Because people are motivated to voluntarily support that which they consider 
to be ?legitimate,? legitimacy is said to explain the stability of social, economic, political, 
and other collective systems. If legitimacy is correlated with stability, then those who are 
interested in building and sustaining a set of relationships, a set of rules, or a distribution 
of power, wealth, prestige, or status are, by that fact, interested in understanding, 
achieving, and maintaining its legitimacy; likewise, those who are interested in changing 
those relationships, rules, and distributions are thereby interested in understanding and 
chalenging their legitimacy. 
                                                
1
 Nicol? Machiaveli, The Prince, 108. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 502 
Theories of rational choice have dominated the fields of political science, 
economics, and to some degre policy studies for several decades, so it is worth briefly 
considering what distinguishes the work that that set of theories does from what theories 
of legitimacy do. Karol Soltan has said that the entire literature on rational-choice theory 
can be summarized in four words: ?People respond to incentives.?
2
 I would argue, albeit 
les succinctly, that the entire literature on legitimacy is a series of variations and 
elaborations on an equaly simple observation: People are motivated by what is right. 
Incentives certainly mater, especialy in the short term, but what people think is right 
matters as well. 
People can be induced to behave in certain ways through the application or 
manipulation of external incentives, but external incentives are costly in both efort and 
resources. Because people are self-motivated to behave in ways that are consistent with 
their own views of what is right and wrong, what is good and bad, and what is virtuous 
and vicious, systems of rules and relationships based on some blend of those views (or on 
habits and norms of behavior that sem not inconsistent with those views) tend to be les 
costly and more stable in the long run. Thus, for example, the problem of collective 
action ? which posits that certain activities, namely those that would benefit some group 
of people but that are costly to initiate, would not take place in the absence of selective 
incentives to initiators ? presents a real puzzle only if one asumes that people merely 
respond to incentives: When a few individuals actualy initiate activities that benefit a 
larger group of people without also selectively benefiting themselves, their behavior 
                                                
2
 Karol Soltan, personal communication, 8 April 2008. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 503 
presents no puzzle to those of us who asume that people are motivated by and are 
wiling to sacrifice for what they think is right. 
Moreover, when policy makers ? not only for states, but for nonstate actors, non-
profit organizations, busineses, and any other type of institution ? asume only that 
people respond to incentives, they might find themselves facing perverse results: The day 
care center that wants parents to pick up their children on time and implements as an 
incentive a policy to fine those who are late might find that the parents no longer consider 
showing up on time to be a moral imperative, since the fine is now equivalent to a fe for 
service (the service being extended-hours daycare), and that latenes actualy increases as 
a result.
3
 Likewise, in a violent conflict over something the contenders consider to be 
sacred rather than material, ofering material incentives in exchange for compromise is 
often taken as a deep insult and can trigger an increase, rather than a decrease, in 
violence.
4
 People might be les wiling to help a friend load furniture into a moving van if 
they are offered money than if they are offered nothing more than gratitude: in one 
scenario it is labor and the pay is perhaps not worth the efort; in the other it is a favor to 
a friend, something offered gladly and without ulterior motive.
5
 ?Teachers incentivized to 
produce higher test scores get higher test scores but not beter-educated students. CEOs 
incentivized to improve the performance of the company?s shares improve the 
                                                
3
 Uri Gnezy and Aldo Rustichini, ?A Fine Is a Price,? Journal of Legal Studies 29, no. 1 (200). 
4
 Jeremy Ginges, et al., ?Sacred Bounds on Rational Resolution of Violent Political Conflict,? Procedings 
of the National Academy of Sciences 104, no. 18 (207): 7357. 
5
 James Heyman and Dan Ariely, ?Efort for Payment: A Tale of Two Markets,? Psychological Science 15, 
no. 11 (2004): 787. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 504 
performance of the company?s shares but not the performance of the company.?
6
 As the 
psychologists who developed self-motivation theory put it (with respect to work 
environments, but equaly applicable in other contexts): 
Humans are inherently motivated to grow and achieve and wil fully 
commit to and engage in even uninteresting tasks when their meaning and 
value is understood. ? Carot and stick approaches to motivation ? lead 
to a heightened focus on the tangible rewards of work rather than on the 
nature and importance of the work itself. Such approaches can create 
short-term productivity increases by controlling people?s behavior, but the 
resulting motivation is of poor quality ? it is unsustainable and ? tends 
to undermine intrinsic interest in work. [Focusing on] and [nurturing an 
interest in] the intrinsic importance of work ? has been shown to link to 
beter performance, especialy in the complex, creative, and heuristic tasks 
that increasingly characterize modern work.
7
 
Or as Mark Twain?s Tom Sawyer once learned: ?Work consists of whatever a body is 
obliged to do, and ? Play consists of whatever a body is not obliged to do.?
8
 
In short, people do respond to incentives, but they also are motivated by what is 
right or what they believe is right. This disertation studied the later motivation in the 
context of complex urban violence, a context that has nothing if not complex sets of 
incentives. 
The study of extreme cases ? cases with extremely high or extremely low values 
on the study variables ? can help to iluminate more general relationships betwen and 
                                                
6
 Bary Schwartz and Keneth Sharpe, ?The Wrong Cure: Beter Incentives, for Doctors or Insurers, Won?t 
Lead to Beter Health Care,? Slate, htp:/ww.slate.com/id/224193/ (acesed 4 August 209). 
7
 Dan N. Stone, Edward L. Deci, and Richard M. Ryan, ?Beyond Talk: Creating Autonomous Motivation 
Through Self-Determination Theory,? Journal of General Management 34 (209): 75. (authors? citations 
omited). 
8
 Mark Twain, The Adventures of Tom Sawyer (New York: Harper & Brothers Publishers, 1920 [1875]), 
19. (emphasis in original); se same quote in James Heyman and Dan Ariely, ?Efort for Payment: A Tale 
of Two Markets.? 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 505 
among the phenomena under study.
9
 Medel?n is an extreme case of high violence, high 
both in its level of violence (it had the highest murder rate in the Western Hemisphere in 
1991 and again in 2002) and in its level of complexity (in terms of the variety of actors, 
their shifting motivations, and the instability of their aliances). But Medel?n is also an 
extreme case of instrumental social relations: social capital and trust among strangers are 
extremely low, and the tendency for expediency or exploitation to be used to achieve 
short-term results at the expense of long-term relationships is extremely high; it?s the 
kind of place for which Imanuel Kant must have imagined his hypothetical Kingdom of 
Ends as a corrective.
10
 As such, it can be argued that, if a role for legitimacy can be found 
in an explanation for violence and stability in a place such as Medel?n, then legitimacy 
can be said to at least partly explain something about violence and stability more 
generaly. Does legitimacy ? or its opposite, ilegitimacy ? add anything to our 
understanding of violence in Medel?n that cannot be adequately explained by incentives? 
If so, what does that imply for theorists of legitimacy or, more importantly, for policy 
makers charged with reducing violence in complex environments? 
This appendix analyzes the concepts of legitimacy and ilegitimacy, discusses 
some isues involved in measuring them in the real world, and introduces the multi-
dimensional, multi-level framework that was used in this study to evaluate the dynamics 
of legitimacy and ilegitimacy in Medel?n (se Chapter 9). It begins with a discussion 
and definitions of legitimacy, ilegitimacy, and related terms and briefly considers their 
relationships to the concepts of loyalty, support, right, duty, imitation, expertise, and 
                                                
9
 Robert K. Yin, Case Study Research: Design and Methods, 47; Stephen Van Evera, Guide to Methods for 
Students of Political Science, 24. 
10
 Immanuel Kant, Groundwork of the Metaphysic of Morals. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 506 
leadership. This discussion exposes not just what legitimacy and ilegitimacy are, but 
what they are said to do: legitimacy motivates compliance, ilegitimacy motivates 
opposition; legitimacy sustains, ilegitimacy impedes (? C.1). Other phenomena or 
behaviors ? agrement, habit, seduction, persuasion, compromise, force, coercion, 
barter, and deception ? can achieve similar outcomes to what legitimacy does; they 
therefore can be mistaken for legitimacy or can be used as alternatives to legitimation 
(? C.2). These ?alternatives? to legitimacy complicate eforts to measure it: since 
legitimacy is a latent (unobservable) phenomenon, one can rarely be fully confident that 
what one is measuring is legitimacy and not something else. To help overcome this 
problem, I have developed a conceptual framework through which key questions about 
the phenomenon being measured might be answered (?? C.3-C.6): Legitimacy of what? 
Legitimacy according to whom? Legitimacy by what proceses? And, Legitimacy by what 
criteria? From these questions comes a discussion of the measurement framework itself 
(? C.7), which lays out the diferent legitimizing characteristics that a role, policy, 
distribution, or structure might have (transparent, credible, justifiable, equitable, 
acesible, and respectful) and the diferent levels at which they might be measured 
(individual beliefs, group behaviors, and system features), and identifies a provisional set 
of indicators that may be evaluated to determine the degre to which nonstate, state, and 
community actors might be considered legitimate. Chapter 9 reported the results of the 
analysis, guided by this framework, of the dynamics of legitimacy and ilegitimacy of 
those actors. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 507 
C.1. What Is Legitimacy, and What Does It Do? 
It was Thucydides whose writings on an ancient war were the first to have 
survived history with an extended discussion of legitimacy ? the concept if not the term 
? intact. In his History of the Peloponnesian War, the Grek historian imagines the 
dialog betwen representatives of the island of Melos, who had wished to stay neutral in 
the war betwen Athens and Sparta, and representatives of Athens, who wanted to take 
over the strategicaly located island and who ofered the Melians a chance to surrender or 
die. The Athenians pointed out that their force was so far superior that resistance would 
be futile and deadly; the Melians offered a series of pragmatic and increasingly 
preposterous (in Thucydides? teling) arguments as to why the Athenians should let them 
remain neutral and not take over their island. Implicit to some of their initial arguments 
was the credible moral claim that atacking or enslaving the Melians would violate 
principles of honor and right action, principles held by Athenians and Melians alike: just 
because the Athenians would easily win the batle did not mean it was right that they 
should wage it. The Athenian retort that ?the strong do what they can and the weak suffer 
what they must? ? a death sentence to the Melians ? was esentialy the same argument 
that Thrasymachus made to Socrates, in Plato?s acount, regarding what justice entails 
and implies.
11
 But the Melians? pragmatic response to the Athenians is in some ways 
more satisfying than Socrates?s airy response to Thrasymachus (and, later, Glaucon). 
They said that if Athens atacked Melos, the other neutral islands would learn about it 
and, fearing the same fate, would band together in self-defense. In short, they argued, 
                                                
11
 Thucydides, The History of the Peloponesian War, trans. Richard Crawley, vol. 6, Great Boks of the 
Western World (Chicago: Encyclop?dia Britanica, 1952 [431 BCE]), book V; Plato, The Republic of 
Plato, trans. Francis MacDonald Cornford (New York: Oxford University Pres, 1945 [370 BCE]), bok I. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 508 
power unchecked and ilegitimate is unsustainable, as it tends to trigger opposition. The 
Athenians responded that they would take their chances. And indeed even after they 
besieged the island, annihilated the men, and enslaved the women and children, the other 
neutral islands did not band together as the Melians had hopefully predicted. (It is 
doubtful that any surviving Melians would have taken any comfort in learning that 
Athens did end up losing the war and its empire anyway.) 
The Melians were implicitly, if ineloquently, trying to tel the Athenian invaders 
something about legitimacy and ilegitimacy that history has long demonstrated, namely 
that legitimacy contributes to long-term stability while ilegitimacy contributes to short-
term instability. They were hoping that, if the Athenians did not buy a moral argument 
about the legitimacy of their proposed actions, they would buy a military argument about 
the consequences of ilegitimate action. Since then, writers interested in the moral 
question, some of whom have cited the Melian dialogs as the first published acount of 
arguments for legitimate action, have focused on legitimacy and its role in maintaining 
stability or, les commonly, on ilegitimacy and its role in undermining stability, while 
writers interested in the military question have focused both on legitimacy as something 
to be earned and ilegitimacy as something to be avoided: In 4
th
 Century BCE Grece, 
Aristotle, concerned with how to live a good life and create a just political system, was 
interested in the legitimacy of constitutions and of the distributions of the rewards of 
social life. In 3
rd
 Century BCE India, Ashoka, concerned with right conduct and the 
stability of his empire, acted on his belief in the legitimizing influence of transparent 
rules, public deliberation, and respect for diferences. In 16
th
 Century Italy, Nicol? 
Machiaveli, concerned with how politicians remain in power, was interested both in how 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 509 
ilegitimacy is avoided and how legitimacy is earned, paying atention, for example, to 
the public behavior of leaders and their efectivenes as leaders within the limits of what 
is aceptably cruel. In 17
th
 Century England, John Locke, concerned with how 
governments should operate, was interested in the sources of legitimacy, concluding that 
the consent of the governed is the most important. In 18
th
 Century France, Jean-Jacques 
Rousseau combined Machiaveli?s concern for the sources of stability with Locke?s 
concern for the sources of legitimacy and concluded that consent of the governed is the 
main source of both. In 19
th
 Century Germany, Karl Marx and Friedrich Engels, 
concerned about the wel-being of the working clases, were interested in the legitimizing 
myths that the ruling clases imposed upon society as a way to maintain disparities in 
wealth and power. And in early 20
th
 Century Germany, Max Weber described the 
proceses through which social order is maintained, linking individual belief to group 
behavior in the first multilevel theory of the legitimation of social orders, and thereby 
bringing the study of legitimacy out of the realm of political philosophy and into the 
realm of social science.
12
 
Most of these and other authors never used the term legitimacy in their writings 
(Weber was the most important exception), but the underlying dynamics they were 
                                                
12
 Aristotle, Politics, trans. Ernest Barker and R. F. Staley, The World?s Clasics (New York: Oxford 
University Pres, 195 [30 BCE]); Ven. S. Dhamika, The Edicts of King Asoka: An English Rendering 
(Barre, Mas.: The Budhist Publication Society and DharmaNet, 1993 [ca. 257 BCE]), 
http:/ww.cs.colostate.edu/~malaiya/ashoka.html (acesed 21 July 2009); Amartya Sen, ?Why 
Democratization Is Not the Same as Westernization: Democracy and Its Global Rots,? The New Republic 
229, no. 14 (2003): 28; Nicol? Machiaveli, The Prince; John Locke, ?An Esay Concerning the True 
Original, Extent, and End of Civil Government,? in Social Contract: Esays By Locke, Hume, and 
Rouseau, ed. Ernest Barker, The World?s Clasics, No. 51 (New York: Oxford University Pres, 1960 
[1690]); Jean-Jacques Rouseau, ?The Social Contract,? in Social Contract: Esays By Locke, Hume, and 
Rouseau, ed. Ernest Barker, The World?s Clasics, No. 51 (New York: Oxford University Pres, 1960 
[1762]); Karl Marx and Friedrich Engels, The German Ideology, trans. W. Lough and C.P. Magil (London: 
Lawrence & Wishart, 1938 [1845]); Max Weber, Economy and Society: An Outline of Interpretive 
Sociology (New York: Bedminster Pres, 1968 [1914]). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 510 
refering to was the same. In fact, the term legitimacy was not commonly used in English 
to describe the set of phenomena with which it is currently asociated until the 20
th
 
Century. It derives from the Latin lex (law) and legitimus (lawful), and from Medieval 
Latin?s legitimare (to make or declare to be lawful) and its past participle legitimatus, 
which entered Middle English around 1494 as legitimat. But the context in which this 
form was used in Middle English was initialy restricted to questions regarding rights to 
inheritance (of goods or titles) by children born outside of legaly, religiously, or socialy 
sanctioned mariages; such a child in a royal family, for example, could not become king 
or queen, or one from a common family could not presume to have the same right as his 
or her half-siblings to inherit their parents? property. The term ilegitimate entered 
English around 1536 to refer to those children born out of wedlock. In the 17
th
 Century, 
legitimate was becoming used more in line with the Latin original, to refer to something 
in acordance with laws or norms (and not just those regulating the rights of children 
born out of wedlock), and by the early 19
th
 Century this usage broadened further to refer 
to acordance with fact (?genuine, real?).
13
 As the laws regulating the rights of 
ilegitimate children were liberalized over time, the emphasis in the term?s usage shifted 
increasingly toward the more general sense (of acordance with laws, facts, rules, norms, 
expectations, and so on) such that by the 20
th
 Century this broader sense came to 
dominate usage. 
When something is or someone acts in acordance with these standards, what is 
the result? In other words, what does legitimacy do? Morris Zeldtich Jr. reviewed 
                                                
13
 Online Etymology Dictionary, s.v. ?legitimate? and ?ilegitimate,? htp:/ww.etymonline.com (acesed 
22 July 2009); The Concise Oxford English Dictionary of Etymology, s.v. ?legitimate,? 
http:/ww.encyclopedia.com/doc/1O27-legitimate.html (acesed 2 July 209). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 511 
24 centuries? worth of literature on legitimacy and, observing in those literatures a broad 
range of circumstances in which it has been found to play a role, summarized what 
legitimacy entails and what it produces: ?legitimacy is always a mater of voluntarily 
acepting that something is ?right,? and its consequence is always the stability of 
whatever structure emerges from the proces.?
14
 To this formulation, most theories of 
legitimacy offer only useful complications. Many of those complications have to do with 
the mechanisms through which legitimacy is said to bring about stability: Some have said 
legitimacy induces voluntary compliance, a mechanism for stabilizing relationships (or 
systems) of authority or dominance. Others have said legitimacy validates certain paterns 
of behavior and thereby provides a predictable regularity, and therefore stability, to social 
structures, political systems, and social orders in general. Others have said it encourages 
political participation or social actions that demonstrate or reinforce support. Stil others 
have said it lowers the barier to entry into markets or the cost of controlling teritory or 
other resources. Some of these mechanisms are discussed presently. 
Weber defined the term domination as ?the probability that a command with a 
given specific content wil be obeyed by a given group of persons?
15
 and said that ?the 
legitimacy of a system of domination may be treated sociologicaly only as the 
probability that to a relevant degre the appropriate atitudes wil exist, and the 
corresponding practical conduct ensue.?
16
 The ?appropriate atitudes,? he argued, have to 
do with people?s beliefs about what makes the system legitimate: the grounds for those 
                                                
14
 Moris Zelditch, Jr., ?Theories of Legitimacy.? 
15
 Max Weber, Economy and Society: An Outline of Interpretive Sociology, 53. 
16
 Ibid., 214. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 512 
beliefs may be rational (?resting on a belief in the legality of enacted rules and the right 
of those elevated to authority under such rules to isue commands?), traditional (?resting 
on an established belief in the sanctity of imemorial traditions and the legitimacy of 
those exercising authority under them?), or charismatic (?resting on devotion to the 
exceptional sanctity, heroism or exemplary character of an individual person, and of the 
normative paterns or order revealed or ordained by him?).
17
 In other words, the presence 
of belief in legitimacy is an indicator of legitimacy. 
The ?corresponding practical conduct? that ensues from these beliefs is not 
merely compliance with a command, acording to Weber, but just about any action that is 
oriented to what he caled a ?valid? social order. The presence of such behaviors is also 
an indicator of legitimacy. Weber argued that many individuals behave in ways that sem 
to derive from a personal belief in certain ?maxims? (norms, values, principles, symbols, 
etc.) about what counts as a duty or as proper behavior; the fact that many such 
individuals behave that way creates a general expectation that behaving in contrary ways 
wil result in social disapproval or legal sanction; ?proper? conduct therefore becomes 
routinized, defined as normal, and, in Weber?s term, validated; and so even if some 
individuals do not themselves believe in the maxims that they believe others believe in, 
those nonbelievers nevertheles orient their own actions toward an understanding that a 
social order based on belief in those maxims does nevertheles exist. In other words, what 
makes a social order valid is the fact that people usualy act as if they believe that others 
                                                
17
 Ibid., 215. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 513 
act as if they believe in the existence of a set of legitimate maxims for duty and proper 
conduct, and this validation of proper conduct is what makes the social order stable.
18
 
Following Weber, Sanford M. Dornbusch and W. Richard Scott used the term 
propriety to refer to personal (individual-level) judgments about the legitimacy of the 
values, norms, symbols, practices, etc. of organizations, and validity to refer to 
organizational (group-level) judgments about what is legitimate; action undertaken to 
support those values, norms, practices, etc. they caled endorsement when undertaken by 
individuals and authorization when undertaken by the group.
19
 Zeldtich and Henry A. 
Walker drew on this terminology to argue that validity is the main proces that maintains 
stability, as Zeldtich explained: 
Validity has a straightforward social influence efect on propriety, and 
therefore has a canceling efect on impropriety. But in addition, it embeds 
the norms, values, beliefs, practices, and procedures of a group in a system 
of social controls, creating expectations of both authorization and 
endorsement if they are violated, which also counteract presures to 
change.
20
 
In short, what Weber and his followers have said legitimacy does ? and most 
subsequent writing on legitimacy has elaborated, complicated, or contested his findings 
? is two things: at the (micro) level of individual interaction, legitimacy generates 
voluntary compliance with commands, and at the (macro) level of the system or social 
order as a whole, it generates and validates a more complex and difuse set of behaviors 
                                                
18
 Ibid., 31. 
19
 Sanford M. Dornbusch and W. Richard Scot, Evaluation and the Exercise of Authority, 1st ed ed. (San 
Francisco: Josey-Bas Publishers, 1975). 
20
 Moris Zelditch, Jr., ?Theories of Legitimacy,? citing Moris Zelditch, Jr. and Henry A. Walker, 
?Legitimacy and the Stability of Authority,? in Advances in Group Proceses, ed. Edward J. Lawler 
(Greenwich, Con.: JAI Pres, 1984), 1. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 514 
that tend to sustain a social order. Legitimacy, in other words, induces voluntary 
compliance and generates social stability. 
In the middle of the 20
th
 Century, the political sociologist Seymour Martin Lipset 
shifted the conversation from the legitimation of social orders to the legitimation of 
political systems. ?Legitimacy involves the capacity of a political system to engender and 
maintain the belief that existing political institutions are the most appropriate or proper 
ones for the society.? Lipset focused specificaly on the legitimacy of states and in 
particular on the legitimacy of democratic states and the complex ways in which stability, 
economic development, efectivenes, and legitimacy afect each other. Like Weber, he 
argued that legitimacy derives in part from beliefs about what counts as valuable, but he 
made an important extension by arguing that the legitimacy of a system derives at least in 
part from the efectivenes of the system: ?The extent to which contemporary democratic 
systems are legitimate depends in large measure upon the ways in which the key isues 
which have historicaly divided the society have been resolved.?
21
 And so Lipset was les 
concerned about the compliance action of legitimacy than about its efects on the stability 
of the overal system.
22
 
In this he was not alone. In the middle of the century, in the context of a Cold War 
in which it was not yet clear whether democratic or Communist systems would prevail, 
most English-language authors who wrote about legitimacy (mainly, political sociologists 
and political scientists) were concerned about questions of political stability, and 
                                                
21
 Seymour Martin Lipset, ?Some Social Requisites of Democracy: Economic Development and Political 
Legitimacy,? American Political Science Review 53, no. 1 (1959): 69. 
22
 Seymour Martin Lipset, Political Man: The Social Bases of Politics, expanded ed. (Baltimore: Johns 
Hopkins University Pres, 1981 [1961]). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 515 
especialy the stability of democracies. Amid that era?s rapid growth in the availability of 
comparative country data, the political scientist David Easton encouraged the 
development of more systematic frameworks to organize the data so that the stability of 
political systems might be studied more scientificaly. Defining politics as the 
?authoritative alocation? of a society?s values (wealth, power, prestige, etc.), Easton 
analyzed legitimacy within a framework of support for such alocations. Such political 
support, he argued, can be offered either to specific political authorities within the system 
or to more difuse structures, such as the political community or the political system as a 
whole.
23
 In other words, like Weber, Easton found that legitimacy has both specific and 
difuse forms (and therefore should be expected to have both specific and difuse efects). 
Those who followed Easton?s lead ? and there were many ? focused as he did 
on the role of political support in sustaining political systems, but much of the empirical 
evidence for their research ended up coming from data about political participation and 
from survey data about trust in government. Russel J. Dalton, for example, studied how 
legitimacy afects participation in advanced industrial democracies. In addition to 
providing empirical support for the multidimensionality of Easton?s framework (i.e. the 
specific and difuse forms of legitimacy), Dalton?s work was an important contribution to 
a string of studies that cumulatively suggested that legitimacy, in addition to inducing 
compliance and generating stability, also increases political participation.
24
 In that line of 
                                                
23
 David Easton, A Systems Analysis of Political Life (New York: Wiley, 1965); David Easton, A 
Framework for Political Analysis (Englewod Clifs, N.J.: Prentice-Hal, 1965). 
24
 Rusel J. Dalton, Democratic Chalenges, Democratic Choices: The Erosion of Political Suport in 
Advanced Industrial Democracies (New York: Oxford University Pres, 204). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 516 
research, participation was conceptualized as a reflection of consent or support for the 
system. 
But this line of research became increasingly disheartening to democrats as the 
century progresed: the traditional measures of political participation were found to be in 
a long, steady decline among Western democracies, as were the scores on opinion 
surveys about trust in government, and these declines were interpreted as a decline in 
legitimacy. But these trends led others to question whether democratic legitimacy itself 
was in retreat or whether the long-understood link betwen legitimacy and stability was 
incorrect; whether legitimacy had been misconceptualized; or whether the long-used 
measures of participation and opinion had simply turned out to be poor indicators.
25
 
To answer these questions, John A. Booth and Mitchel A. Seligson reviewed the 
literatures on declining trust in government and declining citizen participation and 
identified the central chalenge those literatures posed: 
These instances of declining legitimacy with no apparent impact on 
system stability nicely frame the central conundrum of research in this 
field: One might ask, ?Where?s the beef?? What are and where are the 
mising efects of legitimacy?s observed decline? If institutional 
legitimacy has indeed declined so much in recent decades, why have we 
not by now observed at least a few breakdowns of established 
democracies, or more frequent and widespread protests directed at them? 
And why do even the newer democracies, with significantly worse 
performance than developed democracies, appear to enjoy strong popular 
support?
26
 
                                                
25
 John A. Both and Mitchel A. Seligson, The Legitimacy Puzle in Latin America: Political Suport and 
Democracy in Eight Nations (New York: Cambridge University Pres, 209); John A. Both and Mitchel 
A. Seligson, ?Legitimacy and Political Participation in Eight Latin American Countries? (paper presented 
at the Midwest Political Science Meeting, Chicago, 2-4 April 2009); M. Stephen Weatherford, ?Mapping 
the Ties That Bind: Legitimacy, Representation, and Alienation,? Western Political Quarterly Western 
Political Quarterly 44 (191): 251. 
26
 John A. Both and Mitchel A. Seligson, The Legitimacy Puzle in Latin America, 3. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 517 
Their answer to this ?puzzle? was that legitimacy researchers had for too long been 
measuring the wrong thing (namely citizen participation, in the form of voting, writing to 
representatives, etc.) or at least were measuring it incorrectly. Previous researchers, they 
argued, had found that the relationship betwen citizen participation and legitimacy is 
linear, that is, lower legitimacy leads to lower within-system participation and higher 
protest behavior. But Booth and Seligson, looking at the range of participation behaviors 
in Latin America, took a more expansive view of what counts as involvement in politics, 
including such activities as voluntering in civil society organizations, geting involved in 
local government, and so on. What they found was that the relationship betwen 
participation and views of legitimacy is not linear but U-shaped: participation (both 
within the system and non-threatening alternatives outside of the system) is highest 
among those who consider the political system to be most legitimate and also among 
those who consider it to be least legitimate, and lowest among those in the middle. By 
expanding the definition of participation to involvement in politics outside of the official 
channels (e.g. voluntering in civil society, participating in protests, etc.), they 
demonstrated that, in democracies: ?citizens with low support norms can and do work for 
change within the system through elections and campaigns. They also sek alternative 
arenas for participation in civil society, community, or local government. These activities 
do not threaten political system stability.?
27
 
A paralel efort to measure legitimacy started with David Betham, a political 
philosopher, social scientist, and critic of Weber?s conceptualization of legitimacy, 
particularly the way it had been used by subsequent researchers who equated legitimacy 
                                                
27
 John A. Both and Mitchel A. Seligson, ?Legitimacy and Political Participation in Eight Latin American 
Countries.? 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 518 
with belief in legitimacy (he therefore was implicitly criticizing the ?trust in government? 
measures). Betham shifted the focus from trust and participation to a conceptualization 
of legitimacy that he thought would make it both more realistic and more measurable. In 
one aspect of that conceptualization, he defined power as being ?rightful or legitimate? 
when it ?is acquired and exercised acording to justifiable rules? ? justifiable, he 
clarified, in terms of the beliefs of the society in question ? and when it is acquired and 
exercised ?with evidence of consent.? The evidence of consent that he argued should be 
sought was related to the idea that some forms of participation are encouraged by 
legitimacy: ?what is important for [measuring] legitimacy is evidence of consent 
expresed through actions which are understood as demonstrating consent within the 
conventions of the particular society.?
28
 Taking up that theme, Bruce Giley ? whose 
own definition of legitimacy also emphasized action (?a state is more legitimate the more 
that it is treated by its citizens as rightfully holding and exercising political power?) ? 
re-specified Betham?s ?acts of consent? as ?positive actions that expres a citizen?s 
recognition of the state?s right to hold political authority and an aceptance, at least in 
general, to be bound to obey the decisions that result.?
29
 Legitimacy, in other words, is 
not simply a reflection of belief, but is reflected in the acord betwen individual beliefs 
and system features and is indicated by the presence of acts of consent. The Betham and 
Giley formulations pulled together the separate threads of what political sociologists and 
political scientists had been saying about what legitimacy does in or for political systems: 
                                                
28
 David Betham, The Legitimation of Power (Houndsmil: MacMilan Education, 191), 12 (emphasis in 
original). 
29
 Bruce Gilley, ?The Meaning and Measure of State Legitimacy.? 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 519 
it induces voluntary compliance and encourages participation (acts of consent), and by 
doing so, it generates stability. 
Interest in studying the legitimacy of entities other than states, political systems, 
and social orders grew stronger in the last quarter of the 20
th
 Century, particularly in the 
fields of organizational studies and busines management. Summarizing that literature, 
Tatiana Kostova and Srilata Zaheer wrote: ?Scholars have defined organizational 
legitimacy as the aceptance of the organization by its environment and have proposed it 
to be vital for organizational survival and succes.?
30
 Their article was not explicit but by 
acceptance they semed to mean lack of opposition in the form, for example, of 
demonstrations, boycotts, or campaigns to have operating licenses revoked, applications 
rejected, or duties imposed. Survival is a species of stability (so this is consistent with 
previous literature), but what of succes? Again, the authors were not explicit but in the 
case of for-profit busineses, the subject of their study, the organizational literature 
suggests that legitimacy keeps operating costs low and lowers the bariers to entry into 
markets and thereby helps increase profit; for non-profit organizations, presumably 
legitimacy also lowers operating costs, although in their case cost would be measured, 
perhaps, more by the amount of efort or the number of staff-hours or volunter-hours 
than by the amount of money required to fulfil their objectives. 
By the 1980s and 1990s, the work on legitimacy that had been done by 
organizational sociologists was encouraging social psychologists to take a look at the 
                                                
30
 Tatiana Kostova and Srilata Zaher, ?Organizational Legitimacy Under Conditions of Complexity: The 
Case of the Multinational Enterprise,? Academy of Management Review 24, no. 1 (199): 64. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 520 
concept.
31
 The most important relevant work that had been done in that field before then 
had been by Henri Tajfel, who in the 1970s had developed social-identity theory by 
placing research subjects arbitrarily into diferent groups and observing that, even in the 
absence of any relevant diferences in characteristics betwen the groups, people would 
develop favoritism toward members of their own group, caled the in-group, and biases 
against members of the other group, the out-group. A puzzle in social-identity theory 
arose, however, when it became clear that some people had a tendency toward out-group 
favoritism, the opposite of the normal patern. It turned out that social-identity theory 
could acount for this by incorporating facts about social status, stability, and views of 
legitimacy: members of low-status groups who considered the social system to be stable 
and legitimate tended to acept their position in society even though they would be beter 
off under some other arangement.
32
 (In fact, Juan Jos? Linz made the important 
observation that the legitimacy of something could be measured by the degre to which 
the people who do not benefit from that something nevertheles support it.
33
) 
Subsequent work during the 1990s in social-identity theory, social-dominance 
theory, and system-justification theory suggested a variety of mechanisms for how this 
                                                
