Blood and Fire

A book in the series

Latin America Otherwise: Languages, Empires, Nations

Series editors: Walter D. Mignolo, Duke University

Irene Silverblatt, Duke University

Sonia Saldívar-Hull, University of California at Los Angeles

Blood and Fire

La Violencia in Antioquia, Colombia, –

        

Duke University Press Durham and London 

©  Duke University Press
All rights reserved
Printed in the United States of
America on acid-free paper
Designed by Rebecca Giménez
Typeset in Adobe Minion by
Tseng Information Systems, Inc.
Library of Congress Cataloging-
in-Publication Data appear on
the last printed page of this book.
Publication of this book was made
possible by a subvention granted
by the Hull Memorial Publication
Fund of Cornell University.

Frontispiece: Urrao, August .

To heal the wounds of three years
of partisan strife, the parish priest
of Urrao organized a collective
Catholic burial of Liberal and
Conservative casualties of la
Violencia. The priest kept careful
notes in the parish registry of
deaths that were the direct result
of la Violencia and after the
military coup of June , ,
instructed his parishoners to
collect the remains of their dead
relatives scattered outside the town
limits. The coffins are child-sized.
About the Series

History and immigration are changing the principles and assumptions

of Area Studies programs that were set up during the Cold War. Mary
Roldán’s superb study of hegemony and violence in Colombia is not just
another study in which Latin America is the object observed from the
United States. When Roldán explicitly states in the epilogue that ‘‘dur-
ing two long stretches’’ of her investigation, between  and ,
‘‘I lived in my parents’ apartment in downtown Medellín in the heart
of Medellín’s commercial district,’’ she brings to the foreground the
phenomenological and existential dimension of her study. While Area
Studies project the ‘‘objective’’ and ‘‘disembodied’’ perspectives of the so-
cial sciences, Roldán’s investigation builds on the existential and phe-
nomenological while using the scholarly tools of the social sciences. By
enriching her analysis with her personal and emotional investment in
the issues being explored, Roldán works to correct the shortcomings of
Area Studies, particularly those that detach the researcher from the local
history of his or her investigation.
Blood and Fire is an outstanding historical description and interpre-
tation of a fundamental period in the history of Colombia and of Latin
America (–). It is also a theoretical contribution to the under-
standing of the State beyond existing theories, mainly based on paradig-
matic examples of the European State. State building in Latin America
was simultaneous with state building in Europe during the nineteenth
century, but while in Europe many states were imperial, in Latin America
all the states were neo- or postcolonial. In Latin American state build-
ing, the notion of ‘‘internal colonialism’’ is essential, and Roldán makes
good use of it. By so doing, she also inscribes her work in a Latin Ameri-
can tradition of critical social thought that goes back to the late s.
In this regard Roldán introduces a second significant change in relation
to Area Studies. She builds upon the theoretical legacies of critical social
thought to show that Latin America is not only a place for the cultivation
of violence, but a place where critical thought can flourish.
About the Photographs

The inclusion of recent photographs of displacement and violence in a

book about la Violencia—a phenomenon that took place some fifty years
earlier—may seem like a peculiar choice to many readers and so requires
some explanation on the part of the author. When my editor, Valerie
Millholland, first approached me about providing photographs to ac-
company this text, I demurred. Most of the existing images of the period
were ones used to fan partisan hatred by one group against another and
were almost without exception lurid representations that exploited the
victims and titillated the viewer but contributed little to a deeper under-
standing of the complexity and human sorrow of violence. On a research
trip to Medellín in June of , as this book was about to enter into
production, I happened upon an exhibit of works by Jesús Abad Colo-
rado in the recently renovated Museo de Antioquia. I was so moved by
his photographs of the current conflict in Antioquia and by the fact that
nearly all of them were taken of displacements and violence occurring
in the very same towns most affected by violence during the period I
study in this book, that I resolved then and there to approach the pho-
tographer about the possibility of using some of his photographs to ac-
company this text. Little did I know that in addition to being an ex-
tremely gifted visual storyteller, Jésus Abad Colorado wrote narratives
to accompany his photographs that in their basic outlines mirrored al-
most exactly the stories I recount here. It is the hope of both the pho-
tographer and myself that the conscious association of these images of
recent violence in Antioquia with a written narrative of events taking
place half a century earlier will invite readers to draw connections be-
tween past and present violence. Perhaps the anguish of recurrence these
images bring to mind may lead to a greater understanding of the his-
torical roots of conflict in Colombia. That is certainly our wish and
motivation.—Mary Roldán

Jesús Abad Colorado received his journalism degree from the University
of Antioquia in Medellín, Colombia. Between  and May of  he
worked as a photojournalist for the regional daily newspaper, El Colom-
biano. His work has appeared regularly in national magazines and social
research books. He coauthored the book Relatos e Imágenes, El desplaza-
miento Forzado en Colombia, and his photographs have been exhibited
both in Colombia and abroad.
Contents

Acknowledgments xi

Introduction 

. Medellín and Core Municipalities 

. Bajo Cauca, Magdalena Medio, and the Northeast 

. Urabá and Western Antioquia 

. Urrao and the Southwest 

Epilogue 

Appendix A: Tables 

Appendix B: Maps 

Notes 

Bibliography 

Index 
Acknowledgments

Over the years, many individuals and institutions have contributed to

and supported the research and writing of this manuscript. In Colom-
bia I would like to thank the employees at the Archivo de la Goberna-
ción de Antioquia for allowing me to consult the governor’s correspon-
dence and other regional government materials in  and ; the
Centro Jorge Eliécer Gaitán in Bogotá and its Director Gloria Gaitán; the
Fundación Antioqueña de Estudios Sociales in Medellín; the library and
newspaper collection at the Universidad de Antioquia, Medellín; the re-
search collection at the Instituto de Estudios Regionales at the Universi-
dad de Antioquia; the Salón Antioquia in the Bibliotéca Pública Piloto
in Medellín; the Bibliotéca Luis Angel Arango in Bogotá; the Bibliotéca
Nacional in Bogotá; the Archivo de la Alcaldía Municipal de Urrao and
that town’s Parish Registry and Casa de la Cultura; and Froilan Montoya
Mazo’s personal archive in Medellín. Dr. Montoya Mazo, who has since
died, also very kindly introduced me to and obtained interviews for me
with several gaitanista leaders of the Violencia period and with several
Liberal ex-guerrillas. Colombian colleagues and friends too numerous to
name have also provided hospitality, affection, and intellectual guidance,
among them: Jorge Pérez, Maria Mercedes Botero, Alvaro Tirado Mejía,
Jorge Orlando Melo, Victor Alvarez, Beatriz Patiño, Patricia Londoño,
Constanza Toro, Ana Lucía Sánchez, Jesús María Alvarez, Maria Teresa
Uribe de Hincapie, Gonzalo Sánchez, and Mauricio Romero. I could not
have completed the research for this book without the assistance of sev-
eral students from the Universidad de Antioquia, among them: Gloria
Granda, Rodrigo Arango, and Mario Gaviria. Many thanks to Gustavo
Ochoa for his excellent map-making skills and to Fernando Mejía for
many hours of tedious data entry.
Several colleagues and institutions in the United States also supported
my work since its initial emergence as a Ph.D. dissertation at Harvard
University. The Tinker Foundation provided summer research support
during three summers as did the Radcliffe President’s Fund, the Com-
mittee on Iberian and Latin American Studies, the history department,
xii Acknowledgments

and the Sheldon Kennedy Traveling Research Fund at Harvard Univer-

sity. Financial support was also provided by a Fulbright-Hays Doctoral
Dissertation Grant. I could not have asked for a more intellectually de-
manding adviser or mentor than John Womack Jr. He was never satisfied
with easy explanations, always prodded me to push my research and fine-
tune my interpretations further, and, though I sometimes proved stub-
born or resistant to his good advice, I recognize that he shaped the way I
think about history in profound and intangible ways. As an assistant pro-
fessor at Cornell University I received support from the Society for the
Humanities Summer Research Fund, the John T. and Catherine D. Mac-
Arthur Foundation Peace Studies Research Fund, and the Latin American
Studies Summer Faculty Research Fund.
My colleagues at Cornell have been unfailingly supportive. For good
meals, helpful readings, critical thoughts, and kind words, I wish to
thank: Tom Holloway, Walter LaFeber, Sandra Greene, Rachel Weil, Itsie
Hull, Shirley Samuels, Billie Jean Isbell, Tom Volman, Lourdes Benería,
Bill Goldsmith, Barbara Lynch, Debbie Castillo, and Mary Jo Dudley. I
want to single out Tom Holloway for help above and beyond the call of
duty. Tom read my dissertation when I arrived at Cornell, told me what
he thought was good in it and what had to go, then looked me in the eye
and asked, ‘‘Where’s the blood?’’ It took me two years to figure out how
to face the ‘‘blood’’ of Colombian violence and then some more years to
get to the middle of it, but I am very grateful to Tom for forcing me to
face my inner demons. When Tom read the completely revised manu-
script in its entirety and emitted a gruff, ‘‘great stuff,’’ I went off and cried
from sheer relief. I particularly want to thank Catherine LeGrand who
has been the kindest of friends and the most generous of colleagues. She
has given me unfailing intellectual encouragement, provided gently criti-
cal but probing comments of my work, and, whenever I lost faith, applied
cleverly flattering remarks so I would press on. I also wish to thank Lisa
Dundon, Richard Stoller, Jim Brennan, and Jeff Rubin. Michael Jiménez
died before this book went to press. I wish to acknowledge here what a
special being he was and the great honor it was to have been his friend.
Several graduate students have provided a fertile environment for the dis-
cussion of thorny theoretical issues and comparative problems over the
years, among them are Estelle Tarica, Brett Troyan, Leslie Horowitz, Anne
Brophy, Angela Wilson, and Michelle Bigenho.
The writing of this manuscript was made possible by the generosity
 Acknowledgments xiii

of  and their program ‘‘Movilidad de Investigadores,’’ the

Centro de Estudios Regionales Cafeteros y Empresariales () in
Manizales and its Director, Dr. Cesar Vallejo Mejía, and Planeación
Nacional in Bogotá. These three institutions provided the financial sup-
port and time away from teaching necessary to the completion of this
book. Thank you. Valerie Millholland was a patient, encouraging, and
wonderful editor. This book might never have reached the publication
stage without her prodding, and the enormously helpful criticisms of
Duke University Press’s anonymous readers. Whatever errors and omis-
sions remain are solely my responsibility.
Finally, I wish to thank my dear friend, Margarita Crocker, for un-
flagging moral support over many years and my husband, Christopher
London, who is my best intellectual partner and the person who always
sees the point of what I do even when I lose faith. Bearing and raising
children has given me a new perspective on violence and work. I hope
Lucas and Sophia will forgive their mother for remaining glued to a com-
puter for nearly two years and not infrequently declining to go outside
and play.
Blood and Fire
Peque, July 2001. When
a guerrilla commander
told her to flee in the
wake of a paramilitary
attack, the elderly woman
in this portrait refused,
commenting, ‘‘I’ve been
running since 1950.’’
Introduction

For many people violence and Colombia are synonymous. Colombia

(map ), after all, produces the bulk of the coca processed into cocaine
and shipped to the world’s largest consumer of drugs, the United States,
and suffers the crime and corruption that result from this illicit trade.
Colombia is also home to the oldest guerrilla insurgency in the Western
Hemisphere; the country that accounted for half of the world’s kidnap-
pings in ; the place where paramilitaries inscribe bloody messages
on the bodies of their largely peasant victims; a land the U.S. media likes
to refer to as ‘‘twice the size of France’’; a land over which the central state
exerts little authority; and a formal democracy where a handful of elite
families are thought to monopolize control of the media, politics, and
the nation’s (licit) economy. Until recently, the Colombian city consid-
ered to represent the apex of lawlessness was Medellín, the capital of the
northwestern province of Antioquia and, for the better part of two de-
cades, the financial center of a global narcotics enterprise known as the
‘‘Medellín cartel.’’ 1
This book is not directly about narcotics or Colombia’s contemporary
crisis. Instead, it examines the experience of the department of Antioquia
(see map ) during the first seven years (–) of a civil war that was
spurred by a struggle for power between members of the Conservative
and Liberal parties and that has come to be known simply as la Violencia
or ‘‘the Violence.’’ 2 Initially, I did not intend to draw parallels between
the period of la Violencia and contemporary Colombia, but I came to
see that recent and past periods of violence are inextricably intertwined.
I can pinpoint the day I ceased to regard la Violencia as something en-
tirely distinct from current, daily, lived Colombian reality. I was sitting
in my office preparing the last lecture of the spring semester for my sur-
vey course on modern Latin America. In a moment of procrastination I
checked my email. There was a message from a friend in Bogotá—a fel-
low violentólogo at the National University 3—telling me that a colleague
from the University of Antioquia in Medellín had just been assassinated
at point-blank range by three hooded individuals who carried guns with
Map . Colombia. (Source: Charles Bergquist, Labor in Latin America: Compara-
tive Essays on Chile, Argentina, Venezuela, and Colombia [Stanford University
Press, ])
 Introduction 

silencers. My friend had omitted the name of the murdered professor,

but I knew the moment I read the message, with a certainty I cannot ex-
plain, that it was Hernán Henao, a man with whom I had collaborated
for several months on an interdisciplinary seminar devoted to analyzing
violence in Medellín and thinking about peaceful ways to end it.
This was not the first time someone I knew had been killed. During
one particularly horrible period in the early s, it seemed as if there
was a funeral every week, sometimes more, of a professor, journalist, stu-
dent, or human rights advocate. People called each other frequently to
tell their loved ones that they were on their way home, had just arrived at
the office, or were leaving to run an errand because otherwise ordinary
delays were cause for mortal fear. Despite this familiarity with violence,
Hernán’s death plunged me into a deep depression from which it took
months to recover. I wandered the halls of my building that day howling
with pain. I replayed over and over again in my imagination the sight of
Hernán agonizing in a pool of blood in the campus office of the Instituto
de Estudios Regionales (), every inch of which was as familiar to me
as my own house. I remember feeling anger, fear, numbness, disbelief. I
couldn’t think why anyone would kill Hernán, an academic whose life
had been devoted to discussing and anguishing over a way to negotiate a
space for tolerance, mutual respect, and plurality in an increasingly polar-
ized society, but who had never himself advocated violence or taken part
in violent activities. Neither Hernán nor any of the other professors af-
filiated with  believed that the massacres, forcible displacements, or
persistent violations of human rights that take place daily in Colombia
were attributable to a single cause. Hernán and others had reached out
to the victims of violence of the right and left, regardless of ideology, and
offered them solace, education, and programs to help rebuild their lives.
His murder seemed utterly senseless.
In the midst of feeling betrayed and vulnerable, I suddenly realized
the point of terror and how it worked. I mean that I realized it in every
fiber of my body, not as an intellectual abstraction. I had just finished a
preliminary version of this manuscript and felt that I simply couldn’t face
thinking about violence any longer. I fantasized about setting it aside, as if
by doing so I could set aside the reality of violence, too. And then the real-
ization struck me. I knew that even if I could never absolutely establish
the trajectory by which violence had occurred or the exact motivations
behind it, even if I could not swear to the existence of an objective ‘‘truth’’
 Blood and Fire

Map . Department of Antioquia and its municipalities. (Source: Instituto Geo-

gráfico Augustín Codazzi)

about historical events, I nonetheless had to try to trace, with the greatest
precision I could muster, the complicated, murky, sometimes contradic-
tory, and seemingly unrelated events that led to violence. The only way
to overcome my own terror was to refuse to be silenced.
This book is the result of that realization; it is the outcome of a convic-
tion that what has happened in the past is crucial to understanding what is
happening today and that refusing to accept that most violence is incho-
ate, random, or inexplicable is a moral obligation. It is also a small tribute
to the people whose insistence on uncovering unwelcome truths in the
face of extreme threat has been a cause of constant inspiration to me. My
awareness of links between past and present conflict, however, should not
be understood as a belief that violence in Colombia is somehow inherent,
 Introduction 

unique, inevitable, or static. On the contrary, if the case of la Violencia

in Antioquia is at all representative of Colombian violence as a whole,
then what is significant about this study is the discovery of how selective
and concentrated supposedly generalized violence has been, and to what
degree factors such as ethnicity and race, cultural differences, class, and
geography have shaped the evolution, trajectory, direction, and incidence
of violence in Colombia over time. The historical act of glossing la Vio-
lencia as a generalized phenomenon gives short shrift to the memories
of those who refused to take part in violence and to the memories of its
true victims, the thousands of unnamed rural folk who died and whose
voices have been silenced or forgotten. Hernán Henao dedicated himself
to elucidating the causes of violence and the identity of its victims, and
in its own way this book tries to carry that legacy forward.