31
 John T. Jost and Brenda Major, ?Emerging Perspectives on the Psychology of Legitimacy,? in The 
Psychology of Legitimacy: Emerging Perspectives on Ideology, Justice, and Intergroup Relations, ed. John 
T. Jost and Brenda Major (New York: Cambridge University Pres, 201), 3. 
32
 See, for example, Tajfel, Henri, Colin Fraser, and Joseph Maria Franciscus Jaspars, eds. 
The Social Dimension: European Developments in Social Psychology. New York: 
Cambridge University Pres Editions de la Maison des Sciences de L'Homme, 1984, and 
others cited in John T. Jost and Brenda Major, ?Emerging Perspectives on the Psychology of Legitimacy.? 
33
 Juan J. Linz, ?Crisis, Breakdown, and Re-Equilibration,? in The Breakdown of Democratic Regimes, ed. 
Juan J. Linz and Alfred C. Stepan (Baltimore: John Hopkins University Pres, 1978). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 521 
and related proceses operated.
34
 These new developments, along with renewed interest 
in the topic by organizational theorists, encouraged two psychologists, John T. Jost and 
Brenda Major, to organize a conference on the psychology of legitimacy in August 1998 
at Stanford University to se if they could unify some of this work. With contributions 
from social and organizational psychologists, organizational and political sociologists, 
political scientists, and experts in conflict studies and busines management, the 
conference helped pull together some loose threads in theories of legitimacy in at least a 
dozen sub-disciplines.
35
 Most of the chapters in the resulting edited volume, published in 
2001, were collaborative eforts, some chapters were writen by researchers long 
influential in their fields, and the overal efect of the project was a significant advance in 
the unavoidably interdisciplinary field of legitimacy studies.
36
 (And yet some academics, 
apparently unaware of important advances in other disciplines, were stil complaining ? 
several years after that volume was published ? that ?Few books about legitimacy have 
been published in the last decade? and ?This whole field of study needs much firmer 
links to psychology.?
37
) Some of these developments are incorporated in the framework I 
developed for the present study and are discussed under the appropriate headings below. 
                                                
34
 See John T. Jost, Diana Burges, and Christina O. Moso, ?Conflicts of Legitimation Among Self, 
Group, and System: The Integrative Potential of System Justification Theory,? in The Psychology of 
Legitimacy: Emerging Perspectives on Ideology, Justice, and Intergroup Relations, ed. John T. Jost and 
Brenda Major (New York: Cambridge University Pres, 201). 
35
 John T. Jost and Brenda Major, ?Emerging Perspectives on the Psychology of Legitimacy.? 
36
 John T. Jost and Brenda Major, eds., The Psychology of Legitimacy: Emerging Perspectives on Ideology, 
Justice, and Intergroup Relations (New York: Cambridge University Pres, 201). 
37
 Lyn T. White, Legitimacy: Ambiguities of Political Suces or Failure in East and Southeast Asia, vol. 
1, Series on Contemporary China (Hackensack, N.J.: World Scientific, 2005), pp. 3 and 26. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 522 
In practical maters, the concept of legitimacy got a significant boost in Western 
military doctrine during the mid 2000s, when the U.S. occupation of Iraq was chalenged 
by a growing insurgency, and it had become clear that the American stabilization strategy 
was failing. Commanders in the field and researchers back at home started digging for a 
new approach, based on history, experience, and previous studies of insurgency and 
counterinsurgency. A flurry of published articles noted the key role of legitimacy in the 
clasics of counterinsurgency (COIN) doctrine, and some argued in favor of a new U.S. 
COIN strategy with legitimacy at its core (while others argued against it).
38
 The U.S. 
Army and Marine Corps took the raw material of the old Marine Corps field manual on 
?small wars,?
39
 and rewrote, greatly expanded, and jointly published it as a new field 
manual that characterized COIN as a contest not over teritory but over legitimacy
40
: 
?Succesful counterinsurgents support or develop local institutions with legitimacy and 
                                                
38
 See, for example, Colin H. Kahl, ?Coin of the Realm: Is There a Future for Counterinsurgency?,? 
Foreign Afairs 86, no. 6 (207): 169; Edward N. Lutwak, ?Dead End: Counterinsurgency Warfare as 
Military Malpractice,? Harper?s Magazine 314, no. 181 (207): 33; John A. Lyn, ?Paterns of 
Insurgency and Counterinsurgency,? Military Review LXXV, no. 04 (205): 22; Rod Thornton, 
?Historical Origins of the British Army?s Counterinsurgency and Counterterorist Techniques,? in 
Combating Terorism and Its Implications for the Security Sector, ed. Amb. Dr. Theodor H. Winkler, Anja 
H. Ebn?ther, and Mats B. Hanson (Stockholm: Swedish National Defense Colege, 205), 26; John A. 
Nagl, Learning to Eat Soup with a Knife: Counterinsurgency Lesons From Malaya and Vietnam (Chicago: 
University Of Chicago Pres, 205); Robert R. Tomes, ?Relearning Counterinsurgency Warfare,? 
Parameters 34, no. 1 (204): 16; Henry H. Perit, Jr., ?Iraq and the Future of United States Foreign Policy: 
Failures of Legitimacy,? Syracuse Journal of International Law and Commerce 31, no. 2 (204). 
39
 United States Marines Corps, Smal Wars Manual, NAVMC, 2890 (1987 reprint) (Washington: 
Headquarters, United States Marine Corps, 1940). 
40
 Se John A. Nagl, ?The Evolution and Importance of Army/Marine Corps Field Manual 3-24, 
Counterinsurgency,? Foreword to the University of Chicago Pres Edition, and Sarah Sewal, ?A Radical 
Field Manual,? Introduction to the University of Chicago Pres Edition, both in U.S. Army/Marine Corps, 
The U.S. Army/Marine Corps Counterinsurgency Field Manual (Chicago: University of Chicago Pres, 
2007). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 523 
the ability to provide basic services, economic opportunity, public order, and security.?
41
 
Subsequent developments in military doctrine have kept legitimacy front and center as 
wel, and the leson of most of them is that legitimacy brings stability.
42
 
C.1.1. Legitimacy, Loyalty, Suport, Right, and Duty 
If one were to distil the esence of legitimacy from this brief and perhaps 
simplistic review, what might emerge? Legitimacy, to define it succinctly, is worthines 
of support, or in Karol Soltan?s definition, a ?right to loyalty.?
43
 To claim that something 
is legitimate is to give a moral or normative reason (?it is right?) to obey, support, acept, 
imitate, comply with, or refrain from opposing it within some bounded range of activity 
or experience. To say that one should ofer such support, or that something is worthy of 
such support, is diferent from saying that one merely does offer such support. Support 
can be externaly motivated as wel ? it can be coerced or purchased, for example ? but 
loyalty and self-motivated support are what make, for example, a social relationship or a 
governance structure legitimate, stable, and sustainable. To say that somebody or 
something has a right to one?s loyalty or support implies that somebody else has a (self-
recognized) duty or a moral, social, or legal obligation to provide that loyalty or support. 
Legitimation, sometimes caled legitimization, is the proces of granting or gaining 
                                                
41
 United States Department of the Army, Counterinsurgency, Field Manual No. 3-24, Marine Corps 
Warfighting Publication No. 3-33.5 (Washington: Department of the Army and Marine Corps Combat 
Development Comand, 206), p. 5?2. 
42
 United States Department of Defense, Irregular Warfare (IW), Department of Defense Directive, 
3000.07 (Washington: 208), htp:/ww.dtic.mil/whs/directives/cores/pdf/3007p.pdf; United States 
Department of the Army, Stability Operations, Field Manual, No. 3-07 (Washington: Headquarters, 
Department of the Army, 208), htp:/www.fas.org/irp/dodir/army/fm3-07.pdf. 
43
 Karol Soltan, ?Legitimacy and Power,? working paper (Colege Park, Md.: University of Maryland, 
April 205). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 524 
legitimacy, sometimes involving a recategorization of something that was once not 
legitimate as something that is;
44
 to legitimize (or to legitimate) is to grant or gain 
legitimacy. 
Loyalty and legitimacy both ultimately derive from the Latin word lex, meaning 
law, and both words stil retain this meaning in some senses: legitimacy as acordance to 
law or to legal principles, and loyalty as fidelity to one?s legal obligations. Both words 
have evolved beyond their legal origins to include acordance with other types of rules or 
values and other senses of duty. Loyalty, then, is a set of beliefs, felings, and behaviors 
that indicate atachment, devotion, faithfulnes, or fidelity and that help to sustain or 
deepen a relationship. 
Physicaly speaking, to support something is to hold it in place, and by analogy its 
other meanings suggest maintenance or sustenance: to support something or someone is 
to help them stay where they are or get where they want to go. Financial support, for 
example, involves providing the resources to someone who needs them to achieve a goal; 
moral support entails encouragement and expresions of agrement; scientific support 
involves demonstrating that a claim about some phenomenon is consistent with a set of 
already established and acepted standards; political support can involve voluntering for 
a campaign, publicly announcing that one wants a certain politician or party to win or 
remain in power, or circulating unflatering information about opponents; and so on. 
Loyalty and support can be purchased or bartered; feigned loyalty or support can be 
coerced. But when loyalty or support derives from a norm, a value, or a belief that the 
                                                
44
 Herbert C. Kelman, ?Reflections on Social and Psychological Proceses of Legitimization and 
Delegitimization,? in The Psychology of Legitimacy: Emerging Perspectives on Ideology, Justice, and 
Intergroup Relations, ed. John T. Jost and Brenda Major (New York: Cambridge University Pres, 201), 
54. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 525 
recipient (a person, a social order, a set of rules, etc.) deserves that loyalty or support, is 
worthy of it, or has a right to expect it, then it may be credibly claimed that the recipient 
is legitimate. But when one?s opposition derives from a norm, a value, or a belief that the 
recipient is worthy of opposition, or that one has a duty to oppose, then it may be credibly 
claimed that the recipient is ilegitimate (se next section). 
Beyond support and loyalty, which come into play in the literatures on the 
legitimation of power relations and social, economic, and political structures, people 
sometimes speak of legitimacy in terms of a worthines of imitation or perpetuation. 
Consider the English language under French rule: after England was conquered in 1066 
by the French Duke of Normandy Wiliam the Bastard (so named because of his 
ilegitimate birth, and renamed Wiliam the Conqueror for perhaps obvious reasons), 
formerly Anglo-Saxon words became ?vulgarities? (literaly, words of the common 
people) while their French equivalents become ?proper,? ?polite,? or ?educated.? The 
only reason that mention of the word fuck is considered transgresive in ?polite? 
company in English today is its Anglo origin: it was the term common people had used 
for the same set of acts that proper society in French-dominated England had refered to 
as copulation, which was considered a legitimate word, unlike the former. In other words, 
French-dominated high society was considered worthy of imitation; vulgar English 
farmers were not. Some people are said to be ?natural? leaders, ?alpha males,? ?queen 
bees,? or ?trend-seters? ? the kind of people others want to be like, to imitate; they 
atract support semingly efortlesly. Some academic papers are unreadable, filed with 
jargon and long, gramaticaly complex sentences; they are writen that way not merely 
because the authors are bad writers (and they are indeed bad writers) but because they are 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 526 
imitating the writing style of other academics whose works are considered to be part of 
the canon of their field, ?legitimate? (genuine) clasics that had set a standard for 
scholarship that created a halo efect upon nonesential aspects of their works, including 
their bad writing style: somehow, writing in plain English just doesn?t sem ?scholarly,? 
whereas imitating the tortured prose of the masters lends their own work an air of 
credibility. What counts as a ?legitimate? (not manufactured) grievance? What makes 
someone count as a ?legitimate? scientist, a ?legitimate? physician, or, more generaly, a 
?legitimate? expert on some given topic, beyond mere credentials? Why do some people 
simply sem ore believable than others, or more admirable, more worthy of our trust, 
more worthy of our imitating them? The answer is that there is something about them that 
we believe is admirable, good, right, rightful, proper, or virtuous, and that somehow those 
individuals reflect, promote, embody, or symbolize the values that we hold most dear or 
the standards to which we most aspire: we have judged them worthy, somehow, of 
imitation, of perpetuation, and more generaly of our support. 
C.1.2. Ilegitimacy and Oposition 
Ilegitimacy is not merely the absence of legitimacy (i.e. an unworthines of 
support) but a worthines of opposition: to say that something is ilegitimate is to give a 
moral or normative reason to ignore, disobey, reject, or oppose it, actively or pasively. 
An obligation of disobedience or a duty to oppose would both be based on an asumption 
or an argument that that which is to be disobeyed or opposed is ilegitimate. 
To ilustrate the diference betwen legitimacy and ilegitimacy, consider a 
hypothetical legitimacy scale, from zero to 10, in which this multidimensional 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 527 
phenomenon were boiled down to a single score for the purposes of comparing the 
legitimacy of one thing to that of another, or of tracking the legitimacy of something over 
time. (This is not an uncommon way to think about measuring legitimacy, and it has its 
merits in certain comparative pursuits.
45
) Most theorizing on legitimacy takes place in the 
range of one to ten on this hypothetical legitimacy scale, with the higher scores 
representing more legitimacy, lower scores representing les legitimacy, rising scores 
representing legitimation, faling scores representing delegitimation, and a score of zero 
representing neutrality. (Delegitimation is the proces of withdrawing or losing 
legitimacy; to delegitimize, or delegitimate, is to withdraw or lose legitimacy.) 
What the present study has discovered is that much more theorizing needs to take 
place, as it were, ?inside the zero?: that is, theories of legitimacy need to be supplemented 
with independent theorizing about ilegitimacy. To begin with, the hypothetical scale 
might be extended into negative teritory, with the full range being ?10 to +10. In that 
case, it becomes imediately apparent that there is a diference betwen zero and the 
negative numbers, with zero (neutrality) representing not only an absence of legitimacy 
but also an absence of ilegitimacy, negative numbers representing ilegitimacy, with 
scores closer to ?10 representing more ilegitimate and scores closer to ?1 representing 
les ilegitimate. A decline in score below zero would not necesarily represent 
delegitimation but what might be caled ilegitimation. In delegitimation ? representing 
a declining but stil positive score, until it reaches neutrality at zero ? people can 
withdraw support or withdraw all support from something without necesarily taking the 
next step of actualy opposing it. 
                                                
45
 e.g. Bruce Giley, ?The Meaning and Measure of State Legitimacy?; John A. Both and Mitchell A. 
Seligson, The Legitimacy Puzle in Latin America. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 528 
I propose that the word ilegitimize be used to describe the proces of taking that 
next step, that is, going beyond merely withdrawing support and actualy initiating 
opposition, a belief in the rightnes of opposition, a duty to oppose, etc. In other words, 
delegitimation, the proces in which a worthines of support is lost, should be considered 
a separate proces from ilegitimation (or ilegitimization, if a syllable must be added), 
the proces in which a worthines of opposition is confered (neutralization could involve 
either proces, as long as the goal or the end state is neutrality). In short, we need to take 
a closer look at the diference betwen delegitimation and ilegitimation and distinguish 
betwen the dynamics ?at zero? and the dynamics ?below zero.? It is not always clear in 
studies about the efects of delegitimation and what can be done to prevent it that the 
authors have considered whether going to zero is diferent from going below zero.
46
 
Legitimacy is something that generates, among other things, voluntary 
compliance; without legitimacy (or agrement or habit), compliance has to be forced, 
coerced, or bartered, otherwise the result is non-compliance (se ? C.2). But non-
compliance is diferent from opposition. If you command me to bring you a cup of tea, it 
is one thing for me believe that you do not have the right to ask me for it and that I 
therefore do not have a duty to bring it to you (i.e. the request is not legitimate), and 
another thing for me to believe that it is wrong for you to have a cup of tea in the first 
place and that I therefore have a duty to oppose your geting it from anybody (i.e. the 
request is ilegitimate). Legitimacy is a worthines of support (or, in some contexts, of 
                                                
46
 See, for example: Suzane E. Fry, ?When States Kil Their Own: The Legitimating Rhetoric and 
Institutional Remedies of Authority Crises? (Doctoral Disertation, New York University, 205); Joane 
Martin, Mauren Sculy, and Barbara Levitt, ?Injustice and the Legitimation of Revolution: Damning the 
Past, Excusing the Present, and Neglecting the Future,? Journal of Personality and Social Psychology 59, 
no. 2 (1990): 281; Barington Moore, Jr., Injustice: The Social Bases of Obedience and Revolt (White 
Plains, N.Y.: M.E. Sharpe, 1978); Ted Robert Gur, Why Men Rebel (Princeton, N.J.: Princeton University 
Pres, 1970). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 529 
loyalty or imitation) and ilegitimacy is a worthines of opposition. Legitimacy sustains 
and ilegitimacy impedes. In the short term, legitimacy also induces compliance with 
demands and requests, and encourages supportive participation and public action, while 
ilegitimacy induces opposition. Legitimacy lowers the costs, and ilegitimacy raises the 
costs, of sustaining a structure or a relationship. In short, legitimacy induces compliance, 
encourages participation, and lowers costs, and so achieves stability, while ilegitimacy 
induces disobedience, encourages opposition, and raises costs, and so threatens stability. 
Are there other phenomena that can achieve the same? 
C.2. What Else Does What Legitimacy Can Do? 
It is sometimes said that the prototypical legitimacy situation is one in which one 
party makes a demand on another party who does not wish to comply: ?Bring me a cup of 
tea.? ?No.? What tools are available to bring about compliance? Alternatively, when 
someone complies with a command, a demand, an order, or a request, what explains that 
compliance? Another situation said to be prototypical is when somebody who would not 
otherwise benefit from a structure or relationship nevertheles supports it.
47
 When one 
party provides support to another party, what explains that provision? When a set of 
relationships or rules remains stable over long periods, what explains that stability? 
Legitimacy is only one potential explanation, or only one potential tool: people think they 
have a duty to obey, to comply, or not to oppose, or that something is worthy of their 
support, loyalty, participation, or imitation. One of the main findings of this study is that, 
while legitimacy does lower the costs and prolong the stability of teritorial and social 
                                                
47
 Juan J. Linz, ?Crisis, Breakdown, and Re-Equilibration.? 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 530 
control, it is neither necesary nor sufficient to such control in the short term. If that is the 
case, then it must be the case that there are other things that also can encourage or explain 
the stability of teritorial and social control. What might those other things be? The next 
few sections review several key ?alternatives? to legitimacy: agrement, habit, seduction, 
persuasion, compromise, force, coercion, barter, and deception. 
C.2.1. Agrement and Habit 
Legitimacy is a phenomenon that comes into play only amid conflict: conflicts 
over interests, values, distributions, duties, truth claims, etc. Where there is not conflict 
? where there is agrement ? legitimacy is not necesary. If it was your idea to bring 
me a cup of tea in the first place, it is completely unnecesary for me to request or 
command that you bring it to me. Likewise habit: if for many years you have brought me 
a cup of tea every day at 4:00 pm, it is, again, completely unnecesary for me to request 
that you bring it to me this afternoon. The patern has been set, and, as humans are 
creatures of habit, it would take some event or change in situation to break the patern. 
People develop daily, wekly, monthly, and other regular routines, or develop certain 
paterns in the way we think or behave, and we tend to stick with those routines and 
paterns as long as the conditions under which they develop stay within certain bounds. 
Once those bounds are crossed, it stil might take some time for people to respond by 
breaking out of their usual paterns. Yet, powerful as the paterns set by habits are, they 
stil are not based on anything more than precedent; behaviors reflecting deeper beliefs 
are more stable, as Weber observed: 
An order which is adhered to from motives of pure expediency is 
generaly much les stable than one upheld on a purely customary basis 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 531 
through the fact that the coresponding behavior has become habitual. The 
later is much the most common type of objective atitude. But even this 
type of order is in turn much les stable than an order which enjoys the 
prestige of being considered binding, or, as it may be expresed, of 
?legitimacy.?
48
 
C.2.2. Seduction, Persuasion, and Compromise 
?There is only one way under high heaven to get anybody to do anything,? Dale 
Carnegie wrote in the most influential book on influence published in the 20
th
 Century. 
?And that is by making the other person want to do it.? 
Of course, you can make someone want to give you his watch by sticking 
a revolver in his ribs. You can make your employees give you cooperation 
? until your back is turned ? by threatening to fire them. You can make 
a child do what you want it to do by a whip or a threat. But these crude 
methods have sharply undesirable repercussions.
49
 
Many years ago, I read or heard ? where, and from whom, I cannot now remember ? 
that seduction is the art of ?making other people think it was their idea?; we need not go 
into detail about what ?it? entails, but suffice it to say that this definition is the distilation 
of centuries of wisdom about influencing other people: if somebody wants to do 
something, if somebody thinks doing it was his or her own idea, then that person is 
probably going to do it regardles of what you do or do not want them to do. 
It would be useful if there were a word in the English language that described 
esentialy the same phenomenon as the word seduction but without its sexual or imoral 
connotations, which makes some people uncomfortable enough that the word?s mere 
mention can be a distraction from an otherwise serious conversation. Other words come 
                                                
48
 Max Weber, Economy and Society: An Outline of Interpretive Sociology, 31. 
49
 Dale Carnegie, How to Win Friends and Influence People, revised ed. (New York: Pocket Boks, 1982 
[1936]), 18. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 532 
close ? cooptation, persuasion, temptation, influence, atraction, etc. ? but just do not 
sem to adequately capture the dynamic involved in ?making other people think it was 
their idea.? So I wil use the term seduction and trust that the reader is adult enough to 
handle it. Often, to seduce means to tempt or to persuade someone to have sex or to do 
something considered imoral, sometimes using deception to achieve that end. But there 
is no necesary reason that the act in question must involve sex or must transgres social 
norms: people can be seduced into doing something moraly neutral or moraly good, as 
long as it is something that they had not initialy ben inclined to do but changed their 
mind. There are many diferent methods of seduction, ranging from deception and 
flatery, to the deployment of charisma or conformity to what the other values or admires, 
to a longer-term project of relationship-building. The esence of seduction, however, is 
finding a way to get somebody to independently change their mind or believe that it was 
his or her own idea to do the act in question, or that it is something that he or she had 
wanted to do anyway. 
Once that is achieved ? however it be achieved ? the explanation for the act is 
litle diferent from that of agrement: the other party wants to do it and indeed came up 
with the idea to do it, without realizing that the idea had in fact partly originated with the 
seductor. This is diferent from both deception and persuasion, however, as persuasion is 
more explicit, normaly involving an explicit statement by one party of what action is 
desired, followed by argued reasons why the other party should undertake that action, 
often involving reference to self-interest or values or good outcomes and so on. 
Seduction, as noted, might involve outright deceptions ? from lies to elaborate 
manipulations of circumstances or happenstances ? but it need not do so: to the degre 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 533 
deception is involved, to seduce is to deceive more by omision (the failure to state one?s 
intentions outright) than by commision, and it might involve a great deal of truth-teling, 
honest flatery, grooming, or merely listening and expresing interest. (Se ? C.2.4 for 
more on deception.) 
It is instructive to note the diference in methods of seduction betwen two of 
literature?s greatest seductors: the fictional Don Juan, and the historical figure Giacomo 
Girolamo Casanova de Seingalt. In most fictional acounts, Don Juan is so driven to be 
with women that he usualy ends up using false flateries and deceptions that later catch 
up to him: he seduces, but he does not win loyalty in the proces.
50
 Casanova, by contrast, 
was known as a man who knew how to listen, to ask questions, to expres interest, and to 
be intensely atentive: he was succesful, and he won the loyalty of the women he 
seduced.
51
 One man was focused on short-term succes and quantity, the other on long-
term succes and quality (in addition to quantity). The diference in the approaches is the 
diference betwen appreciation and flatery, as Carnegie describes that diference: ?One 
is sincere and the other insincere. One comes from the heart out; the other from the teth 
out. One is unselfish; the other selfish. One is universaly admired; the other universaly 
condemned.? Carnegie?s advice for geting people to want to do what you want them to 
do is either to offer them sincere appreciation, under the logic that more flies are caught 
with honey than with vinegar,
52
 or to arouse an ?eager want? by suggesting, implying, or 
                                                
50
 Se, for example, Jos? Zorila, Don Juan Tenorio, Project Gutenberg (204 [184]), 
http:/ww.gutenberg.org/eboks/5201 (acesed 16 September 209). 
51
 J. Rives Childs, Casanova: A New Perspective, 1st U.S. ed. (New York: Paragon House Publishers, 
1988); Wiliam Bolitho, Twelve Against the Gods: The Story of Adventure (New York: Simon and 
Schuster, 1929). 
52
 In fact, whether honey or vinegar catches more flies turns out to depend on the vinegar. In a home 
experiment, I found that comon household fruit flies were more attracted to balsamic-style vinegar than 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 534 
showing them something that atracts them to the idea of doing what you want them to 
do.
53
 
If two people agre on something, there is no need for legitimacy. Where ends or 
means are not in conflict, where a demand is complied with because the complier wants 
to do it, legitimacy is not the phenomenon that explains the compliance: agrement is. If, 
however, two people do not agre on something ? say one person wants another person 
to do a favor ? there are a number of other strategies, or a number of other explanations, 
that do not directly involve legitimacy. One is compromise, in which each side in the 
disagrement works to find some mutualy aceptable solution that either leaves both 
parties beter off or both parties worse off in ways that are mutualy considered equitable 
or aceptable.
54
 Thre others are considered next. 
C.2.3. Force, Coercion, and Barter 
The diference betwen force and coercion was discussed at length in Appendix B 
(? B.1). Both are forms of instrumental violence: force (instrumental-physical violence) 
uses power or pain to induce (involuntary) behavior, while coercion (instrumental-
psychological violence) uses fear or suffering to induce (technicaly voluntary) behavior. 
Force achieves involuntary compliance simply by making it happen. Coercion achieves 
semi-voluntary compliance by credibly expresing an intent to use or keep using violence 
                                                                                                                                            
to either red-wine vinegar or clover honey (none, however, actualy got caught in any of the liquids), 
thereby demonstrating that, in legitimacy studies, no proposition is as simple, or as setled, as it sems. Se 
also Randal Munroe, ?Flies,? xkcd, http:/xkcd.com/357 (acesed 15 September 209). 
53
 Dale Carnegie, How to Win Friends and Influence People, 29. 
54
 Roger Fisher and Wiliam Ury, Geting to Yes: Negotiating Agrement Without Giving in (Boston: 
Houghton Miflin, 1981). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 535 
as punishment for noncompliance, while also credibly asuring the victim that that 
outcome wil be avoided by complying. 
But while force is always a form of violence, coercion need not be. One can 
coerce by threatening not only violence as punishment for noncompliance but also the 
loss of something of value as punishment for noncompliance. Some forms of coercion, 
such as extortion (like the vacuna protection rackets of Medel?n?s underworld), are 
certainly examples of psychological violence, as they threaten nontrivial losses. 
However, blackmail, for example, would not be considered an example of violence at al, 
because what is threatened would cause only a trivial form of suffering: embarasment. 
But it is a form of coercion, because it stil threatens the loss of something valued, such 
as reputation, or peace of mind, or perhaps the continuation of some valued deception (a 
love afair, a guilty pleasure, etc.). 
Like force and coercion, barter can achieve compliance as wel, but it does so not 
by simply taking something that is valued (force), nor by threatening something that is 
valued (coercion), but by giving something that is valued: barter achieves compliance 
through exchange. If I want your cup of tea, I could simply take it from you using my 
superior strength (force); or I could threaten to break your coffe table or threaten to tel 
your spouse about your afair if you don?t bring me a cup (coercion); or I could ofer you 
$2.49, promise to bring you a cup tomorrow, or ofer you a packet of tea cookies if you 
do bring me a cup (barter). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 536 
C.2.4. Deception 
Deception cuts across al of these categories. Lying to people, manipulating facts 
and situations to make one thing sem like it is another thing, statements of intentional 
omision, and so on are al ways of pretending to use the strategies of compliance 
discussed in the previous thre sections. You can lie to make somebody think there is 
more agrement than there realy is. You can make yourself sem ore atractive to 
somebody than you actualy are (e.g. some people pretend that they want to get maried 
just so that they can get somebody to date them for a short while). You can use lies and 
false facts during the course of persuasion. You can lie about your initial bargaining 
position as a way of geting more than your fair share in a compromise. You can pretend 
that you have more power at your disposal than you do. And you can barter on credit and 
never pay up, or give somebody something that is not what you said it was, such as 
counterfeit money or a bottle of wine that is not as rare as you had claimed. 
C.3. Legitimacy of What? 
Keping al of these ?alternatives? to legitimacy in mind, we can now turn to a 
discussion of legitimacy itself and how it might best be understood and analyzed. A 
proper analysis of legitimacy should begin by identifying or defining the confere and the 
referee. The confere is the entity upon which or upon whom legitimacy is or is not to be 
confered. The referee is the person or group of people who are judging the degre to 
which the confere is or is not legitimate. Too many authors of studies on legitimacy 
neither define the term itself, nor specify the confere about whom they are writing, nor 
explicitly acknowledge whose views of legitimacy are acounted for in their work. For 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 537 
the present study, I have endeavored not to follow that practice. The next section 
discusses the role of the refere; this section discusses conferes, which can be said to be 
the unit of analysis. Identifying the confere answers the question: legitimacy of what? 
For much of the 20
th
 Century, most of the research on political legitimacy or the 
legitimacy of power has taken the state or a government or a political system as the 
appropriate unit of analysis. But things other than states can and do govern (se 
Appendix A), and things other than governments and states can be legitimized: ?Over the 
course of the history of the subject, theories have in fact emerged not only of the 
legitimacy of power and rewards, but also of status, of inequalities in general, of 
procedures and procedural justice, of deviance and social control, of social protest, of 
social change, and in fact of almost any aspect of the structure of social groups,? 
including ?acts, persons, roles, and rules, hence the structure of relations and groups, and 
the groups themselves,? up to and including the social order as a whole.
55
 Social 
structures are usualy defined as the persistent paterns and components of social 
relations at specific times and places (se ? A.1).
56
 This definition is broad enough to 
encompas organizations, borders, the division of labor, states, statelets, the distribution 
of economic and political goods, asociations, regimes, mafias, systems, commands, the 
means of production, and institutions for mariage, education, law, justice, property, and 
the regulation of violence, among many others. Any one of these structures can be subject 
to legitimation, delegitimation, or ilegitimation by the individuals and groups that make 
them up; any one of these structures, in other words, can be conferes of legitimacy. 
                                                
55
 Moris Zelditch, Jr., ?Theories of Legitimacy,? 39 (his citations omited). 
56
 Encyclop?dia Britanica Online, s.v. ?social structure,? htp:/ww.britanica.com (acesed 2 February 
2009). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 538 
For the purposes of this study, I asume that legitimacy can be confered upon a 
role (within a social, political, economic, or cultural structure or relationship), a policy (a 
plan of action, or a statement about what one wil do under what conditions, how one wil 
do it, or to what ends it wil be done; se Appendix A), a distribution (of wealth, power, 
prestige, status, etc. across a defined set of individuals or groups), or more commonly a 
structure (which entails some combination of roles, policies, and distributions). 
To understand the conventional view, taken by many authors who study 
legitimacy, about what counts as a confere, consider the folowing analysis of social 
structures.
57
 Social structures, especialy political structures, can be said to be 
categorizable into thre elements, corresponding to what might be described, colloquialy 
and simplisticaly (but, for the purposes of this ilustration, usefully enough), as the 
rulers, the ruled, and the rules, each of which has any number of subcomponents, any 
one of which can be subject to legitimation, delegitimation, or ilegitimation: 
? The rulers, or powerholders, include state and nonstate entities that hold and 
exercise power within the social structure; examples include political officials, 
policy makers, decision makers, and commites. At a subordinate or 
intermediate level (i.e. betwen rulers and ruled), this category can include the 
powerholders? agents or administrative staf, who play a role sometimes 
similar to that of the powerholders and sometimes similar to that of the groups 
who make up the next category. The legitimacy of rulers can be deficient in 
                                                