La Violencia in Antioquia

Two hundred thousand Colombians are estimated to have died as a result

of violence between  and . Over two million others migrated or
were forcibly displaced from their homes and towns, the majority were
never to return. The impact of la Violencia was so great that it provoked
Colombia’s only twentieth-century military coup and led later to an un-
precedented agreement between the leaders of the Liberal and Conser-
vative parties to alternate control of the presidency and share political
power for nearly twenty years.
Of the Colombian regions hardest hit by violence, Antioquia ranked
third in the total number of violent deaths registered nationally between
 and , as approximately , of the province’s inhabitants are
estimated to have died as a result of the Violence. In  nearly  per-
cent or ,, of Colombia’s total population of ,, lived in
Antioquia. Thus, there was a regional, per capita casualty rate of nearly .
percent over the time period.4 In other words, many deaths occurred in
Antioquia, but because the overall regional populations of other severely
affected provinces were much smaller than Antioquia’s, the impact of
casualties in these other provinces was even more pronounced.5 Antio-
quia also registered the eighth highest number of migrations as a result
of violence in Colombia (, or  percent of the national total of
migrations caused by violence). But, again, in regional terms, the seven
provinces that led the nation in total migrations as a result of violence
 Blood and Fire

Map . Administrative subregions. (Source: Instituto Geográfico Augustín

Codazzi)

had populations significantly smaller than Antioquia’s and therefore ex-

perienced a much higher proportional displacement of their population.6
What makes the case of Antioquia during la Violencia significant is not
the number of casualties or migrations that occurred as a result of vio-
lence, but rather where violence took place in the province and why.
In this book I draw on previously untapped sources such as regional
and municipal government archives, judicial testimony, parish death
records, and interviews to tell a story that echoes the findings of research-
ers tracing the trajectory of violence in other Colombian regions between
 and  and also challenges them. Despite ranking third as the de-
partment most severely affected by violence, Antioquia was not beset by
widespread violence nor was the violence most pronounced or concen-
 Introduction 

Map . Peripheral municipalities. (Source: Instituto Geográfico Augustín

Codazzi)

trated in the coffee-producing municipalities of the southwest as has his-

torically been thought.7 Instead, violence proved most severe in Antio-
quia’s geographically peripheral zones where land tenure, production,
labor, and the state’s authority were markedly different from the domi-
nant paradigm in Antioquia’s centrally settled municipalities. In Antio-
quia, the earliest stage of la Violencia (–) affected in indelible
ways those areas situated in the department’s geographic periphery such
as the tropical lowlands of Urabá, the Bajo Cauca (lower Cauca Valley),
and Northeast and Magdalena Medio (middle Magdalena Valley), but not
Antioquia’s coffee sector or centrally located municipalities. (See maps 
and .)
Violence-related death statistics provide a crude index of the spatial
 Blood and Fire

and temporal dimensions of Antioqueño violence. The total number of

officially registered deaths in Antioquia during the years of la Violencia
varied only slightly between a low of , () and a high of ,
().8 But deaths in three categories: ‘‘homicide,’’ ‘‘unspecified or ill de-
fined,’’ and ‘‘other violent deaths’’ rose significantly between  and
, and then declined until . In  the cumulative total of deaths
encompassed by these three categories peaked at ,, accounting for
nearly  percent of all the deaths registered in that year.9
Death statistics collected by Antioquia’s governor’s office (for inter-
nal purposes, not public dissemination) give a more precise picture of
regional violence.10 Before , the regional government did not keep
a separate statistical record of deaths specifically related to violence, but
government records and interviews with survivors suggest that violence
was largely sporadic between  and  and concentrated in cen-
trally located towns where the total number of violence-related deaths
was low.11 Three quarters (twelve of sixteen) of the officially registered
deaths specifically listed as the direct consequence of violence by the gov-
ernor’s office in , for instance, occurred in centrally located towns.
By , however, the pattern of sporadic, centrally concentrated deaths
shifted. Deaths explicitly deemed the result of violence numbered in the
hundreds by  and were concentrated in Antioqueño towns located in
the furthest southwest (Urrao),12 western Antioquia, and in the far east-
ern portions of the department (the Northeast, Bajo Cauca, and Magda-
lena Medio). Core area towns such as Medellín, the industrial towns near
Medellín (such as Bello or Envigado), the coffee-producing south and
southwest, the near east (oriente) and the immediate north-central subre-
gions, in contrast, reported very few violence-related casualties between
 and .13 In fact, half of the more than four thousand violence-
related regional deaths officially registered between  and May 
took place in just five municipalities (Dabeiba, Puerto Berrío, Urrao,
Cañasgordas, and Remedios), all of them located on Antioquia’s periph-
ery (map ; also see appendix A., A..)
Of all the violence-related deaths tallied by the regional government,
 percent occurred in western Antioquia and Urabá,  percent occurred
in the southwest,  percent in the Magdalena Medio region, and  per-
cent in the northeastern section of Antioquia. With the exception of the
highly populated southwest, all of the areas with the highest percent-
age of casualties were also the least populated in Antioquia. Also, of all
 Introduction 

Map . Deaths due to violence, –. (Source: Instituto Geográfico Augus-

tín Codazzi and Archivo Privado del Señor Gobernador de Antiquia, , vol. ,
‘‘Informe sobre la acción del bandolerismo de  a mayo de ,’’ Medellín,
May )

the officially registered deaths from violence occurring between  and
, half occurred in a single year, . Just one town, Puerto Berrío,
accounted for nearly a quarter of these. The selective and concentrated
nature of violence is even more striking when deaths related to violence
are measured as a percentage of local population. Based on the census
of , only one quarter of  percent of Antioquia’s population suffered
violence-related deaths between  and , but Puerto Berrío in the
Magdalena Medio lost  percent of its population to violence while Cau-
casia in the Bajo Cauca lost nearly  percent of its inhabitants. West-
ern towns such as Urrao, Dabeiba, and Cañasgordas, moreover, lost be-
 Blood and Fire

tween  and  percent of their populations to violence during a three-year

period.
The ‘‘official story’’ of violence represents it as a widespread, generi-
cally partisan phenomenon waged indiscriminately between Liberal and
Conservative rural folk, but the official record uncovers a violence re-
markably limited in scope and far more varied in impulse. How are we to
account for the geographic and temporal specificity of violence-related
deaths in Antioquia? Why were towns located on the margins of the de-
partment the sites of most severe and prolonged violence? Why were the
majority of towns in the coffee heartland (the southwest), which were
equally Liberal and where it has always been supposed that the violence in
Antioquia was centered, so much less violent than towns on the periph-
ery? Is it possible that factors in addition to partisan differences influ-
enced the severity of violence and shaped a more pronounced concen-
tration in specific geo-cultural areas? Did the objectives of violence shift
over time and were they dependent upon factors peculiar to local rather
than generalized national circumstances? If so, how would we have to re-
think our conceptualization of the relationship between partisan politics
and violence in Colombia?
Antioquia was Colombia’s most populated, Conservative, and eco-
nomically influential department at mid-century. The province was also
—and had been for some decades—one of Colombia’s largest regional
producers of coffee for export, the nation’s main producer of gold, and the
national leader in industry, commerce, and finance. Antioqueños were
sometimes less likely to occupy national political office than the inhabi-
tants of other Colombian provinces, but Antioquia’s voters were numer-
ous and the province’s men of capital dominated powerful private pro-
ducer associations such as the National Federation of Coffee Growers
(), the National Federation of Merchants (), and the
National Association of Industrialists (), entities instrumental in
shaping Colombian economic and social policy.
In a country where Liberal and Conservative differences were thought
to define individual identity and to have caused the majority of Colom-
bia’s violent struggles since the nineteenth century, moreover, Antioquia
was perceived as both a political maverick and as reluctant to take up
arms in the name of politics. Indeed, there was little in Antioquia’s past
to suggest that it should have become an area hard hit by partisan vio-
lence during la Violencia. Neither the province of which Medellín is the
 Introduction 

capital nor Medellín itself was associated with violence in the Colombian
imaginary. A stereotype existed of Antioquia and its inhabitants, but it
was one that characterized paisas 14 as the nation’s sharpest businessmen
and pragmatic technocrats, a region of aggressive colonizers who were
also fiercely Catholic. A prolific lot, Antioqueños figured in the national
imagination as the people who opened and peopled Colombia’s south-
western frontier, who came to embody coffee cultivation and culture in
the early twentieth century, and who gave rise to a society characterized
by a sense of strong regional identity, large families, and small property
holders. Many a joke was made targeting regional inhabitants as too ob-
sessed with making money to spare the time to take part in politics. When
forced to choose between going to war over political differences and arriv-
ing at a negotiated solution that would preclude social unrest and allow
business to continue unimpeded, the region’s inhabitants were perceived
as usually opting for the latter. What happened then by mid-century to
make Antioquia an important locus of violence?
To those familiar only with the recent history of Colombia or Antio-
quia, the association of violence with both the country and the region
might seem self-evident. As David Bushnell ruefully notes in the intro-
duction to his recent synthesis of Colombian history, ‘‘Colombia is today
the least studied of the major Latin American countries, and probably
the least understood.’’ 15 In contrast to many of its neighbors, Colom-
bia has rarely suffered from dictatorships, boasted no powerful military,
managed its finances conservatively, and displayed no conflict based on
ethnic differences. Moreover, except for the brief appeal of Liberal popu-
list leader Jorge Eliécer Gaitán in the s and the military govern-
ment of General Gustavo Rojas Pinilla in the mid-s, Colombia rarely
fell victim to the sway of populist or authoritarian politics. By the mid-
twentieth century, the persistence of identification with the same parties
that had oriented individual political affiliation since the nineteenth cen-
tury at the expense of supposedly more modern forms of political ex-
pression reinforced the idea that Colombia was somehow unique and
that there existed no common frame of reference with which to com-
pare events in Colombia to those in the rest of Latin America. This has
relegated the phenomenon of la Violencia to a kind of historical limbo
much written and obsessed about by Colombian specialists but regarded
by other Latin Americanists as an aberration peculiar only to Colombia.
At first glance la Violencia does appear as a throwback to an earlier
 Blood and Fire

age of caudillo civil wars and peasant atavism that confirms the notion
of Colombia as out of step with other ‘‘modernizing’’ nations in the re-
gion. The bulk of the killing during la Violencia took place in rural areas,
and peasants constituted the majority of casualties. Victims were often
tortured, dismembered, and sexually mutilated, and women were fre-
quently raped in front of their families. These conditions alone, however,
are insufficient to distinguish conflict in Colombia from that typical of
the rest of Latin America. But, while national political struggles, per-
sonal feuds, agrarian unrest, and clientelist competition informed con-
flicts in other Latin American societies, these had either taken place in
the nineteenth or early twentienth centuries, involved war with another
nation, or occurred in the context of suppressing an indigenous popu-
lation.16 Alternatively, violence occurring in Latin America in the post-
Violencia years was explained as leftist insurgency or anticommunist
state terrorism waged in defense of national security and democracy.17
There seemed to be no Latin American precedent for a conflict in which
those killing each other were citizens of the same state who attacked
one another because of partisan differences and who did so with a sav-
agery rarely seen outside the context of racially or ideologically moti-
vated wars.18 In other words, what distinguished the Colombian Violencia
from twentieth-century violence occurring elsewhere in Latin America
was that it was fought in terms of mid-nineteenth-century political par-
tisanship not modern political or social objectives. There were of course
comparably brutal and complex cases of civil conflict in other parts of
the world to which la Violencia might be compared, but this required
attributing the same symbolic and innate power to Colombian partisan
differences as that attributed to religious and ethnic and racial differences
present elsewhere.19
But cultural, religious, and ethnic and racial differences did exist in
Antioquia and were fundamental features of how violence unfolded in
the region. Indeed, it is the argument of this study that la Violencia
in Antioquia can only be understood against the backdrop of profoundly
perceived differences between geo-cultural areas internal to the province,
and that these differences were often as critical as, or more so, than par-
tisan factors in determining the intensity, incidence, and trajectory of
violence in the region. To make clear how the Antioqueño experience
of la Violencia differs from historical interpretations of the phenome-
non, and the significance of these differences for the study of violence
 Introduction 

in Colombia more generally, I have divided the remainder of this intro-

duction into three parts. First, I provide a brief overview of Colombian
politics and society in the decades preceding la Violencia. I then sum-
marize the various interpretations and regional case studies that form
the core of Violencia studies from the s to the present in order to
provide a comparative basis for a consideration of the issues raised by
the Antioqueño experience of the Violence. Lastly, I lay out a theoreti-
cal framework for thinking about the relationship between geography,
politics, ethnicity and race, class, and violence and explore the reasons
why these issues, rather than partisan identity alone, shaped the course
of mid-century conflict in Antioquia.

Politics and Society in the Decades before la Violencia

Initial attempts to make sense of la Violencia sought an explanation in the

peculiarities of Colombian political history. Like Liberals and Conserva-
tives elsewhere in nineteenth-century Latin America, Colombian politi-
cal parties were divided into opposing camps of protectionists and free
traders, centralists and federalists, and pro- and anti-clerical feeling. The
significance of specific issues to the determination of individual politi-
cal understanding and comportment differed to some degree from re-
gion to region, depending on the availability of resources, the structure
of land tenure and production, kinship relations, accidents of history,
and myriad other intangibles. An Antioqueño Conservative of moderate
stripe, for instance, might simultaneously embrace both free trade and
federalism (positions more typically associated with the Liberal party)
and yet strongly support the Catholic Church (a position more typical
of pro-clerical Conservatives). What set Colombia’s parties apart from
Liberal and Conservative parties in other Latin American countries, how-
ever, was the Colombian system’s ability to foster a deep identification
between the parties and the vast majority of its citizens.20 The Colombian
parties attracted individuals of all classes, regions, and racial and ethnic
origin and, in the absence of a well-developed sense of national iden-
tity, scholars have argued, party affiliation shaped the average Colom-
bian’s sense of self and belief from the nineteenth through the twentieth
centuries.21 Identification with one of the two parties also persisted in
Colombia long after Liberal and Conservative parties elsewhere in Latin
America disappeared or gave way to multiparty systems.
 Blood and Fire

Policy and ideological differences between Liberals and Conserva-

tives fueled most of the repeated nineteenth-century civil wars for which
Colombia became famous, although the majority of the so-called civil
wars occurring in Colombia before the War of the Thousand Days (–
) might more accurately be described as skirmishes. The ostensible
catalysts of such ‘‘wars’’ were not insignificant—the suppression of con-
vents, the abolition of slavery, the empowerment of artisans, struggles
to seize control of the central government, and so on—but they rarely
engaged more than a small percentage of Colombians in actual physi-
cal combat. Civil war casualties were for the most part also relatively
few, although the destruction and confiscation of property affecting a
particular individual, clan, or interest group could become the basis of
long-standing resentment that cemented partisan identity. In the end,
however, despite a reputation for chronic disorder, nineteenth-century
Colombia does not appear to have been noticeably more violent than
other Latin American countries of the time.22
In  Liberal Rafael Nuñez won control of Colombia’s presidency
and, with the support of the Conservative party revoked many of the
political and social reforms passed during several decades of Liberal
political domination. The revised Constitution of  replaced state au-
tonomy with strict centralism, converted previously elected offices into
a hierarchically determined system of government appointments, estab-
lished literacy requirements for male suffrage in national elections, and
restored the preeminence of the Roman Catholic Church in matters such
as public education.23 A severe downturn in the export price of coffee
during the second half of the s as well as growing discontent among
Liberals over their political exclusion eventually sparked the outbreak
of the War of the Thousand Days, the last and greatest of Colombia’s
nineteenth-century civil conflicts.24 In contrast to the limited engage-
ments characteristic of earlier struggles, the war produced more than
, casualties, a large number of maimed and displaced people, and
the irrevocable loss of Panama.25
There were fears that, if the war were allowed to continue, further ter-
ritorial dismemberment (beyond the already dramatic loss of Panama)
would occur and Colombia’s economic future would be compromised
at the very moment when coffee seemed to promise a way out of eco-
nomic stagnation. Ultimately, these fears converged to bring fighting to
an end. General Rafael Reyes, Colombia’s first twentieth-century military
 Introduction 

ruler and the man behind the elimination of the most exclusionary poli-
cies associated with the Regeneración (as the Nuñez regime was known),
came to power in . Reyes enjoyed the overt support of the moder-
ate faction of the Conservative party known as the Historical Conserva-
tives—many of whom were Antioqueño capitalists—and the tacit sup-
port of many Liberals.26 Reyes institutionalized minority representation
in Colombia’s various legislative bodies and promoted policy initiatives
that proved crucial to the support of domestic industry and the export
economy, especially the coffee sector. Although a combination of factors
led to Reyes’s quick fall from grace, he laid the basis for a period of eco-
nomic expansion within a climate of relative bipartisan cooperation that
characterized what has sometimes been called the pax conservadora of
 to .27
Several aspects of coffee production helped it to emerge as a focus
around which members of both parties and numerous regional interests
could cooperate to set aside the partisan antagonisms that had under-
mined national political stability during Colombia’s first century of inde-
pendent existence. First, by the s significant sectors of the population
of both historically Liberal and Conservative regions were associated with
coffee production or its commercialization. Second, coffee was grown
by both large landowners in the eastern and central regions as well as
by small and medium-sized property holders in the central cordillera
(among them Antioquia and the regions its inhabitants colonized to
the south). Charles Bergquist has argued persuasively that these circum-
stances ensured that ‘‘a large proportion of the Colombian body politic
identified with the political economy of the export-import interests in
control of the government after ’’ and that smallholders ‘‘fully en-
dorsed the liberal political ideology, social conservatism, and pro-export
economic policies of the new order.’’ 28
Despite continued differences between Liberals and Conservatives,
consensus emerged between businessmen and coffee growers from 
to  regarding the importance of and need for state investment in
infrastructure and economic development. During these years many of
the elite leaders of both parties intermarried, attended the same schools,
and dominated regional and national politics.29 The s in particular
witnessed unprecedented private and public expenditure on an ambi-
tious program of public works and education. But investment and eco-
nomic growth did not benefit all Colombians during the heady years
 Blood and Fire

that came to be known as the Dance of the Millions. The Conserva-

tive coalition of coffee growers, export merchants, and industrialists that
had dominated Colombian political fortunes for more than two decades
toppled in  amid rumors of fiscal mismanagement and accusations
that they sacrificed the lives of Colombian workers to U.S. interests dur-
ing the  United Fruit strike in Santa Marta.30
During the presidential election of , the Conservative party split
and lost to the Liberal opposition. The change from one political admin-
istration to another in Colombia typically meant the substitution of one
party’s members for those of the other in patronage jobs and government
positions. When Liberal Enrique Olaya Herrera was elected president
(–), violence broke out in several regions of Colombia where
Liberals unleashed their long-suppressed resentment on the Conserva-
tive opposition. Indeed, while many scholars consider the assassination
of Liberal populist, Jorge Eliécer Gaitán, on April , , as the semi-
nal event that catalyzed la Violencia, the factors that led up to the Liberal
leader’s death and the emergence of severe unrest in its aftermath can
in part be traced to the changes occurring in Colombia during the s
and s.
Industrial employment and unprecedented public works investment
had begun to transform Colombia from a predominantly rural to an in-
creasingly urban country in the early decades of the twentieth century.
In  a third of Colombia’s population was classified as urban whereas
by  nearly half of the nation’s inhabitants lived in urban areas. Rural
migration to cities was only temporarily interrupted by the contraction
of employment during the period of economic recession between 
and .31 The effects of urban growth—pressure on public services,
the increased cost of living, and the emergence of an increasingly vocal
underclass—were felt in cities such as Bogotá, Medellín, Cali, and Barran-
quilla.32 Urbanization thus coincided with both the shift to a period of
Liberal government after nearly fifty years of Conservative rule in 
and the emergence of popular demands for expanded political recogni-
tion and participation. These profound national changes were reflected
in the administration of Liberal Alfonso López Pumarejo who allied with
sectors of his party to shift Liberal policy in a more progessive and so-
cially inclusive direction in .
Alfonso López Pumarejo’s Revolution on the March (–) was a
more modest version of the Cárdenas administration that came to power
 Introduction 

in the same year in Mexico and the progressive Popular Front govern-
ments that sprang up in other parts of Latin America during the s.
López initiated social legislation, abolished literacy requirements for suf-
frage, and extended legal recognition and rights to workers and peas-
ants.33 As he expanded the functions of the state, López also centralized
its power, elevating the state into a mediator between conflicting social
and economic interest groups.34
Agrarian unrest had become acute in several Colombian regions in
the years immediately preceding López’s rise to power.35 In some areas,
colonists hoping to escape the effects of economic downturns in the
s migrated in search of regions with supposedly abundant public
lands only to find that these had been swallowed up by recently estab-
lished large-scale commercial agriculture and cattle ranches. Conflicts
in these areas emerged between landless folk competing with each other
and with powerful capitalist landowners. In other areas, previously un-
organized rural workers mobilized to protest changes in tenancy laws,
dismissals, and poor wages on plantations.36 To resolve the problem of
growing agrarian unrest and to preclude economic disruption in regions
where struggles over land were most severe, López initiated Law  of
. The law declared that property had a social function and sought to
mediate competing claims to public lands while providing titles to those
petitioners who could prove they had resided on and made improve-
ments to the land. López did not intend to undermine the principle of
private property in Colombia nor was it his intent to do away with large
landowning.37 Although agrarian unrest diminished after Law ’s pas-
sage, the land reform law confirmed only a limited number of squatter
claims, making the validity of petitions not initiated before  much
more difficult to prove.38 Reaction to the law, in any case, rested less upon
its actual impact than upon the elite’s perception of its threat.
When taken in conjunction with López’s recognition and legalization
of labor organizations such as the Confederation of Colombian Workers
(), and his introduction of organized labor into the once restricted
arena of elite politics, his social policies fueled resentment among men
of capital like those in Antioquia.39 In addition, López’s toleration of
Communist leaders—many of whom headed important labor unions
(affiliated with the newly created ) in strategic sectors such as oil,
transportation, and mining—led the more reactionary members of both
parties to repudiate the López administration as dangerously radical.40
 Blood and Fire

The nearly hysterical alarm evinced by the nation’s entrepreneurs and in-
dustrialists over López’s championing of working-class interests and his
extension of state authority between  and  formed a critical back-
drop to the vituperative red-baiting that helped incite partisan violence
in the forties and is only understandable when set against the background
of growing capitalist investment and economic expansion taking place in
the decade preceding the outbreak of la Violencia. Colombian industry,
for instance, embarked upon a period of expansion that led it to grow
in real terms at an unprecedented rate of  percent per year between
 and . Nowhere was the impact of industrial growth more clearly
felt than in Antioquia, especially in the industrial hub around Medellín
where textile mills and other light industries formed the core of the local
economy.41
At the end of López’s term, the Liberal party sought out a candidate
who might halt the momentum of López’s revolution and reassure elite
interests. They found their champion in Eduardo Santos, a prominent
businessman and the patriarch of Colombia’s family-owned, largest cir-
culation daily newspaper, the Bogotá-based El Tiempo. During his presi-
dency (–) Santos muzzled labor unrest, put down strikes, and
deflected popular demands so as to curtail the movement of labor his
predecessor had nurtured and encouraged.42 Despite the distrust he gen-
erated among members of the elite, however, Alfonso López remained
a charismatic political leader and he returned to power in  with
the support of the very groups whose interests he had defended during
his first presidency. But López’s second term in office proved a disap-
pointment to his more progressive supporters. Disagreement within the
Liberal party, increasingly fierce Conservative opposition, and the in-
tensification of rural partisan conflict culminated in  with a failed
military coup led by disgruntled army officers.43 When López was finally
forced from office in  and Liberal Alberto Lleras Camargo assumed
the presidency in May, the conservative social trend already apparent in
the later years of Liberal government became more pronounced. One of
Lleras Camargo’s first acts as president was the dissolution of a long and
bitter strike led by the Magdalena Transport Workers (), per-
haps Colombia’s strongest and most militant union, and the only one
with a closed shop.44 Lleras Camargo also implemented Law  of 
regulating collective bargaining agreements in Colombia. While the law
confirmed the social services and benefits labor had won under Alfonso
 Introduction 

López Pumarejo’s Revolution on the March, it also marked a critical shift

in the relationship that had been established between labor, the Liberal
party, and the state in the mid-thirties. The law strictly defined the cri-
teria for a legal strike, outlawing strikes in the sector público, that is, for
workers employed in public works, transportation, communication, and
municipal and state government (the source of most patronage hiring).
These were precisely those sectors of the workforce that were most vocal
and most dependent upon an alliance with the Liberal state for their well-
being.45 Failure to comply strictly with the labor code’s criteria for a strike
became the basis for dismissing workers’ demands, however well inten-
tioned or legitimate they might have been. Popular and working-class
interests, already battered by declining real wages, unemployment, and
harassment, were further weakened by the loss of state advocacy on their
behalf.46
In addition to the growth of the urban population, industrialization,
and the incipient political empowerment of an organized working class,
an emergent middle sector of professional politicians of non-elite origin
had also gradually come to demand greater political participation in the
national political arena during the s and s. Some of these pro-
fessional politicians identified with the program embraced by the parties’
traditional elite leadership, but others used populist appeals and criti-
cism of bipartisan elite rule to expand their electoral support and confirm
their political participation in party directorates and the national govern-
ment. The divide between the political culture of convivialismo (as elite
bipartisan political rule was called) and the new politics of mass inclu-
sion was embodied in the figure of Liberal populist Jorge Eliécer Gaitán.
A dark-skinned man of humble birth, Gaitán symbolized not only the
rise of a growing nonwhite, urban popular core in Colombian society,
but also the rise of non-elite politicians emboldened by the extension of
education and suffrage that had taken place during the previous two de-
cades.47 The urban lower class and the aspirants to political power among
the provincial middle-class or petit bourgeois sectors linked their for-
tunes together to press for an opening of the political sphere. The clash
between the popular forces represented by young, up-and-coming poli-
ticians of both parties and an elite concerned with reasserting the exclu-
sionary, paternalistic rule of pre- Colombia came to a critical climax
in the presidential campaign of .
The Liberal party split over the candidacies of Gabriel Turbay Ayala
 Blood and Fire

(the party’s official candidate) and the dissident, Jorge Eliécer Gaitán,
and lost the election to the moderate elite Conservative, Mariano Ospina
Pérez. Partisan conflicts like those experienced in the early thirties, when
power changed from Conservative to Liberal hands, once more emerged
at the municipal level. Conservatives excluded from participation in gov-
ernment patronage and elected offices during the previous sixteen years
of Liberal hegemony celebrated the defeat of the Liberal opposition with
acts of intimidation and physical harassment in a number of Colom-
bian departments. Although Ospina himself campaigned on a bipartisan
political platform that promised the inclusion of Liberals in his cabinet,
gubernatorial offices, and municipal government positions, his stance en-
countered considerable opposition from extremists within the Conserva-
tive party and the Liberal followers of Gaitán. When the Liberal party won
the congressional elections of , the basis of Ospina’s National Union
compromise dissolved.48 Tensions between the Conservative government
and the opposition escalated steadily from that point on, reaching a cli-
max with the assassination of Gaitán by a mentally disturbed gunman in
Bogotá on April , .
The Bogotazo, as the popular uprising in response to Gaitán’s assas-
sination came to be known, left the nation’s capital a smoldering mass
of ruins; churches and public buildings were transformed into heaps of
rubble; trolley cars were derailed and burned; stores looted; the city’s
sidewalks overflowed with the debris of broken glass and ruined mer-
chandise. Meanwhile, decomposing corpses hurriedly thrown in piles in
Bogotá’s central cemetery seemed to give material testimony to the exis-
tence of an anonymous, dangerous crowd that had captured the elite
imagination and provoked increasing anxiety of an impending attack
upon elite privilege by a ragged, bloodthirsty army of the nation’s ex-
cluded.49 Surrounded by a burning and looted city and unsure of just
how many troops or individuals might come to his defense, Ospina
nonetheless resisted Liberal demands that he hand over power.50 Instead,
the president purged the police of Liberals (many of whom had turned
against the government and collaborated with the rioters), reshuffled the
cabinet and once more attempted to establish a bipartisan government.
The administration also implemented modest reforms of the social secu-
rity system, established price controls on basic food items, and sponsored
a U.S. economic mission to examine the nation’s development policies
and make recommendations on how best to maximize the state’s effi-
 Introduction 

ciency.51 But Ospina’s attempt to shift attention away from partisan issues
to less controversial technocratic matters proved unsuccessful. The Con-
servative party leader, Laureano Gómez, and his followers (known as
laureanistas) led a violent bid for the presidency during  that further
ignited already combustible partisan animosities in Colombia’s country-
side. In the wake of growing incidents of partisan unrest, Ospina Pérez
declared a State of Siege, and in November  the president closed
the congress indefinitely.52 Congress would remain inactive for the next
nine years.
A surreal quality enveloped Colombia between  and . As vio-
lence raged in rural areas and multiple groups under local and regional
leadership terrorized the countryside, in Bogotá, Laureano Gómez ruled
seemingly removed from the din and clamor of widespread strife.53 In
urban areas such as Medellín, moreover, business went on as usual; busi-
ness, in fact, boomed. In  the president of the National Associa-
tion of Industrialists could coolly declare that Colombia’s economy had
never been better, repeating his assertion on the eve of the military coup
in .54 Insisting that violence was in check, denying its severity, and
blaming its existence upon isolated, depraved bandits, the national gov-
ernment seemed oblivious to its inability to assert its authority outside
Bogotá and the nation’s principal cities. By  tentative attempts at bi-
partisan dialogue between the more moderate members of the parties,
many of them representatives of prominent economic interests, were
under way. Several months later a military coup—Colombia’s first and
last during the twentieth century—backed by significant civilian and elite
support put an end to Laureano Gómez’s presidency on June , .
The military dictatorship that came to power under the leadership of
General Gustavo Rojas Pinilla in  and which ruled Colombia until
 initially succeeded in reducing partisan tensions in Colombia.55 The
government pardoned Liberal guerrilla groups and removed some of the
more hated Conservative local leaders who had been in charge of mobi-
lizing paramilitary groups against the Liberal opposition in rural areas.
After a brief respite, however, partisan-motivated violence gave way to
common criminal delinquency, social banditry, and incipient, radical
peasant leagues. Rojas Pinilla’s growing ambition, moreover, frightened
the very civilian elite forces that had initially supported the general’s mili-
tary coup. In  power reverted once more to civilian rule and, in an
unprecedented attempt to simultaneously put an end to violence and
 Blood and Fire

preclude future military intervention, leaders of the Liberal and Con-

servative parties agreed on a power-sharing arrangement known as the
National Front. What had begun as partisan conflict in the countryside
took on a distinctly social and economic cast by the later years of the
s, giving rise in some areas to the nucleus of what would constitute
insurgent, leftist guerrilla groups in the s. It appeared that la Violen-
cia had not ended, but simply evolved.

Interpreting la Violencia

In the s social scientists took up the challenge of understanding la

Violencia and devised numerous theories to explain it. These alternatively
attributed violence in Colombia to conflicts provoked by the transition
from a ‘‘premodern’’ to a ‘‘modern’’ society, to exaggerated aggression
fueled by status deprivation, or to rivalries between patron-client sys-
tems in which peasants blindly followed the dictates of an elite leader-
ship or party boss.56 While the patron-client analysis offered clues to
the seemingly national scope of violence, it failed to explain why, if dis-
putes originating among an elite leadership in Bogotá could incite the
most distant citizen to take up arms, significant areas within Colombia
remained untouched by la Violencia. Other than through some vague
‘‘quasi-religious’’ appeal, how were ideology and party allegiance actually
disseminated and understood?
New scholarship in the s shifted the focus of work on la Violencia
in other directions. The power of the state, the expansion of the political
arena, the rise of new political actors and leaders such as Jorge Eliécer
Gaitán in the decades preceding la Violencia, and the quest for alternative
forms of economic and political mobility were issues increasingly singled
out as playing important roles in the development of the violence.57 As
scholars grounded their research in region-specific studies, moreover, it
became apparent that while partisan conflict provided the initial catalyst
to violence, and perhaps even a seemingly logical framework in which
to understand the intensity of the conflict, reliance on the notion of in-
herited party hatreds was insufficient to account for the divergence and
specificity of violence. La Violencia resembled the Mexican Revolution in
the way that historians might agree that the latter phenomenon was set
off by Porfirio Díaz’s decision not to seek reelection, but they might not
agree on the composition of those fighting, their exact objectives once
 Introduction 

violence got under way, or the long-term implications of the revolution.

La Violencia has similarly proven to be an extraordinarily heterogeneous
and complex phenomenon.58
Indeed, recent studies of la Violencia raise as many questions as they
answer. They reveal, for instance, how little is actually known about the
workings of Colombian politics at the local, regional, or national levels
or about the internal organization of the parties themselves. Were the
parties monolithic? 59 How did understandings of partisan affiliation dif-
fer among individuals belonging to different classes, regions, or ethnic
and racial groups? 60 Was it really true that partisan affiliation took prece-
dence over any other kind of identity in Colombia? 61 If not, what shaped
people’s beliefs, actions, and sense of identity? Even less was known about
the nature of the Colombian state, how strong or weak it was or whether
a central state existed at all. Was power centralized in the state to such a
degree, as some researchers argue, that competition between the parties
for its control could set off national unrest of the scope of la Violencia? 62
Or was the problem just the opposite? Perhaps no central state existed or
it had so tenuous a presence in most areas of the national territory that it
proved helpless to control conflict between omnipresent political parties
when it broke out? 63
Then there were the social and economic implications of la Violen-
cia. Was violence the response of a frightened elite to the mid-twentieth-
century expansion of the Colombian electorate and the rise of middle-
sector politicians? 64 Had the rise of Jorge Eliécer Gaitán and his political
movement introduced class struggle in Colombia? Did gaitanismo repre-
sent a threat to the exercise and workings of traditional politics in Colom-
bia? 65 Did the spread of popular uprisings in the aftermath of Gaitán’s
assassination and their subsequent repression constitute the seeds of a
failed social revolution? 66 Was violence waged under traditional party
banners to deflect attention away from or to justify crushing other latent
sources of conflict such as struggles over land, declining opportunities
for social mobility, and growing worker unrest? 67
Two very influential analyses of the violence posited that la Violen-
cia was the result of excessive partisan clientelism and the growing com-
petition between two monolithic parties to control access to the cen-
tral state. Paul Oquist argued that as the central state grew in the s
competition between Conservative and Liberal leaders to monopolize ac-
cess to the state’s largesse and influence became increasingly urgent. Ac-
 Blood and Fire

cording to Oquist, the struggle to achieve ‘‘hegemonic’’ control of the

Colombian state unleashed violence that led to its partial ‘‘breakdown.’’ 68
French sociologist Daniel Pécaut, on the other hand, argued that the
state’s power to build a sense of national identity or act as a suprapartisan
arbiter of conflict between different sectors of Colombian society had
been eclipsed by the persistence of two ‘‘subcultures.’’ 69 These subcultures
were defined by individual identification with either the Liberal or Con-
servative party. Since only partisan affiliation could guarantee individual
material needs and physical survival, any conflict between the parties in-
evitably resulted in widespread conflict. The use of force, over which the
Colombian state had never achieved complete monopoly, in turn, Pécaut
suggested, became more dispersed among competing corporate interests
as partisan competition to control the state intensified.
Various scholars gave greater empirical precision to the hypotheses of
violence put forth by Oquist and Pécaut. Herbert Braun, for instance, fo-
cused on the urban rather than the rural manifestations of violence, more
specifically, on the prelude to and aftermath of Gaitán’s assassination in
Bogotá on April , . In much greater detail than Pécaut, Braun laid
bare the insular, aristocratic, aloof character of political exchange in pre-
Gaitán Colombia. Braun argued that elite members of both parties co-
incided in their social views and interests, and political decision-making
took place not in congress, but over shots of whiskey at Bogotá’s exclu-
sive gun or jockey clubs.70 The critical question always present in the
minds of Colombia’s elite, and roused to hysterical urgency by Gaitán’s
persona, Braun argued, revolved not around ideological differences but
rather around the issue of how to deal with the lower classes.71 Gaitán
challenged the insularity of gentlemen’s politics precisely by reveling in
his plebeian and mixed-race origins and by manipulating his identifica-
tion with and appeal to the popular classes into a major political move-
ment.72 Braun did not believe, however, that the basis of violence was the
insurrectionary or revolutionary content of Gaitán’s message to the poor.
On the contrary, in Braun’s estimation, Gaitán had a fundamentally petit
bourgeois attitude toward the masses, admonishing them to bathe and
act responsibly and to overcome their socioeconomic condition through
hard work and education, not class struggle.73 Braun suggested that the
overreaction of a dominant class terrified by its own prejudices against
a lower class it had long demonized and its misconception of Gaitán’s
political message led it to dangerously raise the stakes of political ex-
 Introduction 

change. The divisive and vituperative rhetoric employed by the elite had
the unintended effect of promoting and legitimizing violence among the
parties’ nonelite membership rather than reasserting the political system
as it had existed before Gaitán’s mobilization of the popular classes. While
Braun noted that Liberal and Conservative elites were equally opposed
to Gaitán, he blamed Conservatives more than the Liberals for the incep-
tion of violence. Braun argued that Conservative efforts to shore up an
eroding electoral position led the party to unleash violence in order to
recuperate the loyalty of the popular classes, and he implied that Conser-
vatives embraced Christian Socialist rhetoric when addressing workers
only as a political ploy to undermine Gaitán’s movement. While the effort
to substitute Liberals in office was certainly a critical factor in fomenting
violence, Braun may have been too cynical in assuming that the adoption
by some Conservatives of a kind of Christian Socialist position vis-à-vis
workers was nothing more than posturing.74
Braun’s theses were quite compelling, but he limited his study to
Bogotá, leaving unanswered the question of whether or not and in what
manner Gaitán and the reaction he elicited among Bogotá’s politicians
affected the emergence and nature of violence outside the capital. Mean-
while, Gonzalo Sánchez, Carlos Ortíz, and scholars such as Jaime Arocha,
James Henderson, and Darío Fajardo gave specificity and concrete mean-
ing to the abstraction of battles waged in the capital by examining the
day-to-day patterns of violence in several Colombian regions.75 In look-
ing at political culture from the ‘‘bottom’’ up, these scholars also reintro-
duced the relationship between socioeconomic conditions and violence
that had faded from the discussion of la Violencia since the early allusions
to such a link in the days of patron-client analysis.
Gonzalo Sánchez argued that an analysis of Gaitán and his move-
ment was the necessary starting point for understanding la Violencia. Like
Braun, Sánchez also believed that the issue of lower-class mobilization
or political incorporation was at the very heart of la Violencia. In sharp
contradiction to Braun, however, Sánchez insisted that Gaitán had intro-
duced the question of class into the Colombian arena, and that Gaitán’s
movement constituted a first attempt at a revolutionary challenge to
the established Colombian economic and political system. For Sánchez,
April  marked a critical turning point in Colombian history. Answer-
ing the question left in suspense by Braun, Sánchez insisted that Gaitán’s
movement had profoundly affected Colombian society at all levels, con-
 Blood and Fire