57
 Some of this terminology and the conventional conceptualizations they refer to are drawn from David 
Easton, A Systems Analysis of Political Life; Leslie Gren, The Authority of the State (Oxford: Clarendon 
Pres, 198); Jean Hampton, Political Philosophy (Boulder, Colo.: Westview Pres, 197); Rusel J. 
Dalton, Democratic Chalenges, Democratic Choices; Herbert C. Kelman, ?Reflections on Social and 
Psychological Proceses of Legitimization and Delegitimization?; Bruce Giley, ?The Meaning and 
Measure of State Legitimacy.? 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 539 
any aspect of any level of analysis (e.g., local, national, transnational), branch 
(executive, judicial, legislative), or function (e.g., commerce, law 
enforcement, acounting, advising, etc.). 
? The ruled are common people, the members or constituents of the political, 
social, cultural, or religious community who do not hold power, or the 
subjects from whom the holders of power can rightly expect support, loyalty, 
or compliance within established bounds; sometimes they play a role, as noted 
above, as agents or administrative staf of the rulers. Some have high status 
within the structure, others have low status. The legitimacy of the ruled can be 
deficient in their membership (i.e. who is included in or excluded from the 
recognized political, social, cultural, or religious community) or in their 
structure (the distribution of influence, resources, and status among members). 
? The rules are the policies that regulate social life, the systems through which 
they operate, or the bounds of behavior that delimit both one?s proper role in 
the social structure and the claims or demands that powerholders may 
legitimately direct toward their agents, staf, or subjects. Examples include 
regimes, rights, laws, and common or traditional practices that specify and 
regulate obligations and relationships. The legitimacy of the rules can be 
deficient by virtue of the regime or system through which they are expresed, 
the proceses through which they operate, the policies that they produce (e.g., 
laws, regulations), or the outcomes that they generate (e.g., how laws are 
enforced). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 540 
In conventional analyses of legitimacy, the proper unit of analysis ? the answer 
to legitimacy of what? ? is the powerholder (the ruler): that individual or institution wil 
be considered legitimate if it has, or if the political community subject to its power (the 
ruled) believes that it has, a right to perform its role in the social structure or is worthy of 
occupying that role, as long as it came to occupy that role in the acepted way and acts 
within the proper, established bounds of that role (the rules). The esential legitimacy 
situation is conventionaly said to take place when the powerholder makes a demand of 
some sort that requires compliance from subjects: the demand is considered legitimate to 
the degre that the subjects voluntarily comply because they believe the powerholder has 
the right to make the demand and they have a corresponding moral duty to comply with 
it. If they comply out of mere self-interest, their compliance is not based on the 
powerholder?s legitimacy but on some other consideration (e.g. force, coercion, barter; 
see ? C.2). 
This conventional view is analyticaly satisfying and can be useful for 
understanding and influencing situations involving one-way demands and the stability of 
a political regime. But it mises some important dynamics. Legitimacy is not always ? 
perhaps not usualy ? a one-way stret. Not only do subjects have to recognize (and 
treat) powerholders as legitimately holding power; but powerholders also have to 
recognize (and treat) subjects as legitimate members of the community; and both have to 
recognize (and treat) the bounds as a legitimate guide to their relationship. In 
government, for example, rulers can legitimately make demands on the ruled to obey 
laws, but the ruled can legitimately make demands on the rulers to actualy govern them 
? to make and implement policies, to build and manage institutions and networks, and to 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 541 
deliver political goods that protect and improve their lives. Powerholders are also 
members of the political community; a president, for example, is not only a powerholder 
in the role of president but also a subject in the role of citizen: rights and duties cut across 
roles. (Se Appendix A for an extensive discussion of these isues.) It is for this reason 
that I use the terms conferes and referees when discussing judgments made about 
legitimacy: it does not presuppose the identity or the role of the confere: the refere, of 
any sort, makes a judgment about whether to confer legitimacy upon the confere, of any 
sort: The president is a confere in the analysis of the legitimacy of the presidency but a 
refere in the analysis of the legitimacy of the political community. This point becomes 
clearer in the analysis of the dynamics of legitimacy in Medel?n (se Chapter 9). 
C.4. Legitimacy Acording to Whom? 
The referee is the person or group of people who are judging the degre to which 
the confere is or is not legitimate. Identifying the refere answers the question: 
legitimacy according to whom? Referes include both outsiders and insiders. Outsiders 
are people who neither are members of, nor are afected by, the role, policy, distribution, 
or structure in question (i.e. the confere); authors of academic papers writen about 
legitimacy in other places are outsider referes. Insiders are people who are part of the 
structure or relationship in question, are afected by the actions of the entity occupying 
the role, or are afected by the policy or distribution. Insiders can be members of high-
status or low-status groups. 
For simplicity, I define low-status groups or individuals as those people who 
would be ?beter-of,? somehow, under some distribution of wealth, power, prestige, or 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 542 
status other than the one that actualy exists; colloquialy, they are the ones who got the 
short end of the stick. Some low-status individuals support the existing system despite 
their position within it, implicitly ?blaming? themselves or other members of their in-
group for their position in society; others, however, ?blame? the system and either do not 
support it or actively oppose it. ?For many disadvantaged groups, it is dificult to hold 
simultaneous beliefs about the goodnes and legitimacy of the self, the ingroup, and the 
social system,? wrote John T. Jost, Diana Burges, and Christina O. Mosso. In a grand 
synthesis of the varied strands of system-justification theory and related research, they 
elaborated J?rgen Habermas?s concept of a ?legitimation crisis?
58
 to 
cover the myriad ways in which people suffer from and atempt to cope 
with and resolve contradictory needs to (a) fel valid, justified, and 
legitimate as individual actors (ego justification), (b) develop group 
memberships that they and others believe to be worthwhile and legitimate 
(group justification), and (c) preserve a sense that the prevailing system of 
social arangements is fair, legitimate, and justifiable (system 
justification). ? [For] members of high status groups, motives for ego 
justification, group justification, and system justification are consistent 
and complementary, whereas for members of low status groups, these 
motives are often in conflict with one another ? resulting in ambivalence 
[about their status in society], decreased ideological coherence, 
disengagement from the system, partial or total dis-identification with the 
ingroup, [or] individual mobility and group exit.
59
 
Given this discussion, I define high-status group or individuals as those people for whom 
self-, group-, and system-justification motives are not in conflict because they are 
advantaged (relative to low-status members) under the system that exists. 
To evaluate any claim that something is or is not legitimate, it is esential to 
understand who is making the claim. In every society, some groups are favored and 
                                                
58
 J?rgen Habermas, Legitimation Crisis (Boston: Beacon Pres, 1975). 
59
 John T. Jost, Diana Burgess, and Christina O. Moso, ?Conflicts of Legitimation Among Self, Group, 
and System.? 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 543 
protected while others are neglected or mistreated. Such inequities are exaggerated in 
oppresive societies, but they are present in democratic societies as wel. The government 
gives contracts to some people, their family benefits from those contracts, and they find it 
dificult to oppose the government that is benefiting their family, even if it oppreses 
others. Under the Somoza regime in Nicaragua during the 1970s, for example, labor 
leaders, activists, leftists, and many others were routinely arested, tortured, and kiled, 
while many busines leaders, family and friends of the Somozas, political alies, and 
other elites were permited the fredom to profit and thrive. In their daily lives, many 
elite Nicaraguans could go weks without personaly encountering a member of a 
disadvantaged group, aside, perhaps, from household staf; given their range of daily 
experience, it should not be surprising that many elites believed the Somoza government 
was legitimate. It should be equaly unsurprising that a family whose son had 
?disappeared? would have a diferent opinion. Similarly, many Germans before and 
during World War I were not aware of the scale of genocide their Nazi government was 
undertaking; of those who were aware of it, some even approved, believing the Holocaust 
was a legitimate response to the ?Jewish problem.? Normatively, one can state without 
reservation that there is no galaxy in the universe where such a genocide is, in fact, 
legitimate; but positively, we can nonetheles observe that there was, at the time, a 
population within Germany who, however wrongly, did so believe. 
This dynamic, partly captured by the common sentiment that ?where you stand 
depends on where you sit,? was present in abundance in Medel?n. And yet in Medel?n, 
as elsewhere, there were low-status individuals who might have been high-status 
individuals under some diferent system, yet they supported or were neutral toward the 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 544 
system as it existed; as mentioned earlier, the legitimacy of some confere is often said to 
be indicated by the degre to which low-status insiders support it. The distinction 
betwen high-status insiders and low-status insiders is meant to capture these dynamics. 
The distinction betwen insiders and outsiders is meant to correct an important 
pathology of many studies of legitimacy. Many authors ? outsiders to the conferes they 
study ? believe (or employ methods that imply a belief) that legitimacy is an objective 
atribute of powerholders or of certain types of social or political structures, and that 
subjective beliefs about legitimacy are not relevant because, for example, they are too 
easily manipulated. Since legitimacy is an objective atribute, they implicitly (sometimes 
explicitly) argue, judgments about legitimacy are beter based on an objective analysis of 
the degre to which certain conferes conform to those objective standards; therefore, 
conferes have no referes. My response to this line of argument is that, in fact, it is the 
author making this argument who is the refere: any person, whether in a direct 
relationship with the confere under analysis or some academic outsider writing about 
that confere, is acting as refere whenever he or she specifies the standards upon which 
judgments about the confere?s legitimacy are to be based. Hence my use of the term 
outsider referee. 
When measuring the degre to which something is legitimate or ilegitimate, 
whose criteria should be used to make the measurement? The answer depends on whether 
one believes legitimacy resides in the beliefs of individuals, in the behaviors of groups, or 
in the atributes of social structures. There has long been a divide, for example, betwen 
those who asume that legitimacy resides in the objective features of the structure whose 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 545 
legitimacy is being measured and those who asume that legitimacy resides in the 
subjective beliefs of the people subject to its power
60
: 
? The macro or system-level perspective ?takes for granted the epistemic 
asumption that an outside observer, relying on fairly gross aggregate 
evidence, can measure the legitimacy of a political system and rank it in 
comparison with other systems.?
61
 As described in the previous paragraph, the 
macro researcher (an outsider refere) begins by specifying a set of 
supposedly universal criteria for legitimacy, then proceds by measuring the 
degre to which the confere under study mets those criteria. 
? The micro or individual-level perspective relies on insider referes? reported 
opinions about the legitimacy of the confere under study or on observable 
behaviors suggestive of a belief in legitimacy. The insider referes? criteria for 
legitimacy are generaly acepted without question. 
The asumptions underlying both approaches are problematic. The traditional micro 
approach asumes that legitimacy is equivalent to belief in legitimacy. The traditional 
macro approach either asumes the population under study shares the researcher?s own 
normative views about what counts as legitimate, or rejects their opinions on the mater 
as being irelevant. Neither view is correct. The experiences of counterinsurgents, for 
example, demonstrate definitively that diferent populations have diferent opinions about 
the kind of life that is worth taking up arms to defend: ignoring those opinions when 
fighting an insurgency can be deadly. The suggestion that legitimacy is a mater of public 
                                                
60
 M. Stephen Weatherford, ?Measuring Political Legitimacy,? American Political Science Review 86, no. 1 
(192): 149; Bruce Giley, ?The Meaning and Measure of State Legitimacy.? 
61
 M. Stephen Weatherford, ?Measuring Political Legitimacy.? 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 546 
opinion reduces legitimacy to approval ratings and ignores that reported opinions are 
often unreliable indicators of beliefs, and that beliefs themselves often reflect adaptive 
preferences rather than the preferences an individual would have under beter social, 
economic, political, or personal circumstances.
62
 
More recent eforts have been made to avoid some of these problems by 
incorporating both the objective atributes of conferes and the subjective beliefs of 
referes into theories of the determinants of legitimacy and asumptions about where 
legitimacy resides. Typical of this genre is Betham: ?A given power relationship is not 
legitimate because people believe in its legitimacy, but because it can be justified in terms 
of their beliefs? (emphasis in original): 
When we sek to ases the legitimacy of a regime, a political system, or 
some other power relation, one thing we are doing is asesing how far it 
can be justified in terms of people?s beliefs, how far it conforms to their 
values or standards, how far it satisfies the normative expectations they 
have of it. We are making an asesment of the degre of congruence, or 
lack of it, betwen a given system of power and the beliefs, values, and 
expectations that provide its justification.
63
 
To measure this ?congruence,? this approach first determines the referes? normative 
criteria for legitimacy, then measures the degre to which the confere mets those 
criteria. It is a micro approach in that it asumes that the perspective that maters is that of 
the insider referes, but it is macro in that what it measures are the objective atributes of 
the confere in question (specificaly, its congruence with referes? values), not the 
opinions of the referes. 
                                                
62
 Jon Elster, Sour Grapes: Studies in the Subversion of Rationality (New York: Cambridge University 
Pres Editions de la Maison des sciences de l?home, 1983), 109; Amartya Sen, ?Rights and Capabilities,? 
in Resources, Values, and Development, ed. Amartya Sen (Cambridge, Mas: Harvard University Pres, 
1984), 307; Martha Nussbaum, ?Adaptive Preferences and Women?s Options,? Economics and Philosophy 
17 (2001): 67. 
63
 David Betham, The Legitimation of Power, 11. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 547 
In this same spirit, an alternative approach bridges micro and macro, not by 
reconciling objective atributes of conferes with subjective beliefs of referes, but by 
observing and measuring referes? public behaviors, which are asumed to be both a 
reflection of individual subjective beliefs (micro) and a response to objective facts about 
the structure in question (macro). An excelent variant of this approach, incorporating a 
meso or group-level measure, is Giley?s comparative study of the legitimacy of states, in 
which the author uses both atitudinal (micro) and behavioral (meso) indicators to 
develop an index of state legitimacy based on his own (macro) criteria for legitimacy, 
which he derives from the literature.
64
 
The present study takes an approach that esentialy combines these last two 
approaches. It treats legitimacy neither as solely ?a direct property of political 
institutions? nor as solely ?a property of individual psychologies?
65
 but also as a 
measurable feature of the interaction betwen the two. This approach is macro in that it 
provides a ?thin? or universalistic acount of human reasoning about legitimacy and 
measures the objective atributes and behaviors of the diferent entities and agents upon 
whom legitimacy is being confered (or withheld) to se how wel they match insider 
referes? judgments about them; meso in that it recognizes that actual reasoning about 
legitimacy takes place in particular contexts and is expresed in interactions with other 
people and groups, and measures those judgments as expresed in actions; and micro in 
that it measures individual referes? private perceptions and expresions of legitimacy. In 
other words, it measures legitimacy at thre levels of analysis: public attributes, or the 
                                                
64
 Bruce Giley, ?The Meaning and Measure of State Legitimacy.? 
65
 Robert Grafstein, ?The Legitimacy of Political Institutions,? Polity 14, no. 1 (1981): 51. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 548 
publicly measurable features of the conferes (a macro measure); group behaviors, or 
publicly expresed judgments of those atributes (meso); and individual beliefs, or private 
judgments about them (micro). 
C.5. Legitimacy by What Proceses? 
Legitimacy has a substantive component and a procedural component. The 
substantive component is discussed at length in the next section. The procedural 
component of legitimacy can be considered a form or a driver of legitimation, 
delegitimation, or ilegitimation. Many theories of legitimacy make the asumption that 
legitimacy derives from a set of esentialy static conditions, and that if those conditions 
are present, legitimacy emerges. That is not usualy the case. Legitimation proceses are 
esential to geting al constituents of a social structure to acept decisions and support 
outcomes. Diferent social actors have diferent ways, proceses, and methods of 
legitimizing decisions. The substantial content of an arangement may be otherwise 
aceptable, but if it was not derived by fair or localy legitimate proceses it could stil be 
rejected by some parties as ilegitimate.
66
 Because proces can (for the sake of 
convenience) be considered an atribute of the justifiable criterion discussed in the next 
section, it wil not be discussed further here; further analysis wil be left to future 
research. 
                                                
66
 Herbert C. Kelman, ?Reflections on Social and Psychological Proceses of Legitimization and 
Delegitimization?; Tom R. Tyler, ?A Psychological Perspective on the Legitimacy of Institutions and 
Authorities,? in The Psychology of Legitimacy: Emerging Perspectives on Ideology, Justice, and 
Intergroup Relations, ed. John T. Jost and Brenda Major (New York: Cambridge University Pres, 201), 
416. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 549 
C.6. Legitimacy by What Criteria? 
This section addreses the substantive component of legitimacy, the criteria 
against which something can be judged legitimate or otherwise. The vast literatures on 
legitimacy, authority, society, politics, culture, and identity are filed with discussions of 
lists and typologies of the substantive criteria, sources, or factors that contribute to beliefs 
about legitimacy. The most common themes (and key concepts) fal into a few categories: 
? Law. Acordance with law is the oldest sense of the Latin word legitimus, 
dating at least as far back to Cicero in the first century BCE, and legal 
legitimacy continues to enjoy widespread atention today, particularly in 
writings in the philosophy of law but also in political theory generaly. 
Relevant topics include, among others, the rule of law, the laws vs. morals 
debate, sovereignty, and political authority. Law can both be legitimate and 
contribute to legitimacy.
67
 
? Tradition (custom, religion, wisdom, rules of succesion, habit, divine right of 
kings). Legitimacy as conformity with tradition ? the belief that a rule or 
ruler should be obeyed and an institution supported because they have always 
been obeyed and supported ? predates the term ?legitimacy? and probably 
                                                
67
 Max Weber, Economy and Society, 220; Jos? Guilherme Merquior, Rouseau and Weber: Two Studies in 
the Theory of Legitimacy (Boston: Routledge & Kegan Paul, 1980), 2; Friedrich A. von Hayek, Law, 
Legislation, and Liberty: A New Statement of the Liberal Principles of Justice and Political Economy 
(London: Routledge and Kegan Paul, 1982); Neil J. Kritz, ?The Rule of Law in the Postconflict Phase: 
Building a Stable Peace,? in Turbulent Peace: The Chalenges of Managing International Conflict, ed. 
Chester A. Crocker, Fen Osler Hampson, and Pamela R. Aal (Washington: United States Institute of Peace 
Pres, 201), 801; Ian Shapiro, The Rule of Law (New York: New York University Pres, 194); Ane-
Marie Slaughter, A New World Order (Princeton: Princeton University Pres, 204); Abram Chayes and 
Antonia Handler Chayes, The New Sovereignty: Compliance with International Regulatory Agrements 
(Cambridge, Mas.: Harvard University Pres, 197); Stephen D. Krasner, ?Sharing Sovereignty: New 
Institutions for Colapsed and Failing States?; Noah Feldman, What We Owe Iraq: War and the Ethics of 
Nation Building (Princeton: Princeton University Pres, 204), 148; Lon L. Fuler, The Morality of Law, 
Revised ed. (New Haven: Yale University Pres, 197); Jean Hampton, Political Philosophy. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 550 
predates history itself, although the term did not begin to denote acordance 
with custom until the Middle Ages. Traditional societies have long considered 
such things as the counsel of shamans or wise men, rules of succesion, the 
divine right of kings, patriarchy, and the authority of the Church to be 
legitimate features of their cultures. Many aspects of modern societies, too, 
retain this belief: if people have been using a private road for many years and 
the property?s owner one day decides to prevent aces, the public may sue for 
continued aces and can reasonably expect a modern court to cite custom as a 
reason to decide in their favor.
68
 
? Leadership (charisma, moral capital). A leader of heroic, extraordinary, or 
purportedly divine character can confer legitimacy upon an organization, a 
political or social movement, or a religious worldview simply through the 
force and atraction of his or her personality, a quality Weber caled charisma. 
Entirely apart from the inherent justice of their causes, for example, 
Mohandas Gandhi and Martin Luther King Jr. built up, and judiciously spent, 
great reserves of moral capital on behalf of their ultimately succesful social 
movements. Likewise, Mohamad and Jesus Christ atracted followers who 
later founded what would become major world religions. And cults of 
personality have given tyrants such as Kim Il Sung and Saddam Hussein 
sufficient charismatic legitimacy to remain in power for longer periods, and 
perhaps with lower levels of coercion, than the injustice of their actions as 
leaders might otherwise suggest. Nelson Mandela?s stature was a significant 
                                                
68
 Max Weber, Economy and Society, 226; Jos? Guilherme Merquior, Rouseau and Weber, 2. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 551 
contributor to the succes of South Africa?s democratic transition after 
Apartheid. A hero in one endeavor can win a broad following in another 
simply by force of his or her heroism, as has been the case of military heroes 
who have run for elected political ofice; likewise for individuals who come to 
be sen as symbolizing or embodying some important value.
69
 
? Efectivenes (eficiency, order, stability, prosperity, governance). Legitimacy 
as efective control over teritory, or more broadly as efective governance, is 
a common theme in political theory and is closely related to, and for some 
authors identical with, the concept of sovereignty. The governments of 
Myanmar and China, for example, are sovereign over their teritories, and 
much of the international community acts as if they consider those countries 
legitimate, despite evidence of weak or nonexistent consent from the 
governed. Within countries, it is often the case that people are wiling to 
acept coercive forms of rule when they believe the alternative is widespread 
disorder. In places where anarchy reigns, people tend to sek out protective 
asociations and to support whatever group is most capable of preventing 
theft, murder, rape, and armed atack in their community. When a new 
government demonstrates that it lacks the ability to govern, and crime begins 
to rise or the economy begins to fal, some citizens might begin to pine for the 
                                                
69
 Max Weber, Economy and Society, 111; Alejo G. Sison, The Moral Capital of Leaders: Why Virtue 
Maters (Northampton, Mas.: Edward Elgar Publishing, 203); John Kane, The Politics of Moral Capital 
(New York: Cambridge University Pres, 201). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 552 
good old days of aristocracy or dictatorship, opening the door to a strongman 
candidate, while others might begin to protest or otherwise pres for change.
70
 
? Consent (social contract, democracy, self-determination). This is the most 
commonly cited source of legitimacy, whether in writings about electoral 
democracy, participatory democracy, or other forms of ?public reasoning.? It 
captures the idea that power is legitimate to the degre that those afected by it 
have somehow consented to the way it is exercised. It need not derive from a 
system of representative democracy, and especialy not merely from elections. 
What people care about is having a voice, having some say over how their 
lives are regulated. I prefer the term acces rather than consent to reflect this 
broader idea, since much of the literature (in English and Spanish) on consent 
has a strong bias in favor of Western democratic norms.
71
 
? Norms (sociocultural norms, sacred values, principles, hegemony, ideology, 
theology, justice, fairnes, merit, human rights, false consciousnes). Many 
people measure the legitimacy of the social order and their leaders against the 
                                                
70
 Seymour Martin Lipset, Political Man; Seymour Martin Lipset, ?Some Social Requisites of Democracy: 
Economic Development and Political Legitimacy?; Samuel P. Huntington, Political Order in Changing 
Societies (New Haven: Yale University Pres, 1968); Daniel Kaufman, ?Governance Redux: The 
Empirical Chalenge,? in Global Competitivenes Report 203-2004 (Geneva: World Economic Forum, 
2004); Robert I. Rotberg and Deborah L. West, The God Governance Problem: Doing Something About 
it; Noah Feldman, What We Owe Iraq, 74; Robert Nozick, Anarchy, State, and Utopia (New York: Basic 
Books, 1974), 12. 
71
 Thomas Hobes, Leviathan: Parts One and Two (New York: Liberal Arts Pres, 1958 [1651]); John 
Locke, ?An Essay Concerning the True Original, Extent, and End of Civil Government?; David Hume, ?Of 
the Original Contract,? in Social Contract: Esays By Locke, Hume, and Rousseau, ed. Ernest Barker, The 
World?s Clasics, No. 51 (New York: Oxford University Pres, 1960 [1748]); Jean-Jacques Rousseau, 
?The Social Contract?; John Rawls, A Theory of Justice, revised ed. (Cambridge, Mas.: Belknap Pres of 
Harvard University Pres, 199); Robert Nozick, Anarchy, State, and Utopia; Robert Alan Dahl, 
Democracy and Its Critics (New Haven: Yale University Pres, 1989); Amartya Sen, ?Why 
Democratization Is Not the Same as Westernization: Democracy and Its Global Rots?; Alen E. Buchanan, 
Justice, Legitimacy, and Self-Determination: Moral Foundations for International Law, Oxford Political 
Theory (New York: Oxford University Pres, 204). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 553 
norms, values, and principles they hold most dear; the closer the match, the 
greater the legitimacy. Because some norms are pased through generations, 
this origin is closely related to, but not identical with, traditional legitimacy, 
discussed previously. High levels of social capital and civic engagement 
within a community may signal that that community considers itself norm-
legitimate. Some authors, however, argue that perceptions of norm-legitimacy 
exist only because the powerful impose ?hegemonic? or ?bourgeois? values 
upon the broader population to sustain their own advantages. Whatever the 
source of the norms, however, it is stil the norms that are the source of 
legitimacy in this conception. Principles such as justice, fairnes, merit, or 
respect for human rights can come into play, as can religious beliefs and 
political ideologies. Citizens of democracies often vote for political oficials 
who share their values rather than their exact policy preferences, and many 
consider the subsequent policy outcomes, even those going against their 
economic or political interests, to be entirely legitimate. In a transitional 
government, a political oficial lacking either charisma or the sanction of 
tradition may wield influence by conforming to ? or, beter, standing as a 
symbol of ? societal and cultural norms. Ahmad Chalabi had much more 
influence among U.S. officials than he ever did among the Iraqi population, 
partly because he was sen as an outsider whose values were at odds with 
their own.
72
 
                                                
72
 Barington More, Jr., Social Origins of Dictatorship and Democracy: Lord and Peasant in the Making 
of the Modern World (Boston: Beacon Pres, 196); Hary Eckstein, Division and Cohesion in Democracy: 
A Study of Norway (Princeton: Princeton University Pres, 196); Gabriel A. Almond and Sidney Verba, 
The Civic Culture Revisited: An Analytic Study (Boston: Litle, Brown, 1980); David Betham, The 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 554 
? Identity (social identity, identity politics, respect, pride). A judgment of a 
worthines of support sometimes comes simply from in-group identification 
or inter-group dynamics.
73
 
Lists of such criteria appear al over the literatures on legitimacy across several 
academic and policy disciplines, including political science, sociology, psychology, 
philosophy, organizational studies, and military doctrine. For analytic purposes such lists 
can be useful, as they can help identify the intelectual and cultural resources that can be 
brought to bear upon real-world problems. But not always; in a sense, these lists and 
typologies provide neither a ?thick? enough nor a ?thin? enough acount of the sources of 
legitimacy: not thick enough because they cannot be used as a field guide to any 
particular population?s reasoning about legitimacy, and not thin enough because they 
cannot be used as a general guide to human reasoning about legitimacy across diferent 
fields. A thinner acount would provide a more useful starting point from which the 
particular, complex, mesy details can be hatched.
74
 
                                                                                                                                            
Legitimation of Power; Muthiah Alagapa, ed., Political Legitimacy in Southeast Asia: The Quest for 
Moral Authority, Contemporary Isues in Asia and the Pacific (Stanford, Calif.: Stanford University Pres, 
1995); Robert D. Putnam, ?Bowling Alone: America?s Declining Social Capital,? Journal of Democracy 6, 
no. 1 (1995): 65; Ronald Inglehart, Modernization and Postmodernization: Cultural, Economic, and 
Political Change in 43 Societies (Princeton: Princeton University Pres, 197); Theda Skocpol and Moris 
Fiorina, eds., Civic Engagement in American Democracy (Washington: Brookings Institution Pres, 199); 
Bruce Giley, ?The Meaning and Measure of State Legitimacy?; Scot Atran, Robert Axelrod, and Richard 
Davis, ?Sacred Bariers to Conflict Resolution,? Science 317, no. 5841 (207): 1039; Jeremy Ginges, et al., 
?Sacred Bounds on Rational Resolution of Violent Political Conflict.? 
73
 John T. Jost and Brenda Major, The Psychology of Legitimacy: Emerging Perspectives on Ideology, 
Justice, and Intergroup Relations; David John Frank and John W. Meyer, ?The Profusion of Individual 
Roles and Identities in the Postwar Period,? Sociological Theory 20, no. 1 (202): 86; Judith A. Howard, 
?Social Psychology of Identities,? Annual Review of Sociology 26 (200): 367; Jan E. Stets and Peter J. 
Burke, ?Identity Theory and Social Identity Theory,? Social Psychology Quarterly 63, no. 3 (200): 224; 
The Social Dimension: European Developments in Social Psychology. 
74
 Michael Walzer, Thick and Thin: Moral Argument At Home and Abroad; Bruce Gilley, ?Thick Or Thin? 
An Empirical Intervention.? 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 555 
Such an acount is presented here. This thinner acount, which was used as the 
basis for this study?s analytic framework, was intended to demonstrate the very human 
reasoning that underlies judgments and behaviors about what counts as right or wrong, as 
worthy or unworthy to support, as a duty to comply or to oppose, and so on. By 
developing this thin acount, I was able to capture and incorporate into a single, simple 
framework the broad range of sentiments and motives underlying the criteria for 
legitimacy identified in the existing literatures. This framework can provide a useful 
starting point from which the particular, complex, mesy details of the dynamics of 
legitimacy in the real world can later be uncovered. If nothing else, it can provide a 
baseline for the types of questions that should be asked for any study of legitimacy: no 
fewer questions than are suggested by the framework, but no limits to the questions that 
can be asked beyond that. 
This thin acount identifies six criteria that can be considered causal indicators for 
legitimacy: transparent, credible, justifiable, equitable, accesible, and respectful. 
? Transparent and credible describe the most basic criteria that motivate people 
to support a confere, because together they make life predictable: people 
want to know, for example, what the rules are, and to know that they can be 
enforced non-arbitrarily, even if they don?t agre with the rules, and even if 
they don?t benefit from the rules. These two ?predictability? criteria do not 
actualy confer legitimacy; they are more like contributory or background 
conditions, necesary but not sufficient, that make it possible for people to live 
their daily lives and plan out personal projects within the given constraints. To 
put it another way: predictability wil not generate legitimacy (and voluntary 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 556 
support), but its significant absence wil very likely generate ilegitimacy (and 
active opposition). 
? The justifiable criterion tends to be the central component of most acounts of 
legitimacy, as it captures the values people hold most dear ? their judgments 
about what is right (in acord with valuable norms or rules), good (in acord 
with valuable outcomes), proper (in acord with valuable proceses), and 
admirable (representative of their values) ? and therefore worthy of their 
support or loyalty. Justifiability is a mater of connecting something about the 
confere to something that the refere considers right, good, proper, 
admirable, or otherwise valuable. 
? The equitable criterion reflects ideas about fairnes: people want to be asured 
that inequalities are justified and that, if they have les of something than 
someone else has, it is for a good reason. It is important to recognize that 
equity is diferent from equality: equity implies that there is some standard 
against which the justice of a distribution is measured. People who are not 
wel-off in a society often nevertheles consider that society?s arangements to 
be legitimate, believing that those who are beter-of have gotten that way by 
merit or by right and that those who are worse-off lack the same merits or 
rights. This is implicit in Linz?s conception of legitimacy, that those who 
would be beter-of under another arangement nevertheles support the 
current arangement. When those who are worse-of believe that they are 
worse-off only because of the unfairnes of the beter-off, then that is an 
indication that the society?s arangements are not legitimate. Things can be 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 557 
unequaly distributed without there being any adverse influence on the level of 
legitimacy, as long as that inequality is not unjustified acording to the 
standards of the community. In every community, there are people who have, 
for example, more connections in their social network than others have, and 
al else equal they wil therefore tend to have more influence within the 
community. Whether that inequality is acepted by those with les influence 
wil depend both on the sources of that inequality (e.g. the fact that some 
people are more gregarious than others, or that wealthy or physicaly atractive 
people wil tend to atract more people, for example) and on views about 
whether those sources are valid as a basis for inequality (e.g. some people 
might judge others? higher levels of influence to be valid if it is due mainly to 
their social skils and not to their wealth or beauty; for others, it simply might 
not mater). Likewise, every community has people with more money than 
others; whether that inequality is acepted by those who are poorer wil 
depend both on the sources of that inequality (e.g. inheritance, hard work, 
connections, theft, etc.) and on views about whether those sources are valid as 
a basis for inequality (e.g. in communities without moral prohibitions against 
stealing from outsiders, disparities might be justified by an aceptance that 
some people are simply beter thieves than others; in communities where hard 
work both is valued and reliably leads to greater wealth, disparities might be 
justified by an aceptance that some people simply work harder than others). 
Also, equalities that are unjustified can be considered inequitable: if two 
people have the same level of influence over certain policy questions, that is 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 558 
an equality; but if one of them has far greater knowledge about the substance 
of the isue in question than the other and yet has merely equal influence, then 
that, arguably, is an inequity. 
? The accesible criterion captures much of the literature on consent as the basis 
for legitimacy, but goes beyond what many authors consider to be a strictly or 
traditionaly democratic basis; regardles of the specific system of consent or 
public reason, what people want is some asurance that they have a voice, 
some say in how the things that afect them operate; what people want is some 
sort of aces to the system. 
? Finaly, the respectful criterion captures the literatures on human dignity and 
pride: consistently disrespectful treatment, even if everything else is justified, 
equitable, and acesible, tends to create tension with people?s desire and 
ability to be loyal or offer support, or tends to be so demoralizing as to make 
the question of support nearly irelevant. 
These six criteria together represent a rather thin conception of legitimacy: 
everybody can agre in principle that, for example, the rules regulating political and 
social relations should be transparent, credible, justifiable, equitable, acesible, and 
respectful. Actualy measuring legitimacy, however, is a mater of measuring what these 
adjectives mean in a real-world context acording to real-world values and standards. 
What they mean wil difer depending on whether the confere is a role, a policy, a 
distribution, or a structure, as discussed in the next section. 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 559 
C.7. Measurement Framework 
Among the greatest chalenges to measuring a latent (unobservable) phenomenon 
such as legitimacy is measurement validity, the question of whether one can be certain 
that one is, in fact, measuring legitimacy and not something else. To support 
measurement validity in this study, I combined the approaches described in the last 
several sections into a more comprehensive framework than has been used in the past.
75
 
This framework measures the legitimacy of a particular confere by, first, identifying 
both the outsider refere (usualy the author of the study or the agency sponsoring the 
study) and the high-status and low-status insider referes whose views or behaviors are 
relevant to questions about the confere?s legitimacy; and, second, incorporating a 
measurement strategy that identifies an efect indicator (proxy) and a set of causal 
                                                
75
 The framework described here is a modified and expanded version of one that I had previously 
developed for analyzing legitimacy. Se Robert D. Lamb, ?Measuring Legitimacy in Weak States.? 
Table C-1. A Framework for Measuring Legitimacy 
Legitimacy of what? 
(conferee: role, policy, distribution, or structure) 
Insider refere 
According to whom? 
Low-status insiders High-status insiders 
Outsider 
referee 
Measurement 
strategy 
Type of 
indicator (criteria) 
Individual 
beliefs 
Group 
behaviors 
Individual 
beliefs 
Group 
behaviors 
Public 
attributes 
Proxy Var
0A
 Var
0B
 Var
0C
 Var
0D
 Var
0E
 
Causal
1
 Var
1A
 Var
1B
 Var
1C
 Var
1D
 Var
1E
 
Causal
2
 Var
2A
 Var
2B
 Var
2C
 Var
2D
 Var
2E
 
    ? ? ? ? ? ? 
Causal
n
 Var
nA
 Var
nB
 Var
nC
 Var
nD
 Var
nE
 
 
 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 560 
indicators that can be measured at the levels of individual belief, group behavior, and 
public atribute (se Table C-1). 
Most authors who atempt to measure legitimacy acknowledge that they cannot be 
certain that it is legitimacy and not something else that they are measuring. Those who 
asert that legitimacy resides primarily in individual belief acknowledge the dificulties 
of recal and other biases inherent in measuring opinions. Those who asert that 
legitimacy resides primarily in group behaviors acknowledge the dificulty of 
determining whether certain behaviors derive from belief rather than coercion. And those 
who asert that legitimacy resides in the objective characteristics of the structure under 
study acknowledge that their outsider judgment of the system?s legitimacy may wel 
difer from that of insiders. Yet these authors draw the data or observations that underlie 
their studies of legitimacy from only one or, at best, two of these levels of analysis 
(micro, meso, or macro). But human life is lived at multiple levels simultaneously, and 
legitimacy, being a human phenomenon, is a multi-level phenomenon. 
A proper analysis of legitimacy, therefore, should identify indicators across 
multiple level of analysis, and this framework is intended to be used to look for evidence 
at al thre levels of analysis. By doing so, it provides a higher degre of certainty about 
what is being measured: if individuals say they believe some structure to be legitimate, 
and if groups act as if they believe that structure to be legitimate, and if that structure has 
characteristics that suggest it operates legitimately, then it is very dificult (albeit not 
impossible) to argue that legitimacy is not at work in the structure; but if one of those 
levels does not agre with the others, that suggests that something other than legitimacy 
is at play (coercion, for example). 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 561 
Furthermore, this framework does not measure only proxies for legitimacy, nor 
does it measure only causal indicators: rather, it measures both a proxy variable and six 
causal indicators (transparent, credible, justifiable, acesible, equitable, and respectful). 
Legitimacy is a latent variable. It cannot be measured directly, so variables need 
to be chosen to measure it indirectly. A variety of strategies is available for measuring 
latent variables,
76
 but only two wil be considered here. Bollen and Lennox distinguish 
betwen ?indicators that influence, and those influenced by, latent variables,? caling the 
former causal indicators and the later efect indicators.
77
 
Causal indicators ? also caled constitutive or composite indicators
78
 ? 
collectively determine the latent variable; that is, they collectively constitute a measure of 
legitimacy: 
 L
1
 = ?
1
 X
1
 + ?
2
 X
2
 + .. + ?
Q
 X
Q
 + ?
1
 (9?1) 
where L
1
, is the latent variable, legitimacy; X is an indicator in a composite that includes 
Q indicators; ? is the coeficient of X, or the efect that X has on legitimacy; and ?
1
 is the 
disturbance term.
79
 By definition a legitimate confere is one that mets the refere?s 
criteria for support, and this approach is useful when indicators are available that can 
measure those criteria. 
                                                
76
 Keneth A. Bolen, ?Latent Variables in Psychology and the Social Sciences.? 
77
 Keneth A. Bolen and Richard Lenox, ?Conventional Wisdom on Measurement,? 305. 
78
 Bruce Giley, ?The Meaning and Measure of State Legitimacy.? 
79
 This and the efect-indicator equation are simplified versions of equations (2) and (1), respectively, in 
Keneth A. Bolen and Richard Lenox, ?Conventional Wisdom on Measurement: A Structural Equation 
Perspective?; folowing them, no subscripts are used to index individuals, but unlike them, nor are any used 
to index the indicators. Se Ibid., 305. n. 303 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 562 
Efect indicators ? also caled proxy or substitutive indicators ? are measures of 
phenomena that come about as a consequence of the action of the latent variable; that is, 
the efect indicator is a proxy for the latent variable: 
 Y = ?L
1
 + ? (9?2) 
where L
1
 is again the latent variable, legitimacy; Y is the proxy indicator; ? is the 
coeficient of L, or the efect that L has on Y; and ? is the measurement eror asociated  
 with the indicator.
80
 In this case, an efect indicator would measure something that is a 
result of a structure?s having or lacking legitimacy, such as the size of the internal secret 
police (large size indicates low legitimacy) or rates of voluntary payment of taxes (high 
rates indicate high legitimacy).
81
 
Table C-2 lists the causal criteria and suggests how basic questions might be 
formulated based on the type of confere (it does not list structure, because structure 
encompases roles, role-holders, policies, and distributions); for conceptual 
completenes, it additionaly includes a generic proxy indicator. This framework and 
                                                
80
 Ibid. 
81
 Ibid; Bruce Giley, ?The Meaning and Measure of State Legitimacy.? 
Table C-2. Criterial Measures of Legitimacy 
Confere Type: 
Is the ? 
Indi-
cator 
type: 
Criteria: 
Is the 
conferee 
? 
role or role- 
holder ? 
policy ? distribution ? 
proxy legitimate? worthy of suport? worthy of suport? worthy of suport? 
causal transparent? identifiable? known? known? 
causal credible? capable? enforceable? stable? 
causal justifiable? right, god, proper, 
or virtuous? 
right, god, proper, 
or admirable? 
right, god, proper, 
or admirable? 
causal equitable? fair? fair? fair? 
causal accessible? accountable? amendable? amendable? 
causal respectful? civil? elevating? elevating? 
 
 
Lamb, Robert D. | Appendix C. Legitimacy and Ilegitimacy 563 
these criteria were used as a guide to the sorts of questions to keep in mind as I read 
reports and studies and interviewed experts and residents. Given the dificult research 
environment in Medel?n, not al of these questions could be answered. But the efort 
yielded a more detailed analysis of the dynamics of legitimacy across diferent groups 
than would have been possible using existing frameworks (se Chapter 9). That analysis 
was meant both as an exposition of the dynamics of legitimacy in Medel?n and an 
ilustration of the application of this framework to a real-world context. 
 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 564 
Apendix D. Research Design and Hypotheses 
The primary objectives of this study were to describe and to explain the paterns 
of violence in Medel?n since 1984; secondary objectives were to identify the causal 
mechanisms driving those paterns and to determine what role, if any, legitimacy played. 
The research questions were: (1) What have the overall patterns of violence been during 
the past 25 years? (2) What has caused those patterns to change? And (3) What role did 
legitimacy or ilegitimacy play in those changes? To answer these questions, multilevel, 
multidimensional measurement frameworks were developed for violence and legitimacy. 
Quantitative and qualitative data were drawn from published time series, published 
literature, and interviews with experts and residents. The case was studied over five time 
periods since 1984, using an embedded-case design to review evidence for both the city 
as a whole and for a sector within the city caled Caicedo La Siera. 
The following sections describe the motivation behind this study and the method 
and data used to cary it out (?? D.1-D.3). After defining the symbols to be used, this 
appendix discusses and analyzes the hypotheses that were tested, rejected, and ultimately 
synthesized regarding the relationship betwen legitimacy and violence (?? D.4-D.7). 
D.1. Motivation 
Some of the most presing and chalenging security threats in countries at al 
levels of development have been given strange labels such as ?iregular,? ?asymmetric,? 
and ?open-source,? and are said to emerge from ?failing? (?fragile,? ?lawles,? ?hollow,? 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 565 
?paper?) states, ?ungoverned? areas, ?feral? cities, and other ?complex? security 
environments, which are often understood to include some otherwise wel governed 
places such as the cities of North America and Europe.
1
 While it is understood that some 
important threats do continue to come from strong states with organized militaries, it is 
increasingly recognized that some of the most dificult chalenges emerge in places that, 
due to conflicts and weak or iresponsible governance, are hospitable to harmful activities 
by gangs, insurgents, terorists, criminal enterprises, and trafickers in arms, humans, or 
drugs (?ilicit activity? and ?ilicit actors? for short).
2
 Yet the policies and strategies 
deployed against these chalenges are oftentimes inefective (or counterproductive), 
                                                
1
 ?Iregular?: United States Special Operations Comand and United States Marines Corps, Irregular 
Warfare (IW) Joint Operating Concept (JOC), Version 1.0 (Washington: United States Department of 
Defense, 207), htp:/www.fas.org/irp/dodir/dod/iw-joc.pdf. ?Asymetric?: Robert M. Casidy, Major, 
U.S. Army, ?Why Great Powers Fight Smal Wars Badly,? Military Review LXX, no. 05 (200): 41. 
?Open-source? warfare and ?holow? states: John Rob, Brave New War: The Next Stage of Terorism and 
the End of Globalization (Hoboken: John Wiley & Sons, 207). ?Failing? states: Stephen D. Krasner, 
?Sharing Sovereignty: New Institutions for Colapsed and Failing States,? International Security 29, no. 2 
(204): 85; Robert I. Rotberg, ?The New Nature of Nation-State Failure,? Washington Quarterly 25, no. 3 
(202): 85; Ren? Lefort, ed., UNESCO Courier?When the State Fails: A Survival Guide (Paris: United 
Nations Educational, Scientific and Cultural Organization, 201). ?Fragile? states: United States Agency 
for International Development (USAID), Fragile States Strategy (Washington: 205), 
http:/ww.usaid.gov/policy/205_fragile_states_strategy.pdf; United Kingdom Department for 
International Development, Why We Ned to Work More Efectively in Fragile States (London: 205). 
?Lawles? states: James Dao and Eric Schmit, ?A Nation Chalenged: Strategy: U.S. Ses Batles in 
Lawles Areas After Afghan War,? The New York Times, 8 January 202. ?Paper? states: Ken Menkhaus, 
?Beneath the Surface of Paper States,? in UNESCO Courier?When the State Fails: A Survival Guide 
(Paris: United Nations Educational, Scientific and Cultural Organization, 201). ?Feral? cities: Richard J. 
Norton, ?Feral Cities,? Naval War Colege Review 56, no. 4 (203): 97. ?Complex? chalenges: Consortium 
for Complex Operations, http:/coportal.org (acesed 10 October 2008). 
2
 A very smal sampling of relevant works includes: Ane L. Clunan and Harold Trinkunas, eds., 
Ungoverned Spaces? Teritorial Statehod, Contested Authority and Softened Sovereignty (forthcoming); 
David Kilculen, ?New Paradigms for 21st Century Conflict,? eJournal USA: Countering the Terorist 
Mentality, published electronicaly by U.S. Department of State, htp:/usinfo.state.gov/journals 
/itps/0507/ijpe/kilcullen.htm (acesed 18 October 208); Angel Rabasa, et al., Ungoverned Teritories: 
Understanding and Reducing Terrorism Risks (Arlington, Va.: RAND Project Air Force, 207), 
http:/ww.rand.org/pubs/monographs/MG561; John Robb, Brave New War: The Next Stage of Terorism 
and the End of Globalization; Dale E. Lichtblau, et al., ?Analyzing Adversaries as Complex Adaptive 
Systems,? monograph, IDA Papers P-3868 (Alexandria, Va.: Institute for Defense Analysis, October 
2006); John Rapley, ?The New Middle Ages,? Foreign Affairs 85, no. 3 (206): 95; United States 
Department of Defense, Quadrenial Defense Review Report (Washington: 206); Richard J. Norton, 
?Feral Cities.? 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 566 
especialy, as happens often, when undertaken in ways that undermine the internal 
legitimacy of the people and institutions responsible for addresing them. There is a sense 
among many in the development and defense communities that building legitimacy 
should therefore be a central component of eforts to counter ilicit activity.
3
 
The dificulty with this emphasis on legitimacy-building is that few people can 
say with confidence that they know how legitimation actualy works in such 
environments. Doctrines claiming that legitimation contributes to stabilization are 
generaly based on research and experiences that find a general correlation betwen the 
two phenomena, but they are mostly silent about the causal mechanisms supposedly 
linking legitimacy to stability. Most scholarly studies, meanwhile, asume that conflicts 
take place only betwen a state and a competitor, focus on the legitimacy of only the 
state, or operate at the level of general theory, providing scant guidance on how to 
actualy influence populations who live where no state governs, where the state governs 
only very weakly, where state actors are despised, or where conflicts involve more 
complex mixes of armed actors. Before deploying legitimacy-building policies to addres 
security isues at the local, national, or regional levels, policy-makers should have a 
stret-level understanding of how atitudes about legitimacy develop, how they change, 
and how those changes afect eforts to stabilize areas that are plagued by violence and 
contested or controlled by ilicit armed actors. 
This study focused on legitimacy at the stret level, specificaly on the micro- and 
multi-level dynamics of legitimacy in Medel?n, Colombia. Medel?n was a nearly ideal 
case for this study. For more than two decades, violence in and around the city has been a 
                                                
3
 Robert D. Lamb, Ungoverned Areas and Threats From Safe Havens. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 567 
devastating product of complex conflicts among stret crews, gangs, militias, 
paramilitaries, insurgents, trafickers, mafias, and government forces; high levels of 
violence, intimidation, corruption, and a breakdown in the rule of law have facilitated 
dangerous ilicit activities whose efects have been felt throughout the country and 
beyond. Global trends suggest that such environments are likely to become increasingly 
prevalent.
4
 In Medel?n, however, key indicators of violence and criminality declined 
dramaticaly betwen 2003 and 2007, when its homicide rate, for example, fel below that 
of some U.S. cities.
5
 The barios that experienced some of the most dramatic changes are 
smal enough to permit a detailed look at the proceses, atitudes, and events that drove 
them. It was an important case, therefore, for insight into the proceses that produce 
violence or stability in complex urban environments, and to test propositions specificaly 
about the role of legitimation, delegitimation, and ilegitimation ? especialy now that 
the homicide rate has returned to 2003 levels and continues to rise. A fine-grained 
explanation of the paterns of violence in Medel?n during the past 25 years could help 
policy makers facing complex urban violence ? in Medel?n and elsewhere ? to 
understand and influence the proceses through which that violence might be reduced. 
                                                
4
 One such trend is the growing diffusion of authority, identities, and loyalties away from the state, toward 
other types of institutions and other levels of governance, which greatly increases the complexity of the 
governance environment (se Appendix A). Another is urbanization: half of the world?s population is now 
urban, and that proportion keps growing. Samuel K. Moore and Alan Gardner, ?Megacities By the 
Numbers,? IEE Spectrum 44, no. 6 (207): 24; Stanley D. Brun, Jack Francis Wiliams, and Donald J. 
Zeigler, eds., Cities of the World: World Regional Urban Development, 3rd ed. (Lanham, Md.: Rowman & 
Litlefield, 203). These trends suggest that the most important security chalenges might increasingly be 
complex urban conflicts, something U.S. policy makers, strategists, and defense analysts have said they do 
not yet have a handle on. For an excelent introduction and discusion of complex urban security 
environments, see Richard J. Norton, ?Feral Cities.? 
5
 Medel?n en Cifras [Medel?n by the Numbers] (Medel?n: Alcald?a de Medel?n, 207); Ralph Rozema, 
?Urban DR-Proceses: Paramilitaries and Criminal Networks in Medel?n, Colombia.? 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 568 
This study was designed, therefore, to explain the changes in the paterns of violence 
there and to extract useful policy recommendations from the explanation. 
D.2. Method: Single-Case Study with In-Case Variation 
This disertation is the report of a case study of the paterns of violence in 
Medel?n, Colombia, betwen 1984 and the first six months of 2009, a period during 
which violence increased dramaticaly, fluctuated for some years, declined dramaticaly, 
and then rose again. The objectives of the study were, first, to describe the paterns of 
violence, something that, surprisingly, had not been done for the entire period for the full 
range of forms of violence; second, to explain the paterns of violence; and third to 
determine what particular role legitimacy or ilegitimacy did or did not play in that 
explanation. 
The case was defined as changes in the patterns of violence in Medel?n from 
1984 to 2009. The case was not changes in homicide rates, because violence takes 
multiple forms: While a drop in murder rates is an important policy goal, that drop can 
come about through undesirable mechanisms (e.g. consolidation of mafia control) and 
can be acompanied by a rise in other forms of violence (e.g. forced displacement, 
kidnapping, etc.). Normaly, therefore, the policy goal is an improvement in the overal 
patern of violence rather than in only one of its aspects. 
An embedded-case design was chosen as the best method to capture the 
microdynamics and the multi-level dynamics of legitimacy and violence.
6
 Because life 
                                                
6
 Robert K. Yin, Case Study Research: Design and Methods, revised ed. (Newbury Park, Calif.: Sage 
Publications, 1989). 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 569 
takes place at multiple levels but is experienced most imediately at the micro level, the 
case was studied at two levels of analysis: the city of Medel?n as a whole, and a sector of 
Medel?n caled Caicedo La Siera. To capture within-case variation, both levels were 
studied over five periods in recent history (yielding the equivalent of ten subcases), 
divided acording to whether homicides (as a proxy for violence) were declining or 
rising: the period 1984 to 1992 experienced a dramatic increase in homicides at al thre 
levels of analysis; the period 1992 to 1998 witnesed fluctuations in the magnitude of 
homicides but with an overal downward trend; this was followed by another spike in 
violence betwen 1998 and 2003; the period 2003 to 2007 experienced a dramatic decline 
in homicides at al thre levels; and, finaly, beginning in 2007 or 2008 homicides began 
an ominous rise that continued into 2009. That design made it possible to compare the 
thre periods in which homicides were rising with each other and with the two periods in 
which homicides were faling; and to compare the two declining periods with each other 
and with the thre rising periods. Other forms of violence turned out either to track the 
paterns in homicides reasonably wel or to go against them in ways that did not 
invalidate the division of time. 
The multilevel, multidimensional measurement frameworks introduced earlier 
(se Appendices B and C) were used as a systematic way to analyze and organize the 
quantitative and qualitative data that were collected. For each subcase, I tried to find 
evidence about what forms of violence were present and how they were changing (i.e. 
rising or faling in both absolute terms and relative to one another). Also for each 
subcase, I tried to find evidence related to the legitimacy or the ilegitimacy of 
community actors, nonstate armed actors, and state actors. In addition to quantitative 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 570 
data, qualitative data, and general impresions regarding changes in violence and 
legitimacy, I sought substantive information about the history of teritorial control, the 
quality of governance, and the extraction of resources, especialy ilicit resources, in each 
subcase. 
As noted in the introduction to this chapter, data could not be found for al of the 
variables, or for al of the subcases, that both measurement frameworks caled for. Using 
these frameworks, however, makes it possible for the researcher to be honest to his or her 
audience about exactly what the analysis was and was not based on, and helps to flag 
mising data to guide future data-collection eforts. While the mising data in this study 
did pose a serious obstacle to what could have been ? in theory ? a much more 
comprehensive analysis, this study nevertheles was the first even to atempt a 
comprehensive, multilevel, multidimensional analysis of the paterns of violence and the 
dynamics of legitimacy in a complex urban environment such as Medel?n. As such, its 
findings provide a richer explanation of the causal mechanisms among teritorial control, 
governance, legitimacy, and violence in Medel?n than had previously been found (or 
sought), and therefore provides a sound foundation upon which future research on the 
dynamics of complex urban violence may be based ? and more importantly, upon which 
sound policy might be designed. 
Policy makers do not have the luxury of puting of dificult policy decisions 
about complex policy isues just because reliable data are not available for the methods 
that could be designed to study them. (Se Appendix A for a discussion of the growing 
complexity of the policy-making environment.) If, in the real world, policy design for 
some isues must necesarily be impresionistic due to data and methodological 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 571 
constraints, then social science should contribute to beter policy design by ensuring that 
it is, at minimum, systematicaly impresionistic. For the present study of complex urban 
violence, I have endeavored to organize and analyze the relevant data and impresions as 
systematicaly as data and methodological constraints permited. 
D.3. Data Sources 
Short of ethnography, the most important sources of data for a study about the 
dynamics of legitimacy and the experience of violence necesarily must come from 
interviews with human beings teling their life stories, felings, beliefs, impresions, and 
opinions. In addition to 26 formal interviews that I did with residents of Caicedo La 
Siera or experts on violence and legitimacy in Colombia, Medel?n, and Caicedo La 
Siera, plus countles informal conversations with people from a broad diversity of 
backgrounds, I was lucky to have found a number of autobiographies and interviews that 
provided rich primary-source materials: 
? Patricia Nieto, ed., Jam?s Olvidar? Tu Nombre [I Wil Never Forget Your 
Name] (Medel?n, Colombia: Alcald?a de Medel?n, 2006). This edited volume 
contains autobiographical acounts of life in Caicedo La Siera from the late 
1980s to the mid-2000s, including personal and family histories of people 
displaced by violence in the countryside, writen by the protagonists as part of 
a series of writing workshops led by the editor, a journalist. 
? La Siera: Urban Warfare in the Barrios of Medel?n, Colombia, DVD, 
documentary film writen and directed by Scott Dalton and Margarita 
Martinez (Medel?n, Colombia: Human Rights Watch, 2005). This 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 572 
documentary was filmed in Caicedo La Siera and includes filmed interviews 
with a commander (?dison) and a fighter (Jes?s) of the Metro Bloc 
paramilitary group. The interviews were taken during the group?s victorious 
war against leftist militias and its subsequent defeat by rival paramilitaries of 
the Cacique Nutibara Bloc. 
? Arleison Arcos Rivas, ?Ciudadan?a Armada: Aportes a la Interpretaci?n de 
Procesos de Defensa y Aseguramiento Comunitario en Medel?n: El Caso de 
las Milicias Populares? [Armed Citizenry: Contributions to the Interpretation 
of Community Protection and Defense Proceses in Medel?n: The Case of the 
People?s Militias] (Master?s thesis, Universidad de Antioquia, 2005). In the 
appendices to this political-studies master?s thesis are transcripts of the 
author?s interviews with the founder (Hugo) and a commander (alias 
?Fernando?) of the November 6-7 Militia (M-6&7: Milicias 6 y 7 de 
Noviembre), which controlled upper Caicedo La Siera. 
? Juan Diego Alzate Giraldo, ?Alg?n D?a Recuperaremos la Noche: La 
Construcci?n de la Amenaza y el Miedo en Bario Caicedo?Las Estancias? 
[Someday We?l Get Back the Night: The Construction of Threat and Fear in 
Bario Caicedo?Las Estancias] (Undergraduate thesis, Universidad de 
Antioquia, 2005). This anthropology undergraduate thesis includes extended 
quotes from the author?s interviews with residents of bario Las Estancias in 
lower Caicedo La Siera. 
? Alonso Salazar J., No Nacimos Pa? Semila: La Cultura de las Bandas 
Juveniles de Medel?n [Born to Die: The Culture of Youth Gangs in Medel?n], 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 573 
5? ed. (Bogot?: Centro de Investigaci?n y Educaci?n Popular?CINEP, 1991). 
The author, at the time a student of communications but today the mayor of 
Medel?n, compiled and edited these autobiographical acounts of life in 
Medel?n?s youth gangs in the late 1980s based on his extensive interviews 
with the subjects. 
? Gilberto Medina Franco, Una Historia de las Milicias de Medel?n (Historia 
Sin Fin) [A History of the Militias of Medel?n (A Never-Ending Story)] 
(Medel?n, Colombia: Instituto Popular de Capacitaci?n?IPC, 2006), 
http:/ww.clacso.org.ar/biblioteca (acesed 10 April 2009). This 
autobiographical acount of the history of Medel?n?s militias was writen in 
fragments by a militant until 1994, when he demobilized, and was compiled 
and edited after his death by people who had known him. 
Violence data came from both oficial government sources and from various non-
governmental organizations and academic institutions. The main government sources 
were: 
? the office of the Government Secretariat (Secretar?a de Gobierno de 
Medel?n), the administrative and record-keeping arm of the city of Medel?n; 
? the national public prosecutor?s office (Fiscal?a General de la Naci?n); 
? the Department of National Statistics Administration (DANE: Departamento 
Administrativo Nacional de Estad?stica), the Colombian census bureau, which 
publishes vital statistics (estad?sticas vitales); 
? the national police department (Polic?a Nacional), which has subunits in the 
country?s departments and cities, and so is sometimes cited as Polic?a 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 574 
Metropolitana del Vale de Abur? (the Aburr? Valey Metropolitan Police); 
and 
? the National Institute of Forensic Medicine and Sciences (Instituto Nacional 
de Medicina Legal y Ciencias Forenses), Colombia?s medical examiner, 
which operates the country?s morgues (most commonly cited as Medicina 
Legal). 
The Secretar?a de Gobierno and the Fiscal?a build their data sets from the same source: 
public reports and prosecutions of crimes. DANE derives its data from multiple 
governmental sources, including Medicina Legal and its own surveys. The Polic?a 
Nacional?s dataset is derived from crime reports; it does not make city-level data 
available at the bario level, but does release data by comuna, or ward (the four barios 
that make up Caicedo La Siera are part of Comuna 8). Medicina Legal?s data are based 
on forensic analyses of corpses that enter the country?s morgues; its data for Medel?n are 
disaggregated by date, time of day, and stret addres of incident, making it possible to 
derive bario-level data for almost the entire study period. 
In addition to government sources, many NGOs colect data, although the quality 
of their data, the transparency of their methods, and their wilingnes to share vary 
widely: 
? National Labor School (Escuela Nacional Sindical): Human Rights Data Bank 
(Banco de Datos de Derechos Humanos); 
? Popular Education and Research Center (CINEP: Centro de Investigaci?n y 
Educaci?n Popular): Human Rights and Political Violence Data Bank (Banco 
de Datos de Derechos Humanos y Violencia Pol?tica); 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 575 
? People?s Training Institute (IPC: Instituto Popular de Capacitaci?n): IPC 
Database (Base de Datos del Instituto Popular de Capacitaci?n); and 
? Conflict Analysis Resource Center (CERAC: Centro de Recursos para el 
An?lisis de Conflictos): Conflict Data Base (Base de Datos de Conflicto). 
Bario-level homicide data from Medicina Legal were derived by Maria Victoria 
Llorente of the Center for Economic Development Studies (CEDE: Centro de Estudios 
sobre Desarollo Econ?mico) of the Universidad de los Andes in 2006 and were used in 
this study with her permision (I derived the bario-level data for 2007 and 2008). 
Finaly, survey data were useful for tracking trends in, for example, opinions 
about political actors and levels of fear over time. Most public opinions surveys in 
Medel?n do not collect data by bario, or even by comuna, but by zone, even though the 
surveys were conducted in the houses of the participants (and the survey-takers would 
therefore have been able to collect that data).
7
 The two most important sources were: 
? Galup Colombia Ltda., Gallup Poll Bimestral: Pol 62 (Bogot?: Galup 
Colombia, 2008); and 
? Encuesta de Percepci?n Ciudadana 2008 [Public Perception Survey 2008] 
(Medel?n, Colombia: Medel?n C?mo Vamos, 2008), 
http:/ww.medelincomovamos.org. 
                                                
7
 Piedad Patricia Restrepo R., personal comunication (Interview No. 2), 26 February 209. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 576 
D.4. Symbols 
The following abbreviations and symbols wil be used throughout this appendix: 
? ? causation (i.e. the left variable causes the right variable) 
? ? condition (i.e. the left variable enables the right variable) 
? ? positive fedback (i.e. the left and right variables are mutualy interdependent, 
alternating betwen cause and efect) 
? ? an increase in the value of the variable that folows 
? ? a decrease in the value of the variable that folows 
? ? observed change in the variable it modifies 
? ? a continuation of the cycle from the beginning of the equation 
f(y,z) ? read: ?f as a function of, asociated with, or with respect to, y and z? 
x
y
 ? read: ?x of y? (example: L
v
 is ?legitimacy of violence?), except for t
r
,c
r
,and r
t
 
X ? variable; other factors 
L ? legitimacy 
L(t
r
,c
r
,j,e,a,r
t
) ? legitimacy as a function of, respectively: transparency, credibility, 
justifiability, equity, acesibility, or respect 
I ? ilegitimacy 
I (t
r
,c
r
,j,e,a,r
t
) ? ilegitimacy as a function of transparency, credibility, etc. 
P ? policies or policy events initiated by state or nonstate actors, including ilicit armed 
actors 
G ? quality of governance 
G(C
S
) ? quality of governance with respect to social control 
V ? overal paterns of violence in Medel?n 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 577 
V
s
 ? self-inflicted violence 
V
i
 ? undiferentiated interpersonal violence 
V
if
 ? interpersonal-intrafamilial violence 
V
ic
 ? interpersonal-communal violence 
V
c
 ? undiferentiated collective violence 
V
cp
 ? collective-political violence 
V
ce
 ? collective-economic violence 
V
cs
 ? collective-social violence 
D ? conflict (D=dispute, which can be violent or nonviolent) 
R ? resources 
T ? teritory 
C
R
 ? control of resources (e.g. primary commodity exports, ilicit markets, extortion 
rackets, etc.) but not necesarily the teritory containing them 
C
T
 ? control of teritory, including the resources within it 
C
S
 ? social control or, more generaly, social order 
D(C
R
) ? conflict (dispute) over control of resources 
D(C
T
) ? conflict over control of teritory 
D(C
S
) ? conflict over social control 
v(D(C
R
) ? violence asociated with a conflict over control of resources 
v(D(C
T
)) ? violence asociated with a conflict over control of teritory (including the 
resources in it) 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 578 
D.5. Model Hypotheses 
No study wil be considered credible by a reader if author and reader do not share 
a common understanding of the key terms used in the study?s report. Here and throughout 
this report, terms are defined to be sure that there is no mistaking their meaning. A theory 
is a ?causal law (?I have established that A causes B?) or a causal hypothesis (?I surmise 
that A causes B?) together with an explanation of the causal law or hypothesis that 
explicates how A causes B?; a law is an ?observed regular relationship betwen two 
phenomena?; and a hypothesis is ?conjectured relationship betwen two phenomena.? An 
explanation is a statement about the causal mechanisms through which one phenomenon 
causes another, or a set of ?causal laws or hypotheses that connect the cause to the 
phenomenon being caused, showing how causation occurs. (A causes B because A causes 
q, which causes r, which causes B)?
8
 (se intervening phenomenon, below). 
The purpose of this study was not only to describe (se Chapter 8) but also to 
explain the changes in the paterns of violence, ?V, in Medel?n and, in the proces, to 
discover what role any changes in perceptions and expresions of legitimacy, L, played in 
those changes in violence. In this study, I focused primarily on increases and reductions 
in interpersonal-communal violence, V
ic
, and in the thre forms of collective violence, 
political, social, and economic, respectively, V
cp
, V
cs
, and V
ce
, because insufficient data 
were available to make any meaningful observations about the other categories; the bias 
in the analysis is toward the physical manifestations of those categories, for the same 
reason, but as some good data do exist (along with substantial anecdotal evidence) for 
their psychological manifestations, it should be asumed that the use of terms such as 
                                                
8
 Stephen Van Evera, Guide to Methods for Students of Political Science (Ithaca, NY: Cornel University 
Pres, 197), 9. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 579 
collective-economic violence or interpersonal-communal violence, in the absence of a 
modifier, is meant to include both physical and psychological forms. 
Eforts to describe or explain violence in Colombia had been almost exclusively 
partisan afairs before the publication of La Violencia en Colombia, by Orlando Fals-
Borda, ?lvaro Guzm?n, and Luis Uma?a in 1962, a non-partisan look at a broad range of 
violent actors and acts of violence in the country.
9
 What litle was published in the two 
decades that followed were generaly either expositions of various violent actors or 
partisan-flavored (e.g. pro-establishment or Marxist) explanations of the causes of the 
country?s civil war.
10
 In the 1980s, amid the explosion of violence and new types of 
violent actor countrywide, interest in the topic also exploded, and with it the quality and 
variety of the scholarship (which is not to say that this was not also acompanied by an 
explosion of cras polemics). Yet most works continued either to focus exclusively on 
political violence or to atribute other categories of violence to political actors rather than 
recognize that the mid-1980s were experiencing the beginnings of a convergence of 
motives, types of violence, and types of violent actor: narcotrafickers hiring guerilas to 
guard coca crops, narcotrafickers using military-grade weapons and tactics normaly 
asociated with insurgencies, politicians hiring gangs to asasinate political opponents, 
paramilitaries masacring or displacing peasants suspected of guerila influence then 
taking their land for economic gain, narcotrafickers forming self-defense groups and 
                                                