stituting ‘‘a national insurrection, which, particularly outside Bogotá,

laid bare the enormous creative capacity of the masses for revolutionary
action.’’ 76 Although the ‘‘revolution’’ failed because it lacked coordina-
tion and because Colombia’s elite cohered against it, everyone now had
a glimpse of what class war might be like. Sánchez argued that the after-
math of Gaitán’s assassination triggered a violent reaction and retrench-
ment by Colombia’s elite, first against Gaitán’s followers and then, as
Conservatives gained power, by Conservatives against Liberals, unions,
agrarian leagues, and any other group that might represent a threat to the
status quo.
According to Sánchez, once the threat of social revolution from below
was suppressed by elite coercion, what followed, at least during the first
phase of la Violencia from  to , was a period of violence charac-
teristic of that experienced during Colombia’s nineteenth-century civil
wars. The ultimate impact of this period of la Violencia was to reinforce
old party identities and the strength of gamonales—the local bosses or
power brokers—within the parties.77 The tenor of violence changed, how-
ever, when in  armed popular groups in the cattle frontier of the
Llanos split into those led by Liberals and those under Communist direc-
tion. By the end of the first phase of la Violencia, Sánchez argued, partisan
violence had given way to violence that had little to do with disputes be-
tween Colombia’s Liberal and Conservative parties.78 Some former Lib-
eral guerrillas, in turn, became the nucleus for Colombia’s contemporary
leftist guerrillas.
Gonzalo Sánchez recognized (along with other scholars) that the ob-
jectives and nature of violence could vary from region to region depend-
ing on the economic conditions and social arrangements in each. Fajardo,
Arrocha, and Sánchez theorized that violence in cattle frontiers such as
the Llanos, for instance, was most likely to shift away from traditional
to more radical objectives. Coffee-producing towns, in contrast, evinced
partisan but not revolutionary violence because, unlike cattle frontiers,
coffee towns were nationally integrated through commercial, political,
and social networks. Both Sánchez and Fajardo drew a further conclu-
sion that violence coincided with the emergence of large agribusiness
haciendas.79
Regional examinations of the course of violence suggested important
differences in the day-to-day workings of politics outside the capital and
the factors that influenced variations in the experience and trajectory of
 Introduction 

la Violencia. By the s, the notion of a single, blanket interpretation of

violence gave way to the acknowledgment that violence had many mani-
festations and meanings. Local conditions appeared to be the most sig-
nificant factor in determining the nature of violence and its objectives.
In the most thorough regional study to date, Carlos Ortíz Sarmiento fo-
cused precisely on local issues in his examination of the development of
violence in Quindío.80 Like Sánchez, Pécaut, and Oquist, Ortíz acknowl-
edged the importance of party identity in shaping the course of violence,
but he also noted the influence of an individual’s place of birth, kinship
relations, municipal loyalty, and cross-party relationships.81 While Ortíz
acknowledged the weak presence of the state, he disagreed with Oquist
who assumed that the absence of the state necessarily provoked a vac-
uum of authority. Instead, Ortíz showed how local political arrangements
and beliefs were not automatically affected by national developments.82
Ortíz laid bare the nexus of local political understandings and behavior
that operated at the regional level, arguing that faraway disputes between
vaguely recognized national leaders were unimportant unless they co-
incided with local struggles over revenues, boundaries, and patronage.
Ortíz also focused his attention on determining exactly who held power
and how they used it at the vereda and municipio levels.83 Rather than as-
suming that patron-client relations worked from the top down, he dem-
onstrated convincingly how these were also constantly renegotiated from
the bottom up. Bogotá and the municipio were connected through in-
tricate, dynamic links between gamonales and national politicians. The
currency of political adhesion was patronage and votes.
Although introduced by Pécaut, Sánchez, and Braun, the concept of
the professional politician who gained power and challenged the elite in
the mid-twentieth century was brought down to local terms by Ortíz.
Gaitán, for instance, after an initially lukewarm reception, gradually at-
tracted Liberal support in Quindío, but in the aftermath of his assassina-
tion had faded from the region’s politics.84 La Violencia emerged around
the intrusion of ‘‘outsiders’’ with ambitions to become local power bro-
kers—policemen and mayors appointed to municipios by the central and
regional government in  whose presence disturbed webs of local
power—rather than as a result of Gaitán’s death.
In exploring the alliances and confrontations that emerged in the
s and s, Ortíz also traced the complexities of individual alle-
giances, while rooting these within a framework of economic and politi-
 Blood and Fire

cal changes in the region and the nation. Such changes spawned tensions
not only between classes or individuals but also between villages, towns,
regions, and the national government.85 Ortíz concluded that violence
was due less to an umbilical relationship between dominant national poli-
ticians and their obedient followers in the localities, and more to the pres-
ence of specific social actors operating within a particular context who
might or might not choose to capitalize upon national ideologies and
movements to achieve local objectives and satisfy local aspirations.86
Where then does the case of Antioquia fit within the broader frame-
work of regional studies of violence? In what ways does the experience of
Antioquia during la Violencia confirm or challenge the findings of studies
of violence for Colombia as a whole?
One of the central premises of this book is that violence in Antioquia
was intimately linked to struggles waged between the regional and the
central states and between the regional state and its peripheral inhabi-
tants over the right to determine political, social, economic, and cul-
tural practices. Mid-century violence, moreover, was built on latent, un-
resolved conflicts in the areas where it was most intense and cannot be
understood outside the context of broader structural issues and transfor-
mations affecting Colombia as a whole. While no single analytical frame-
work can adequately capture the multiplicity of reasons why violence did
or did not occur in specific localities, a close reading of individual inci-
dents of violence in Antioquia can bring to the surface multiple, lived
realities that are crucial to a reconstruction of violence and its motiva-
tions and that continue to shape the geo-specific incidence of violence
in contemporary Colombia. A regional study of the heterogeneous ex-
perience of local violence thus enables us to explore how the meanings of
concepts such as the state, partisan affiliation, clientelism, regional iden-
tity, and citizenship were contested and redefined in historically contin-
gent ways by different sectors of society at different times and in different
places.
La Violencia was—and violence in Colombia continues to be—about
state formation and reformation. The process of state formation occurred
and was fought out at multiple sites among diverse, dynamic sectors and
produced varying outcomes. How local and regional participants experi-
enced the effects of state formation and how they responded to these
varied in relation to specific and subjective individual and collective posi-
tionings within the region and the nation. The emergence of violence in
 Introduction 

Antioquia was therefore not the result of a monolithic, coherent, top-

down dissemination of inherited party hatreds or the result of central
strategy or mandate. Rather, the escalation of partisan conflict between
Colombia’s two parties provided the catalyst for latent regional and local
conflicts to come to the fore in the s and created unprecedented
opportunities for previously marginalized sectors to pursue divergent
struggles in their pursuit of power. Not all Antioqueños experienced their
relationship to the state or the parties in the same way. Indeed, multiple
realities coexisted in Antioquia. How individual Antioqueños negotiated
the complexities and challenges of mid-century change and why these
negotiations were expressed most violently in peripheral areas form an
essential aspect of the individual stories that make up the larger narra-
tive of regional violence in this book. In short, for the areas most severely
affected, la Violencia did not represent the culmination or apex of a his-
tory of partisan hatreds so much as it marked a critical stage in an evolv-
ing history of regional state and identity formation. In the peripheral
areas that formed the central locus of conflict, la Violencia represented a
fundamental struggle—and ultimate failure—to impose a hegemonic re-
gional project of rule predicated on notions of cultural, ethnic, and racial
difference.

A Theoretical Framework for Thinking

about the State and Clientelism

In a recent anthology examining the impact of the Mexican Revolution

and the formation of the Mexican state, Derek Sayer and Phillip Corrigan
suggest that the state may either be conceptualized as a ‘‘thing,’’ a tan-
gible, fixed entity where power is believed to reside or, more dynamically,
as a ‘‘claim.’’ In the latter instance, the state represents an attempt to ‘‘give
unity, coherence, structure, and intentionality to what are in practice fre-
quently disunited, fragmented attempts at domination.’’ 87 For Sayer and
Corrigan that which we call ‘‘the state’’ is subject to constant change and
renegotiation. To study it requires abandoning the notion that there exists
an already defined, fully operational apparatus in which power is cen-
tered. Instead, Sayer and Corrigan suggest, the study of the state is the
study of how ruling practices are developed and exercised over time. The
central issue of inquiry thus becomes how political power is constructed
and naturalized, the effects of this naturalization, and the ways in which
 Blood and Fire

those that the state supposedly dominates also shape the practice of poli-
tics. ‘‘Performances,’’ Sayer and Corrigan argue, constitute a crucial di-
mension of the power that represents itself as ‘‘the state’’ and us as mem-
bers of the ‘‘body politic’’: ‘‘it is the exercise of power pure and simple that
itself authorizes and legitimates; and it does this less by the manipulation
of beliefs than by defining the boundaries of the possible.’’ 88
Thus, the mundane rituals of obtaining a driver’s license, observing
the speed limit, and paying taxes are what construct power and, over
time, legitimize it. Sayer identifies the institutionalization of such rituals
as the coercive aspect of the organized exercise of power. But coercive
practices may also enable power. People may seize upon the obligations
or forms imposed by the state to do things that were not envisioned by
the framers of those forms. The state, moreover, also ‘‘incorporate[s] ele-
ments of counter-hegemonic cultures’’ in the interests of advancing some
other agenda or as a mechanism of achieving ‘‘legitimacy.’’ Indeed, ‘‘the
hegemony of the state is also exactly what is most fragile about the state,
precisely because it does depend on people living what they much of the
time know to be a lie.’’ 89
How does a theoretical formulation of the state as a dynamic and con-
tested process help us to understand the relationship between the state
and violence in Colombia? First, by treating the state as a claim that is
constantly being constructed and negotiated rather than as an immu-
table, ahistorical thing, it is possible to consider the existence of com-
peting claims or states and the role such competition may have played
in the development of violence. A struggle between two competing state
claims, or two hegemonic projects, if you will, is precisely what I argue
occurred in Antioquia in mid-century. There existed, on the one hand,
a regional claim whose ruling practices—constructed over the course
of the late-nineteenth and early-twentieth centuries—were characterized
by a suprapartisan, pragmatic, technocratic rule, and an emphasis on ma-
terial development at the expense of rigid partisan ideology.90 It was an
elite-led, paternalistic form of rule in which popular participation was
limited, but it promised some social protection, education, employment,
mobility, public investment, and development at a time when the cen-
tral state was not yet in a position to guarantee these. In return, the re-
gional state demanded of its citizens conformity (or the illusion of it,
the ‘‘performance’’ to which Sayer alludes) to a specific set of values such
as Catholic ritual observance, marriage, work discipline, capitalism, and
 Introduction 

political moderation. A bargain emerged in Antioquia that guaranteed

a modicum of order and regularity of rule, but it operated only where
the values the regional state claimed to embody found material form,
that is, where access to property ownership or mobility was possible, an
extensive nuclear family structure existed, and a strong sense of Catholi-
cism operated. The linchpin of ‘‘order’’ was the prioritization of regional
and economic interests over partisan differences and the containment of
the expression of partisan differences within boundaries that would not
challenge the regional status quo.
The strength of the regional state claim precluded the emergence or
viability of a central state claim in Antioquia until the s. One could
even argue that the politics of convivialismo, based on the alliance of cof-
fee producers and merchants that characterized the period from  to
, represented a moment in which conflict between the central and re-
gional state projects was minimized because these were one and the same.
The period between  and —one of the few periods in Colom-
bian history when Antioqueño politicians occupied the presidency and
played a visible role in national politics—may be read as a moment in
which Antioqueño elites attempted (but failed) to remake Colombia in
their own idealized image. Changes in Colombia’s suffrage law, the rise of
the Liberal party to national power, and the expansion of what until then
had been a weak and largely ineffective central state, however, brought
into competition and conflict the regional and national projects of rule.
Indeed, one of the most important effects of this competition between
distinct state claims was that partisan clientelism threatened to eclipse
a suprapartisan or bipartisan regional model of the state. The regional
state model of rule mediated inclusion in the state through patron-client
relations embedded in economic associations (the Federation of Coffee
Growers, for instance), kinship relations, shared local origin, and the ap-
pearance of satisfaction of an idealized regional regime of cultural confor-
mity, but not necessarily, or only secondarily, through either the Liberal
or Conservative parties.91
The limited significance of partisan clientelism in Antioquia before the
s was due mainly to the availability of economic avenues of mobility
whose access was not primarily or exclusively dependent upon partisan
affiliation (such as coffee production, mining, and commerce). It was due
as well to the persistence of a regional vision of government as techni-
cal management rather than what elite Antioqueños disdainfully referred
 Blood and Fire

to as politiquería, or politicking. This was not altruism on the elite’s part

per se, but rather the result of fearing class conflict more than privileg-
ing partisan loyalty. The dominance of private over public investment or,
rather, the complexly intertwined nature of public and private spheres
in Antioquia limited the mobilization and integration of regional clien-
talist networks through the central state.92 Antioqueños who relied on
state employment were obviously part of partisan patron-client machines
(teachers, municipal employees, and public works personnel), but these
were often regulated by regional rather than national dictates, and pa-
tronage positions constituted a small percentage of overall employment
in the region before .
A regional tradition of not overly privileging partisan affiliation in the
distribution of even state-determined employment was so strong, more-
over, that it could still be found operating even in the midst of la Violen-
cia when partisan competition over patronage distribution became most
acute. Antioqueño Conservatives reminded their Conservative governor
in  of an implicit agreement not to ‘‘take political reprisals against
workers and lower-ranking employees since the individuals [Liberals]
who are detrimental to the government’s party are those that hold high
ranking positions.’’ 93 Even those most sympathetic to Conservative ex-
clusionary rule insisted to the governor that he ‘‘use all the means at
[his] disposal to ensure that men who fulfill their social obligations not
be deprived of work or thrown into the street just because they oppose
our political creed, since most of these men are fathers.’’ 94 At the apex
of la Violencia, the maintenance of a gendered social system of family-
based capitalist integration took precedence over partisan concerns in
Antioquia. In areas of the economy (agriculture, ranching, and mining)
where capitalists belonged to different parties but shared economic inter-
ests, moreover, hiring was neither contingent upon partisan affiliation
nor upon voting for a particular party.95 Indeed, the apparent indifference
with which the average Antioqueño approached the question of partisan
politics was significant enough to prompt acute concern among politi-
cians anxious to replace the regional model of suprapartisan rule with
that of partisan clientelism during la Violencia. They despaired publicly
of ‘‘the excessive insistence of our working people on simply economic
affairs’’ and lamented that ‘‘Antioquia’s human groups play so small a role
in the struggle between the parties.’’ 96 This tendency in Antioquia con-
trasted with that of other Colombian departments where employment
 Introduction 

and survival had been dependent upon public or state patronage hiring
and was indexed to patron-client networks mediated through the parties
since the nineteenth century, long before the emergence of an identifiably
important central state.
The claim of the central state contrasted sharply with that of the re-
gional state. The central state project was not predicated upon confor-
mity to a specific set of cultural, economic, or social values in the way
that the regional state model of rule was. Participation in the central
state was technically open to any adult male simply by virtue of having
been born on Colombian territory. Despite the more inclusive nature of
the central state ruling project, however, the state’s inability to consis-
tently enforce its presence at either the regional or municipal level di-
minished its appeal among Antioqueños. While the central state’s prom-
ise of labor and social legislation, land reform, and expanded political
participation was certainly embraced by regional inhabitants, the central
state’s inability to make good on its promises undermined its potential
base of support. Reluctance to identify with the project of the central state
was particularly pronounced among Antioqueños residing in centrally
located areas where the regional state exerted a strong presence and re-
sponded with reasonable agility to local demands and needs. For, while
the regional state ruling project was predicated on conservative notions
of ‘‘respectability’’ and ‘‘social conformity,’’ Antioquia’s political leaders
were in some respects economically and socially progressive. They could
be tiresomely paternalistic, but they were aggressive builders of schools,
factories, health facilities, and roads. Access to the benefits of paternalistic
rule, moreover, was not predicated primarily on shared partisan affilia-
tion. Discipline and a willingness to work were held in far higher esteem
than partisanship, while access to individual mobility (though not in-
clusion in the elite) was based on the appearance of cultural conformity
and merit. Antioquia’s elite was not egalitarian nor did the hegemonic
bargain implied by an exchange of education, employment, and limited
political access in return for apparent compliance represent an equal ex-
change between regional inhabitants and their leaders. But, the ‘‘bargain’’
struck between central core inhabitants and Medellín’s elite did represent
an exchange, one that typically had a better chance of being partially ful-
filled than did comparable exchanges between the central state and local
citizens. Such an attitude contrasted sharply with the governing style of
other Colombian regions or even the central state. In centrally settled
 Blood and Fire

areas, moreover, conformity to regional state ruling practices and values

guaranteed the enabling power that Sayer notes is the flip side of coercion.
Inhabitants in core municipalities could parlay their conformity to re-
gional ruling practices into demands that the regional state take their calls
for political recognition and inclusion seriously. Further, they could de-
mand—and expect—that the regional state prevent violence from jeop-
ardizing local economic prosperity or the status quo regardless (in most
cases) of their partisan affiliation.
The national state model of rule, in contrast, was most appealing to
those sectors of Antioqueño society who in the s benefited from the
central state’s expanded control of patronage and recently achieved regu-
latory functions. The growth of the central state coincided with the Lib-
eral party’s rise to power. Thus, members of the Liberal lower-class and
emergent middle-class politicians in Antioquia were initially integrated
into the central state’s ruling project through the expansion of state em-
ployment and the recognition and co-optation of organized labor.97 The
central state project also appealed to Antioqueños left out of the regional
state model of political rule, that is, the majority of the inhabitants resid-
ing in the region’s periphery, including important sectors of unionized
labor employed in foreign-owned industries such as mining and oil pro-
duction. These sectors, in addition to identifying with the Liberal Party
or parties friendly to the left wing of the Liberal party (such as the Com-
munist or Socialist parties) defied or failed to conform to the cultural
values which underpinned the regional state ruling project. The failure
to reproduce the values associated with Antioqueñidad barred peripheral
inhabitants from participation in the hegemonic bargain that governed
relations between central core inhabitants and the regional authorities.
Peripheral areas, moreover, were ones where the regional state was either
absent, weak, or present only as a repressive force.
In sum, clientelism and the competition over the state did indeed play
central roles in the definition of violence in Antioquia as they did else-
where in Colombia during la Violencia, but the reasons why they did are
specific to Antioquia and must be understood at the regional level. In the
areas where the regional state was strong and enjoyed legitimacy, partisan
violence never threatened the status quo and was largely avoided or was
mediated in nonviolent ways. In the areas where the regional state’s re-
lationship with the local citizenry was hostile and intermittent, partisan-
based clientelist networks and a central state project clashed with supra-
 Introduction 

partisan clientelism and a regional ruling project, provoking a violent

conflagration that precluded the possibility of mediation. It was in geo-
graphically peripheral areas where the regional and central state claims
and their respective clientelist networks came into severe competition
and formed a significant catalyst to violence.