9
 Orlando Fals-Borda, ?lvaro Guzm?n, and Luis Uma?a, La Violencia en Colombia [Violence in 
Colombia] (Bogot?: Tercer Mundo Editores, 1962). 
10
 Jaime Arturo G?mez Corea, et al., ?Estado del Conocimiento Sobre la Violencia Urbana en Antioquia 
en la D?cada de los Noventa,? [?The State of Knowledge about Urban Violence in Antioquia in the 
1990s,?] in Balance de los Estudios Sobre Violencia en Antioquia, ed. Pablo Emilio Angarita Ca?as 
(Medel?n, Colombia: Municipio de Medel?n y Editorial Universidad de Antioquia, 201); Carlos M. Ortiz 
S., ?Los Estudios Sobre la Violencia en las Tres ?ltimas D?cadas?; Luis Fernando Medina, ?A Critique of 
?Resource-Based? Theories of Colombia?s Civil War,? An?lisis Pol?tico No. 61 (208): 44. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 580 
alying themselves with paramilitaries, and so on. The widespread asumption in the 
literature that the country?s violence was primarily atributable to politics in the context 
of civil war ? and derived either from mere ?banditry? or from the underlying social 
conditions driving the conflict ? was an eror that is understandable given the country?s 
history, but it was clearly incomplete. 
This state of afairs began to change in 1986 when then-President Virgilio Barco 
Vargas appointed a national commision of respected academics ? who came to be 
known as the first of Colombia?s ?violentologists? ? to study the causes of the country?s 
violence. The commision?s report was released in August 1987 and its main impact was 
to recognize that Colombia suffered not from ?violence? but from ?violences? of many 
diferent types, involving many diferent actors, and with many diferent and complex 
causes, including, among others, what it termed a ?culture of violence? that lowered the 
normative bariers to its use in Colombian society.
11
 
Since then, studies of violence in Colombia ? the only country where 
?violentology? has emerged as a distinct field of scholarship ? have proliferated, 
offering no smal number of causal explanations. One critical review of the literature on 
violence in the area found 43 articles, 39 books, 21 book chapters, 71 undergraduate 
theses, 11 masters? theses, and one doctoral thesis on the causes and consequences of 
violence in Antioquia alone ? and those are just the ones published in Colombia, in 
Spanish, betwen 1953 and 1999; 76 of those works focused specificaly on violence 
                                                
11
 Comisi?n de Estudios sobre la Violencia, Colombia: Violencia y Democracia: Informe Presentado al 
Ministerio de Gobierno [Colombia: Violence and Democracy: Report Presented to the Ministry of 
Government] (Bogot?: Universidad Nacional de Colombia, 1987). 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 581 
Medel?n.
12
 In most of these, legitimacy played either no role in the explanation or only 
an implicit role; in fact, very few orks have looked explicitly and systematicaly at the 
role of legitimacy in Medel?n?s violence. The few that have done so were considering 
primarily the legitimacy of the state (ignoring the dynamics of legitimacy in micro-
teritories and statelets where the state was absent) and in some of those it is clear that by 
?legitimacy? the authors actualy meant ?democracy.?
13
 The only recent work looking 
specificaly at the dynamics of legitimacy in Medel?n focused only on one conflict in one 
sector of the city, and it underspecified what was meant by legitimacy: did not define it, 
did not specify the legitimacy of what, nor legitimacy according to whom, nor legitimacy 
by what criteria (se Appendix C).
14
 This, in fact, is a pathology of most studies in which 
legitimacy plays a key role, and one I hoped to avoid in the present study. Moreover, 
almost al existing eforts to explain violence have looked only at what has caused 
increases in violence. It would be more iluminating to study what drives both the 
increases and the reductions in violence, across diferent categories of violence, in 
Medel?n, and this is what I atempted here as wel. 
                                                
12
 Pablo Emilio Angarita Ca?as, Balance de los Estudios Sobre Violencia en Antioquia. 
13
 Alfonso Monsalve Sol?rzano, Legitimidad y Soberan?a en Colombia 1958-2003 [Legitimacy and 
Sovereignty in Colombia 1958-2003], 1a ed., Serie Estudios Pol?ticos (Medel?n, Colombia: Editorial 
Universidad Pontificia Bolivariana, 204); Marco Palacios, Betwen Legitimacy and Violence: A History of 
Colombia, 1875-2002, trans. Richard Stoler, Latin America in Translation/en Traduci?n/Em Tradu??o 
(Durham: Duke University Pres, 206); Malcolm Deas, ?Interpretation of the Recent Colombian 
Violence,? in The Legitimization of Violence, ed. David Ernest Apter (Washington Square, N.Y: New York 
University Pres, 197); Eduardo Posada Carb?, ?Ilegitimidad? del Estado en Colombia: Sobre los Abusos 
de un Concepto [?Ilegitimacy? of the State in Colombia: On the Abuses of a Concept], 1a ed., Libros de 
Cambio (Bogot?: Alfaomega Colombiana: Ideas para la Paz, 203). 
14
 Pablo Emilio Angarita Ca?as, et al., ?Orden, Seguridad y Legitimidad,? [?Order, Security, and 
Legitimacy,?] in Din?micas de Guera y Construci?n de Paz: Estudio Interdisciplinario del Conflicto 
Armado en la Comuna 13 de Medel?n, ed. Pablo Emilio Angarita Ca?as, H?ctor Galo, and Blanca In?s 
Jim?nez Zuluaga (Medel?n, Colombia: Universidad de Antioquia?INER, Universidad de Medel?n, 
Corporaci?n Regi?n, y Instituto Popular de Capacitaci?n?IPC, 208), 107. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 582 
To identify the possible role that legitimacy might have played in the changes 
observed in the paterns of violence in Medel?n, I identified six logical categories of 
hypotheses derived from the five types of ?conjectured relationships? that can exist 
betwen any two given phenomena. In this case, the relationship in question was betwen 
L (legitimacy) and V (violence) and the question was, what kind of phenomenon was L? 
The possibilities are as follows: 
? A causal phenomenon, or a cause, is the ?phenomenon doing the causing?; in 
general theory, it is the independent variable. Hypothesis: changes in 
legitimacy caused changes in the paterns of violence (?L??V). 
? A caused phenomenon, or an effect, is the outcome or output of a system or 
proces: the ?phenomenon being caused?; in general theory, it is the 
dependent variable. Hypothesis: changes in the paterns of violence caused 
changes in legitimacy (?V??L). 
? An intervening phenomenon is the causal mechanism through which a cause 
causes an effect: it is ?caused by the causal phenomenon and cause[s] the 
[caused] phenomenon?; in general theory, it is an intervening variable. To 
explain a hypothesis, one need only to identify its intervening variables: that 
discovery constitutes the explanation. Hypothesis: legitimacy was the means 
through which a policy, policy event, or other factor caused changes in the 
paterns of violence (P??L??V or X??L??V). 
? An antecedent phenomenon or antecedent condition is a phenomenon ?whose 
presence activates or magnifies the causal action of the causal and/or 
[intervening] phenomena?; in general theory, it is a condition variable, an 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 583 
interaction term, an assumption, or, if its emergence is fairly sudden, a 
trigger. Hypothesis: Legitimacy magnified or enabled a policy, policy event, 
or other factor to cause the changes in the paterns of violence (?L?P??V or 
?L?X??V).
15
 
? A fedback phenomenon is esentialy an intervening phenomenon whose 
causal and caused phenomena are one and the same; put another way, it is a 
causal phenomenon in one moment and a caused phenomenon in the next. 
Such fedback loops or fedback proceses are manifestations of the 
bidirectional causality or mutual interdependence common in nonlinear or 
dynamic systems (sometimes caled dynamical systems), in which ?the value 
of al the variables at a given time depend on the values of these variables at 
the imediately preceding time, [which in turn] determine the state of the 
system at the succeding point in time,?
16
 often leading to unexpected or 
semingly inexplicable behavior of the system. Positive fedback or positive 
loops magnify the causal action of the variables involved and so ?acelerate 
basic proceses and bring some of them to ceilings at which they rest?; 
negative fedback loops, by contrast, ?provide counteracting forces, which 
sometimes lead to a stable equilibrium, sometimes to oscilation [betwen or 
among multiple equilibria], and sometimes to chaos?
17
 (a highly complex but 
                                                
15
 The terms and quotes of the first four bulets are from Stephen Van Evera, Guide to Methods for Students 
of Political Science, 7. 
16
 Robin R. Valacher and Andrzej Nowak, ?The Emergence of Dynamical Social Psychology,? 
Psychological Inquiry 8, no. 2 (197): 73. 
17
 Robert A. Haneman, Randal Colins, and Gabriele Mordt, ?Discovering Theory Dynamics By 
Computer Simulation: Experiments on State Legitimacy and Imperialist Capitalism,? Sociological 
Methodology 25 (195): 1. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 584 
non-random patern). Hypothesis: Legitimacy and violence were in a fedback 
relationship, so changes in one caused changes in the other, which in turn 
caused changes in the first, and so on (?L??V), most likely with some policy, 
policy event, or other factor involved as wel (P??L??V?P or 
X??L??V?X). 
The six sets of model hypotheses, therefore, are: 
(1) not legitimacy, (2) legitimacy as a causal 
phenomenon, (3) legitimacy as a caused 
phenomenon (efect), (4) legitimacy as an 
intervening phenomenon, (5) legitimacy as an 
antecedent phenomenon, and (6) legitimacy as 
fedback (se Table D-1). The correct explanation 
turned out to involve ilegitimacy more than 
legitimacy, and ilegitimacy operated as an 
intervening phenomenon in a larger positive-fedback cycle. Nevertheles, it is useful to 
review the logical hypotheses, to make explicit the possibilities that had been considered 
and rejected, the beter to state the study?s conclusions with clarity and confidence. 
Model Hypothesis 1: Not Legitimacy 
The most prominent school of thought in this category is what one critic cals the 
?supply-side? theory of civil war, so designated because of the emphasis it places on the 
supply of resources as the primary driver of civil war; Luis Fernando Medina has 
systematicaly (and in my view succesfully) refuted the validity of the findings and 
Table D-1. Six Model Hypotheses 
Legitimacy?s 
Relationship 
 to Violence 
 
Model 
Hypothesis 
 
 
No. 
none X??V 1 
cause ?L??V 2 
effect ?V??L 3 
intervening 
phenomenon 
X??L??V 
P?L?V 
4 
antecedent 
phenomenon 
?L?X??V 
?L?P?V 
5 
feedback 
phenomenon 
X??L??V?X 
P?L?VP 
6 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 585 
recommendations of most supply-side explanations of Colombia?s violence by criticizing 
their results and policy recommendations as esentialy non sequiturs to the evidence they 
produce.
18
 But it pays to consider the possibility that some beter-designed studies might 
find a reasonable explanation that excludes al legitimacy dynamics. 
The touchstone of this school of thought is a 2000 paper by Paul Collier and Anke 
Hoefler, ?Gred and Grievance in Civil War,? which found that civil wars are explained 
beter by ?gred? than by ?grievance.?
19
 Their explanation of conflict, however, does 
involve an indicator of the presence of ilegitimacy: grievance, in the role of intervening 
variable; therefore their model is discussed with the fourth set of hypotheses. Other works 
in this school that sem on their face not to find a role for legitimacy also turn out, upon 
closer examination, to include indicators of legitimacy in their explanations and are 
discussed under the appropriate headings.
20
 
One can imagine an argument that atributes increases in certain categories of 
violence, particularly economic violence, to conflicts over resources or teritory, conflicts 
caused by the mere availability (or perhaps triggered by the discovery) of those resources, 
without the intervention of any indicators for legitimacy or ilegitimacy: 
 R??v
ce
(D(C
R
)), or (H-1.1) 
 R??v
ce
(D(C
T
)), (H-1.2) 
                                                
18
 Luis Fernando Medina, ?A Critique of ?Resource-Based? Theories of Colombia?s Civil War.? 
19
 Paul Colier and Anke Hoefler, ?Gred and Grievance in Civil War,? working paper, Policy Research 
Working Paper 235 (Washington: World Bank Group, May 200), htp:/econ.worldbank.org/external 
/default/main?pagePK=64165259&theSitePK=469372&piPK=64165421&menuPK=6416093&entityID=
000094946_00060205420011. 
20
 e.g. Mauricio Rubio, ?Ilegal Armed Groups and Local Politics in Colombia,? Journal of Drug Isues 35, 
no. 1 (2005): 107. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 586 
which is read to say that the availability of (usualy ilicit) resources (such as drug 
markets, extortion rackets, or smuggling routes), R, has caused an increase in collective-
economic violence, ?V
CE
( ), asociated with disputes over control of either just those 
resources, D(C
R
), or the teritory in which those resources are available (or are ?lootable,? 
in some studies), D(C
T
). While this amounts to making the trivial observation that 
resources are necesary for violent conflicts over resources, it is silent on whether 
resources or the discovery of resources are sufficient to generate such conflicts: to 
disprove it, one needs only to identify a place in Medel?n where such resources were 
present but violent conflicts over them were not. And, indeed, there have been many such 
places in Medel?n in the past 25 years; an example would be the upper-class 
neighborhood of El Poblado, which had many wealthy busineses, many of which were 
extorted by nonstate armed actors and so they counted as a resource; but El Poblado did 
not suffer nearly the levels of violence as the city?s peripheral barios. An additional 
explanation, therefore, is needed for why violent conflicts have broken out in some 
barios but not in others. The presence or discovery of ilicit resources may wel turn out 
to play an important role, but it is not decisive. 
Some authors suggest that violence derives from a desire for security (i.e. in-
group defense against atacks by out-groups) or from the absence of efective social or 
political mechanisms to keep normal conflicts from becoming violent. The types of 
violence asociated with this set of explanations tend to be either interpersonal-communal 
or collective-social, and their cause tends to be atributed to either a breakdown in social 
order, a breakdown in the rule of law, or, as noted earlier, a culture of violence. This set 
of explanations can be iluminated somewhat by the basic statistics for homicides in 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 587 
Medel?n betwen 1990 and 2002. During that period, 39.6 percent of al murders took 
place in barios clasified as ?low? and ?low-low? on the city?s socioeconomic scale, and 
46.2 percent took place in ?middle? and ?middle-low? barios, indicating a low quality of 
life. In 89.7 percent of cases only one person was kiled during the homicide event, and in 
93.0 percent of those cases nobody else was even injured, suggesting that this was not 
organized violence ? homicides were neither civilian casualties (?collateral damage?) 
nor military casualties in urban warfare, nor indiscriminate violence due to terorism ? 
but rather a more difuse phenomenon, smal-scale atacks, primarily one-on-one, not 
centraly directed. In only 15.4 percent of cases could a motive for the murder could be 
identified, but the most common motives that could be identified were revenge (7.0 
percent of al cases) and armed robbery (4.5 percent); the motives underlying the rest of 
the cases (84.6 percent in total) are unknown, which makes it dificult to identify the 
most important causes of homicides more generaly.
21
 Stil, these descriptive statistics are 
strongly suggestive ? but no by means conclusive ? of a dynamic in which no political, 
social, or religious authority has been capable of preventing violence, punishing violence, 
or instiling norms of self-control. 
Many authors, in fact, have observed that the majority of homicides in Colombia 
and Medel?n in particular have had nothing directly to do with the country?s civil war or 
with the batles over micro-teritories that have been the city?s most visible forms of 
violence: most have simply been examples of interpersonal-communal violence, related 
to common crime and private grudges. Sa?l Franco Agudelo found higher rates of violent 
                                                
21
 Marleny Cardona, et al., ?Homicidios en Medel?n, Colombia, Entre 190 y 202: Actores, M?viles y 
Circunstancias.? 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 588 
deaths in Colombia in regions with higher indicators of impunity.
22
 Francisco E. Thoumi 
atributed the rise of narcotraficking (and its asociated violence) to a culture and social 
order that instils no significant self-generated behavioral controls, or as he put it: in 
Colombia, ?the rules of law are weak, social capital is poor, and solidarity and trust are 
lacking?.
23
 The most general form of this explanation would be as follows: 
 ?C
S
??V, (H-1.3) 
which could be interpreted as indicating either that low levels of social control (or a low 
quality of social order) has led to high levels of violence, or that a decline in social order 
has caused an increase in violence, usualy interpersonal-communal or collective-
economic violence. No social order can fully contain violence, of course, but this 
explanation is saying more than that. It is recognizing that violence results from a 
breakdown either in the social proceses through which ?conduct is defined as desirable 
or undesirable, approved or deviant, permited or prohibited,?
24
 whether that be through 
some system of active social control, through which norms are constructed and socialized 
by the institutions of society (family, school, church, media, recreation, etc.) or through 
some system of reactive social control, through which violations of those norms are 
punished, either formaly by criminal justice institutions or informaly by social 
                                                
22
 Sa?l Franco Agudelo, El Quinto, No Matar: Contextos Explicativos de la Violencia en Colombia [The 
Fifth, Thou Shalt Not Kil: Explanatory Contexts of Violence in Colombia], 1a ed. (Bogot?: IEPRI 
Universidad Nacional & Tercer Mundo Editores, 199), cited in Jaime Arturo G?mez Corea, et al., 
?Estado del Conocimiento Sobre la Violencia Urbana en Antioquia en la D?cada de los Noventa,? 175. 
23
 Francisco E. Thoumi, ?Introduction,? Journal of Drug Isues 35, no. 1 (205): 1; se also Francisco E. 
Thoumi, ?The Colombian Competitive Advantage in Ilegal Drugs: The Role of Policies and Institutional 
Changes,? Journal of Drug Isues 35, no. 1 (205): 7. 
24
 ?Mediante el control social se definen las conductas deseables e indeseables, conformes o desviadas, 
l?citas o prohibidas ?? Hernando Le?n Londo?o Ber?o, Ricardo Le?n Molina L?pez, and Juan David 
Posada, ?Pol?tica Criminal y Violencia Juvenil,? [?Crime Policy and Juvenile Violence,?] in Balance de los 
Estudios Sobre Violencia en Antioquia, ed. Pablo Emilio Angarita Ca?as (Medel?n, Colombia: Municipio 
de Medel?n y Editorial Universidad de Antioquia, 2001), 31. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 589 
institutions and social presures (shaming, mocking, etc.).
25
 (Se ? 10.1 for a broader 
discussion of social control.) Yet this explanation risks being true by definition, as 
violence is almost always considered the best indicator of a breakdown in social order. 
To avoid its being trivial, one would need to say more than this. 
A special form of Model H-1.3 says more, and it is perhaps the single most 
common explanation for the violence Medel?n has experienced over the past 25 years: 
the absence of the state: 
 ?G(C
T
, C
S
)??V, (H-1.4) 
which is to say that a low quality of governance with respect to teritorial and social 
control (or that a decline in government control, interpreted here as control by state 
actors) has been the main cause of Medel?n?s high levels of (or increases in) violence. 
This suffers from a logical lapse, however: the absence of the state cannot be a cause of 
violence. It would be more acurate to say that in some places the state has not prevented 
violence or maintained order and nobody else has either. At best, the absence of the state 
has been a background condition that might have activated some other set of factors, X, to 
cause the violence: 
 ?G(C
T
, C
S
) ? X??V. (H-1.5) 
What could X be? One possibility is that the state?s absence has simply enabled existing, 
nonviolent disputes over social and teritorial control to become violent: X = D(C
T
)+D(C
S
), 
but to disprove that it is sufficient to show that, during some periods or in some places 
where the state has had no presence, normal, nonviolent disputes have not habitualy 
turned violent. And there have, in fact, been such places in Medel?n, particularly at the 
                                                
25
 Ibid. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 590 
very beginning of the study period when some ?stateles? barios (e.g. upper Caicedo La 
Siera) had not yet been taken over by gangs and social order was maintained by the 
communities? own structures of social control, such as the family, the church, nosey 
neighbors, and so on. 
If one recognizes that states are not the only kind of entity capable of maintaining 
order, then one may reinterpret ?G(C
T
, C
S
) to represent, not necesarily the state?s 
absence, but also the absence or insufficiency of other structures of social control, or 
perhaps a decline in the quality of governance with respect to social and teritorial 
control. In other words: a breakdown in social order or the rule of law. With this, a much 
more credible explanation for violence begins to emerge: 
 ?G(C
T
,C
S
)?[D(C
T
)+D(C
S
)]??v(D(C
T
),D(C
S
)), (H-1.6) 
which is read to mean that a breakdown in social order or the rule of law in some places 
has removed the restraints that had kept prexisting disputes over teritory or social 
structures nonviolent, and the removal of those restraints has made it possible for those 
disputes to become violent. This is plausible, but it raises the question of what has caused 
the breakdown in the rule of law; it?s like saying the valey flooded because the levee 
broke: wel, what caused the levee to break? If we want to keep legitimacy and 
ilegitimacy out of the explanation, it would make sense to reach back to the beginning of 
this discussion: perhaps the presence or discovery of ilicit resources is to blame: 
 R?[D(C
T
)+D(C
S
)]??G(C
T
, C
S
)?[D(C
T
)+D(C
S
)]??v(D(C
T
),D(C
S
)), (H-1.7) 
which looks more complicated than it realy is: the presence or discovery of ilicit 
resources, R, has exacerbated, ?, prexisting disputes over teritorial and social control, 
[D(C
T
)+D(C
S
)], beyond the ability of local governance structures to manage them, leading 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 591 
to a breakdown in either social order or the rule of law, ?G(C
T
, C
S
), that has acelerated, ?, 
the loss of the remaining restraints on those conflicts, [D(C
T
)+D(C
S
)], and without those 
restraints those conflicts have become violent, ?v(D(C
T
)+D(C
S
)). This, in fact, is a 
plausible explanation. But it might be shown to be incomplete if evidence were found to 
indicate that the breakdown in the rule of law has come about through the action of 
something other than the presence or discovery of ilicit resources or the presence of 
prexisting conflicts. Model H-1.7 was the first hypothesis to be tested. 
One thing that al of the hypotheses discussed so far have in common is that they 
represent eforts to explain increases in or the onset of violence. But what explains the 
declines in violence that have taken place in Medel?n as wel, such as those experienced 
during the mid-1990s and the mid-2000s? It cannot be simply that the causes 
disappeared: ilicit resources have continued to be exploited even after the violence 
asociated with their exploitation them has declined. It might make sense to say that 
violence driven by conflict has declined when the conflict has been resolved, and while 
this has been true in cases of negotiated setlements, it is also simple enough to identify 
cases in which the underlying conflict was not so much resolved as won: one side has 
defeated the other in the conflict over resources (or the teritory containing those 
resources) and so the violence declined because the war was now over. This is consistent 
with Stathis Kalyvas?s theory that violence in civil war is largely a function of teritorial 
control, with areas of partialy contested control tending to be the most violent, areas of 
full teritorial control tending to be the least violent, and areas of equaly contested 
control faling somewhere in betwen.
26
 That theory, however, was not intended to apply 
                                                
26
 Stathis N. Kalyvas, The Logic of Violence in Civil War (New York: Cambridge University Pres, 206). 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 592 
to non-civil-war environments, such as the complex urban violence that is the subject of 
the current study. However, it is worth considering its merits in this case. This argument 
would be a variant of Model H-1.6 ? when normal disputes over teritorial control can 
no longer be governed, they turn violent ? but with the vertical arows reversed and a 
lower emphasis on governance. The simplified general form would be 
 ?C
T
??v(D(C
T
)), (H-1.8) 
which is to say that an increase (beyond some unspecified level) in teritorial control by 
one actor (i.e. that actor?s victory) has led to a decline in violence asociated with the 
dispute over teritorial control. Asking what enables the increase in teritorial control is 
equivalent to asking what causes victory in war: a mercantilist might suggest that 
superior resources caused the increase in control, but one can imagine other scenarios as 
wel ? superior strategy, superior morale, luck, etc. ? and some of those wil be 
considered in later hypotheses. 
One popular explanation has it that the most important declines in violence have 
come when state actors have entered pacts of convenience with the major nonstate armed 
actors: the nonstate actors would keep violence below a certain threshold so the state 
actors could take credit for good governance and good policy, and in exchange the state 
actors would forgive the nonstate actors? past crimes and ignore their future crimes. 
There are two basic variants, the conspiratorial and the analytical, but they difer mainly 
with respect to what the speaker believes counts as evidence.
27
 Structuraly they are 
identical: 
                                                
27
 Luis Fernando Quijano Moreno, personal comunication (Interview No. 5), 16 March 209; Francisco 
Guti?rez San?n and Ana ar?a Jaramilo, ?Crime, (Counter-)Insurgency and the Privatization of Security: 
The Case of Medel?n, Colombia,? Environment and Urbanization 16, no. 2 (204): 17. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 593 
 ?V?[?$C
T
 
+?$G]?P(?D(C
T
),?G(C
S
,C
T
))??V, (H-1.9) 
which explains that high levels of violence have raised the costs of teritorial control and 
governance for al actors, [?$C
T
 
+?$G], creating an incentive for a policy (the pact), P ( ), 
in which the state actor agres not to chalenge the nonstate actor?s control over teritory 
(and, implicitly, the ilicit resources in it), ?D(C
T
), in exchange for the nonstate actor?s 
enforcing social order within that teritory, ?G(C
S
,C
T
), resulting in a (short-term) decline 
in violence, ?V. (Clearly there are other efects of the pact ? for example, the costs of 
control and governance would decline as wel ? but in this model those are by-products 
of the main dynamic leading to the decline in violence, and so are not represented.) 
As for al good conspiracy theories, the absence of evidence of any pacts, and 
indeed, any evidence of their absence, is often taken as proof of their existence: people 
asume that there were pacts to decrease violence, and then trot out as evidence of those 
pacts the fact that the violence fel. Rather than beg the question, however, I took it 
seriously: Have there been explicit pacts? Or even, have there been implicit pacts ? that 
is, without communicating directly, did the state and nonstate actors recognize their 
mutual interest in ending the contest for teritorial control? Evidence for the existence of 
such pacts would tend to support this explanation, an easy mater in the case of the 
various peace negotiations of the mid-1990s, but somewhat more dificult in the more 
conspiratorial acount of the paramilitary demobilizations of the mid-2000s. To reject 
this explanation, however, it is sufficient to demonstrate that the state actors continued to 
contest the nonstate actors? control, or that the nonstate actors were unwiling or unable 
to reduce violence directly. And here the evidence is mixed, as wil be discussed at length 
in Chapters 4 and 6. Model H-1.9 was the second hypothesis to be tested. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 594 
One final explanation not involving legitimacy is worth pursuing. Sometimes 
refered to as the ?self-cleaning oven? theory of violence,
28
 it argues that when the 
perpetrators and victims of homicides in an area are al members of, say, waring gangs 
and their deaths take place in the context of their gang war, then sooner or later the kiling 
wil stop simply as a mater of atrition: the kilers wil keep kiling each other off until 
there are no kilers left ? problem solved. (Technicaly, there would have be at least one 
kiler left, unles his victory celebration involved suicide.) The problem with this theory 
(which could be modeled as V??V) is that it never happens that way: aside from the fact 
that innocent people are nearly always kiled in addition to the gangsters (and therefore 
any policy, based on this theory, of ignoring the situation would likely be considered 
unjustifiable and unfair), it also is often the case that the gang-war violence generates 
more violence as surviving family members or friends of victims often join the war to get 
revenge, leading to a spiral of increasing violence. 
Model Hypothesis 2: Legitimacy as Causal Phenomenon 
This set of hypotheses suggests that some change in legitimacy ? in the 
legitimacy of something, acording to somebody, by some set of criteria ? was the 
primary driver of the changes in paterns of violence, and that other considerations either 
were irelevant, were dominated by legitimacy, or were themselves triggers of 
legitimacy: 
 ?L??V, or (H-2.1) 
 X?L??V, (H-2.2) 
                                                
28
 Popularized by an of-hand comment made by a police-officer character in Clockers, DVD, film directed 
by Spike Le (1995). 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 595 
which is to say that changes in legitimacy caused changes in violence without other 
intervening phenomena (Model H-2.1), or existing atitudes about legitimacy caused the 
change in violence but they were not activated until the appearance of some other factor 
(Model H-2.2). 
However, there realy is only one realistic argument to be made for how 
legitimacy might be such a direct cause of changes in violence, and it comes down to the 
question of the legitimacy of what: in this case, the answer is the legitimacy of violence 
itself, or to be more precise, the legitimacy or ilegitimacy of the use of diferent forms of 
violence against diferent types of people under diferent circumstances. In a useful 
discussion of the legitimation of violence, Robin M. Wiliams Jr. noted that most of the 
violence that takes place in the world is considered by its perpetrators and their 
supporters not as ?deviant behavior, not disapproved conduct, but virtuous action in the 
service of applauded values?; four general strategies, he argued, are employed to 
legitimize the use of violence: 
? recategorization, by redefining violence or certain violent acts as ?not 
violence?; 
? rationalization, by refering to ?extenuating circumstances? (e.g. insanity) or 
?aceptable reasons?; 
? validation, by claiming violence is instrumental to protecting social values or 
achieving socialy valued goals; or 
? obligation, by valuing violence as ?a means to destroy evil? or a means to 
demonstrate virtue or social values.
29
 
                                                
29
 Robin M. Wiliams, Jr., ?Legitimate and Ilegitimate Uses of Violence: A Review of Ideas and 
Evidence,? in Violence and the Politics of Research, ed. Wilard Gaylin, Ruth Macklin, and Tabitha M. 
Powledge, The Hastings Center Series in Ethics (New York: Plenum Pres, 1981), 23. Wiliams?s terms 
are, respectively, neutralization, justifiable, rightful, and ?obligatory or dutiful?; I have changed the terms 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 596 
There is a risk that this kind of explanation might be trivial: that violence results when 
people decide it?s okay to use violence. That raises the question of what made people 
decide al of a sudden that it?s okay to use violence (in which case legitimacy would not 
be the causal phenomenon but an intervening phenomenon, and that is discussed as the 
fourth set of hypotheses). Stil, atitudes about virtue, social values, social goals, and 
?evil? ? and whatever situations activate them ? are among the strongest candidates for 
an explanation involving a direct causal link betwen legitimacy and violence. 
A non-trivial form of this explanation, therefore, can be derived from Model H-
2.2: certain atitudes about the legitimacy of violence are already present but are 
esentialy dormant until something triggers their manifestation. For example, in a 
community with very conservative, traditional values, people might consider it legitimate 
to threaten, atack, or even kil people who violate or represent a threat to those values ? 
usualy drug dealers, prostitutes, homosexuals, vagrants, minorities, and other members 
of low-status groups. But if such people are not present in the community in any numbers 
that sem threatening, then the community might ignore them, or maybe perpetrate 
relatively low levels of violence against them. If, however, more of these low-status 
individuals start ariving, they might surpas the high-status groups? threshold of 
tolerance, triggering a backlash against them in the form of atacks or threats. This is 
extremely common throughout the world. The violence might be rationalized by a lone 
vigilante who recognizes, perhaps, that threatening and kiling people is wrong but 
excuses it in this instance as the only solution to the perceived threat; or the violence 
                                                                                                                                            
for clarity and to ensure that his use of justifiable as a strategy of legitimizing violence is not mistaken for 
my use of justifiable as a criterion for legitimacy more generaly, and that his use of neutralization is not 
mistaken for my use of the same term to refer to the process in which something becomes neither legitimate 
nor ilegitimate. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 597 
might actualy be validated by the community, as some members form ?self-defense? or 
?civil-defense? groups ? often litle beter than death squads ? as a noble means to 
protecting the community against the deviants and criminals who, they believe, simply 
deserve to die or disappear. In fact, there is substantial anecdotal evidence to suggest that 
these things have actualy happened in Medel?n (e.g. the formation of bario self-defense 
groups) and in Colombia more generaly (e.g. the corruption of paramilitaries by its 
asociation with narcotraficking): 
 X?L
V
??V, (H-2.3) 
which says that some event, X, has activated atitudes about the legitimacy of violence, 
L
V
, to trigger some change in the levels or types of violence, ?V, in Medel?n. Model H-
2.3 was the third hypothesis to be tested. 
Model Hypothesis 3: Legitimacy as Caused Phenomenon 
The discussions of criteria and causal indicators for legitimacy provide some 
background to this set of hypotheses (se Appendix C). The legitimacy framework 
described there gives six general descriptors of phenomena that tend to contribute to 
beliefs and behaviors that are strongly suggestive of the presence of legitimacy: the two 
descriptors of predictability (transparent and credible) plus justifiable, equitable, 
acesible, and respectful. How they contribute to legitimacy and the question of whether 
they count as actual criteria for legitimacy or merely as enablers are addresed in those 
discussions. Here, I take it as given that whatever factors contribute to the emergence of 
legitimacy or ilegitimacy do so through the action of one or more of these six criteria. 
For example, people might be loyal to a particular politician because he talks about his 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 598 
religious beliefs in the same way they do or because his life story and his rhetoric are 
inspirational, regardles of whether his actual policies benefit them in other ways: he is 
worthy of their support (i.e. their support is justifiable) because he embodies or promotes 
something they value. Or, people in a dangerous neighborhood who might otherwise have 
been inclined to share information with the police suddenly stop doing so after hearing an 
officer utter an ethnic slur: he showed them disrespect, and even though the police patrols 
had been keeping the peace, the community now demands that they stop ? or goes 
farther and actualy atacks police when they enter ? even if the result is more violence 
and a lower quality of life. The findings of a scientific paper are likely to be considered 
unworthy of support ? it should not be believed ? if its methods and data are not 
adequately described, because without that transparency nobody has a basis on which to 
judge the findings. A government is likely to lose political support if it proves itself to be 
incapable of achieving its stated policy objectives, because people want to be able to trust 
that what a government says wil happen, wil happen: the loss of credibility leads to a 
loss of support at least in part because of the loss of predictability. 
If violence somehow causes a change in legitimacy, therefore, it must do so 
through the action of one or more of these six criteria. The most logical candidates are 
credible and justifiable: 
? An increase in violence makes life unpredictable: if you cannot get to work, 
go shopping, etc. without the risk or fear of geting mugged, raped, or kiled, 
then you cannot reliably plan out your day or cary out long-term projects. As 
a consequence of the violence, the people whose role is to keep the peace wil 
lose their credibility and any policies that had been implemented with the 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 599 
intention of controlling violence wil likewise lose their credibility. A loss of 
credibility tends to contribute to a loss of legitimacy. The reverse is true as 
wel: succes in controlling violence wil tend to contribute to the credibility 
of the role or policy that achieved it, and the role or policy wil tend to be 
considered more worthy of support as a result. 
? Justifiability is a mater of connecting something about the confere to 
something that the refere considers right, good, admirable, valuable, etc. As 
suggested in the discussion of the legitimation of violence in the last section, 
validation and obligation are two strategies that people sometimes employ to 
make violence not merely aceptable but valuable, admirable, even noble. 
Once violence ? of a certain type, under certain circumstances, against 
certain categories of people ? is considered valuable or noble within a certain 
community, then the people or groups within that community who engage in 
violence of that type, under those circumstances, against those people wil 
tend to be considered worthy of support in that endeavor. This is a very 
common dynamic in gangs, for example, in which violence against rival gangs 
is considered noble and obligatory, and in those gangs the most warlike 
gangsters wil tend to be considered the most admirable and the most worthy 
of, say, leadership. Similarly, people for whom nonviolence is a supreme 
virtue wil tend to find any policy involving violence or any perpetrator of 
violence to be worthy of their opposition, and many policies or promoters of 
nonviolence to be worthy of their support. 
Modeling these is a fairly straightforward mater: 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 600 
 ?V??L(c
r
), or (H-3.1) 
 ?V??L(j), (H-3.2) 
although one can complicate both in any number of ways, for example by identifying the 
many diferent ways that the paterns of violence might change, an exercise undertaken 
for al of the other hypotheses in this section, and one that wil not, therefore, be 
elaborated here. Models H-3.1 and H-3.2 were the fourth and fifth hypotheses to be 
tested. 
Model Hypothesis 4: Legitimacy as Intervening Phenomenon 
There are two ways legitimacy can intervene betwen a causal phenomenon and 
violence: either as an efect of a phenomenon and then the direct cause of the change in 
violence, or as an efect of a phenomenon, and then as the cause of some other factor that 
is itself the direct cause of the change in violence: 
 X??L
V
??V, or (H-4.1) 
 X
1
??L?X
2
??V. (H-4.2) 
Elaborating on the discussion of the second set of hypotheses, what distinguishes the two 
is the legitimacy of what: only the legitimacy of violence can act as a direct cause of 
changes in violence, whereas the legitimacy of any other type of confere can appear 
earlier in the chain of causality. 
In the first version ? legitimacy as an efect of one phenomenon but then the 
most proximate cause of violence ? some significant change in some factor causes 
people to change their behaviors and views of the legitimacy of violence such that the 
bariers to its use either rise or fal. For example, in a neighborhood where crime is 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 601 
rampant, some particularly heinous crime might push people beyond the limit of their 
tolerance, and they might decide that the only solution is to fight fire with fire, by 
forming a self-defense group to atack the drug dealers, rapists, extortionists, etc., and 
rationalizing the violence as the only available means to protecting the community. A 
longer-term version of this explanation might involve the delegitimation or ilegitimation 
of certain forms of violence over time: for example, the practice of social cleansing ? 
kiling low-status people defined as deviant or dangerous ? might enjoy a community?s 
support for many years, but as the number and variety of victims grows it might start 
making more and more people uncomfortable to the point where it not merely loses 
support (delegitimation) but actively triggers opposition (ilegitimation). Or paramilitary 
violence against certain populations ? leftist political parties, vilagers in guerila-
controlled areas, radical students ? might be tolerated by a majority of the population as 
long as it sems to be the only way insurgents might be defeated, but if it emerges that 
the paramilitaries behind those kilings are profiting economicaly from the practice (e.g. 
masacring vilagers to get their land) or that their behavior is now more of a threat to 
society than that of the insurgents, then that particular form of collective-political 
violence might lose its previous justification and become ilegitimate, forcing the 
paramilitaries to change their strategy (e.g. from masacres to more targeted kilings). 
Alternatively, as people?s confidence in the government rises, they might begin to believe 
that the state security forces should have a monopoly on violence, delegitimizing 
collective political or social violence by nonstate actors in the proces. There is 
substantial anecdotal evidence to suggest that many of these things have actualy 
happened in Medel?n (e.g. the formation of bario self-defense groups) and in Colombia 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 602 
more generaly (e.g. the coruption of paramilitaries by its asociation with 
narcotraficking, and their change of strategy after a delegitimation). As shown above, 
this explanation takes the following general form: 
 X??L
V
??V, (H-4.1) 
which says that some event or situation, x, has caused people to change their atitudes and 
behaviors about the legitimacy of violence, ?L
V
, which in turn has led to a change in the 
levels or types of violence, ?V, in Medel?n. Model H-4.1 was the sixth hypothesis to be 
tested. 
The second way legitimacy can be an intervening phenomenon, as noted, is for it 
to appear earlier in a longer chain of causation, as shown earlier: 
 X
1
??L?X
2
??V, (H-4.2) 
which says that some factor, X
1
, has caused a change in atitudes about legitimacy, ?L, 
which has resulted in a change in some other factor, X
2
, that is itself the proximate cause 
of the change in violence, ?V. 
The Collier-Hoefler ?gred vs. grievance? model of conflict, introduced briefly 
in the discussion of the first set of hypotheses (and adapted here as a model of violence 
due to conflict), is a prime example of this type of argument: they found that rebelions 
are not initiated by people who are unhappy with some facet of the way they are 
governed (grievance), but rather, ?opportunities for primary commodity predation cause 
conflict, and ? the grievances that this generates induce diasporas to finance further 
conflict? (gred)
30
: 
 R??V(D(C
R
))?I ?R?V(D(C
R
, I))?I ?, (H-4.3) 
                                                
30
 Paul Colier and Anke Hoefler, ?Gred and Grievance in Civil War.? 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 603 
which is read to mean that the discovery or availability of certain types of resources, R, 
has generated violence asociated with conflicts over those resources (gred), generating 
among the ?aggrieved? victims a duty of opposition (i.e. ilegitimacy), I, which has 
atracted resources from abroad, ?R, that have exacerbated the violent conflict (now over 
both resources and grievances), ?V(D(C
R
,I)), which has generated further grievances as 
the cycle, ?, repeated. Model H-4.3 was the seventh hypothesis to be tested. 
Model Hypothesis 5: Legitimacy as Antecedent Phenomenon 
There are few situations in which an exogenous change in legitimacy has 
activated some other factor?s ability to afect violence. The most promising, however, 
would be an anti-violence policy of some sort that had been tried repeatedly without 
succes, but suddenly succeded because now those who were implementing it had the 
backing of some key population: 
 ?L?P??V. (H-5.1) 
An example might be inteligence-gathering. In urban warfare (or iregular warfare more 
generaly), in countering organized crime, and in many other situations that result in 
violence, an important factor of a succesful strategy is reliable human inteligence: 
knowing who is a member of the armed group, knowing the group?s habits or where they 
hide their weapons, knowing their smuggling routes for ilicit trade, and so on. One side 
in the fight, say the state, can be trying for years to recruit inside informants without 
succes; sometimes this is because the state actors are considered outsiders, even 
enemies, by those on the inside, somebody not worth working with. But if something 
changes and atitudes about the state?s legitimacy change ? or if atitudes about the 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 604 
legitimacy of the insiders who control the area change ? recruiting informants might 
suddenly become easier for the state. To se if there were any policies that experienced 
breakthroughs as a result of people?s changing atitudes about the legitimacy of nonstate 
actors or state actors in Medel?n, Model H-5.1 was the eighth hypothesis to be tested. 
Model Hypothesis 6: Legitimacy as Fedback 
Here, the changes in violence are said to have been caused by dynamic 
interactions among legitimacy, violence, and some other factor: 
 ?V??L?X??V, (H-6.1) 
or any number of variations on this model. It might have worked as follows, for example: 
Some convergence of diferent trends and events, X, tipped the balance of local 
judgments about legitimacy (of some state or nonstate actors) to a point where certain 
anti-violence policies (e.g. targeting selected perpetrators of violence) could finaly 
succed, ?L?P??V; this preliminary succes initiated virtuous cycles in which the 
relative stability generated some legitimacy, ?L, which both facilitated policy and was 
enhanced by the policy it facilitated, ?L?P, while the policy itself both lowered violence 
and was made easier to implement as a result of the stability it created, P??V, which 
itself was further legitimizing, V??L, and so on, ?, as follows: 
 X??L?P??V??L?P??V??L ?, or (H-6.2) 
 P??L?P??V??L?P??V??L ?, (H-6.3) 
which says the same thing but that, instead of some ?convergence of diferent trends and 
events? starting the cycle, it was some particular set of policies or policy events that 
shifted people?s judgments about legitimacy and began the cycle (e.g. a local kiling that 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 605 
led to an atitude of ?enough?; the failure of Pastrana?s despeje strategy; police stepped 
up patrols in the barios and kept residents safe). For example, it is possible that some set 
of policies by a militia in a bario (e.g. seting up neighborhood-watch patrols, helping 
people buy groceries, chasing drug dealers of of stret corners, circulating moralistic 
propaganda, etc.) shifted those bario residents? views of the relative legitimacy of the 
militants versus the gangs who had been in control (e.g. the militants sem to be 
commited so it might be worth supporting them, while the gangsters truly are a menace 
to the community and should be opposed), leading to a breakthrough in local fundraising 
and human inteligence that made it possible to succed in their anti-violent-crime 
campaigns (i.e. it lowered interpersonal-communal violence, but actualy increased 
collective-social violence in the short-term), which greatly increased their credibility 
acording to the community, which further supported the policies, and so on. Diferent 
variations are possible, but Model H-6.3 was the ninth and final hypothesis to be tested. 
D.6. Rejected Hypotheses 
The introductory chapter to this study discussed the six model hypotheses, into 
which a total of 21 hypotheses found in the literature were categorized, 12 of which were 
imediately rejected for either logical lapses or the presence of fairly obvious 
contradictory evidence. That left nine plausible hypotheses to test: two from the first of 
the six categories, one from the second, two from the third, two from the fourth, one from 
the fifth, and one from the sixth. Each of these hypotheses was analyzed and rejected in 
whole or in part, and parts of those hypotheses that were not wholy rejected were 
synthesized into the two models that constitute the major explanations of the paterns of 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 606 
violence in Medel?n: one that explains most of the increases in violence, and one that 
explains most of the decreases in violence. The nine hypotheses that were tested, and the 
two models that were synthesized, are represented in Table D-2. In the sections that 
follow, the reasons for rejecting the nine hypotheses are provided, and the two 
synthesized models are explained. 
Hypothesis i 
Hypothesis i, claiming legitimacy played no role, argued that the presence or 
discovery of ilicit resources, R, exacerbated prexisting disputes over teritorial and 
social control, [D(C
T
)+D(C
S
)], beyond the ability of local governance structures to manage 
them, which led to a breakdown in either social order or the rule of law, ?G(C
T
, C
S
), which 
Table D-2. Nine Rejected Hypotheses and Two Synthesized Models 
No. Rejected Hypothesis Old No. 
i R?[D(C
T
)+D(C
S
)]??G(C
T
, C
S
)?[D(C
T
)+D(C
S
)]??V(D(C
T
),D(C
S
)) H-1.7 
ii ?V?[?$C
T
 +?$G]?P(?D(C
T
),?G(C
S
, C
T
))??V H-1.9 
iii X?L
V
??V H-2.3 
iv ?V???L(C
r
) or ?V???L(C
r
) H-3.1 
v ?V??L(j) H-3.2 
vi X??L
V
??V H-4.1 
vii R??V(D(C
R
))?I?R?V(D(C
R
, I))?I ? H-4.3 
viii ?L?P??V H-5.1 
ix P??L?P??V??L?P??V??L ? H-6.3 
 Synthesized Models  
I ?G(C
T
:t
r
,c
r
)??I(t
r
,c
r
)??$(C
T
)?D(C
T
)??v
c
(D(C
T
))?G(C
T
, C
S
)??V
ic
??G(C
T
, C
S
) ? 
II ?C
T
??v
c
(D(C
T
))?G(C
T
,C
S
)??V
ic
??G(C
T
, C
S
)  
   
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 607 
accelerated the loss of the remaining restraints on those conflicts, [D(C
T
)+D(C
S
)], and 
without those restraints those conflicts became violent, ?v(D(C
T
)+D(C
S
)). 
This was a plausible explanation, but it was rejected because aces to resources 
simply was not the main driver of al of the city?s violent disputes: some groups wanted 
to control teritory not because (or not only because) they wanted to control its resources, 
but because they wanted to protect their families and neighbors from violence (e.g. some 
stret crews and militias), or as part of a larger political strategy (e.g. some militias and 
the political paras), although many were indeed interested mainly in the resources (e.g. 
the drug gangs and narco paras). Moreover, it did not acount for interpersonal-
communal violence, only collective violence. Stil, this hypothesis did correctly identify 
disputes over teritorial control and problems of governance as significant factors that 
influenced both each other and much of the city?s violence, and it was clear that resources 
were needed to get or keep teritorial control and so would have had to play some role. So 
those parts of this hypothesis were worth keeping in mind. 
Hypothesis i 
Hypothesis i also claimed that legitimacy played no role. It argued that high 
levels of violence raised the costs of teritorial control and governance for the nonstate 
armed actors who controled the teritories and for the state actors who were supposed to 
be controlling them, [?$C
T
+?$G]; the high costs created an incentive for a policy (usualy 
a negotiated agrement of some sort), P ( ), in which the state actors agred not to 
chalenge the nonstate actors? control over teritory (and, implicitly, the ilicit resources 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 608 
in it), ?D(C
T
), in exchange for the nonstate actors? enforcing social order within that 
teritory, ?G(C
S
,C
T
), resulting in a (short-term) decline in violence, ?V. 
As Part I demonstrated, many such agrements were negotiated explicitly and 
some implicit agrements existed as wel; but this hypothesis was rejected because, in the 
vast majority of cases, these agrements fel apart prety quickly as one or more parties to 
the agrements almost invariably broke their promises (se Chapters 4-6). Moreover, 
most declines in violence had nothing to do with agrements. Finaly, it was prety clear 
that legitimacy and ilegitimacy did play some role in the paterns of violence (the urban 
militias made this point explicitly), and these first two hypotheses did not acount for that 
at al. Stil, this second hypothesis correctly identified the costs of teritorial control as 
something that somehow interacts with governance, violence, and the need for resources, 
and so that was something else to keep in mind as wel. 
Hypothesis ii 
Hypothesis ii claimed that legitimacy was a causal phenomenon and argued that, 
in a given teritory, some event, X, activated dormant atitudes about the legitimacy of 
certain types of violence, L
V
, which lowered the moral barier to their use, which 
increased the frequency of their use, ?V. 
This was a common proces in Medel?n: in communities with very conservative, 
traditional values, many people considered it legitimate to threaten, atack, or kil people 
who violated or represented a threat to those values ? usualy drug dealers, prostitutes, 
homosexuals, vagrants, minorities, and other members of low-status groups. And 
increases in the number of such low-status people into certain barios and increases in 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 609 
transgresive behaviors there did indeed trigger violent backlashes. Moreover, it was 
clear that the nonstate armed actors who controlled teritory had their own views about 
the circumstances under which it was legitimate to use violence (e.g. to protect teritory), 
and violence often was triggered by those very circumstances (e.g. an increase in 
atempts to infiltrate their teritory). 
In fact, this hypothesis could be acepted as valid if we were to think of it as an 
auxiliary proces that explains individual decisions to use violence. As a broader 
explanation, however, it was incomplete, because it suggested that atitudes about the 
legitimate use of violence were static (and were merely triggered by changing 
circumstances), when in many cases both the circumstances and the atitudes changed 
over time. (Hypothesis vi, however, did acount for such changes.) 
Hypothesis iv 
Hypothesis iv claimed that legitimacy and ilegitimacy were caused phenomena, 
specificaly that they emerged as a consequence of violence: As violence rose, ?V, life 
became unpredictable (i.e. if you cannot get to work, go shopping, etc. without the risk or 
fear of geting mugged, raped, or kiled, then you cannot reliably plan out your day or 
cary out long-term projects). As a consequence, the people whose job was to keep the 
peace lost their credibility, or any policies that had been implemented with the intention 
of controlling violence likewise lost their credibility, and this loss of credibility tended to 
undermine residents? motivation to support those in power, ?L(C
r
). 
The reverse could be claimed as wel: succes in controlling violence, ?V, tended 
to contribute to the credibility of the role or policy that achieved it, and the role or policy 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 610 
tended to be considered more worthy of support as a result, ?L(C
r
). But this hypothesis 
failed to acount for the fedback that would logicaly occur in this proces, especialy in 
the more extreme cases in which violence led to a loss of credibility that did not merely 
undermine support but actualy sparked opposition, and oftentimes that opposition 
involved violence. It also does not adequately acount for the efect of violence on 
governance, the rule of law, or social order, any of which could, in fact, have been an 
intervening phenomenon through which violence undermined legitimacy. 
Hypothesis v 
Hypothesis v, also claiming that legitimacy was a caused phenomenon, argued 
that violence, ?V, either directly legitimized or directly neutralized or ilegitimized its 
perpetrators, ?L, or that nonviolence directly legitimized nonviolent conferes: Wherever 
violence under certain circumstances was considered valuable or noble within certain 
communities, then people or groups within those communities who engaged in violence 
under those circumstances tended to be considered worthy of support in that endeavor. 
This was a very common dynamic in gangs, for example, in which violence against rival 
gangs was considered noble or obligatory, and in those gangs the most warlike gangsters 
tended to be considered the most admirable and the most worthy of, say, leadership. 
Similarly, this hypothesis argued, people for whom nonviolence was a supreme virtue 
tended to find any policy involving violence or any perpetrator of violence to be worthy 
of their opposition, and many policies or promoters of nonviolence to be worthy of their 
support. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 611 
The only real problem with this was that it failed to identify what else might have 
generated the legitimacy or ilegitimacy in any given instance: both phenomena have 
multiple causes, and this identifies only one. At best, this could be considered an 
auxiliary proces to more general proceses of legitimacy-building. But it was rejected as 
the best-fit explanation. 
Hypothesis vi 
Hypothesis vi, which claimed that legitimacy was an intervening phenomenon, 
argued that some event or situation, x, caused people to change their atitudes and 
behaviors about the legitimacy of violence, ?L
V
, which in turn led to a change in the 
levels or types of violence, ?V. In a short-term version of this hypothesis, some 
significant change in some factor caused people to change their behaviors and views of 
the legitimacy of violence such that the bariers to its use either rose or fel. In many 
neighborhood where crime was rampant, for example, it often happened that some event 
(e.g. a particularly heinous crime) pushed residents beyond the limits of their tolerance, 
so they formed self-defense groups to atack the drug dealers, rapists, extortionists, etc. In 
a longer-term version, certain forms of violence were neutralized or ilegitimized over 
time: for example, the practice of social cleansing ? kiling low-status people defined as 
deviant or dangerous ? enjoyed the support of many communities for many years, but as 
the number and variety of victims grew it started making more and more people 
uncomfortable to the point where it not merely lost support (neutralization) but actively 
triggered opposition (ilegitimation). Similarly, militia and paramilitary violence against 
certain populations was long tolerated by many in Medel?n as long as it semed to be the 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 612 
only way crime could be controlled or insurgents defeated. But after it emerged that the 
militias or paramilitaries behind those kilings were profiting economicaly from the 
practice, or after their behavior became a greater threat to residents than their victims had 
been, then that particular form of collective-political violence lost its previous 
justification and become ilegitimate. Similarly, as people?s confidence in the government 
rose during the 2000s, they began to believe that the state security forces should have a 
monopoly on violence, delegitimizing collective political or social violence by nonstate 
actors in the proces. 
The only real problem with this set of explanations was that it was incomplete; it 
did a beter job than Hypothesis ii of describing an important auxiliary proces that 
acounted for the efects on violence of both changing circumstances and changing 
atitudes about legitimacy. But neither the short-term nor the long-term versions 
adequately acounted for interactions with the isues of governance, rule or law, and 
social order that clearly played a part in the majority of cases. 
Hypothesis vii 
Hypothesis vi, which also claimed that legitimacy was an intervening 
phenomenon, argued that the discovery or availability of certain types of resources, R, 
generated violence asociated with conflicts over those resources, generating among the 
victims a duty of opposition (i.e. ilegitimacy), I, which atracted resources from 
outsiders, ?R, that exacerbated the violent conflict (now over both resources and 
grievances), ?V(D(C
R
,I), which generated further grievances as the cycle repeated. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 613 
The only real problem with this hypothesis was its emphasis on resources as the 
main driver of violent conflicts, when, as was pointed out in the criticism of the first 
hypothesis, most of the conflicts realy revolved around teritorial control, which often 
was desired as a means of controlling resources within teritories, but not always. The 
most important contribution this explanation made, however, was its recognition that 
Medel?n?s violence requires a dynamic, not a static, explanation: some factors afected 
other factors, which afected other factors, which afected the initial factors, and so on. It 
also identified some of the factors that ended up being involved in most of the city?s 
violence. 
Hypothesis vii 
Hypothesis vii claimed that legitimacy was an antecedent phenomenon. It argued 
that some exogenous change in legitimacy, ?L, activated the ability of some policy or 
strategy (that was already being tried), P, to somehow finaly work and thereby afect 
violence, ?V. And in fact there were some strategies ? by both state and nonstate armed 
actors ? that experienced breakthroughs as a result of people?s changing atitudes about 
the legitimacy of nonstate armed actors or state actors in Medel?n. An important set of 
examples were strategies of infiltration by competing nonstate armed actors: when 
outside armed actors wanted to control some teritory, they knew they needed aces to 
insiders to be able to take over, but it was often the case that they would try for many 
months or many years to get that aces without succes ? until those who had been 
controlling the teritory delegitimized themselves to insiders, and those insiders were 
suddenly open to outside help. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 614 
This hypothesis correctly recognized that atitudes about legitimacy and 
ilegitimacy afected diferent actors? ability to take or keep teritorial control, but it 
failed to identify any mechanisms through which that might take place. Moreover, its 
explanatory power was limited because the proces it described did not sem common. 
Hypothesis ix 
Hypothesis ix, which claimed that legitimacy was a fedback phenomenon, argued 
that some particular set of policies or policy events, P, tipped the balance of local 
judgments about legitimacy (of some state or nonstate actors) to a point where certain 
anti-violence policies (e.g. targeting selected perpetrators of violence) could finaly 
succed, ?L?P??V; this preliminary succes initiated virtuous cycles in which the 
relative stability generated some legitimacy, ?L, which both facilitated policy and was 
enhanced by the policy it facilitated, ?L?P, while the policy itself both lowered violence 
and was made easier to implement as a result of the stability it created, P??V, which 
itself was further legitimizing, V??L, and so on. The clearest example of such virtuous 
cycles was the peace camps, in which urban guerilas set up neighborhood-watch patrols, 
helped people buy groceries, chased drug dealers of of stret corners, circulated 
moralistic propaganda, and so on, policies that shifted those bario residents? views of the 
relative legitimacy of the militants versus the gangs who had been in control, leading to a 
breakthrough in local fundraising and human inteligence that made it possible to succed 
in their anti-violent-crime campaigns, which greatly increased their credibility acording 
to the community, who further supported the policies, and so on. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 615 
This is a compeling but ultimately limited explanation, for two reasons: first, 
while many anti-violent-crime campaigns did lower interpersonal-communal violence, 
the campaigns themselves also usualy increased collective-social violence in the short-
term; and second, more generaly, this proces did take place, but not teribly often 
outside of the peace camps. More commonly, violence fel only after one side or another 
won a war over teritorial control. 
D.7. Synthesized Models 
While none of the rejected hypotheses semed to have enough evidence to support 
it as the best overal explanation, pieces of many of them did sem to be relevant. The 
chalenge, therefore, was to determine how they might fit together to acount for most of 
the relevant dynamics. 
An overal look at the city?s paterns of violence (Chapter 8) and dynamics of 
legitimacy (Chapter 9) yielded thre key observations that semed to be common to most 
instances in which violence rose or fel in the city?s micro-teritories and statelets: 
1. Teritorial control has reduced violence. 
2. The cost of teritorial control has been positively correlated with ilegitimacy. 
3. Ilegitimacy has been positively correlated with unpredictability in the manner 
and outcomes of governance. 
A synthesis of these findings yielded the two synthesized models (se Table D-2), which 
explain the relationship betwen teritorial control and violence and acount for the role 
of ilegitimacy. The first model explains most increases in collective and interpersonal-
communal violence: 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 616 
 ?G(C
T
:t
r
,c
r
)??I(t
r
,c
r
)??$(C
T
)?D(C
T
)??V
c
(D(C
T
))?G(C
T
, C
S
)??V
ic
??G(C
T
, C
S
) ? (I) 
Model I explains that when those in control of teritory have governed in a manner 
characterized by a lack of transparency and credibility, ?G(C
T
:t
r
,c
r
), they have made it 
very likely that the people they governed would decide they should be opposed because 
they had made life intolerably unpredictable, ?I(t
r
,c
r
), and this ilegitimation has raised 
the cost of teritorial control, ?$(C
T
), making feasible a contest for control by rivals, 
D(C
T
), and thereby triggering a collective-violent conflict over control of teritory, 
?V
c
(D(C
T
)), which has complicated eforts to govern in a manner consistent with the rule 
of law and social order, G(C
T
, C
S
), which has removed the restraints on (and therefore has 
enabled) an increase in interpersonal-communal violence, ?V
ic
, which further exacerbated 
and was further exacerbated by problems of governance, social order, or the rule of law, 
?G(C
T
, C
S
), which caused the cycle to start over again. Ilegitimacy, then, was an 
intervening phenomenon in a positive-fedback cycle. 
What about the relationship betwen teritorial control and declines in collective 
and interpersonal-communal violence? Here, the Kalyvas model
31
 prevailed, with one 
elaboration: 
 ?C
T
??v(D(C
T
)?G(C
T
, C
S
)??V
ic
??G(C
T
, C
S
) (II) 
Model I explains that an increase in teritorial control by one actor (i.e. that actor?s 
victory), ?C
T
, has led to a decline in collective violence asociated with the dispute over 
teritorial control, ?v(D(C
T
), making it possible to govern in a manner consistent with the 
rule of law and social order, G(C
T
, C
S
), and therefore to beter control interpersonal-
                                                
31
 Stathis N. Kalyvas, The Logic of Violence in Civil War. 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 617 
communal violence, ?V
ic
, a decline that both further enabled and was further enabled by 
improvements in governance, social order, or the rule of law, ?G(C
T
, C
S
). 
A narative description of these findings is as follows. Within Medel?n, control 
over micro-teritories has required resources (e.g. money, weapons, communication, 
transportation) and people (e.g. recruits, financiers, facilitators, denouncers, informants, 
etc.) to counter opposition from external rivals (i.e. external to the micro-teritory), from 
internal rivals, and from rivals to the internal alies of the controlling group. For resource-
rich groups (for example, a mafia-backed gang with aces to narcotraficking income), 
legitimacy has not been necesary to hold teritory because whatever support was needed 
for these fights could be purchased or coerced using resources they already had. To 
maintain control, therefore, the controlling group neded only to engage in a strategy of 
ilegitimacy-avoidance. However, when that strategy has failed ? when their abuses, 
moral transgresions, or losses of credibility have caused life in the micro-teritory to 
become intolerably unpredictable ? then the level of moraly driven non-compliance and 
outright opposition has risen, increasing both the number of rivals and the level of 
resources required to coerce or purchase needed support. This has provided strategic 
opportunities for existing external and internal rivals to recruit more people and resources 
into their own fight against the controlling group, making a real contest for control 
tenable. As a consequence, the controlling group and its rivals then engaged directly in 
collective violence against each other and each other?s supporters, or alowed themselves 
to be drawn into private grudges through false denunciations. Moreover, as the 
controlling group focused on defense against rivals, it necesarily shifted resources and 
atention away from maintaining general order in their micro-teritories. Inatention to 
Lamb, Robert D. | Appendix D. Research Design and Hypotheses 618 
order entailed a breakdown in social order and the rule of law that created an atmosphere 
of impunity for common crime asociated with interpersonal-communal violence, which 
both fed back into problems of governance and further damaged the credibility and 
therefore the legitimacy of the controlling group. Thus have both collective violence and 
interpersonal-communal violence risen as a consequence of contests over control of 
micro-teritories (Model I). 
When one side or another in such contests has managed to prevail and take (or 
retake) full control over the micro-teritory, it no longer had a need for violence against 
rivals or rivals? supporters. As a consequence, collective violence asociated with the 
conflict has falen, and the controlling group could again dedicate atention and resources 
to protecting internal alies and maintaining order, so interpersonal-communal violence 
asociated with the breakdown in social order or the rule of law has falen as wel, which 
has fed back into improvements in governance in a virtuous cycle. Thus have both 
collective violence and communal violence falen as a consequence of control over 
micro-teritories (Model I). 
 