Ethnicity, Culture, and Core and Periphery Violence

In her work on frontiers and peasant protest, Catherine LeGrand sug-

gested a close connection between the areas where la Violencia was most
severe in the s and those experiencing land conflicts in the s
and s. Regional studies of la Violencia confirmed that a relationship
existed between land issues and violence, but the coincidence between
conflicts over land and partisan violence was considered to have occurred
primarily in coffee-producing areas.98 A very clear correlation exists be-
tween the areas experiencing the most severe violence and those where
land struggles had occurred in the s and s in Antioquia.99 How-
ever, the incidence of land and labor restructuring did not typically occur
in Antioquia’s coffee towns.
One explanation for the apparent discrepancy in experience between
Antioquia and other regions in Colombia may have to do with the loca-
tion of frontiers and regional migration patterns. While coffee fron-
tiers were still open and recent migration from a variety of departments
characterized provinces where violence was also severe such as Valle
del Cauca, Tolima, and Viejo Caldas (the contemporary departments of
Caldas, Quindío, and Risaralda), in the thirties and forties, the coffee
frontier in Antioquia had effectively closed by the early part of the twenti-
eth century. The market for land in Antioquia’s traditional coffee belt was
relatively static, while coffee lands in places like northern Valle, Tolima,
Quindío, and Caldas were still volatile. Another factor that shaped the
different geographic focus of otherwise similar conflicts in Antioquia and
other departments identified as coffee producers had to do with ethnic
and cultural differences between the different groups colonizing areas
where violence was most pronounced.100
The peripheral areas experiencing the most acute violence in Antio-
quia shared several features which distinguished them in important ways
from the settlement and dominant production and land-tenure patterns
evident in the central or core zones of Antioquia. All peripheral areas
 Blood and Fire

bordered on departments perceived to be ethnically and culturally quite

different from Antioquia or at least from an ‘‘imagined’’ Antioqueño
ideal. Urrao bordered on the Chocó, the Pacific-lowland province with
the greatest Colombian population of African descent. Bajo Cauca bor-
dered on what was then the province of Bolívar (today Bolívar, Sucre,
and Córdoba), an area connected to the Afro-Caribbean coast. Urabá
bordered on both the Chocó and Panama and opened out to the Carib-
bean Sea, while the northeast and Magdalena Medio towns bordered on
the Magdalena River and the provinces of Santander, Cundinamarca, and
Boyacá. (The latter two departments being important areas of indigenous
settlement.) Ironically, since Antioquia was known for its colonizing
vigor, all of these areas (with the exception of Urrao) were colonized in
large part by non-Antioqueño migrants from the Afro-Caribbean coast,
Chocó, Bolívar, and Santander.
Before the s, Antioquia’s peripheral zones (all but Urrao’s tropical,
lowland areas) had held little attraction for Antioqueño settlers from the
mountain valleys and highlands of centrally settled municipios. Regional
myth attributed the reluctance of Antioqueños to venture into the north-
western and northeastern areas of their department to fears of the insalu-
brious climate and the presence of ‘‘wild’’ inhabitants in outlying areas.101
The myth that such places were empty but for a few scattered barbari-
ans and intrepid miners had long dominated the regional imagination.
Or, as former president Carlos E. Restrepo bluntly put it when explaining
why the Bajo Cauca and Urabá were unappealing sites for colonists from
the central region, ‘‘Antioqueños, like the Swedes and the British, can
only work where they establish their homes [hogar] [and] home cannot
flourish where malaria exists.’’ 102 Even before Carlos E. Restrepo penned
these lines in , however, Antioqueños shifted their colonization route
away from the traditional southern coffee belt and toward distant west-
ern and eastern sections of their homeland and the lower Cauca Valley.103
Colonists from centrally settled municipios arrived in peripheral areas
as squatters hoping to work what they assumed were nearly unlimited
public lands, or as cowboys, miners, and public works personnel on the
railroad and newly begun, state-financed road projects. When it became
apparent that many so-called tierras baldias were actually claimed by
large capitalist concerns, conflicts between squatters, colonists, mining
companies, and large landowners ensued.104
Unlike the inhabitants of centrally settled towns, the majority of non-
Antioqueño residents in peripheral zones were unwilling to behave in
 Introduction 

ways that made reciprocity between the regional state and the people of
the core municipalities possible. The regional state’s historic absence or
intermittent presence only as a punitive force in such areas also meant
that none of the infrastructure, public investment, and institutional pres-
ence that integrated peoples in the central zone to the regional state was
present. The construction of a relationship of hostility and distrust be-
tween the regional authorities and the periphery was intimately related
to the historically colonial relationship forged between core and periph-
ery. Local inhabitants of peripheral areas viewed Medellín and migrants
from core municipios as arrogant interlopers who considered themselves
both ‘‘whiter’’ and more civilized than non-Antioqueño migrants, while
the authorities and inhabitants of Antioquia’s traditionally settled areas
dismissed the inhabitants of the periphery as everything they perceived
themselves not to be: lazy, unruly, promiscuous, irreligious, and shifty.
The periphery was linked to disorder in the minds of the regional au-
thorities and centrally settled inhabitants and was thought to be in need
of morality and control (by force if necessary). In contrast, the center was
perceived as absolutist and exclusionary by peripheral inhabitants, and
responsive only to demonstrations of local defiance and violent threat.
But mutual distrust and antipathy between the core and periphery in
Antioquia existed long before the advent of la Violencia and alone would
have been insufficient to catalyze intense violence in peripheral areas.
However, the construction of stereotypes of cultural difference gained
new importance when peripheral areas emerged as Antioquia’s most eco-
nomically dynamic and valuable in the decades of la Violencia. On the
one hand, the stereotype of the periphery as a site of chronic misrule be-
came a justification for the regional state’s refusal to engage in the politics
of negotiation and compromise characteristic of the state’s interactions
with residents of central areas. On the other hand, local perceptions of
the regional state as a colonial and repressive force legitimized the use of
defiance by inhabitants of the periphery to counter regional attempts to
impose partisan hegemonic control.
To understand the nature of violence in peripheral regions within
Antioquia during la Violencia, then, one must acknowledge the inequali-
ties of power embedded in colonialism.105 Like colonies and imperial
metropoles everywhere, this relationship was steeped in fantasies of ex-
tractive wealth, political domination, and cultural subordination. The
latter were expressed and rooted in a regional discourse historically based
upon hierarchies of cultural difference that segregated Antioquia into
 Blood and Fire

centrally located areas perceived to conform to a regional value system

and peripheral areas perceived to deviate from it. In this study I define
‘‘centrally based municipios’’ as those located in the southwestern coffee
region—the embodiment of Antioqueño values and comportment—the
north as far as Yarumal, the south to Abejorral, and the east as far as Santo
Domingo. A town such as Urrao on the border between the southwest-
ern coffee district and the western towns of Dabeiba, Frontino, and the
region of Urabá, would occupy an intermediate position, a buffer zone
between the values of Antioqueñidad and external threats to the integ-
rity of regional identity and order. The Magdalena River Valley, the lower
Cauca Valley, the mining regions of the northwest and the southeastern
areas of Antioquia (San Luis, Cocorná), in contrast, formed part of the
unstable zone I refer to as the ‘‘periphery.’’
Labels such as non-Antioqueño, ‘‘costeño,’’ ‘‘negro,’’ and ‘‘cosmo-
polita,’’ that is, nonwhite, were used to legitimize marginalization or ex-
clusion and were coded to a series of attributes or patterns of behavior
that might or might not characterize peripheral inhabitants but which
had come to constitute a frame of reference that Antioquia’s core in-
habitants, authorities, and elite used to describe the ‘‘other.’’ 106 These be-
haviors typically included sexual partnerships that took the form of free
union (rather than Catholic marriage), physical impermanence (seasonal
migration, transience, vagabondage), collective cultivation (rather than
privately owned plots), a tendency to embrace dissident political move-
ments, and the practice of folk rather than institutionalized religion.107
These were attributes believed to contradict and endanger the ideals as-
sociated with regional identity or Antioqueñidad.
In most instances, difference was conflated with deviance, criminality,
and corruption: the ‘‘other’’ threatened the stability of Antioqueño iden-
tity, authority, and prosperity. ‘‘Cultural competence,’’ or the satisfaction
of norms of ‘‘respectability,’’ was in turn linked to ‘‘cultural milieu,’’ the
idea that ‘‘racial and national essences could be secured or altered by
the physical, psychological, climatic, and moral surroundings in which
one lived.’’ 108 In spatial and ideological terms, Medellín and the cen-
trally located towns over which it governed met the criteria of Antio-
queñidad. These were areas defined in official discourse as being peopled
by ‘‘individuals of noble race, strong, healthy, valiant and hardworking,
the birthplace of liberators and heroes.’’ 109 Peripheral or frontier towns
located in the northwest (Urabá), Bajo Cauca (Caucasia), and Magda-
lena Medio (Puerto Berrío, Maceo) were, in contrast, tropical lowland
 Introduction 

areas of African, Indian, or non-Antioqueño migration and settlement

(see appendix A.). The inhabitants of these areas were imagined by cen-
trally based Antioqueños as ‘‘sickly,’’ ‘‘full of indolence,’’ and full of ‘‘a
passionate nature and inconstancy, superstitious in spirit, [and] predis-
posed to fetishism and anarchy.’’ 110 More importantly, peripheral areas
were not characterized by the presence of a smallholder landowning tra-
dition, a strong local church that could broker local interests, a resident
elite linked to powerful producer associations, or political representatives
integrated into the bipartisan networks of Antioqueño rule.
The regional state project of hegemony, constructed and deployed by
Antioquia’s men of capital and political leaders, was based on the mainte-
nance of hierarchies of difference.111 The appeal of such norms extended
beyond the elite. Tropes of supposed cultural difference could be de-
ployed, as they were during la Violencia, by the central zone peoples of
lower-class origin to justify homicide, usurpation of property, and rape
against the ‘‘coastal peoples’’ (costeños) and ‘‘revolutionaries’’ (revolucio-
narios) of a similar social level. The existence of an ‘‘other’’ was also used
to construct and reinforce the central zone inhabitant’s sense of positive
identity (‘‘I am this because I am not that’’) and deployed as a mecha-
nism of negotiation when bargaining for recognition or political inclu-
sion with the regional elite. The ruling forms embedded in the regional
hegemonic project thus enabled non-elites to legitimize violence in the
name of protecting regional values or Antioqueñidad.
In suggesting the existence of a regional hegemonic project, and the
use of stereotypes of cultural difference to further it, I am not arguing that
there really existed any distinction between peripheral and central areas
and their values, or that even where observable differences of production,
organization, and belief were present, that these were static or inherent.
Rather, I am marking the construction and manipulation of dynamic,
profound, and widespread preconceptions of difference and identity and
signaling their political and social repercussions. The geographer James
Parsons once noted that ‘‘contemporary Antioquia was shaped out of an
initial mixing of Spaniards, Indians and Black slaves,’’ and yet Antio-
queños embraced ‘‘an ethnological heresy by which the inhabitants refer
to themselves as the Antioqueño race [la raza Antioqueña].’’ 112 Belief in
themselves as members of a separate race (defined by the norms of re-
spectability I have already mentioned) was ‘‘firmly rooted in popular
consciousness,’’ Parsons argued, even though by the s, when he con-
ducted his study of Antioqueño colonization, regional censuses showed
 Blood and Fire

that ‘‘the preponderance of mixed blood . . . stood in flagrant contradic-

tion of the assertion that Antioquia is a province of whites.’’ 113
Among the core municipalities of Antioquia there emerged over time
a sense of region and regional identity comparable to what Benedict
Anderson has dubbed the ‘‘imagined community’’ that he argues en-
abled the emergence of nations and national identity.114 In Antioquia,
however, cultural distinctions were constructed and deployed to char-
acterize particular areas within the region and imbue them with sym-
bolic meaning as part of a larger process of constructing regional identity
and power against both peripheral populations and the larger Colombian
nation. Communities create boundaries and oppositions against which
their identity may form and be marked. For Antioquia’s elite and political
authorities, the limits of a regional community were drawn around spaces
that were long-held objects of desire. These were areas of strategic im-
portance, characterized by natural resources and economic potential to
enrich and extend Antioqueño power, but which, for various reasons, had
historically proven difficult to control or resistant to Antioqueño cultural,
political, and economic domination.115 Peripheral areas were the sites in
which the parameters of regional identity and authority were fought over
and shaped and where violence became endemic and widespread.116
An adequate understanding of la Violencia in Antioquia thus requires
the exploration of characteristics and conditions that transcend parti-
san considerations. What was the ethnicity of the people in conflict?
What cultural patterns and expectations ruled their behavior? For how
long had the zone been occupied, and where did the settlers come from?
Where was the village located in relation to the older core of the depart-
ment? What were relations between Liberals and Conservatives in the
area before the emergence of la Violencia? What state institutions oper-
ated in the area, through what agents, and to what effect? What were the
predominant economic activities in the area, and how were local inhabi-
tants connected to them? What relative degree of Antioqueñidad can be
ascribed to the groups in conflict? An examination of these and other ele-
ments, their interaction, and their evolution through time, occupies the
body of this work.

The Book’s Organization

In the first chapter I trace the attempts, from  to , by emer-
gent Conservative middle-sector politicians to conservatize Liberal mu-
 Introduction 

nicipalities and to replace Liberal workers in state employment posi-

tions with Conservative followers.117 I also examine why public works
employment and competition to control patronage positions in Antio-
quia became so critical to the consolidation of electoral fortunes in the
s. When conservatization efforts met with limited success, sectors of
the regional state who were followers of Conservative leader Laureano
Gómez created paramilitary forces (contrachusmas) and deployed them
into peripheral areas where the state had little support but where state-
determined patronage posts were most concentrated. The official use of
systematic violence provoked a rupture within the regional Conserva-
tive party that surfaced in the conflict between a regional suprapartisan
political tradition and the new politics of national, state-sponsored, par-
tisan hegemony. Chapters , , and  are, respectively, detailed analyses
of the evolution and impact of violence in three peripheral zones: Urabá
and western Antioquia; Urrao and the Southwest; and eastern Antioquia
(the Bajo Cauca, Magdalena River Valley, and the Northeast). These were
areas where armed Liberal guerrillas emerged to resist the Conservative
national government between late  and . I examine how the re-
gional and national states responded to violence in each of these areas,
where the state’s monopoly over force was transferred to paramilitary
organizations, the reasons why this happened, and the long-term impli-
cations of such a course of action. I also explore the differences between
different guerrilla groups from region to region, how partisan conflict
intersected with latent tensions over land, labor, and resources in some
areas, and explore the factors that impeded the possible mediation of vio-
lence. I argue that violence in peripheral areas was largely the product
of concerted and systematic harassment waged by selected regional au-
thorities rather than the ‘‘natural’’ outgrowth of partisan conflicts among
local residents. In other words, the regional state and its forces were the
primary instigators of violence on the periphery, and their object was not
just the establishment of partisan hegemony but the forcible imposition
of Antioqueñidad. Local resistance to the regional state was thus waged
not only along partisan lines but also involved struggles over the right
to cultural self-determination and the articulation of alternative concep-
tions of citizenship and identity. The epilogue concludes with a reflection
on the relationship between current states of ‘‘disorder’’ in Antioquia and
la Violencia, in particular the consolidation of private and paramilitary
forms of terror in contemporary Colombia.
Pavarando, municipality of Murindo,
Urabá. April 1998. Government officers
visit this displaced persons’ camp.
1. Medellín and Core Municipalities

Violence in Antioquia developed in two major stages over a seven-year

period. The first stage began in  and ended in , the second began
in  and lasted until the military toppled the Conservative govern-
ment of Laureano Gómez on June , . During the first phase, vio-
lence revolved around three central objectives: achieving Conservative
electoral domination in municipal, regional, and national elections; re-
placing Liberals in state-determined patronage or administrative posi-
tions; and crushing organized labor affiliated with the Confederation of
Colombian Workers ().
Violence during the first stage was selective and sporadic; it was aimed
primarily at Liberals employed by the state and at towns where the level of
integration between the municipal and regional government suggested
that officially executed harassment, fraud, and intimidation were likely
to win an electoral advantage for the Conservative party. The towns most
severely affected by electoral violence were thus concentrated in cen-
trally settled or core municipalities where the Conservative party was well
represented, local government was linked through a variety of offices and
mechanisms to regional administrative power structures, and local resi-
dents reproduced most closely the values associated with Antioqueñidad.
This early phase of violence initially differed very little from the age-
old struggles over offices and votes typical of Colombian political com-
petition since the nineteenth century. Two issues, however, shifted the
course of partisan violence and considerably intensified its potential re-
verberations: the assassination of Liberal populist Jorge Eliécer Gaitán
in April  and the creation of state-endorsed Conservative civilian
police forces (contrachusmas) in the aftermath of the murder. Further,
when official intimidation failed to wrest electoral victories or jobs away
from the opposition in early , the measures adopted by selective sec-
tors of the regional government and the Conservative party in their quest
for political domination became more extreme. The escalation of parti-
san violence provoked severe dissension within the regional Conservative
and Liberal parties and revealed the existence of a threshold or bound-
 Blood and Fire

ary beyond which many Antioqueños were unwilling to go in the pursuit

of partisan objectives. The intensification of state-directed violence also
exacerbated tensions between municipal and regional authorities over
issues of prerogative and jurisdiction and, ultimately, prompted Liberals
in peripheral areas to take up arms against the state.
This chapter lays out the complicated and, at various points, contested
path by which partisan conflict evolved into armed confrontation be-
tween the state and the opposition and attempts to uncover the ways in
which nonpartisan issues were implicated in Liberal/Conservative con-
flict. La Violencia intersected with and magnified a series of latent and
unresolved struggles concerning the power of the state, economic devel-
opment, and class relations in Antioquia. As familiar partisan tensions
threatened to develop in directions that challenged the social, economic,
and political status quo, a still-influential elite sector intervened to medi-
ate the consequences of violence in core municipalities. By  it became
evident, however, that violence waged over elections and appointments
had evolved in focus and location and transcended the parameters of tra-
ditional partisan conflict.