Lamb, Robert D. | Bibliography 619 
Bibliography 
Acosta, Orlando and Jorge Iv?n Gonz?lez. ?Globalizaci?n: Una Aproximaci?n desde la 
Evoluci?n Biol?gica y los Sistemas Complejos Auto-Organizativos.? 
[?Globalization: A Perspective from Evolutionary Biology and Self-Organizating 
Complex Systems.?] An?lisis Pol?tico No. 61 (2007): 101-21. 
Agence France-Prese. ?Venganzas de ?Narcoparamilitares? Disparan Homicidios en 
Medel?n.? MetroLatinoUSA. 13 July 2009. 
Agencia de Prensa IPC. ?Medel?n, Sin Rumbo Claro en Materia de Seguridad.? Am?rica 
Latina en Movimiento. Published electronicaly by Latin American Information 
Agency (ALAI: Agencia Latinoamericana de Informaci?n). 
http:/alainet.org/active/29137 (acesed 28 August 2009). 
??. ?Asesinan a Otro Testigo en Proceso contra Alias ?Mem?n?.? [?Another Murder of 
a Witnes against Alias ?Mem?n?.?] Ud. Reportero. Published electronicaly by 
Tera Colombia. http:/ww.tera.com.co/noticias/articulo/html/acu17137-ud-
reportero-asesinan-a-otro-testigo-en-proceso-contra-alias-memin.htm (acesed 
22 August 2009). 
??. ?En Medel?n, Bosque Del Bario La Siera Es Un Cementerio.? [?In Medel?n, the 
Woods in Bario La Siera Are a Cemetery.?] Published electronicaly by Instituto 
Popular de Capacitaci?n (IPC). 
http:/ww.ipc.org.co/page/index.php?option=com_content&task=view&id=1362
&Itemid=368 (acesed 27 April 2009). 
Lamb, Robert D. | Bibliography 620 
??. ?Nueva Captura en la Corporaci?n Democracia.? [?Another Arest in Corporaci?n 
Democracia.?] VerdadAbierta.com. Published electronicaly by Fundaci?n Ideas 
para la Paz (FIP) and Semana. http:/ww.verdadabierta.com/web3/conflicto-
hoy/50-rearmados/1371-la-corporacion-democracia-acorralada-por-la-justicia 
(acesed 22 August 2009). 
Alagappa, Muthiah, ed. Political Legitimacy in Southeast Asia: The Quest for Moral 
Authority. Contemporary Isues in Asia and the Pacific. Stanford, Calif.: Stanford 
University Pres, 1995. 
Alison, Graham. ?The Impact of Globalization on National and International Security.? 
In Governance in a Globalizing World, edited by Joseph S. Nye and John D. 
Donahue, 72-85. Washington, D.C.: Brookings Institution Pres, 2000. 
Almond, Gabriel A. and Sidney Verba. The Civic Culture Revisited: An Analytic Study. 
Boston: Litle, Brown, 1980. 
Alonso Espinal, Manuel A., Jorge Giraldo Ram?rez, and Diego Jorge Siera. ?Medel?n: 
El Complejo Camino de la Competencia Armada.? [?Medel?n: The Complex 
Course of Armed Competition.?] In Parapol?tica: La Ruta de la Expansi?n 
Paramilitar y los Acuerdos Pol?ticos, edited by Mauricio Romero, 109-64. 
Bogot?: Intermedio/Corporaci?n Nuevo Arco Iris, 2007. 
Alzate Giraldo, Juan Diego. ?Alg?n D?a Recuperaremos la Noche: La Construcci?n de la 
Amenaza y el Miedo en Bario Caicedo?Las Estancias.? [?Someday We?l Get 
Back the Night: The Construction of Threat and Fear in Bario Caicedo?Las 
Estancias.?] Undergraduate thesis, Universidad de Antioquia, 2005. 
Lamb, Robert D. | Bibliography 621 
Amnesty International. Colombia: The Paramilitaries in Medel?n: Demobilization Or 
Legalization? London: Amnesty International, 31 August 2005. 
http:/ww.amnesty.org/en/library/info/AMR23/019/2005. 
Angarita Ca?as, Pablo Emilio, ed. Balance de los Estudios Sobre Violencia en Antioquia. 
[An Evaluation of Studies about Violence in Atioquia.] Medel?n, Colombia: 
Municipio de Medel?n y Editorial Universidad de Antioquia, 2001. 
??. ?Comentario a la Investigaci?n ?Estado del Arte Sobre Pol?tica Criminal y 
Violencia Juvenil?.? [?Comments on the Study ?The State of the Art in Crime 
Policy and Juvenile Violence?.?] In Balance de los Estudios Sobre Violencia en 
Antioquia, edited by Pablo Emilio Angarita Ca?as, 68-79. Medel?n, Colombia: 
Municipio de Medel?n y Editorial Universidad de Antioquia, 2001. 
Angarita Ca?as, Pablo Emilio, H?ctor Galo, and Blanca In?s Jim?nez Zuluaga, eds. 
Din?micas de Guera y Construcci?n de Paz: Estudio Interdisciplinario del 
Conflicto Armado en la Comuna 13 de Medel?n. [Dynamics of War and 
Construction of Peace: An Interdisciplinary Study of the Armed Conflict in 
Medel?n?s Comuna 13.] 1a ed. Medel?n, Colombia: Universidad de Antioquia?
INER, Universidad de Medel?n, Corporaci?n Regi?n, y Instituto Popular de 
Capacitaci?n?IPC, 2008. 
Angarita Ca?as, Pablo Emilio, Blanca In?s Jim?nez Zuluaga, H?ctor Galo, Clara In?s 
Atehort?a Aredondo, Hernando Le?n Londo?o Ber?o, Luz Amparo S?nchez, 
Gonzalo Medina P?rez, Luz Dary Ruiz Botero, and Mario Elkin Ram?rez Ortiz. 
?Orden, Seguridad y Legitimidad.? [?Order, Security, and Legitimacy.?] In 
Din?micas de Guera y Construcci?n de Paz: Estudio Interdisciplinario del 
Lamb, Robert D. | Bibliography 622 
Conflicto Armado en la Comuna 13 de Medel?n, edited by Pablo Emilio Angarita 
Ca?as, H?ctor Galo, and Blanca In?s Jim?nez Zuluaga, 107-42. Medel?n, 
Colombia: Universidad de Antioquia?INER, Universidad de Medel?n, 
Corporaci?n Regi?n, y Instituto Popular de Capacitaci?n?IPC, 2008. 
Arcos Rivas, Arleison. ?Ciudadan?a Armada: Aportes a la Interpretaci?n de Procesos de 
Defensa y Aseguramiento Comunitario en Medel?n: El Caso de las Milicias 
Populares.? [?Armed Citizenry: Contributions to the Interpretation of Community 
Protection and Defense Proceses in Medel?n: The Case of the People?s 
Militias.?] Master?s thesis, Universidad de Antioquia, 2005. 
Aricapa Ardila, Ricardo. Comuna 13: Cr?nica de Una Guera Urbana. [Comuna 13: 
Chronicles of an Urban War.] 2a ed. Medel?n: Editorial Universidad de 
Antioquia, 2005. 
Aristotle. Politics. Translated by Ernest Barker and R. F. Staley. The World?s Classics. 
New York: Oxford University Pres, 1995 [330 BCE]. 
Atehort?a Aredondo, Clara In?s. ?Caracterizaci?n del Desplazamiento Forzado 
Intraurbano: Medel?n 2000-2004.? [?Characterizing Intra-Urban Forced 
Displacement: Medel?n 2000-2004.?] Master?s thesis, Universidad de Antioquia, 
2007. 
Atran, Scott, Robert Axelrod, and Richard Davis. ?Sacred Bariers to Conflict 
Resolution.? Science 317, no. 5841 (2007): 1039-40. 
Avenda?o, Mary Luz. ?En la Mira de Delincuentes.? [?In the Sights of Criminals.?] El 
Espectador. 10 August 2008. 
Lamb, Robert D. | Bibliography 623 
Balesteros, Andr?s, Jorge A. Restrepo, Michael Spagat, and Juan F. Vargas. ?The Work 
of Amnesty International and Human Rights Watch: Evidence From Colombia.? 
Working paper. Documentos de CERAC 003639. Bogot?: Centro de Recursos 
para el An?lisis de Conflictos, 1 February 2007. http:/ideas.repec.org/p/col 
/000150/003639.html (acesed 22 August 2009). 
Bardis, Panos D. ?Violence: Theory and Quantification.? Journal of Political & Military 
Sociology 1, no. 1 (1973): 121-46. 
Barero Tapias, Jos? Ricardo. ?Las Milicias Populares y el ELN Como Casos de Estudio: 
El Anclaje Cultural del Conflicto Pol?tico.? [?People?s Militias and the ELN as 
Case Studies: The Cultural Anchor of the Political Conflict.?] Revista Javeriana 
(2001): 213-38. 
Betham, David. The Legitimation of Power. Houndsmil: MacMilan Education, 1991. 
Besan?on, Marie. Good Governance Rankings: The Art of Measurement, WPF Reports 
36. Cambridge, Mas.: World Peace Foundation, 2003. 
Bolitho, Wiliam. Twelve Against the Gods: The Story of Adventure. New York: Simon 
and Schuster, 1929. 
Bollen, Kenneth A. ?Latent Variables in Psychology and the Social Sciences.? Annual 
Review of Psychology 53 (2002): 605-34. 
Bollen, Kenneth A. and Richard Lennox. ?Conventional Wisdom on Measurement: A 
Structural Equation Perspective.? Psychological Bulletin 110, no. 2 (1991): 305-
14. 
Lamb, Robert D. | Bibliography 624 
Booth, John A. and Mitchel A. Seligson. ?Legitimacy and Political Participation in Eight 
Latin American Countries? (paper presented at the Midwest Political Science 
Meting, Chicago, 2-4 April 2009). 
??. The Legitimacy Puzzle in Latin America: Political Support and Democracy in 
Eight Nations. New York: Cambridge University Pres, 2009. 
Brauer, Jurgen, Alejandro G?mez-Sorzano, and Sankar Sethuraman. ?Decomposing 
Violence: Political Murder in Colombia, 1946-1999.? European Journal of 
Political Economy 20, no. 2 (2004): 447-61. 
Brunn, Stanley D., Jack Francis Wiliams, and Donald J. Zeigler, eds. Cities of the 
World: World Regional Urban Development. 3rd ed. Lanham, Md.: Rowman & 
Litlefield, 2003. 
Buchanan, Alen E. Justice, Legitimacy, and Self-Determination: Moral Foundations for 
International Law. Oxford Political Theory. New York: Oxford University Pres, 
2004. 
Cardona, Marleny, H?ctor Iv?n Garc?a, Carlos Alberto Giraldo, Mar?a Victoria L?pez, 
Clara Mercedes Su?rez, Diana Carolina Corcho, Carlos Hern?n Posada, and 
Mar?a Nubia Fl?rez. ?Homicidios en Medel?n, Colombia, Entre 1990 y 2002: 
Actores, M?viles y Circunstancias.? [?Homicides in Medel?n, Colombia, from 
1990 to 2002: Victims, motives and circumstances.?] Cadernos de Sa?de P?blica 
21, no. 3 (2005): 840-51. 
Carnegie, Dale. How to Win Friends and Influence People. revised ed. New York: Pocket 
Books, 1982 [1936]. 
Lamb, Robert D. | Bibliography 625 
Casidy, Robert M., Major, U.S. Army. ?Why Great Powers Fight Smal Wars Badly.? 
Military Review LXX, no. 05 (2000): 41-53. 
Cebalos Melguizo, Ramiro. ?The Evolution of Armed Conflict in Medel?n.? Latin 
American Perspectives 116, no. 28 (2001): 110-31. 
Chayes, Abram and Antonia Handler Chayes. The New Sovereignty: Compliance with 
International Regulatory Agrements. Cambridge, Mas.: Harvard University 
Pres, 1997. 
Childs, J. Rives. Casanova: A New Perspective. 1st U.S. ed. New York: Paragon House 
Publishers, 1988. 
CIDEAL. La Violencia en Medel?n y Colombia: Iniciativas Para la Soluci?n del 
Conflicto (1980-2004). [Violence in Medel?n and Colombia: Initiatives for 
Solving the Conflict (1980-2004).] Medel?n, Colombia: CIDEAL?Pastoral 
Social?AECI, 2005. 
Clawson, Patrick and Rensselaer W. Le. The Andean Cocaine Industry. 1st ed. New 
York: St. Martin?s Pres, 1996. 
Clunan, Anne L. and Harold Trinkunas, eds. Ungoverned Spaces? Teritorial Statehood, 
Contested Authority and Softened Sovereignty., forthcoming. 
Coase, R.H. ?The Nature of the Firm.? Economica 4, no. 16 (1937): 386-405. 
Collier, Paul and Anke Hoefler. ?Gred and Grievance in Civil War.? Working paper. 
Policy Research Working Paper 2355. Washington: World Bank Group, May 
2000. http:/econ.worldbank.org/external/default/main?pagePK=64165259 
&theSitePK=469372&piPK=64165421&menuPK=64166093&entityID=0000949
46_00060205420011. 
Lamb, Robert D. | Bibliography 626 
Comisi?n de Estudios sobre la Violencia. Colombia: Violencia y Democracia: Informe 
Presentado al Ministerio de Gobierno [Colombia: Violence and Democracy: 
Report Presented to the Ministry of Government.] Bogot?: Universidad Nacional 
de Colombia, 1987. 
Commision on Human Security. New York: Commision on Human Security, 1 May 
2003. http:/ww.humansecurity-chs.org. 
C?mo Vamos en Seguridad: Informe de Indicadores Objetivos?Seguridad, 2004-2007. 
[How We?re Doing on Security: Objective Indicators Report?Security, 2004-
2007.] Medel?n, Colombia: Medel?n C?mo Vamos, 2008. 
http:/ww.medelincomovamos.org. 
Comuna 13, la otra Versi?n: Caso Tipo No. 2, Banco de Datos de Violencia Pol?tica. 
[Comuna 13, the Other Version: Case No. 2, Political Violence Data Bank.] 
Noche Niebla: Panorama de Derechos Humanos y Violencia Pol?tica en 
Colombia. Bogot?: Centro de Investigaci?n y Educaci?n Popular?CINEP and 
Justicia y Paz, 2003. http:/ww.nocheyniebla.org/node/46 (acesed 2 April 
2009). 
Concha Sanz, Tom?s Ernesto. ?El Caso de la Milicias o Acuerdo de Santa Elena.? [?The 
Case of the Militias, or the Santa Elena Acords.?] In Experiencias de 
Intervenci?n en el Conflicto Urbano, edited by Ana Daza A., 213-19. Medel?n, 
Colombia: Alcald?a de Medel?n, 2001. 
Correa Ram?rez, John Jaime. ?Memorias de Pilos y Violencias: Bario Castila, Medel?n 
1950-2000.? [?A Memoir of Crooks and Violence: Bario Castila, Medel?n 
1950-2000.?] In Identidades, Localidades y Regiones: Hacia Una Mirada Micro e 
Lamb, Robert D. | Bibliography 627 
Interdiciplinaria, edited by Renzo Ram?rez Baca and ?lvaro Acevedo Tarazona, 
125-51. Medel?n, Colombia: La Careta Editores, 2007. 
Cruickshank, Douglas. ?Death of a Drug Lord: In ?Kiling Pablo,? Mark Bowden Details 
the 16-Month Game of Cat and Mouse That Finaly Took Down Medel?n Cartel 
Founder Pablo Escobar?With the Help of the U.S. Government.? Salon. 
http:/archive.salon.com/news/feature/2001/05/24/bowden/print.html (acesed 13 
April 2009). 
Cruz Rodr?guez, Edwin. ?Los Estudios Sobre el Paramilitarismo en Colombia.? [?Studies 
of Paramilitarism in Colombia.?] An?lisis Pol?tico No. 60 (2007): 117-34. 
Dahl, Robert Alan. Democracy and Its Critics. New Haven: Yale University Pres, 1989. 
Dalton, Russel J. Democratic Challenges, Democratic Choices: The Erosion of Political 
Support in Advanced Industrial Democracies. New York: Oxford University 
Pres, 2004. 
La Siera: Urban Warfare in the Barrios of Medel?n, Colombia. DVD. Documentary 
Film Writen And Directed by Scott Dalton and Margarita Martinez. Medel?n, 
Colombia: Human Rights Watch, 2005. 
Dao, James and Eric Schmit. ?A Nation Chalenged: Strategy: U.S. Ses Batles in 
Lawles Areas After Afghan War.? The New York Times. 8 January 2002. 
Deas, Malcolm. ?Interpretation of the Recent Colombian Violence.? In The 
Legitimization of Violence, edited by David Ernest Apter. Washington Square, 
N.Y: New York University Pres, 1997. 
Lamb, Robert D. | Bibliography 628 
DeShazo, Peter, Tanya Primiani, and Philip McLean. Back From the Brink: Evaluating 
Progres in Colombia, 1999-2007. Center For Strategic And International Studies 
A Report Of The Americas Program. Washington: CSIS, 2007. 
Dhamika, Ven. S. The Edicts of King Asoka: An English Rendering. Bare, Mas.: The 
Buddhist Publication Society and DharmaNet, 1993 [ca. 257 BCE]. 
http:/ww.cs.colostate.edu/~malaiya/ashoka.html (acesed 21 July 2009). 
Diamond, Jared. ?Vengeance Is Ours: What Can Tribal Societies Tel Us About Our 
Ned to Get Even?? The New Yorker (2008). 
DiCanio, Margaret. Encyclopedia of Violence: Frequent, Commonplace, Unexpected. 
New York: iUniverse, 2004. 
Dornbusch, Sanford M. and W. Richard Scott. Evaluation and the Exercise of Authority. 
1st ed ed. San Francisco: Jossey-Bas Publishers, 1975. 
Duncan, Gustavo. Los Se?ores de la Guera: De Paramilitares, Mafiosos y Autodefensas 
en Colombia. [Warlords: On Paramilitaries, Mafiosi, and Self-Defense Forces in 
Colombia.] Bogot?: Planeta, 2006. 
Duque, Luis Fernando. Antecedentes y Evoluci?n del Programa de Convivencia 
Ciudadana de Medel?n. [Antecedents and Evolution of the Citizens Coexistence 
Program of Medel?n.] Medel?n, Colombia: Alcald?a de Medel?n, 2000. 
??, ed. La Violencia en el Valle de Aburr?: Caminos Para la Superaci?n. [Violence in 
the Aburr? Valley: Ways to Overcome It.] 2a ed. Medel?n, Colombia: Previva?
Prevenci?n de la Violencia y Otras Conductas de Riesgo, Universidad de 
Antioquia, and Area Metropolitana del Vale de Aburr?, 2009. 
Lamb, Robert D. | Bibliography 629 
??, ed. La Violencia en el Valle de Aburr?: Su Magnitud y Programa Para Reducirla. 
[Violence in the Aburr? Valley: Its Magnitude and a Program to Reduce It.] 
Medel?n, Colombia: Previva?Prevenci?n de la Violencia y Otras Conductas de 
Riesgo, Universidad de Antioquia, and Area Metropolitana del Vale de Abur?, 
2005. 
??, ed. Pol?tica P?blica Para la Promoci?n de la Convivencia y la Prevenci?n de la 
Violencia en el Valle de Aburr?: 2007-2015. [Public Policy for the Promotion of 
Peaceful Coexistence and the Prevention of Violence in the Aburr? Valley: 2007-
2015.] 2a ed. Medel?n, Colombia: Previva?Prevenci?n de la Violencia y Otras 
Conductas de Riesgo, Universidad de Antioquia, and Area Metropolitana del 
Vale de Aburr?, 2007. 
Easton, David. A Framework for Political Analysis. Englewood Clifs, N.J.: Prentice-
Hal, 1965. 
??. A Systems Analysis of Political Life. New York: Wiley, 1965. 
Eckstein, Harry. Division and Cohesion in Democracy: A Study of Norway. Princeton: 
Princeton University Pres, 1966. 
El Tiempo. ?Fiesta por la Paz en Bario de Medel?n.? 23 March 1998. 
Elster, Jon. Sour Grapes: Studies in the Subversion of Rationality. New York: Cambridge 
University Pres Editions de la Maison des sciences de l?homme, 1983. 
Encuesta de Percepci?n Ciudadana 2008 [Public Perception Survey 2008]. Medel?n, 
Colombia: Medel?n C?mo Vamos, 2008. http:/ww.medelincomovamos.org. 
Epstein, Joshua, John D. Steinbruner, and Miles T. Parker. ?Modeling Civil Violence: An 
Agent-Based Computational Approach.? Working paper. Center on Social and 
Lamb, Robert D. | Bibliography 630 
Economic Dynamics Working Papers 20. Washington: The Brookings Institution, 
January 2001. 
Fals-Borda, Orlando, ?lvaro Guzm?n, and Luis Uma?a. La Violencia en Colombia. 
[Violence in Colombia.] Bogot?: Tercer Mundo Editores, 1962. 
Feldman, Noah. What We Owe Iraq: War and the Ethics of Nation Building. Princeton: 
Princeton University Pres, 2004. 
Fisher, Roger and Wiliam Ury. Geting to Yes: Negotiating Agrement Without Giving 
In. Boston: Houghton Miflin, 1981. 
Franco Agudelo, Sa?l. El Quinto, No Matar: Contextos Explicativos de la Violencia en 
Colombia. [The Fifth, Thou Shalt Not Kil: Explanatory Contexts of Violence in 
Colombia.] 1a ed. Bogot?: IEPRI Universidad Nacional & Tercer Mundo 
Editores, 1999. 
??. ?Comentarios a la Investigaci?n ?Estado del Arte Sobre Violencia Urbana en 
Antioquia?.? [?Comments on the Study ?State of the Art in Urban Violence in 
Antioquia?.?] In Balance de los Estudios Sobre Violencia en Antioquia, edited by 
Pablo Emilio Angarita Ca?as, 185-92. Medel?n, Colombia: Municipio de 
Medel?n y Editorial Universidad de Antioquia, 2001. 
Frank, David John and John W. Meyer. ?The Profusion of Individual Roles and Identities 
in the Postwar Period.? Sociological Theory 20, no. 1 (2002): 86-105. 
Fry, Suzanne E. ?When States Kil Their Own: The Legitimating Rhetoric and 
Institutional Remedies of Authority Crises.? Doctoral Disertation, New York 
University, 2005. 
Lamb, Robert D. | Bibliography 631 
Fuler, Lon L. The Morality of Law. Revised ed. New Haven: Yale University Pres, 
1977. 
Galup Colombia Ltda. Gallup Poll Bimestral: Poll 62. Bogot?: Galup Colombia, 2008. 
Garc?a M?rquez, Gabriel. News of a Kidnapping. Translated by Edith Grossman. New 
York: Penguin Books, 1998. 
Rodrigo D: No Futuro. DVD. Directed By Victor Manuel Gaviria. Medel?n, Colombia: 
Liberty Multimedia, 1989. 
Giley, Bruce. ?Thick Or Thin? An Empirical Intervention.? Critical Review of 
International Social and Political Philosophy 10, no. 1 (2007): 87-98. 
??. ?The Meaning and Measure of State Legitimacy: Results for 72 Countries.? 
European Journal of Political Research 45, no. 3 (2006): 499-525. 
Gilpin, Robert. Global Political Economy: Understanding the International Economic 
Order. Princeton, N.J.: Princeton University Pres, 2001. 
Ginges, Jeremy, Scott Atran, Douglas Medin, and Khalil Shikaki. ?Sacred Bounds on 
Rational Resolution of Violent Political Conflict.? Procedings of the National 
Academy of Sciences 104, no. 18 (2007): 7357-60. 
Giraldo Ram?rez, Jorge. ?Conflicto Armado Urbano y Violencia Homicida: El Caso de 
Medel?n.? [?Urban Armed Conflict and Homicidal Violence: The Case of 
Medel?n.?] Urvio: Revista Latinoamericana de Seguridad Ciudadana No. 5 
(2008): 99-113. 
Gneezy, Uri and Aldo Rustichini. ?A Fine Is a Price.? Journal of Legal Studies 29, no. 1 
(2000). 
Lamb, Robert D. | Bibliography 632 
G?mez Correa, Jaime Arturo, Luz Mar?a Agudelo Su?rez, Tiberio ?lvarez Echeveri, 
Marleny Cardona Acevedo, Alfredo De los R?os De los R?os, H?ctor Iv?n Garc?a 
Garc?a, Carlos Alberto Giraldo Giraldo, Alfonso Gushiken Miyahui, Mar?a 
Victoria L?pez L?pez, Mauricio Alzate Cata?o, and Ela Cecilia Casas Rojas. 
?Estado del Conocimiento Sobre la Violencia Urbana en Antioquia en la D?cada 
de los Noventa.? [?The State of Knowledge about Urban Violence in Antioquia in 
the 1990s.?] In Balance de los Estudios Sobre Violencia en Antioquia, edited by 
Pablo Emilio Angarita Ca?as. Medel?n, Colombia: Municipio de Medel?n y 
Editorial Universidad de Antioquia, 2001. 
Gonz?lez, Fern?n. ?El Fen?meno Pol?tico de ?lvaro Uribe V?lez: ?De D?nde Proviene la 
Legitimidad de Este L?der Elegido por Segunda Vez Como Presidente?? [?The 
Political Phenomenon of ?lvaro Uribe V?lez: Elected President for the Second 
Time, Where Does This Leader?s Legitimacy Come From??] Instituto de 
Investigaci?n y Debate sobre la Gobernanza [Institute for Research and Debate on 
Governance]. Published electronicaly by Equipo de Investigadores del Centro de 
Investigaciones y Educaci?n Popular (CINEP). http:/ww3.institut-
gouvernance.org/es/analyse/fiche-analyse-245.html (acesed 8 February 2009). 
Grafstein, Robert. ?The Legitimacy of Political Institutions.? Polity 14, no. 1 (1981): 
51-69. 
Gren, Leslie. The Authority of the State. Oxford: Clarendon Pres, 1988. 
Gugliotta, Guy and Jef Len. Kings of Cocaine: Inside the Medel?n Cartel, an 
Astonishing True Story of Murder, Money, and International Corruption. New 
York: Simon and Schuster, 1989. 
Lamb, Robert D. | Bibliography 633 
Guilermoprieto, Alma. ?Exit el Patr?n.? The New Yorker 69, no. 35 (1993): 72. 
Gurr, Ted Robert. Why Men Rebel. Princeton, N.J.: Princeton University Pres, 1970. 
Guti?rez San?n, Francisco and Ana Mar?a Jaramilo. ?Crime, (Counter-)Insurgency and 
the Privatization of Security: The Case of Medel?n, Colombia.? Environment and 
Urbanization 16, no. 2 (2004): 17-30. 
Habermas, J?rgen. Legitimation Crisis. Boston: Beacon Pres, 1975. 
Hamond, Alen L. Which World? Scenarios for the 21st Century. Washington: Island 
Pres, 1998. 
Hampton, Jean. Political Philosophy. Boulder, Colo.: Westview Pres, 1997. 
Hanneman, Robert A., Randal Collins, and Gabriele Mordt. ?Discovering Theory 
Dynamics By Computer Simulation: Experiments on State Legitimacy and 
Imperialist Capitalism.? Sociological Methodology 25 (1995): 1-46. 
Hayek, Friedrich A. von. Law, Legislation, and Liberty: A New Statement of the Liberal 
Principles of Justice and Political Economy. London: Routledge and Kegan Paul, 
1982. 
Held, David, Anthony McGrew, David Goldblat, and Jonathan Peraton. Global 
Transformations: Politics, Economics and Culture. Stanford, Calif.: Stanford 
University Pres, 1999. 
Heyman, James and Dan Ariely. ?Efort for Payment: A Tale of Two Markets.? 
Psychological Science 15, no. 11 (2004): 787-93. 
Hobbes, Thomas. Leviathan: Parts One and Two. New York: Liberal Arts Pres, 1958 
[1651]. 
Hoffman, Bruce. Inside Terorism. New York: Columbia University Pres, 1998. 
Lamb, Robert D. | Bibliography 634 
Hooghe, Liesbet and Gary Marks. ?Unraveling the Central State, But How? Types of 
Multi-Level Governance.? American Political Science Review 97, no. 2 (2003): 
233-43. 
Howard, Judith A. ?Social Psychology of Identities.? Annual Review of Sociology 26 
(2000): 367-93. 
Hume, David. ?Of the Original Contract.? In Social Contract: Esays By Locke, Hume, 
and Rousseau, edited by Ernest Barker. New York: Oxford University Pres, 1960 
[1748]. 
Huntington, Samuel P. Political Order in Changing Societies. New Haven: Yale 
University Pres, 1968. 
Huynh-Quan-Suu, Corinne and Service de Traduction de l?Union Europ?enne. 
??tymologie du Terme ?Gouvernance?.? [?Etymology of the Term 
?Governance?.?] Gouvernance dans l?EU. Published electronicaly by 
Commision europ?ene. http:/ec.europa.eu/governance/docs/doc5_fr.pdf 
(acesed 20 April 2009). 
Hylton, Forest. ?Medel?n: The Peace of the Pacifiers.? NACLA Report on the 
Americas. Published electronicaly by North American Congres on Latin 
America (NACLA). http:/nacla.org/node/4459 (acesed 13 April 2009). 
??. ?Medel?n?s Makeover.? New Left Review 44 (2007): 70-89. 
Inglehart, Ronald. Modernization and Postmodernization: Cultural, Economic, and 
Political Change in 43 Societies. Princeton: Princeton University Pres, 1997. 
Lamb, Robert D. | Bibliography 635 
Instituto Popular de Capacitaci?n (IPC). ?Situaci?n de Violencica, Conflicto Urbano y 
Derechos Humanos en Medel?n.? Observatorio de Derechos Humanos y 
Conflicto Armado (2004). 
Jackman, Mary R. ?License to Kil: Violence and Legitimacy in Expropriative Social 
Relations.? In The Psychology of Legitimacy: Emerging Perspectives on Ideology, 
Justice, and Intergroup Relations, edited by John T. Jost and Brenda Major, 437-
67. New York: Cambridge University Pres, 2001. 
Jentleson, Bruce W. ?The Ned for Praxis: Bringing Policy Relevance Back in.? 
International Security 26, no. 4 (2002): 169-83. 
Jost, John T., Diana Burges, and Christina O. Mosso. ?Conflicts of Legitimation Among 
Self, Group, and System: The Integrative Potential of System Justification 
Theory.? In The Psychology of Legitimacy: Emerging Perspectives on Ideology, 
Justice, and Intergroup Relations, edited by John T. Jost and Brenda Major. New 
York: Cambridge University Pres, 2001. 
Jost, John T. and Brenda Major. ?Emerging Perspectives on the Psychology of 
Legitimacy.? In The Psychology of Legitimacy: Emerging Perspectives on 
Ideology, Justice, and Intergroup Relations, edited by John T. Jost and Brenda 
Major, 3-30. New York: Cambridge University Pres, 2001. 
??, eds. The Psychology of Legitimacy: Emerging Perspectives on Ideology, Justice, 
and Intergroup Relations. New York: Cambridge University Pres, 2001. 
Kahl, Colin H. ?Coin of the Realm: Is There a Future for Counterinsurgency?? Foreign 
Afairs 86, no. 6 (2007): 169-76. 
Lamb, Robert D. | Bibliography 636 
Kalyvas, Stathis N. The Logic of Violence in Civil War. New York: Cambridge 
University Pres, 2006. 
??. ?The Ontology of ?Political Violence?: Action and Identity in Civil Wars.? 
Perspectives on Politics 1, no. 3 (2003): 475-94. 
Kane, John. The Politics of Moral Capital. New York: Cambridge University Pres, 
2001. 
Kant, Imanuel. Groundwork of the Metaphysic of Morals. Translated by H. J. Paton. 
New York: Harper & Row, 1967 [1785]. Reprint, 1958. 
Kasfir, Nelson. ?Dilemas of Popular Support in Guerila War: The National Resistance 
Army in Uganda, 1981-86? (paper presented at the Laboratory in Comparative 
Ethnic Proceses (LiCEP) 6, Dartmouth College, University of California-Los 
Angeles, November 2002). 
Kaufmann, Daniel. ?Governance Redux: The Empirical Chalenge.? In Global 
Competitivenes Report 2003-2004. Geneva: World Economic Forum, 2004. 
Kaufmann, Daniel, Aart Kray, and Pablo Zoido-Lobat?n. ?Governance Maters.? 
Working paper. World Bank Policy Research Working Papers 2196. Washington: 
World Bank, October 1999. http:/ww.orldbank.org/wbi/ 
governance/pubs/govmaters.html. 
Kelman, Herbert C. ?Reflections on Social and Psychological Proceses of 
Legitimization and Delegitimization.? In The Psychology of Legitimacy: 
Emerging Perspectives on Ideology, Justice, and Intergroup Relations, edited by 
John T. Jost and Brenda Major, 54-73. New York: Cambridge University Pres, 
2001. 
Lamb, Robert D. | Bibliography 637 
Keohane, Robert O. and Joseph S. Nye. ?Introduction.? In Governance in a Globalizing 
World, edited by Joseph S. Nye and John D. Donahue, 1-41. Washington, DC: 
Brookings Institution Pres, 2000. 
Kilcullen, David. ?New Paradigms for 21st Century Conflict.? eJournal USA: Countering 
the Terorist Mentality. Published electronicaly by U.S. Department of State. 
http:/usinfo.state.gov/journals/itps/0507/ijpe/kilcullen.htm (acesed 18 October 
2008). 
Kooiman, Jan. Modern Governance: New Government-Society Interactions. Newbury 
Park, Calif.: Sage, 1993. 
Kostova, Tatiana and Srilata Zaheer. ?Organizational Legitimacy Under Conditions of 
Complexity: The Case of the Multinational Enterprise.? Academy of Management 
Review 24, no. 1 (1999): 64-81. 
Krasner, Stephen D. ?Sharing Sovereignty: New Institutions for Collapsed and Failing 
States.? International Security 29, no. 2 (2004): 85-120. 
Kraul, Chris. ?Medelin Has Undergone Renaisance: City Has Cleaned Up and 
Homicides Down 90 Percent.? The Boston Globe. Published electronicaly by 
Boston.com. http:/ww.boston.com/news/world/latinamerica/articles/2009/03 
/29/medelin_has_undergone_renaisance (acesed 29 March 2009). 
Kritz, Neil J. ?The Rule of Law in the Postconflict Phase: Building a Stable Peace.? In 
Turbulent Peace: The Challenges of Managing International Conflict, edited by 
Chester A. Crocker, Fen Osler Hampson, and Pamela R. Aal, 801-20. 
Washington: United States Institute of Peace Pres, 2001. 
Lamb, Robert D. | Bibliography 638 
Krug, Etienne G., Linda L. Dahlberg, James A. Mercy, Anthony B. Zwi, and Rafael 
Lozano, eds. World Report on Violence and Health. Geneva: World Health 
Organization (WHO), 2002. 
Kurtz-Phelan, Daniel. ?The Mathematician of Medel?n.? Newsweek. 11 November 2007. 
??. ?Mr. Nice Guy.? New Republic 230, no. 18 (2004): 11-14. 
Lamb, Robert D. ?Measuring Legitimacy in Weak States? (paper presented at the 
Graduate Student Conference on Security, Georgetown University, Washington, 
18 March 2005). 
??. Ungoverned Areas and Threats From Safe Havens. Final Report Of The 
Ungoverned Areas Project. Washington: Under Secretary of Defense for Policy, 
2 January 2008. http:/ww.cism.umd.edu/papers/display.php?id=306. 
Launay, Claire. ?La Gobernanza: Estado, Cuidadan?a Yrenovaci?n de Lo Pol?tico: 
Origen, Definici?n e Implicaciones del Concepto en Colombia.? [?Governance: 
State, citizenship and political renewal: Origin, definition and implications of the 
concept in Colombia.?] Controversia no. 185 (2005): 91-105. 
Leavel, Hugh Rodman and E. Gurney Clark. Preventive Medicine for the Doctor in His 
Community: An Epidemiologic Approach. 2nd ed. New York: Blakiston Division, 
1958. 
Clockers. DVD. Film Directed by Spike Le. 1995. 
Lefort, Ren?, ed. UNESCO Courier?When the State Fails: A Survival Guide. Paris: 
United Nations Educational, Scientific and Cultural Organization, 2001. 
Leftwich, Adrian. ?Governance, Democracy and Development in the Third World.? 
Third World Quarterly 14, no. 3 (1993): 605-24. 
Lamb, Robert D. | Bibliography 639 
Lesig, Lawrence. Code and Other Laws of Cyberspace. New York: Basic Books, 1999. 
Lichtblau, Dale E., Brian A. Haugh, Gregory N. Larsen, and Tery Mayfield. ?Analyzing 
Adversaries as Complex Adaptive Systems.? Monograph. IDA Papers P-3868. 
Alexandria, Va.: Institute for Defense Analysis, October 2006. 
Linz, Juan J. ?Crisis, Breakdown, and Re-Equilibration.? In The Breakdown of 
Democratic Regimes, edited by Juan J. Linz and Alfred C. Stepan. Baltimore: 
John Hopkins University Pres, 1978. 
Lipset, Seymour Martin. Political Man: The Social Bases of Politics. expanded ed. 
Baltimore: Johns Hopkins University Pres, 1981 [1961]. 
??. ?Some Social Requisites of Democracy: Economic Development and Political 
Legitimacy.? American Political Science Review 53, no. 1 (1959): 69-105. 
Locke, John. ?An Esay Concerning the True Original, Extent, and End of Civil 
Government.? In Social Contract: Esays By Locke, Hume, and Rousseau, edited 
by Ernest Barker. New York: Oxford University Pres, 1960 [1690]. 
Londo?o Ber?o, Hernando Le?n, Ricardo Le?n Molina L?pez, and Juan David Posada. 
?Pol?tica Criminal y Violencia Juvenil.? [?Crime Policy and Juvenile Violence.?] 
In Balance de los Estudios Sobre Violencia en Antioquia, edited by Pablo Emilio 
Angarita Ca?as. Medel?n, Colombia: Municipio de Medel?n y Editorial 
Universidad de Antioquia, 2001. 
Luttwak, Edward N. ?Dead End: Counterinsurgency Warfare as Military Malpractice.? 
Harper?s Magazine 314, no. 1881 (2007): 33-42. 
Lynn, John A. ?Paterns of Insurgency and Counterinsurgency.? Military Review 
LXXV, no. 04 (2005): 22-27. 
Lamb, Robert D. | Bibliography 640 
Machiaveli, Nicol?. The Prince. Translated by Luigi Rici and Eric Reginald Pearce 
Vincent. New York: Mentor/Penguin, 1935 [1513]. 
Manrique Hern?ndez, Rub?n Dar?o. ?Caracter?sticas del Suicidio al Final del Siglo Xx en 
Medel?n.? [?Suicide Characteristics at the End of the 20th Century in Medel?n.?] 
Revista CES Medicina 14, no. 1 (2000): 53-59. 
Martin, Gerard, Diego Corrales, Rub?n Fern?ndez, Laura Vila, and Roberto Luis 
Jaramilo. Medel?n: Transformaci?n de Una Ciudad. [Medel?n: Transformation 
of a City]. Medel?n, Colombia: Alcald?a de Medel?n and Banco Interamericano 
de Desarollo, 2009. 
Martin, Joanne, Mauren Scully, and Barbara Levit. ?Injustice and the Legitimation of 
Revolution: Damning the Past, Excusing the Present, and Neglecting the Future.? 
Journal of Personality and Social Psychology 59, no. 2 (1990): 281-90. 
Mart?nez Escalon (AP), Margarita. ?Asesinan a Fundador del Paramilitarismo en 
Colombia.? La Prensa: El Diario de los Nicarag?enses. 31 May 2002. 
Mart?nez Escalon, Margarita. ?As? Filmamos la Siera.? [?This Is How We Filmed La 
Siera.?] El Malpensante. http:/ww.elmalpensante.com/index.php 
?doc=display_contenido&id=671 (acesed 5 August 2009). 
Marx, Karl and Friedrich Engels. The German Ideology. Translated by W. Lough And 
C.P. Magil. Edited by Roy Pascal. London: Lawrence & Wishart, 1938 [1845]. 
Matard-Bonucci, Marie-Anne. Histoire de la Mafia. [A History of the Mafia.] Bruxeles: 
Editions Complexe, 1994. 
Lamb, Robert D. | Bibliography 641 
McIlwaine, Cathy and Caroline O.N. Moser. ?Violence and Social Capital in Urban Poor 
Communities: Perspectives From Colombia and Guatemala.? Journal of 
International Development 13, no. 7 (2001): 965-84. 
Mearsheimer, John J. The Tragedy of Great Power Politics. New York: Norton, 2001. 
Medel?n en Cifras [Medel?n by the Numbers]. Medel?n: Alcald?a de Medel?n, 2005. 
??. Medel?n: Alcald?a de Medel?n, 2006. 
??. Medel?n: Alcald?a de Medel?n, 2007. 
Medina Franco, Gilberto. Una Historia de las Milicias de Medel?n (Historia Sin Fin). [A 
History of the Militias of Medel?n (A Never-Ending Story)]. Medel?n, Colombia: 
Instituto Popular de Capacitaci?n?IPC, 2006. htp:/bibliotecavirtual.clacso.org.ar 
/ar/libros/colombia/ipc/historiamilicias.pdf (acesed 10 April 2009). 
Medina, Luis Fernando. ?A Critique of ?Resource-Based? Theories of Colombia?s Civil 
War.? An?lisis Pol?tico No. 61 (2008): 44-57. 
Melguizo, Ramiro Cebalos and Francine Cronshaw. ?The Evolution of Armed Conflict 
in Medelin: An Analysis of the Major Actors.? Latin American Perspectives 28, 
no. 1 (2001): 110-31. 
Menkhaus, Ken. ?Beneath the Surface of Paper States.? In UNESCO Courier?When the 
State Fails: A Survival Guide. Paris: United Nations Educational, Scientific and 
Cultural Organization, 2001. 
Merquior, Jos? Guilherme. Rousseau and Weber: Two Studies in the Theory of 
Legitimacy. Boston: Routledge & Kegan Paul, 1980. 
Lamb, Robert D. | Bibliography 642 
Ministerio de Defensa Nacional. PSCD: Pol?tica de Consolidaci?n de la Seguridad 
Democr?tica. [Policy for the Consolidation of Democratic Security.] Bogot?: 
Ministerio de Defensa Nacional, Rep?blica de Colombia, 2007. 
?Montevideo Convention on the Rights and Duties of States? (international treaty signed 
at the Seventh International Conference of American States, Montevideo, 
Uruguay, 1933). 
Moore, Barington, Jr. Social Origins of Dictatorship and Democracy: Lord and Peasant 
in the Making of the Modern World. Boston: Beacon Pres, 1966. 
??. Injustice: The Social Bases of Obedience and Revolt. White Plains, N.Y.: M.E. 
Sharpe, 1978. 
Moore, Samuel K. and Alan Gardner. ?Megacities By the Numbers.? IEE Spectrum 44, 
no. 6 (2007): 24-25. 
Moreno Bedoya, Roberto Armando. Conflicto Urbano y Derechos Humanos en Medel?n: 
Balance desde Diferentes Sectores Sociales 2002. [Urban Conflict and Human 
Rights in Medel?n: An Asesment of Diferent Social Sectors, 2002.] 1a ed. Re 
Lecturas, No. 26. Medel?n, Colombia: Instituto Popular de Capacitaci?n (IPC) de 
la Corporaci?n de Promoci?n Popular, 2003. 
Morgenthau, Hans Joachim. Politics Among Nations: The Struggle for Power and Peace. 
Edited by Kenneth W. Thompson. Brief ed. New York: McGraw-Hil, 1993. 
Moser, Caroline O.N. ?Urban Violence and Insecurity: An Introductory Roadmap.? 
Environment and Urbanization 16, no. 2 (2004): 3-16. 
Lamb, Robert D. | Bibliography 643 
Moser, Caroline O.N. and Cathy McIlwaine. ?Latin American Urban Violence as a 
Development Concern: Towards a Framework for Violence Reduction.? World 
Development 34, no. 1 (2005): 89-112. 
Mu?oz Restrepo, C?sar Augusto, ed. Que los Arboles Dejen Ver el Bosque: Derechos 
Humanos en Antioquia?2005. [That the Tres Might Let the Forest See: Human 
Rights in Antioquia?2005.] Medel?n, Colombia: Instituto Popular de 
Capacitaci?n (IPC) de la Corporaci?n de Promoci?n Popular, 2006. 
Nagl, John A. Learning to Eat Soup with a Knife: Counterinsurgency Lesons From 
Malaya and Vietnam. Chicago: University Of Chicago Pres, 2005. 
Na?m, Mois?s. Ilicit: How Smugglers, Traffickers, and Copycats Are Hijacking the 
Global Economy. 1st Anchor Books ed. New York: Anchor Books, 2006. 
Naranjo Giraldo, Gloria. Medel?n en Zonas. Medel?n, Colombia: Corporaci?n Regi?n, 
1992. 
National Gang Inteligence Center. Gang-Related Activity in the U.S. Armed Forces 
Increasing, Inteligence Asesment. 12 January 2007. http:/ww.stripes.com 
/07/feb07/gangs/ncis_gangs.pdf. 
Neira, Armando. ??S? Nacimos Pa? Semila?.? [??Not Born to Die?.?] Semana. 26 January 
2004. 
New York Times. ?Sentimental Philanthropy.? 29 January 1852. 
??. ?Washington ? the Approaching Meting of the Virginia Legislature ? the 
Charges of Corruption Against Kansas Members of Congres ? an Unorganized 
Teritory ? the Georgia Case in Washington ? Mr. Hear and Mr. Sawyer of 
South Carolina.? 7 February 1870. 
Lamb, Robert D. | Bibliography 644 
Nore?a Betancur, Herman Eduardo. ?Los Paramilitares en Medel?n: La 
Desmovilizaci?n del Bloque Cacique Nutibara: Un Estudio de Caso.? 
[?Paramilitaries in Medel?n: The Demobilization of the Cacique Nutibara Bloc: A 
Case Study.?] Master?s thesis, Universidad de Antioquia, 2007. 
Norton, Richard J. ?Feral Cities.? Naval War College Review 56, no. 4 (2003): 97-106. 
Nozick, Robert. Anarchy, State, and Utopia. New York: Basic Books, 1974. 
Nussbaum, Martha. ?Adaptive Preferences and Women?s Options.? Economics and 
Philosophy 17 (2001): 67-88. 
Nye, Joseph S. Soft Power: The Means to Succes in World Politics. 1st ed. New York: 
Public Afairs, 2004. 
Ocampo M?nera, Fernel. Hablemos de Grandes Valores de Tango y Algo M?s. [Let?s 
Talk about Tango?s Greatest Figures and More.] Manizales, Colombia: Editorial 
Manigraf, 2005. 
Ortiz S., Carlos M. ?Los Estudios Sobre la Violencia en las Tres ?ltimas D?cadas.? 
[?Studies of Violence in the Last Thre Decades.?] Bolet?n Socioecon?mico 
CIDSE, No. 2425 (1992). 
Osborne, David and Ted Gaebler. Reinventing Government: How the Entrepreneurial 
Spirit Is Transforming the Public Sector. Reading, Mas.: Addison-Wesley Pub. 
Co., 1992. 
Palacio-Acosta, Carlos, Jenny Garc?a-Valencia, Johanna Diago-Garc?a, Claudia Zapata, 
Juan Ortiz-Tob?n, Gabriel L?pez-Cale, and Mar?a L?pez-Tob?n. ?Characteristics 
of People Commiting Suicide in Medel?n, Colombia.? Revista de Salud P?blica 
7, no. 3 (2005): 243-53. 
Lamb, Robert D. | Bibliography 645 
Palacios, Marco. Betwen Legitimacy and Violence: A History of Colombia, 1875-2002. 
Translated by Richard Stoller. Latin America in Translation/en Traducci?n/Em 
Tradu??o. Durham: Duke University Pres, 2006. 
P?rez, Elizabeth. ?Dos Muertos Que Marcaron Mi Vida.? [?Two Deaths that Marked Me 
for Life.?] In Jam?s Olvidar? Tu Nombre, edited by Patricia Nieto, 20-29. 
Medel?n, Colombia: Alcald?a de Medel?n, 2006. 
Perit, Henry H., Jr. ?Iraq and the Future of United States Foreign Policy: Failures of 
Legitimacy.? Syracuse Journal of International Law and Commerce 31, no. 2 
(2004). 
Peters, B. Guy. The Future of Governing. 2nd ed. Lawrence: University Pres of Kansas, 
2001. 
Plato. The Republic of Plato. Translated by Francis Macdonald Cornford. New York: 
Oxford University Pres, 1945 [370 BCE]. 
Porch, Douglas and Mar?a Jos? Rasmussen. ?Demobilization of Paramilitaries in 
Colombia: Transformation Or Transition?? Studies in Conflict & Terorism 31, 
no. 6 (2008): 520-40. 
Posada Carb?, Eduardo. ?Ilegitimidad? del Estado en Colombia: Sobre los Abusos de UN 
Concepto. [?Ilegitimacy? of the State in Colombia: On the Abuses of a Concept.] 
1st ed. Libros de Cambio. Bogot?: Alfaomega Colombiana: Ideas para la paz, 
2003. 
Putnam, Robert D. ?Bowling Alone: America?s Declining Social Capital.? Journal of 
Democracy 6, no. 1 (1995): 65-78. 
Lamb, Robert D. | Bibliography 646 
Rabasa, Angel, Steven Boraz, Peter Chalk, Kim Cragin, Theodore W. Karasik, Jennifer 
D.P. Moroney, Kevin A. O?Brien, and John E. Peters. Ungoverned Teritories: 
Understanding and Reducing Terorism Risks. Arlington, Va.: RAND Project Air 
Force, 2007. http:/ww.rand.org/pubs/monographs/MG561. 
Race, Jefrey. War Comes to Long an: Revolutionary Conflict in a Vietnamese Province. 
Berkeley: University of California Pres, 1972. 
Rapley, John. ?The New Middle Ages.? Foreign Afairs 85, no. 3 (2006): 95-103. 
Rawls, John. A Theory of Justice. revised ed. Cambridge, Mas.: Belknap Pres of 
Harvard University Pres, 1999. 
Rend?n Rend?n, Yoni Alexander. Comuna 13 de Medel?n: El Drama del Conflicto 
Armado. [Comuna 13 in Medel?n: The Drama of the Armed Conflict]. Medel?n, 
Colombia: Hombre Nuevo Editores, 2007. 
Restrepo E., Juan Diego. ??Por Qu? Crece la Violencia en Medel?n?? Semana. 11 
August 2008. 
??. ?Desenterar la Verdad de las Laderas de Medel?n.? [?Unearthing the Truth on the 
Outskirts of Medel?n.?] Semana. 21 November 2008. 
??. ?Con Palizas Se Impone el Control Social en Comunas de Medel?n.? [?Social 
Control Imposed with Beatings in Medel?n?s Comunas.?] Published 
electronicaly by Instituto Popular de Capacitaci?n (IPC). htp:/ww.ipc.org.co 
/page/index.php?option=com_content&task=view&id=800&Itemid=368 
(acesed 28 August 2009). 
Reyes, Leady Jhoana. ?Mi Diario.? [?My Diary.?] In Jam?s Olvidar? Tu Nombre, edited 
by Patricia Nieto, 52-59. Medel?n, Colombia: Alcald?a de Medel?n, 2006. 
Lamb, Robert D. | Bibliography 647 
Rhodes, R.A.W. ?The New Governance: Governing Without Government.? Political 
Studies XLIV (1996): 652-67. 
Ria?o-Alcal?, Pilar. Dwelers of Memory: Youth and Violence in Medel?n, Colombia. 
New Brunswick, N.J: Transaction Publishers, 2006. 
Rinc?n P., Rafael, ed. Antioquia, Fin de Milenio: ?Terminar? la Crisis del Derecho 
Humanitario? [Antioquia at the End of the Milennium: Wil the Humanitarian 
Crisis End?] Medel?n, Colombia: Instituto Popular de Capacitaci?n (IPC) de la 
Corporaci?n de Promoci?n Popular, 1999. 
Robb, John. Brave New War: The Next Stage of Terorism and the End of Globalization. 
Hoboken: John Wiley & Sons, 2007. 
Romay, H?ctor. Dicionario de Lunfardo y Terminolog?a Popular. [Dictionary of 
Lunfardo and Folk Terminology]. Buenos Aires: Bureau Editor, 2004. 
Romero, Mauricio, ed. Parapol?tica: La Ruta de la Expansi?n Paramilitar y los Acuerdos 
Pol?ticos. [Parapolitics: The Course of Paramilitary Expansion and Political 
Acords.] Bogot?: Intermedio/Corporaci?n Nuevo Arco Iris, 2007. 
Romero, Simon. ?Mayor of Medel?n Brings Architecture to the People.? The New York 
Times. 15 July 2007. 
Rose, Charlie. ?A Conversation with Sergio Fajardo.? television interview (20 February 
2009), The Charlie Rose Show. http:/ww.charlierose.com/view/interview 
/10098 (acesed 26 August 2009). 
Rosenau, James N. and Ernst Oto Czempiel, eds. Governance Without Government: 
Order and Change in World Politics. New York: Cambridge University Pres, 
1992. 
Lamb, Robert D. | Bibliography 648 
Rotberg, Robert I. ?The New Nature of Nation-State Failure.? Washington Quarterly 25, 
no. 3 (2002): 85-96. 
??. ?The Failure and Collapse of Nation-States: Breakdown, Prevention, and Repair.? 
In When States Fail: Causes and Consequences, edited by Robert I. Rotberg. 
Princeton, N.J.: Princeton University Pres, 2004. 
??, ed. When States Fail: Causes and Consequences. Princeton, N.J.: Princeton 
University Pres, 2004. 
Rotberg, Robert I. and Deborah L. West. The Good Governance Problem: Doing 
Something About it, WPF Reports 39. Cambridge, Mas.: World Peace 
Foundation, 2004. http:/ww.orldpeacefoundation.org 
/WPF39.GovernanceRanking.pdf (acesed 15 September 2009). 
Rousseau, Jean-Jacques. ?The Social Contract.? In Social Contract: Esays By Locke, 
Hume, and Rousseau, edited by Ernest Barker, translated by Gerard Hopkins. 
New York: Oxford University Pres, 1960 [1762]. 
Rozema, Ralph. ?Urban DR-Proceses: Paramilitaries and Criminal Networks in 
Medel?n, Colombia.? Journal of Latin American Studies 40, no. 3 (2008): 
423?52. 
Rubio, Mauricio. ?Ilegal Armed Groups and Local Politics in Colombia.? Journal of 
Drug Isues 35, no. 1 (2005): 107-30. 
Salazar J., Alonso. No Nacimos Pa? Semila: La Cultura de las Bandas Juveniles de 
Medel?n. [Born to Die: The Culture of Youth Gangs in Medel?n.] 5? ed. Bogot?: 
Centro de Investigaci?n y Educaci?n Popular?CINEP, 1991. 
Lamb, Robert D. | Bibliography 649 
Salomon, Frank and Stuart B. Schwartz. South America. Vol. 3, The Cambridge History 
of the Native Peoples of the Americas. Cambridge: Cambridge University Pres, 
1999. 
Santamar?a Delgado, Carolina. ?Bambuco, Tango and Bolero: Music, Identity, and Clas 
Struggles in Medel?n, Colombia, 1930-1953.? PhD disertation, University of 
Pitsburgh, 2006. 
Scheling, Thomas C. Arms and Influence. New Haven: Yale University Pres, 1966. 
Schwartz, Bary and Sharpe, Kenneth. ?The Wrong Cure: Beter Incentives, for Doctors 
Or Insurers, Won?T Lead to Beter Health Care.? Slate. http:/ww.slate.com 
/id/2224193/ (acesed 4 August 2009). 
Semana. ?La Maldici?n de Ca?n: Qu? Llev? a Vicente Casta?o a Mandar Matar a Su 
Hermano Carlos?? [?The Mark of Cain: What Led Vicente Casta?o to Order the 
Kiling of His Brother Carlos??] 26 Agosto 2006. 
Semana. ?La Cacer?a de ?Doblecero?.? [?The Hunt for ?Double Zero?.?] 29 September 
2003. 
Sen, Amartya. ?Rights and Capabilities.? In Resources, Values, and Development, edited 
by Amartya Sen, 307-24. Cambridge, Mas: Harvard University Pres, 1984. 
??. ?Why Democratization Is Not the Same as Westernization: Democracy and Its 
Global Roots.? The New Republic 229, no. 14 (2003): 28-36. 
Shapiro, Ian. The Rule of Law. New York: New York University Pres, 1994. 
Sison, Alejo G. The Moral Capital of Leaders: Why Virtue Matters. Northampton, Mas.: 
Edward Elgar Publishing, 2003. 
Lamb, Robert D. | Bibliography 650 
Skocpol, Theda and Morris Fiorina, eds. Civic Engagement in American Democracy. 
Washington: Brookings Institution Pres, 1999. 
Slaughter, Anne-Marie. A New World Order. Princeton: Princeton University Pres, 
2004. 
Snyder, Jack. ?One World, Rival Theories.? Foreign Policy (2004): 52-62. 
Sol?rzano, Alfonso Monsalve. Legitimidad y Soberan?a en Colombia 1958-2003. 
[Legitimacy and Sovereignty in Colombia 1958-2003]. 1a ed. Serie Estudios 
Pol?ticos. Medel?n, Colombia: Editorial Universidad Pontificia Bolivariana, 
2004. 
Soltan, Karol. ?Legitimacy and Power.? Working paper. College Park, Md.: University of 
Maryland, April 2005. 
Spencer, Herbert. ?Behavior and Ceremony.? The New York Times. 27 January 1878. 
Steinbruner, John D. Principles of Global Security. Washington: Brookings Institution 
Pres, 2000. 
??. The Cybernetic Theory of Decision: New Dimensions of Political Analysis. 
Princeton, N.J.: Princeton University Pres, 2002. 
Steinbruner, John D. and Nancy Galagher. ?Constructive Transformation: An 
Alternative Vision of Global Security.? Daedalus 133, no. 3 (2004): 83-103. 
Stets, Jan E. and Peter J. Burke. ?Identity Theory and Social Identity Theory.? Social 
Psychology Quarterly 63, no. 3 (2000): 224-37. 
Stone, Dan N., Edward L. Deci, and Richard M. Ryan. ?Beyond Talk: Creating 
Autonomous Motivation Through Self-Determination Theory.? Journal of 
General Management 34 (2009): 75-91. 
Lamb, Robert D. | Bibliography 651 
Sullivan, John P. and Elkus, Adam. ?Plazas for Profit: Mexico?s Criminal Insurgency.? 
Smal Wars Journal. Published electronicaly by Smal Wars Foundation. 
http:/smalwarsjournal.com/blog/2009/04/plazas-for-profit-mexicos-crim/ 
(acesed 10 May 2009). 
Tajfel, Henri, Colin Fraser, and Joseph Maria Franciscus Jaspars, eds. The Social 
Dimension: European Developments in Social Psychology. New York: 
Cambridge University Pres Editions de la Maison des Sciences de L'Homme, 
1984. 
Thompson, E. Maunde. ?Review.? The English Historical Review 1(3), no. 3 (1886): 
568-74. 
Thornton, Rod. ?Historical Origins of the British Army?s Counterinsurgency and 
Counterterorist Techniques.? In Combating Terorism and Its Implications for 
the Security Sector, edited by Amb. Dr. Theodor H. Winkler, Anja H. Ebn?ther, 
and Mats B. Hansson, 26-44. Stockholm: Swedish National Defense College, 
2005. 
Thoumi, Francisco E. ?Introduction.? Journal of Drug Isues 35, no. 1 (2005): 1-5. 
??. ?The Colombian Competitive Advantage in Ilegal Drugs: The Role of Policies and 
Institutional Changes.? Journal of Drug Isues 35, no. 1 (2005): 7-26. 
Thucydides. The History of the Peloponnesian War. Translated by Richard Crawley. 
Edited by Robert Maynard Hutchins and Mortimer J. Adler. Vol. 6, Great Books 
of the Western World. Chicago: Encyclop?dia Britannica, 1952 [431 BCE]. 
Lamb, Robert D. | Bibliography 652 
Tily, Charles. ?War Making and State Making as Organized Crime.? In Bringing the 
State Back in, edited by Peter B. Evans, Dietrich Rueschemeyer, and Theda 
Skocpol, 169-87. New York: Cambridge University Pres, 1985. 
Tomes, Robert R. ?Relearning Counterinsurgency Warfare.? Parameters 34, no. 1 
(2004): 16-28. 
Tricker, R. Ian. International Corporate Governance. Englewood Clifs, N.J.: Prentice 
Hal, 1984. 
Twain, Mark. The Adventures of Tom Sawyer. New York: Harper & Brothers Publishers, 
1920 [1875]. 
Tyler, Tom R. ?A Psychological Perspective on the Legitimacy of Institutions and 
Authorities.? In The Psychology of Legitimacy: Emerging Perspectives on 
Ideology, Justice, and Intergroup Relations, edited by John T. Jost and Brenda 
Major, 416-36. New York: Cambridge University Pres, 2001. 
U.S. Army/Marine Corps. The U.S. Army/Marine Corps Counterinsurgency Field 
Manual. Chicago: University of Chicago Pres, 2007. 
United Kingdom Department for International Development. Why We Ned to Work 
More Efectively in Fragile States. London: 2005. 
United Nations Development Programe. Human Development Report 2000: Human 
Rights and Human Development. New York: Oxford University Pres, 2000. 
United States Agency for International Development (USAID). Fragile States Strategy. 
Washington: 2005. http:/ww.usaid.gov/policy/2005_fragile_states_strategy.pdf. 
Lamb, Robert D. | Bibliography 653 
United States Department of Defense. Iregular Warfare (IW), Department of Defense 
Directive 3000.07. Washington: 2008. http:/ww.dtic.mil/whs/ 
directives/corres/pdf/300007p.pdf. 
United States Department of Defense. Quadrennial Defense Review Report. Washington: 
2006. 
United States Department of Defense. Building Partnership Capacity: QDR Execution 
Roadmap. Washington: 2006. 
United States Department of the Army. Stability Operations, Field Manual No. 3-07. 
Washington: Headquarters, Department of the Army, 2008. http:/ww.fas.org/irp 
/doddir/army/fm3-07.pdf. 
United States Department of the Army. Counterinsurgency, Field Manual No. 3-24, 
Marine Corps Warfighting Publication No. 3-33.5. Washington: Department of 
the Army and Marine Corps Combat Development Command, 2006. 
United States Marines Corps. Small Wars Manual, NAVMC 2890 (1987 reprint). 
Washington: Headquarters, United States Marine Corps, 1940. 
United States Special Operations Command and United States Marines Corps. Iregular 
Warfare (IW) Joint Operating Concept (JOC), Version 1.0. Washington: United 
States Department of Defense, 2007. http:/ww.fas.org/irp/doddir/dod/iw-
joc.pdf. 
Uprimny Yepes, Rodrigo and Mar?a Paula Safon San?n. ?La Ley de ?Justicia y Paz?: 
?Una Garant?a de Justicia y Paz y de No Repetici?n de las Atrocidades?? [?The 
?Justice and Peace? Law: A Guarante of Justice and Peace and No Recurrence of 
Atrocities??] Revista Foro No. 55 (2005): 49-62. 
Lamb, Robert D. | Bibliography 654 
Uribe, Amanda. ?Tres Sucesos Amargos.? [?Thre Biter Events.?] In Jam?s Olvidar? Tu 
Nombre, edited by Patricia Nieto, 31-43. Medel?n, Colombia: Alcald?a de 
Medel?n, 2006. 
Valacher, Robin R. and Andrzej Nowak. ?The Emergence of Dynamical Social 
Psychology.? Psychological Inquiry 8, no. 2 (1997): 73-99. 
Van Evera, Stephen. Guide to Methods for Students of Political Science. Ithaca, NY: 
Cornel University Pres, 1997. 
Vargas, Ricardo. ?State, Esprit Mafioso, and Armed Conflict in Colombia.? In Politics in 
the Andes: Identity, Conflict, Reform, edited by Jo-Marie Burt and Philip Mauceri, 
107-25. Pitsburgh, Penn.: University of Pitsburgh Pres, 2004. 
V?lez Rivera, Ramiro Alberto, Gladis Stela L?pez Balesteros, Rommel Cabrera Sosa, 
Alirio Enrique Rodr?guez Vega, and Luis Gabriel Montero Rodr?guez. 
Governabilidad Local en Medel?n: Configuraci?n de Teritorialidades, 
Conflictos y Ciudad. [Local Governance in Medel?n: The Configuration of 
Teritorialities, Conflicts, and the City.] Medel?n, Colombia: Escuela Superior de 
Administraci?n P?blica?ESAP, 2004. 
VerdadAbierta.com. ?Las Auc.? Paramilitares y Conflicto Armado en Colombia 
[Paramilitaries and Armed Conflict in Colombia.] Published electronicaly by 
Fundaci?n Ideas para la Paz (FIP) and Semana. http:/ww.verdadabierta.com 
/web3/victimarios (acesed 31 July 2009). 
??. ??Rambo,? Fidel Casta?o Gil.? Paramilitares y Conflicto Armado en Colombia 
[Paramilitaries and Armed Conflict in Colombia]. Published electronicaly by 
Fundaci?n Ideas para la Paz (FIP) and Semana. 
Lamb, Robert D. | Bibliography 655 
http:/ww.verdadabierta.com/web3/victimarios/los-jefes (acesed 30 July 
2009). 
??. ??Don Berna,? Diego Fernando Murilo Bejarano.? Paramilitares y Conflicto 
Armado en Colombia [Paramilitaries and Armed Conflict in Colombia]. 
Published electronicaly by Fundaci?n Ideas para la Paz (FIP) and Semana. 
http:/ww.verdadabierta.com/web3/victimarios/los-jefes (acesed 13 April 
2009). 
??. ?El Nacimiento de las Autodefensas Unidas de Colombia (1997-2002).? [?The 
Expansion: Birth of the United Self-Defense Forces of Colombia (1997-2992).?] 
Paramilitares y Conflicto Armado en Colombia [Paramilitaries and Armed 
Conflict in Colombia]. Published electronicaly by Fundaci?n Ideas para la Paz 
(FIP) and Semana. http:/ww.verdadabierta.com/web3/la-historia (acesed 
31 July 2009). 
Vila Mart?nez, Marta In?s, Luz Amparo S?nchez Medina, and Ana Mar?a Jaramilo 
Arbel?ez. Rostros del Miedo: Una Investigaci?n Sobre los Miedos Sociales 
Urbanos. [Faces of Fear: A Study of Urban Social Fears]. Medel?n, Colombia: 
Corporaci?n Regi?n, 2003. 
Walt, Stephen M. ?International Relations: One World, Many Theories.? Foreign Policy 
(1998): 29-45. 
Walzer, Michael. Thick and Thin: Moral Argument At Home and Abroad. Notre Dame: 
University of Notre Dame Pres, 1994. 
Lamb, Robert D. | Bibliography 656 
Weatherford, M. Stephen. ?Mapping the Ties That Bind: Legitimacy, Representation, and 
Alienation.? Western Political Quarterly Western Political Quarterly 44 (1991): 
251-76. 
??. ?Measuring Political Legitimacy.? American Political Science Review 86, no. 1 
(1992): 149-66. 
Weber, Max. Economy and Society: An Outline of Interpretive Sociology. Edited by 
Guenther Roth and Claus Witich. New York: Bedminster Pres, 1968 [1914]. 
White, Lynn T. Legitimacy: Ambiguities of Political Succes Or Failure in East and 
Southeast Asia. Vol. 1, Series on Contemporary China. Hackensack, N.J.: World 
Scientific, 2005. 
Wiliams, Robin M., Jr. ?Legitimate and Ilegitimate Uses of Violence: A Review of 
Ideas and Evidence.? In Violence and the Politics of Research, edited by Wilard 
Gaylin, Ruth Macklin, and Tabitha M. Powledge, 23-45. New York: Plenum 
Pres, 1981. 
Wiliamson, Oliver. ?Corporate Governance.? The Yale Law Journal 93, no. 7 (1984): 
1197-230. 
World Bank. Sub-Saharan Africa: From Crisis to Sustainable Growth. Washington: The 
World Bank, 1989. 
??. Governance and Development. Washington: The World Bank, 1992. 
??. ?Consolidated Matrix of Both Internal and External Governance Data Sets and 
Instruments.? Excel spreadsheet, Published electronicaly by The World Bank 
Group. ww.orldbank.org/wbi/governance (acesed 20 April 2009). 
Lamb, Robert D. | Bibliography 657 
Yin, Robert K. Case Study Research: Design and Methods. revised ed. Newbury Park, 
Calif.: Sage Publications, 1989. 
Zelditch, Morris, Jr. ?Theories of Legitimacy.? In The Psychology of Legitimacy: 
Emerging Perspectives on Ideology, Justice, and Intergroup Relations, edited by 
John T. Jost and Brenda Major, 33-53. New York: Cambridge University Pres, 
2001. 
Zelditch, Morris, Jr. and Henry A. Walker. ?Legitimacy and the Stability of Authority.? 
In Advances in Group Proceses, edited by Edward J. Lawler, 1-25. Grenwich, 
Conn.: JAI Pres, 1984. 
Zorrila, Jos?. Don Juan Tenorio. Project Gutenberg. 2004 [1844]. 
http:/ww.gutenberg.org/ebooks/5201 (acesed 16 September 2009).