Conservative Clientelism and Electoral Violence

On August , , moderate Conservative, Mariano Ospina Pérez was

sworn in as Colombia’s president after an intensely contested electoral
battle. Two Liberal candidates, Liberal populist Jorge Eliécer Gaitán and
the official Liberal party candidate, Gabriel Turbay, split Colombia’s Lib-
eral majority and brought to an end sixteen years of uninterrupted Lib-
eral control of the presidency. In his native state of Antioquia where
Conservative voters still slightly outnumbered Liberals, Ospina proved a
popular candidate among the region’s businessmen, coffee growers, and
industrialists who formed the core of the region’s elite. Local capital-
ists hoped that Ospina’s victory might signal a return to the politics of
bipartisan compromise, a restoration of the pre- elite alliance that
many remembered nostalgically as an era of social peace unmarred by
the political demands of the lower classes or emergent professional politi-
cians. Antioquia was a province where career politicians were viewed with
suspicion and where elite Liberals and Conservatives had constructed
and promoted an image of themselves as selfless statesmen whose politi-
cal participation was motivated by civic duty, not vile ambition.1 Ospina’s
 Medellín and Core Municipalities 

lack of direct experience in electoral politics, his impeccable business cre-

dentials, and his ideological moderation appealed to a regional bourgeoi-
sie that had grown wary of what they perceived as increasingly extreme
tendencies in both parties and the rising influence of parvenu political
ambitions.
Among other Colombian Conservatives, however, Ospina Pérez was
less popular. Indeed, his candidacy had been the result of reluctant
compromise within the national party leadership, many of whose fol-
lowers would have preferred to have made the brilliant but controversial
national party chief, Laureano Gómez, their official candidate. Gómez
had made a name for himself as a fierce nationalist, an outspoken ad-
mirer of Franco’s Spain and a critic of Communism, an unabashed de-
fender of Hispanic values, and a corporatist for whom the increasing
secularization of modern society and mass participation in politics were
anathema. The Conservative party chief was also a master orator, an ideo-
logical incendiary whose followers revered him with an intensity that
bordered on religious fervor, but his intransigence elicited strong re-
actions from Antioquia’s Conservatives as well as the Liberal opposi-
tion. Gómez’s often mystical and violent rhetoric particularly frightened
Antioquia’s Conservative moderates who feared the repercussions of pur-
suing the hegemonic partisan objectives the party leader and his followers
embraced. Temporarily thwarted from the presidency by moderates in his
party, Gómez resigned himself to ensuring that those loyal to him would
exert instead a powerful presence in local and regional politics during
Mariano Ospina Pérez’s administration.
In contrast to members of the regional elite, Antioqueño non-elite
politicians, many of whom formed part of an incipient middle sector of
first-generation professionals who had never previously held office, re-
pudiated moderation.2 The middle sector believed bipartisan or supra-
partisan arrangements concentrated power in the hands of a small,
wealthy minority and relegated ideological differences between the par-
ties to a secondary plane in the interests of promoting common eco-
nomic objectives. Many middle-sector Conservatives considered Gómez
a powerful ally in their ploy to undermine the region’s bipartisan ap-
proach to politics. The Conservative leader had publicly attacked con-
vivialismo, the system of elite bipartisan cooperation that shaped Colom-
bian politics from  to  and in which Antioquia’s elite had played
a central role. Indeed, Gómez implicitly blamed the Conservative party’s
 Blood and Fire

fall from national power in  on the willingness of Antioquia’s elite

to sacrifice ideology in the interests of economic modernization. When
members of Antioquia’s Conservative party collaborated with the Lib-
eral government of Eduardo Santos (–), for instance, Laureano
Gómez excoriated them mercilessly as mercenaries interested only in de-
fending their economic interests at the expense of both party doctrine
and non-elite party members.3 Antioquia’s middle-sector Conservatives,
who sought to distinguish themselves from the elite, adeptly invoked
Gómez’s inflammatory anti-Liberal and anti-Communist rhetoric, took
up the banners of militant Catholicism, and equated the politics of com-
promise with a utilitarian conception of society and a lack of party fervor.
Ideological differences were then deployed to justify wresting control of
public patronage positions away from the opposition and the reins of
regional power away from moderates in both parties.
Three significant changes took place during the period of Liberal rule
between  and  that sharpened distinctions between elite and
emergent political forces within the parties and made competition for
control of state patronage after  an unprecedented component of
electoral victories in Antioquia. First, the restructuring of capital and
production following the  stock market crash and the ensuing Great
Depression prompted migration from coffee-producing and traditional
agricultural areas such as those located immediately to the south and
east of Medellín toward the provincial capital and new areas of colo-
nization on Antioquia’s periphery. The specter of social conflict arose
in newly colonized areas as squatters, recent migrants, foreign-owned
companies, and regional capitalists competed for control in areas where
political mechanisms of expression and infrastructure were weak or non-
existent. Simultaneously, young, provincial aspirants to power migrated
to Medellín in search of educational, economic, and political opportu-
nities in unprecedented numbers. These developments were not neces-
sarily or directly a consequence of the rise of the Liberal party to power,
but the temporal coincidence of the Conservative Party’s political decline
and an increase in social and economic dislocation within Antioquia pro-
duced the appearance of a causal relationship, particularly in the eyes of a
nervous elite worried about populist and radical threats to their political
control.
Second, when Liberal Alfonso López Pumarejo’s first administration
(–) passed legislation expanding the size of Colombia’s electorate
 Medellín and Core Municipalities 

and recognizing and institutionalizing the political importance of orga-

nized labor, the tenor of regional political debate changed. Before the im-
plementation of López’s legislative changes, class-based appeals directed
specifically at the needs or interests of the lower and especially the urban
working class had simply not constituted an important component of
political discourse in Antioquia. The newly enhanced presence of Com-
munist and left Liberal leadership in labor unions, and the dissemination
of political programs directed specifically at workers and the disenfran-
chised poor, based on their identity as members of a distinct class rather
than on their regional identity as suprapartisan Antioqueños, brought
into sharp relief the distinctions between ‘‘right’’ and ‘‘left.’’ 4 The acrimo-
nious character of political debate between members of the two parties
increased, as did divisions within the parties between an elite alarmed by
López’s reforms irrespective of their partisan affiliation and middle- and
lower-class sectors in both parties, some of whom viewed ideologically-
based partisan differences as an opportunity to press for political inclu-
sion.
Finally, the central state’s power and attributions grew under López
Pumarejo. As the state’s power expanded, the Liberal party worked to
create links between the party and the state that would enable Liberal
brokers to control the distribution of public patronage jobs, access to
government intercession, and public monies. As new ministries such as
Labor (trabajo) emerged, and as individuals organized into labor unions
affiliated with the Confederation of Colombian Workers, the importance
of brokers who could control lower-class voters, especially urban voters,
increased dramatically.5 The expansion of public works and patronage
jobs that accompanied the electoral and political changes introduced by
López prompted excluded regional Conservatives to associate partisan-
determined patronage hiring as a policy identified with—and crucial to
the maintenance of—Liberal domination. By the eve of the Conservative
party’s return to power in , control of the central state’s largesse and
the monopoly of patronage distribution thus emerged to a previously un-
precedented degree as crucial preconditions of political success in Antio-
quia. This phenomenon drove a sizable wedge between middle-sector
politicians anxious to build a political machine like that constructed by
their Liberal counterparts and a regional Conservative party elite anxious
to restore the status quo of the pre- era.
The generation of men who had governed Antioquia’s political for-
 Blood and Fire

tunes from the turn of the twentieth century to  was not primarily
or exclusively dependent upon political careers for its place in society
or political influence. Instead, these men combined political leadership
with their roles as captains of industry, financiers, and coffee export-
ers and growers. They relied on familial connections and reputations as
nonpartisan ‘‘statesmen’’ committed to technocratic and development-
oriented policy-making to cement their political claims. Antioquia’s elite
was more concerned with keeping social unrest at bay than in monopo-
lizing government positions; they attended the same schools, shared the
same professions, and married into one another’s families across party
lines. Despite differences in partisan affiliation (which were often the re-
sult of arbitrary or serendipitous family traditions dating back to the in-
dependence era and the nineteenth century) many members of Antio-
quia’s elite shared a worldview. This worldview was shaped by a shared
education under the Jesuits at schools such as San Ignacio in Medellín or
at the public Universidad de Antioquia and the regional School of Mines.
The shared experience of studying under the Jesuits may also partially
account for the frequency of anti-Communist discourse, as well as the
invocation of the ideals of social justice, as promulgated in papal encycli-
cals such as Leo XIII’s Rerum Novarum, among Antioquia’s elite political
leaders of both affiliations.6
The existence of a relatively cohesive regional approach to politics
among Antioquia’s elite leaders born before  ensured that no definite
line divided the public from the private sector while the regional bour-
geoisie ruled the department. In moments of crisis a governor such as
Camilo C. Restrepo (Liberal) might appeal to other members of the elite
from both sides of the partisan divide, for instance, to bail out the depart-
mental treasury and avoid a rebellion by unpaid public-sector workers,
without putting up any other collateral than his word as a gentleman—
an action largely inconceivable in any other Colombian department. Al-
though Antioquia’s elite belonged to both parties and had philosophi-
cal differences, Liberal and Conservative elites shared an understanding
of government as a technocratic endeavor, and their central concern
was maintaining the status quo. Furthering economic development and
maintaining social stability defined the regional political agenda.
In contrast to the bipartisan elite, emergent middle-sector politicians
were neither wealthy nor well connected. Nor was the emergent middle
sector inscribed in networks of familial, business, educational, profes-
 Medellín and Core Municipalities 

sional, or social associations with the opposition to the same degree that
the elite was. The only means open to the middle sector in its quest
for political power, moreover, was to generate votes, preferably among
a newly enfranchised, lower-class population of urban voters. Biparti-
sanship was of course antithetical to such an enterprise. Unlike the re-
gional elite who once depended on deference, paternalism, and kin-
or geographically-based clientelist networks to attract voters, emergent,
middle-sector politicians had to exaggerate, not underplay, partisan dif-
ferences in order to distinguish themselves from politicians of compa-
rable social extraction in the opposition. In the thirties and forties, as
never before in Antioquia, politics revolved around the struggle to amass
voter support and to exchange this for patronage positions and subsidies
provided by an expanding central state.
The coincidence of demographic change and the suffrage reforms in-
stituted under López thus made political competition more acute after
. Medellín’s population quintupled between  and , rising from
, inhabitants in  to , in , and on to , in .7
By , Antioquia’s urban areas were growing at a far quicker rate than
the region’s rural areas. While the province’s number of urban inhabitants
grew  percent between  and , the number of rural residents in-
creased by only  percent during the same period.8 As Medellín grew,
so did the number of potential voters and provincial migrants in search
of education and a chance to break into regional politics.9 The sudden
growth in voters, the demographic shift to the city from the countryside,
the expansion of the role and influence of the central state, and the pas-
sage of favorable laws under the Liberal administration of Alfonso López
Pumarejo all conspired to up the ante (and the possibility of conflict) in
the Antioqueño political arena by .
It would be misleading, however, to suggest that naked ambition and
opportunism alone shaped the urgency with which Conservative middle-
sector politicians tackled the problem of public patronage positions and
electoral contests in Antioquia after . Real ideological differences
existed between some Liberals and Conservatives, and the Conservative
preoccupation with the moral and political impact of radical, materialist
ideologies on Colombian politics and society also influenced the zealous-
ness with which the extremists attacked the Liberal opposition. Industri-
alization and the birth of an identifiably urban working class were recent
phenomena in Medellín. The repeated booms and busts of the  to
 Blood and Fire

 period and the social and economic dislocations to which these gave
rise formed the backdrop of most emergent politicians’ political coming
of age. Although in real terms the Communist party posed little real
threat in Antioquia (or Colombia), the presence of real social discontent
and the rapid economic transformation over the previous decade enabled
right-leaning Conservatives to capitalize on inchoate anxieties present in
Colombian society. Some of these Conservatives (men such as Belisario
Betancur or José Mejía y Mejía) sympathized with working people and
advocated socially progressive measures based in Catholic social teach-
ings even though they repudiated class struggle. Others (such as Dionisio
Arango Ferrer) considered the ‘‘popular classes’’ inherently unruly and
believed that only educated men were fit to determine the destiny of the
nation. For this group of Conservatives any measure that might under-
mine ‘‘tradition’’ was automatically characterized as radical agitation and
the prelude to social revolution. The threat of a Communist-led revolu-
tion could be conveniently used to discredit both the Liberal opposition
and their own party’s bipartisan elite. ‘‘Communism’’ and ‘‘Communist’’
became catch-all terms that could be trotted out against anything or any-
one who appeared to challenge the status quo, including working women
in short skirts, the demand for a wage increase, or modernist overtones
in regional painting.
The young men who came to form the middle sector were drawn into
and shaped by public debates concerning the future of capitalism, the ap-
peal of radical ideas among discontented workers, and the ‘‘breakdown’’
of traditional values and morality produced by modernity. In the s,
Medellín’s universities and newspapers became ideological battlefields,
and public demonstrations led by workers, populists, and fascist gangs
competed for the city’s space and the loyalties of the young. The educa-
tional reforms introduced under Liberal president Alfonso López in ,
moreover, contributed to the growing ideological and social divide be-
tween Liberal and Conservative youth in the region. Unlike their elite
leaders, Conservative and Liberal middle-sector politicians did not at-
tend the same secondary schools or universities. When López Pumarejo
redefined the national educational curriculum and incorporated Antio-
quia’s School of Mines, where engineers—the epitome in regional elite
eyes of the ideal profession for the region’s leaders—had traditionally
studied, into the National University system, he alienated many Antio-
queño Liberals and Conservatives. Moderates from both parties resented
 Medellín and Core Municipalities 

the subordination of a regional educational establishment to a national

system and the central state’s effort to dictate educational policy with-
out regional input, but more importantly they rejected the insinuation
of partisan concerns in public institutions. López meant to make pub-
lic schools and universities the sites of a Liberal and ‘‘modern’’ ethic,
and he hired Liberals to further his objective. The consciously partisan
nature of López’s reforms spurred an exodus of Conservative students
toward explicitly Catholic or Conservative educational institutions after
. The establishment of the Catholic Universidad Pontificia Bolivari-
ana in Medellín and the Javeriana in Bogotá in  by prominent advo-
cates of Acción Católica and anti-Communists in the wake of the Span-
ish Civil War attracted middle-sector Conservatives who in another era
would have attended Antioquia’s public universities. López’s conscious
effort to purge the public schools of Conservative teachers, and his use
of education as a medium of patronage through which he might ad-
vance the fortunes of his popular adherents, moreover, marked a con-
scious break with the politics of convivialismo embraced not only by his
predecessor, Liberal Enrique Olaya Herrera, but by moderate Conser-
vative administrations associated with Antioqueño leadership between
 and . As such, López’s educational reforms were explicitly in-
tended to end the elite monopoly on education which through coopera-
tion ‘‘had neutralized the opposition and excluded the pueblo [masses]
from politics.’’ 10
López’s reforms had a paradoxical effect on Antioqueño Conserva-
tives. On the one hand, the region’s bourgeoisie interpreted López’s re-
pudiation of the old convivialismo agreement not to meddle in educa-
tion as a betrayal of gentlemen’s politics.11 Up-and-coming Conservatives
from non-elite backgrounds, on the other hand, tended to agree that edu-
cation, like any other aspect of government patronage, was fraught with
political meaning and, as such, should be shaped by partisan ideals. The
shift of these young people away from public schools became symbolic of
their belief that if they remained in public schools dominated by Liberal
ideas, they would forfeit their identity as Conservatives and any future
claim to represent the ideological vanguard of the party. The attitude of
middle-sector youth was also influenced by efforts at the municipal level
of Acción Católica, the lay and religious organization devoted to spark-
ing a reawakening of Catholic piety and countering the noxious effects of
growing secularization brought on by modernity. Acción Católica vocif-
 Blood and Fire

erously called for a return to Catholic education and morals, thundered

against coeducational establishments, and condemned López’s reforms.12
The shift away from Antioquia’s public secondary and university
establishments by non-elite Conservatives also represented the middle
sector’s explicit rejection of the bourgeois ethic of technical education.
Like López, young, non-elite, Conservative aspirants to power effectively
understood that the regional elite’s privileging of technical skill over par-
tisan considerations exluded non-elites from access to political power.13
By skillfully manipulating the terms of political ‘‘loyalty’’ and ‘‘legiti-
macy,’’ mavericks could discredit suprapolitical bourgeois educational
ideals as treason against the party and even as a threat to morality, so-
cial order, and the defense of religion. In the context of economic reces-
sion, popular mobilization, and global unease over Communism, such
an attack amassed considerable force. In a sense, López’s reforms gave
middle-sector Conservative youth an incomparable justification for the
more critical aspects of their rebellion against the technocratic ideal of
bipartisan bourgeois leadership.
Conservative provincials who were excluded from the genteel dining
rooms of the Club Unión or the editor’s office of the region’s elite-led
daily newspaper, El Colombiano, took over the hallways and classrooms
of local universities to air their grievances publicly and to hone their
skills as orators and leaders during the sixteen-year period of Liberal rule.
They forged relationships with non-elite members of their own party and
joined alternative newspapers (El  de Abril, La Defensa), militant Catho-
lic associations (La Cruz de Malta, Haz Godo Masculino, Alianza Para La
Fé), and informal drinking circles where politics, ethics, and the nation’s
destiny were parsed and reshaped amid late-night tangos and aguardiente
(liquor) in the bars and cafés of Medellín’s bohemian sector, Guayaquil.14
These experiences forged a sense of partisan and generational identity
and left an indelible mark on the participants, many of whom rose to
occupy positions of political power by the mid-s.15
Moreover, by , stark disparities between social sectors were evi-
dent in Medellín. The elite, whose profits and production were expand-
ing, and who could afford to celebrate their wealth by building elaborate
art deco dwellings in upscale residential neighborhoods such as Prado,
contrasted sharply with an urban working class that found it increasingly
difficult to make ends meet. By the late s, one-tenth of  percent of
Colombia’s population controlled  percent of all the profits created by
 Medellín and Core Municipalities 

what one contemporary observer termed ‘‘the highest earning period in

the history of the [Colombian] economy.’’ 16 At the same time, the worst
salary years for workers in Antioquia occurred between  and ,
while the average growth of employment dropped from . percent be-
tween  and  to . percent between  and .17 From the late
s through ,  percent of an Antioqueño worker’s salary went
to pay for food.18 A study of railroad workers, a relatively privileged sec-
tor of Antioqueño labor, discovered in  that the majority of workers’
families suffered from chronic malnutrition, and in  only  percent
of Medellín’s workers received wages above subsistence.19
Widespread misery does not by itself necessarily provoke social un-
rest nor threaten the status quo, but it does make those in power ner-
vous, especially those who have only recently achieved it. The politicians
who came to office in  were men who balanced on a tightrope be-
tween their current professional ‘‘respectability’’ and a not-too-distant
past as provincial, petit bourgeois supplicants who were providentially
saved from rustic destinies by party-provided scholarships and support.
As they surveyed their surroundings, the middle-sector politicians pro-
jected their fears of economic and social unrest onto potential rabble-
rousers and ‘‘Communist’’ elements that they supposed were poised to
infiltrate and subvert Antioquia’s masses.
The impression that a radical uprising was imminent was confirmed
for Antioqueño Conservatives and some elite Liberals when the dissi-
dent Liberal leader, Jorge Eliécer Gaitán, gained the electoral support of
miners, port, and oil workers in peripheral towns such as Zaragoza, Cau-
casia, and Puerto Berrío in the  presidential election. Three of the
’s most militant unions (the oil, mining, and dock and river trans-
port unions) were based in towns where Gaitán won between  per-
cent and  percent of the vote.20 The areas of Gaitán’s support coincided
with those hardest hit by the economic changes of the previous decade.
Mining employment declined nearly  percent between  and ,
and over , agricultural jobs disappeared during the same years in
Antioquia.21 Conflicts between colonos (squatters) and landowners that
had resulted in the violent expulsion of thousands of peasants in the s
and early s, moreover, occurred precisely in the municipalities where
dissident or radical political movements enjoyed the greatest support.
The implications of an intersection between mobilized workers and
squatters, on the one hand, and a vibrant, dissident political leader in
 Blood and Fire

a moment of historical uncertainty and change, on the other, were not

lost on anxious Conservative observers. Gaitán’s supporters along the
Magdalena River were far more militant and organized than were indus-
trial workers based in Medellín and, as such, were not necessarily reliable
indicators of the presence of a subversive substrata within Antioquia’s
working class. But the defiance of workers in peripheral regions in the
s served to give a tangible focus to Laureano Gómez’s dire warnings
of an imminent threat from the Left. While some levelheaded regional
Conservatives recognized the difficult economic conditions that charac-
terized the lives of rural and urban workers in the region and responded
to unrest with proposals for reform, for the majority of recent political
appointees, repression and red-baiting proved more expedient political
options. The latter exploited local anxieties of further economic decline,
pitted ‘‘responsible’’ urban industrial workers—many of whom were af-
filiated with the Jesuit-led Union of Antioqueño Workers (Unión de Tra-
bajadores Antioqueños, or )—against their ‘‘revolutionary’’ -
affiliated counterparts and invoked the need to defend God and nation
against revolution. These visceral appeals were in turn used to justify the
deployment of coercion against recalcitrant Liberal voters and the harass-
ment and dismissal of members of the opposition from state patronage
positions.

The Rise of the Middle Sector to Positions of Power

Governors and their administrative subordinates played an extraordi-

narily important role in the promotion of partisan violence in Antio-
quia between  and . Their powers of appointment over mu-
nicipal mayors, departmental policemen, regional public works boards,
contractors, inspectors, and supervisors ensured that the regional chief
and his cronies could severely disrupt the conduct of municipal gov-
ernment through the calculated appointment of selected loyalists to
local offices. Mariano Ospina Pérez appointed two kinds of Conserva-
tives as Antioquia’s governors between  and : moderates such as
Fernando Gómez Martínez and more intransigent Conservatives either
overtly affiliated with Gómez or sympathetic to him such as José María
Bernal, Eduardo Berrío González, and Dionisio Arango Ferrer. The char-
acter and political sympathies of these men marked provincial politics
and the development of la Violencia in important ways.
 Medellín and Core Municipalities 

José María Bernal was Ospina’s first appointee to the office of gover-
nor and served from August  to November . An engineer and
businessman, Bernal occupied a kind of intermediate position between
Antioquia’s traditional bourgeois leadership and an emergent Conserva-
tive professional political class. He was an ardent Catholic and a right-
leaning Conservative but did not explicitly define himself as a follower
of Laureano Gómez, or laureanista. Bernal and Eduardo Berrío, the gov-
ernor’s hand-picked secretary, were products of San Ignacio, Medellín’s
Jesuit-run boys school, had participated in Acción Católica, contributed
to the newspaper La Defensa, and worked together at various times in
the private sector.22 The two men shared an interest in consolidating
their political careers, restoring the regional Conservative party’s elec-
toral strength, and capturing the patronage and electoral opportunities
lost to the Liberal opposition over the previous sixteen years.
The governor used the months between Ospina’s inauguration and
the regional and municipal elections scheduled for March and October
 to register voters, appoint trusted Conservatives to key regional
offices, and gradually rearrange the distribution of employment in Antio-
quia’s most easily dominated municipios.23 The towns experiencing the
earliest incidents of partisan violence were thus those in which Conser-
vative voters predominated or where politics were competitive (that is,
divided between nearly equal numbers of Liberal and Conservative con-
stituents). These towns were concentrated in the coffee zone, the indus-
trial belt around Medellín, and in predominantly Conservative towns in
the north, east, and west.24 (See map .) Numerous factors influenced
the choice of municipalities slated for conservatization by the regional
governor and his supporters. Those likeliest to be targeted were densely
populated and, in comparison with other parts of Antioquia, possessed
of considerable wealth. After Medellín, these municipalities accounted
for the largest percentage of regional budget expenses, national subsidies,
and loans.25 Affluence or economic potential was manifested through
lucrative public works contracts to build roads, sewage systems, aque-
ducts, schools, and hospitals and constituted an important source of
patronage-determined hiring (see Appendix B.). These towns were also
well integrated into the party system through municipal committees and
the presence of state representatives such as mayors, police inspectors,
and judges. Unlike peripheral municipalities where the regional state’s
presence was weak and the level of integration between local and regional
 Blood and Fire

Map . Electoral Violence, . (Source: Instituto Geográfico Augustín Codazzi

and Colombia, Departamento Administrativo Nacional de Estadística)

government was tenuous, the centrality of core municipalities guaran-

teed that structures were already in place through which political influ-
ence could be exercised. Finally, regional hard-liners chose areas where
the Conservative party already had a significant presence as their targets
for electoral intimidation. Partisan numerical advantage was then used to
effectively pressure into submission nearby towns with Liberal majorities
where the number of Conservative voters would have been insufficient to
turn the electoral tide alone (see table ).
Direct attempts to reduce the electoral power of the opposition in
overwhelmingly Liberal towns such as Santa Bárbara, Amagá, Venecia,
or Angelópolis in the southwestern coffee zone were avoided. Instead,
the regional authorities concentrated on converting towns around these
 Medellín and Core Municipalities 

Table 1. Political tendencies of towns with electoral violence,

Assembly Elections of March , 
Political Dissident Conser-
Municipio Region tendency * Liberals Liberals vatives

Santo Domingo Oriente Conservative

 ,

Jericó Suroeste Conservative
  , 
Ituango Norte Conservative 

 ,
Amagá Suroeste Liberal ,

 

Yolombó Nordeste Liberal    ,
Segovia Nordeste Liberal  


Pueblorrico Suroeste Liberal 




Ebejico Central Liberal  


Olaya Occidente Liberal
 
 

Itagüí Central Competitive ,
  

Bello Central Competitive ,
 
 ,

Fredonia Suroeste Competitive , 

,
Envigado Central Competitive ,


 ,
Sopetran Occidente Competitive 
 
Valdivia Norte Competitive  

Maceo Magdalena Competitive 
  

Bolívar Suroeste Competitive 
, , 
Andes Suroeste Competitive 
, ,
Caramanta Sur Competitive
  

(Source: Colombia, , Anuario Estadístico de Antioquia, Años , , , Apén-
dice / ‘‘Estadística Electoral’’)
* based on voting patterns prior to 

Liberal bastions through a strategy of intermittent cross-municipal ha-

rassment.26 Policemen and mayors typically stripped individuals of their
cédulas, or national identity cards, without which they were ineligible to
vote; physically barred citizens from approaching the registration tables
set up in the months immediately preceding elections; and harassed the
opposition with verbal and physical abuse. Once towns such as Cara-
manta—disadvantageously bordered by solidly Conservative towns such
as Támesis and Jardín—or Pueblorrico—caught between Conservative-
leaning Andes and Jericó—were brought into the orbit of Conservative
influence, political pressure was exerted against towns where Liberals
represented a majority such as Tarso and Amagá. The towns closest to
areas of solid Liberal domination that were gradually conservatized be-
 Blood and Fire

tween  and , were, in turn, later used to extend the radius of
Conservative control between  and .27
The carefully calculated use of scare tactics by public employees and
members of the regional Conservative party in towns where the party al-
ready had some base of support initially failed to achieve an overwhelm-
ing victory for local Conservatives. Of the nineteen towns that reported
electoral violence during , six were (or had been until ) solidly
Liberal, ten were (or had been until ) competitive and three were
solidly Conservative. Six of the nineteen towns cast a significant per-
centage of their votes for Gaitán in the legislative assembly (asamblea)
elections of March  despite repeated harassment from government
officers, while two towns that had not voted for Gaitán in March voted
for local gaitanistas (followers of Gaitán) in the municipal council (con-
cejo) elections of October .28 Miners and road workers, who consti-
tuted an important presence wherever support for Gaitán was statisti-
cally significant, formed the vanguard of Liberal resistance to Conser-
vative electoral intimidation in many of these towns. Voting for Gaitán
was also far greater in towns where Liberals were heavily outnumbered
by Conservatives than in the rest of Antioquia’s municipios, including
those where Liberals predominated.29 In fact, wherever Conservatives
were a clear majority, Liberals relegated to the minority position voted
in droves for Gaitán in what amounted to a clear act of protest and re-
sistance against the regional government and, in some cases, against the
elite-dominated Liberal party directorate and its mandates. In Abejorral,
for instance, where Conservatives had historically dominated local elec-
tions and where electoral pressure did little to increase an already consoli-
dated Conservative advantage, Liberals nonetheless delivered  percent
of their votes for Gaitán in the Assembly election of March , while 
percent of the Liberal votes in equally Conservative Cocorná were also
cast for Gaitán. The Liberal inhabitants of Toledo in the north who made
up only  percent of the town’s electorate cast all their votes for Gaitán
(see table ).
The results of the March  elections proved disappointing to Con-
servative hard-liners. Conservatives handily won majorities in the south-
ernmost corner of Antioquia from Andes up through Jericó, Támesis,
Pueblorrico, and Amagá, areas where many inhabitants already belonged
to the party, but they absolutely failed to destroy Liberal control over
most historically Liberal towns. Only four towns where Liberals were
 Medellín and Core Municipalities 

Table 2. Conservative towns with heavy Liberal-dissident voting,

Assembly Elections, March , 
Gaitanistas
Liberals Gaitanistas Conservatives as Percentage of
Municipality Region Number % Number % Number % Liberal Total

Cocorná Oriente     ,


Guatapé Oriente 

   
Angostura Norte     ,  
Nariño Sur 
  , 
Peñol Oriente     ,


Campamento Norte



 
  
Yarumal Norte   

, 

Toledo Norte


   

Liborina Occidente
  
 ,
  
Ituango Norte 

  
, 

Abejorral Sur     ,  
Anorí Nordeste  

   

(Source: Colombia, , Anuario Estadístico de Antioquia, Años , , , Apéndice /,
‘‘Estadística Electoral’’)

or had been an important presence before —Olaya, Pueblorrico,

Tarso, and Yolombó—were completely conservatized, although harass-
ment lowered Liberal returns throughout the region.30
Although regional Conservative attempts to use official forces in the
quest to achieve electoral domination produced unremarkable results,
individuals such as José María Bernal and Eduardo Berrío González con-
tinued to deploy public forces for partisan purposes. Indeed, the first in-
stances of local violence in Antioquia emerged as a direct consequence
of state-directed harassment rather than from internal divisions between
local members of the two parties. The Liberal party committee in the
southern coffee town of Caramanta (which had boasted a slight Lib-
eral majority before the Conservative party came to power in ) was
the first Antioqueño town to accuse the regional authorities of insti-
gating local partisan violence. The governor had replaced Caramanta’s
native-born mayor with a Conservative appointee from another part of
Antioquia and used him to ensure an electoral victory for the govern-
ment’s party. When confronted with Caramanta’s complaint, the gov-
ernor retorted that his replacement of the native-born Liberal mayor
 Blood and Fire

by a nonnative Conservative had been guided by ‘‘purely administra-

tive’’ concerns, and he scathingly reminded local Liberals that in the de-
termination of appointments ‘‘there existed no Conservatives or Lib-
erals, but only citizens.’’ 31 Yet, as quickly became apparent, the right to
control state patronage and shape local appointments was heavily in-
fluenced by partisan and personal considerations as was the response
of the regional government when faced with municipal complaints re-
garding violence promoted by public employees or official regional
policies.
For instance, wealthy Liberals in another coffee town, Titiribí, where
some of the region’s largest coffee haciendas were located, repeatedly
complained to the governor of disputes arising from the municipal coun-
cil election of .32 They warned the regional authorities in Medellín
that if partisan tensions were allowed to proceed unchecked, they might
provoke a violent reaction that could have serious economic conse-
quences in the region as a whole. When Governor Bernal ignored Titi-
ribí’s complaints, members of the local elite simply bypassed the governor
and went straight to the president instead. They appealed to Ospina Pérez
as an elite, Antioqueño native son, a former businessman like themselves
who could appreciate the dangerous economic consequences of using
the state’s officers to pursue a partisan policy. Alarmed as they hoped he
would be, the president pressured Antioquia’s governor to stop harassing
Titiribí’s Liberal constituents.33
Local reactions to officially sanctioned electoral intimidation and the
regional government’s response to local grievances increasingly inter-
sected with older tensions between municipal and regional inhabitants
regarding the centralization and usurpation of power by the regional
state. The way in which regional and central governments chose to re-
spond to complaints of violence such as those made by Titiribí’s Liberal
property holders also brought to the surface latent tensions within Antio-
quia’s parties. In the case of Titiribí, for instance, the president shared
the same preconceptions as local Liberal coffee growers. Like them, he
balked at the use of partisan-motivated violence that might affect elite
fortunes or jeopardize a critical sector of the regional economy. The gov-
ernor, in contrast, had little in common with the town’s elite and placed
the possible economic repercussions of partisan-motivated policies con-
siderably below his priority as a middle-sector party loyalist to ensure
Conservative electoral victories. The case of Titiribí illustrates the inher-
ent differences that emerged during la Violencia between a regional-elite
 Medellín and Core Municipalities 

political tradition that privileged economic development above partisan

issues and the ideologically inspired politics of middle-sector regional
Conservatives. But Titiribí’s experience also laid bare the uneasy relation-
ship between the president and his regional middle-sector subordinates.
While the president’s direct intercession ensured that Titiribí’s grievances
were redressed, the recalcitrant governor whose authority the president
overrode remained firmly ensconced in office.
In Antioquia,  was not a particularly violent year, but it was a
year in which municipalities increasingly gave voice to and repudiated
regional policies intended to shape local electoral politics and the de-
termination of appointments to local offices. Two cases involving local
complaints of partisan-motivated violence during the year illustrate the
character of clashes between municipalities and the regional government
and the variety of official responses to incidents of municipal unrest. Two
months after Liberals in Titiribí filed their complaint, disgruntled towns-
people in the sparsely populated, predominantly black frontier town of
Cáceres in the Bajo Cauca region also reported incidents of partisan-
motivated violence, but they were ignored by both the regional and cen-
tral governments. Cáceres, like Titiribí, was an overwhelmingly Liberal
town, but unlike Titiribí or other Liberal southwestern coffee towns, the
inhabitants of Cáceres possessed little leverage with which to influence
either the regional or national government. The town’s voters were few,
poorly organized, and historically had displayed a high rate of electoral
abstention.34 Cáceres also lacked brokers (such as elite coffee producers)
who might have influenced the central government, through the media-
tion of a private producer association, to force a regional intervention
on its behalf. Thus, when Governor Bernal’s appointment of a Conser-
vative mayor in Cáceres provoked local outrage as it had in Caramanta
and Titiribí, it won Bernal a mild rebuke from the central government
but effected no change in regional policy. Governor Bernal could and did
defend his refusal to replace the offending town mayor by shifting blame
away from the issue of partisan discrimination and toward the town’s
peripheral geographic location and general poverty: ‘‘I am not bent on
maintaining ignorant elements in the office of municipal mayor,’’ Bernal
caustically responded to a central government reprimand, ‘‘the reality
is that for those areas, climates, and salaries, one cannot obtain terribly
competent elements. The governor’s office would like to send a whole
mayor [todo un alcalde], but can only send  pesos worth of a mayor
[that is, a second-rate mayor].’’ 35
 Blood and Fire

Violence in economically important towns where Conservatives were

numerous, in turn, elicited yet a different response from the central
and regional governments.36 The southwestern town of Fredonia was the
largest volume producer of coffee in Antioquia. The regional railroad ran
an expensive trunk line between Medellín and the town only to service
coffee and cattle haciendas owned by members of the regional bourgeoi-
sie. Evenly divided between Liberal and Conservative voters, Fredonia
tended to give a slight local majority to whatever party happened to be in
power in Bogotá.37 In , however, Liberals dominated the town coun-
cil. They voted the salaries and controlled the appointment of several
local government officers and the town’s police. When Governor Bernal
appointed a Conservative mayor to ensure Fredonia’s shift back to a Con-
servative majority in the  elections, the town council refused to co-
operate with him. In fact, the local police hired by the council resisted the
mayor’s attempt to usurp the council’s authority and shot a local Con-
servative leader when he interceded on the mayor’s behalf.
Members of a locally influential Conservative political family and
other highly placed regional politicians mobilized immediately to im-
press upon the governor the severity of the threat to public order in the
town.38 Governor Bernal rapidly deployed ten departmental guards and
a police lieutenant to reestablish peace in the municipality, but when the
departmental police interceded, a confrontation ensued, leaving seven
individuals gravely injured and a member of Antioquia’s Conservative
party directorate dead. A flurry of preoccupied telegrams was exchanged
between the governor, the president of Colombia and Fredonia’s munici-
pal officials. Both Bogotá’s Conservative and Liberal dailies, El Siglo and
El Tiempo, reported extensively on the happenings in Fredonia, and, in
response to criticisms of gubernatorial mishandling made by both Lib-
erals and Conservatives, the governor offered the president his resigna-
tion.
Violence in Fredonia elicited a response from both the governor and
the president in a way that partisan violence affecting elite Liberal inter-
ests in Titiribí and Caramanta had not. The otherwise defiant Conser-
vative governor’s willingness to step down when violence in Fredonia
was determined to be at least partially his responsibility, moreover, at-
tests to the superior influence exerted by Conservative elite interests on
both the regional and national governments. But these cases also point to
the significance of well-established institutional mechanisms for ensuring
 Medellín and Core Municipalities 

the smooth functioning of accountability and power between municipal

and regional authorities. Despite differences in their partisan make-up,
both Fredonia’s and Titiribí’s experiences were typical of the trajectory of
partisan violence within sectors of core or traditional settlement such as
Antioquia’s coffee zone. The combined effect of strategic economic inter-
ests and established networks of political influence enabled citizens to
pressure either the regional or central government into taking their con-
cerns seriously regardless of their political affiliation. In contrast, both
the regional and central governments ignored local complaints of par-
tisan violence waged in Cáceres. In addition to being overwhelmingly
Liberal, Cáceres had no way of mobilizing a collective protest against
the state in a way that would guarantee a receptive audience. It was eco-
nomically valuable, but its wealth was in the hands of absentee land-
lords and foreign-owned mining concerns, while those who inhabited the
town lacked the institutional intermediaries and political infrastructure
to make their grievances count.
Decisions made to resolve incidents of violence in Antioquia’s munici-
pios were thus based at least in part on factors that transcended the parti-
san composition of a particular municipal electorate. More than whether
a town was Liberal or Conservative, the location and composition (both
ethnic and partisan) and the presence of powerful brokers whose inter-
ests either the regional or central government could not afford to ignore
determined official responses to local incidents of violence. Such con-
siderations become an important factor in determining why partisan vio-
lence in towns with Liberal majorities could evolve in fundamentally dis-
tinct ways.

Violence by Other Means:

The Substitution of Liberals on Public Boards

The effort to achieve electoral victories for the Conservative party in 
was accompanied by a concerted attempt to dislodge the Liberal oppo-
sition from its dominant position in public sector employment. Middle-
sector Conservatives also sought to alter the composition of crucial mu-
nicipal advisory bodies and regional boards that regulated public works
bidding, hiring, and investment. What was at stake was control of an in-
creasingly lucrative and electorally significant chunk of state-determined
patronage.
 Blood and Fire

Conservative municipal committees in towns such as Bello, a densely

populated, industrial, working-class town within Medellín’s metropoli-
tan hub, were among the first to attempt to use the rise of their party
to positions of regional authority to gain control of public patronage.
Bello’s local Conservative committee members demanded that the re-
gional governor intercede and manage the election of the Concejo de
la Cooperativa de Municipalidades (Municipal Cooperatives Board) in
order to guarantee a majority of Conservative representatives. Majority
control of the board would finally enable local Conservative politicians
to usurp and build upon state-sponsored jobs and contracts long domi-
nated by members of the opposition. A year later, in , local Conserva-
tives congratulated the governor for ‘‘eliminating . . . the evil cabal [rosca]
that . . . used its power [on the board] to pursue hateful and exclusionary
policies, engaging in abuses with the membership’s dues.’’ 39 The gover-
nor’s behind-the-scenes manipulation of elections to the cooperative’s
board in Envigado, another important industrial town with a Liberal ma-
jority, also succeeded in guaranteeing local Conservatives an edge over
the opposition. The newly elected board members promptly cut off funds
and interrupted important public works contracts that were already in
process and would have benefited Liberal political brokers and workers.40
Governors also intervened during la Violencia to rearrange the com-
position of administrative boards governing the railroad, customs, pub-
lic roads, and the Social Security Institute, although this meddling that
targeted elite members of the opposition sometimes proved difficult to
accomplish. Only a cataclysmic event such as the death of Gaitán enabled
Governor Dionisio Arango Ferrer to finally rid himself of the presence of
Captain Julián Uribe Gaviria, Antioquia’s Liberal party leader, a former
governor of the department, and the president of the regional commit-
tee of the Federation of Coffee Growers who sat on Antioquia’s powerful
Junta de Rentas, or Customs Revenue Board.41 Even so, Uribe Gaviria was
not removed without a struggle. Within days of his dismissal the minis-
ter of government, Darío Echandía, formally complained that such un-
warranted substitutions were unlawful and demanded Uribe Gaviria’s re-
instatement. But Antioquia’s governor defended his actions by invoking
Law  of  which held that departmental officers of which the board
members of the Junta de Rentas formed a part, could not simultaneously
hold party offices, and that, in any case, Liberals continued to be repre-
sented on the Junta.42 The governor’s action stuck and the central govern-
 Medellín and Core Municipalities 

ment was forced to back down in the first of what would be many show-
downs between regional and national authorities over appointments and
administrative powers.
If some modicum of protection was exercised to prevent the subor-
dination of regional material progress to the vagaries of partisan com-
petition and to limit wholesale dismissals of members of the elite from
influential regional boards during la Violencia, very little protection was
extended to workers. Indeed, organized labor, especially that employed
by the state, was one of the earliest sectors of Antioquia’s population to
feel the effect of regional, state-directed, partisan violence. In addition to
wishing to replace Liberal workers with Conservative ones, Conservative
hard-liners also targeted public sector workers for harassment in order to
crush particular unions, specifically those affiliated with the  such as
Antioquia’s municipal employees, miners, road, railroad, public utilities,
oil, and port workers.43
Conservative attacks on the Communist and gaitanista leadership and
rank and file that made up the  initially met with little resistance from
Antioquia’s men of capital, including those professing a loyalty to the
Liberal party. The regional bourgeoisie had made explicit their lack of
sympathy for organized labor in general, but particularly for so-called
‘‘agitators’’ and ‘‘Communists’’ since the days of Alfonso López Puma-
rejo’s first presidential administration (–). In fact, when interim
Liberal President Alberto Lleras cracked down on the Communist-led
Magadalena Port Workers Union () in  and replaced ,
union members with scab labor, Liberal Antioqueño businessmen ap-
plauded.44 The most serious opposition to López’s pro-labor measures
within Colombia’s Liberal ranks came, in fact, from Antioquia’s business-
men and industrialists. Thus, when Conservative Maríano Ospina Pérez
struck at the ’s influence by granting legal recognition to a compet-
ing national labor federation, the Catholic-inspired Union of Colombian
Workers (), his actions elicited little opposition from Antioquia’s em-
ployers, regardless of their partisan affiliation. It was in Antioquia, the
nation’s leading industrial and mining center after all, where dissident
Liberal leader Gaitán’s regional followers had long lamented, ‘‘the weight
of oligarchic strength [was] strongest’’ in Colombia.45 Antioquia’s bour-
geoisie even opposed modest attempts to require employers to increase
their already low contribution to the Social Security Institute (Instituto
de Seguro Social) insisting that any expansion in state bureaucratic power
 Blood and Fire

constituted an unconstitutional attempt to diminish or wrest power of

determination over social policy from private and regional hands.
The most militant sectors of labor in Antioquia thus anticipated that
the rise of Ospina and his supporters to positions of power would in-
tensify the state’s reactions to labor unrest. They would be right. Shortly
after the president’s inauguration and his announcement of José María
Bernal’s appointment as governor, regional unions initiated a wave of
strikes to protest conditions in foreign-owned mining companies and
the rising cost of living. Rank-and-file petroleum workers in Remedios,
Antioquia, and Barrancabermeja, Santander, employed by Shell and
Tropical Oil and affiliated with the Colombian Petroleum Workers Union
struck in October and November of , while the Federation of Antio-
queño Workers () to which municipal employees, teachers, and
public road workers belonged, threatened to produce a list of demands
if the petitions of the region’s slaughterhouse workers were not met. As a
form of political protest against the elite leadership of the Liberal party,
moreover, miners, oil workers, and road workers in Antioquia voted
overwhelmingly for gaitanista candidates in towns such as Betulia, Cau-
casia, Dabeiba, Ebéjico, Péque, Remedios, Turbo, Segovia, Zaragoza, Titi-
ribí, and Puerto Berrío in the March  elections.46
Gaitán threatened to call a general strike shortly after the March elec-
tions if Liberal victories in Asamblea and congressional elections were
not respected by the central government.47 In April the  voted to
make Gaitán’s threat a reality and struck to protest the unlawful dis-
missal of union members and the use of police surveillance and repres-
sion against workers. But Medellín’s industrialists (of both parties) and
the region’s Catholic-inspired Union of Antioqueño Workers ()
refused to back the general strike, and Gaitán himself, after having issued
the call for a strike, refused to openly support the . 48 When in May
Antioquia’s railroad workers participated in the general strike to protest
the growing use of official violence against workers, declining pay, and
deteriorating social conditions, the regional government declared the
strike illegal and seized the opportunity to dismiss railroad workers it
deemed ‘‘inconvenient.’’ 49
Members of the  were clearly among the earliest objects of Con-
servative attack, but the only workers to consistently be made targets
of Conservative harassment and dismissals were railroad and municipal
workers, that is, individuals who held positions dependent upon political
 Medellín and Core Municipalities 

patronage and whose employer was the state. It was these workers whom
Conservatives in the regional administration blamed for the party’s in-
ability to wrest control of strategic town councils during the  elec-
tions. In contrast, the regional government was far more cautious when
dealing with oil workers and miners who were not part of the state’s pa-
tronage rolls. The governor’s reaction to labor unrest at Shell Oil’s Casabe
camp in Remedios is a case in point. In August  Shell’s manage-
ment telegraphed the governor urgently requesting twenty departmental
policemen to protect the company from a labor uprising that threatened
to take over the camp, but the regional government refused, insisting that
it would be unseemly to defend a foreign company against Colombian
workers.50 Taking a nationalist line, Secretary of Government (lieutenant
governor) Eduardo Berrío suggested the company hire off-duty police-
men as other foreign-owned companies did rather than expect the re-
gional government to underwrite Shell Oil’s security needs.51
The governor’s response to Shell’s management did not emanate from
any sort of personal empathy for oil workers. Indeed, the governor had
more than once derided them as revolutionaries and Communists. In-
stead, Berrío was concerned that it might appear as if the regional govern-
ment were kowtowing to the company by deploying departmental police-
men to repress workers, and he did not want to risk being blamed for
possible fuel shortages that might result from worker protests. Oil pro-
duction in Antioquia had tripled between  and , and the revenue
generated by an obligatory tax on it was used to pay for regional public
works projects.52 Oil was thus too strategic and lucrative a commodity for
the regional government to risk jeopardizing its production even when
ideological principles were involved. In refusing Shell’s request for de-
partmental aid the governor was slyly angling to force Shell Oil’s hand,
to convince the company’s management of the wisdom of opening em-
ployment in the camp to Conservative workers and Conservative political
influence while avoiding the impression that the government was inter-
fering to benefit a foreign-owned enterprise.
The Casabe labor uprising spurred considerable anxiety among Antio-
quia’s regional authorities, who increasingly worried about the presence
of ‘‘provocateurs’’ among the ranks of Antioquia’s labor force by late .
But regional authorities channeled such concerns toward public sector
workers, not the miners or oil workers who were the likeliest sources of
agitation. Consequently, the regional and central governments embarked
 Blood and Fire

upon a series of measures intended to restrict the mobility of labor in

Medellín and other areas boasting a heavy presence of state employees.53
Among other tactics, the authorities commissioned inquiries to measure
the partisan composition and sympathies of workers in the railroad, road
construction, and the municipal public works programs while simulta-
neously increasing the number of guards patrolling railroad dormitories.
The regional authorities also enforced curfews and authorized searches
of workers suspected to be carrying arms or engaged in ‘‘subversive’’ ac-
tivities. The departmental police under the direct authority of the re-
gional government emerged as the preferred medium of state repression
directed at public sector workers. It was the departmental police, for in-
stance, who were used to crush the railroad strike in May of .
The routine use of policemen to repress and harass labor was not a
phenomenon introduced during la Violencia nor a tactic exclusive to
Conservative authorities in Antioquia, but it undoubtedly escalated and
provoked greater resistance than had ever been true during earlier ad-
ministrations. Regardless of partisan affiliation, for instance, earlier re-
gional authorities had made a habit of deploying police agents to break
up strikes and protect private property against the interests of labor
since around . This had fed a widespread sense of hostility between
workers and policemen that went back several decades. But when ani-
mosity between workers and police took on partisan overtones, the ten-
sions between these two groups rose to new heights. Finally, the as-
sassination of Gaitán in April  marked a new low in police and
labor relations in Antioquia as the rate of attacks led by laureanistas
in the regional government against public sector workers rose dramati-
cally and as private forms of violence were introduced to complement
the repressive activities of the police against public sector and industrial
workers.

The Failure of Conservatization and the Escalation of Violence

Despite the use of officially sanctioned harassment against selective Lib-

eral workers and towns during the first two years of Conservative rule in
Antioquia, surprisingly few incidents of local violence actually took place
in the region. Fewer still were reported in either the press or to the gov-
ernor’s office during the months between the October  elections and
the assassination of the Liberal leader, Jorge Eliécer Gaitán, in Bogotá on
 Medellín and Core Municipalities 

April , . The department’s relative calm was shattered, however, by

Gaitán’s murder.
Gaitán had not been particularly popular in Antioquia before his
death. In contrast to cities such as Cali or Bogotá where support for
Gaitán had been significant (up to  percent of the urban vote, for in-
stance), less than  percent of Medellín or Antioquia’s electorate had
voted for him in the  presidential election. Antioqueño support for
the Liberal dissident leader had begun to pick up by the congressional
and regional assembly elections of March , however, and a number
of Gaitán’s local followers won seats on Medellín’s town council in the
October  election. Still, Gaitán’s limited and selective popularity in
Antioquia as a whole probably explains why Medellín and most of Antio-
quia escaped the more destructive effects of popular protest in the wake of
his death. The primary target of armed popular violence in Medellín was
the Conservative newspaper, La Defensa. The newspaper’s offices were set
afire to punish it for being the headquarters of right-wing Catholic and
laureanista Conservatives who were blamed by popular sectors of the Lib-
eral party for both Gaitán’s death and the general escalation of partisan
violence in Colombia. When compared with the effect of popular rage in
Bogotá, where nearly half of the urban center was destroyed in less than
twenty-four hours, however, the extent of damage inflicted in Medellín
appears almost slight. Not so the reaction of the regional authorities.
A prominent labor lawyer in Medellín remembers that after Gaitán’s
assassination, ‘‘labor activists and leaders’’ and those ‘‘who sympathized
with the left’’ were ‘‘indiscriminately imprisoned.’’ 54 When the city’s jail
cells were filled to capacity, the authorities comandeered public schools
and, having exhausted the available space in those as well, constructed a
makeshift prison camp in the municipal bull ring. The majority of those
held were ‘‘railroad workers, particularly those who worked the line be-
tween Medellín and Puerto Berrío.’’ 55 The authorities chose to detain
those from areas of Antioquia where violent demonstrations followed
Gaitán’s assassination and where, in some cases, revolutionary juntas had
taken control of municipal governments. These were towns in which
miners, oil, road, and port workers constituted a significant demographic
and political presence. Most, although not all, of the sites of popular vio-
lence were located on Antioquia’s periphery. The railroad workers em-
ployed on the line between Medellín and Puerto Berrío emerged as the
conduits between provincial and urban rebellion.
 Blood and Fire

Gaitán’s assassination was felt most strongly in the port town of Puerto
Berrío, the oil camps in Yondó, Remedios, southwestern towns such as
Andes, Bolívar, Anzá, and Urrao, and in western towns such as Buri-
ticá, Peque, and Turbo.56 All of these were towns where public works
projects employing road and railroad workers were present, and these
workers played important roles in leading popular protests. Arson and
the looting of commercial establishments and government offices such
as the courthouse, mayor’s office, and customhouse took place in Puerto
Berrío, around the train station in Bolombolo, Venecia, and in the Shell
Oil camp in Casabe. Acts of sedition were attempted in Dabeiba, Cáceres,
Cisneros, Peque, Titiribí, Venecia, Fredonia, Concepción, Envigado, and
Santo Domingo. Six of these towns were located in the southwest and
near east, not the periphery, but all were characterized by the strong pres-
ence of organized labor, especially road workers, railroad personnel, and
miners.57 All of these were also towns in which a significant sector of the
local Liberal electorate had voted for the gaitanista slate in the regional
assembly and national congressional elections of October .
The southwestern town of Bolívar emerged as a focal point of vio-
lence. A narrow Liberal majority had characterized it until , but the
town was conservatized between the election for town council in Octo-
ber  and the presidential election of . One of the first measures
proposed by Bolívar’s mayor in the wake of Gaitán’s assassination was
the right to substitute three Conservative workers for three  members
from the local road working crew who were deemed to be ‘‘subversives.’’
Local Conservatives then volunteered their services to the government
with arms and vehicles to meet the challenges posed to state authority
by workers up in arms.58 The presence of ready and willing local Conser-
vative volunteers and Bolívar’s strategic location—it’s proximity to both
the department of Chocó and public sector workers in nearby south-
western towns—made it an ideal site for counterrevolutionary Conser-
vative organization. These organizations, incorporating both local police
and Conservative civilian volunteers (contrachusma) were deployed to
quell Liberal uprisings in other areas. Bolombolo, a village in Venecia,
where Liberals were powerful and where a large concentration of truck
drivers, road crews, and railroad personnel worked and resided, for ex-
ample, became the immediate target of Conservative posses from Bolívar
in the wake of Gaitán’s death.59 When road workers from Quibdó and
Carmen del Atrato in the Chocó rose up in arms in reaction to Gaitán’s
 Medellín and Core Municipalities 

assassination, seized Antioqueño citizens, destroyed telegraph lines, and

substituted their own supporters in public works jobs, moreover, it was
counterrevolutionary forces based in Bolívar who were sent to put these
disturbances down.60 Although Governor Arango Ferrer authorized the
creation of civilian police corps and handed out official arms to selec-
tively recruited Conservative volunteers elsewhere in the department,
Bolívar seems to have been one of the first testing grounds in Antioquia
for the emergent contrachusma.61
The aftershocks of Gaitán’s death were also acutely felt in areas where
militant workers not employed by the state were present. Indeed, these
workers appear to have been emboldened by the brief period of revo-
lutionary takeovers of some town governments and the general chaos
caused by Gaitán’s assassination. Striking oil workers at the Shell Oil
Casabe camp used the excuse of public disorder as a bargaining chip
to demand the right to determine the political affiliation of candidates
for the offices of inspector and judge as a condition of putting down
their arms.62 Demands such as these that were accompanied by the actual
use or threat of violence fed a growing paranoia in the Conservative
government. Regional and national authorities suspected conspiratorial
plots to overthrow Conservative office holders everywhere and suspected
organized labor of spearheading them. Hence the confidential warning
secretly telegraphed by the governor admonishing his subordinates to
keep an eye on explosives, transport, communications systems, radio sta-
tions, airports, munitions warehouses, aqueducts, electrical plants, fuel
reserves, telephones, and oil pipelines.63 Months after Gaitán’s assassi-
nation, President Ospina Pérez continued to urge all governors to re-
main alert to the possibility of future subversive movements or orders
for strikes to be carried out by  affiliates. He also exhorted governors
to maintain a close level of coordination with local military command-
ers in order to have troops stationed in a timely fashion in the principal
provincial cities.64
While few workers in Medellín and its industrial hub ultimately mo-
bilized to protest the slain political leader’s demise, the regional authori-
ties capitalized on the incidence of revolutionary protest in other crucial
Antioqueño towns to justify the general harassment and dismissal of state
employees and workers everywhere. Conservative authorities pointed to
popular insurrections in certain towns as incontrovertible evidence of a
long-predicted Communist plot to seize control of Colombia. The force
 Blood and Fire

of this argument resonated with some members of the region’s Liberal

elite, who in subsequent years referred to popular outbursts of rage in
the wake of Gaitán’s death as a ‘‘copious criminal hemorrage’’ that justi-
fied the use of repression.65 Members of Medellín’s Liberal oligarchy such
as Carlos Uribe Echeverri, for instance, threw their support behind Lau-
reano Gómez after Gaitán’s assassination, lamenting publicly that little
effort to combat Communism could be expected from his own party.66
While elite Liberals did not endorse the use of partisan policies to dis-
criminate against fellow party members, state-directed repression against
organized labor, especially that perceived to be under Communist leader-
ship, certainly enjoyed the tacit endorsement of many members of the
Liberal bourgeoisie in Antioquia.
Ospina Pérez understood the essentially conservative nature of his
regional compatriots in both parties and the deep anxiety that popu-
lar mobilizations of any sort awakened in them, although he may have
misjudged the choleric temper of the man he chose to restore order in
his native department. To preempt the spread of popular insurrection
in Antioquia, Mariano Ospina Pérez roused Dionisio Arango Ferrer out
of bed at two in the morning the day after Gaitán’s assassination and
named him Antioquia’s governor.67 Part of the crowd of Jesuit-trained,
middle-sector politicians, Arango Ferrer gradually forged a reputation as
the Conservative party’s troubleshooter. He was the man the party relied
upon when public order matters reached critical proportions. Arango,
like other members of the middle sector, opposed bipartisanship and
was not reluctant to use force to maintain order. Where he differed from
the younger, middle-sector politicians who surrounded him in the re-
gional bureaucracy not only during his  term as governor but later
in  and  when he resumed the post, was his clear antagonism to
populism. Arango believed that only the ‘‘intelligent’’ men of the party
could keep insubordination at bay, and he was extremely distrustful of
the pueblo.
An hour after the president’s summons, Dionisio Arango Ferrer strode
into the crowd milling around the Parque de Berrío to reclaim the gov-
ernor’s seat from what he disdainfully referred to as ‘‘the mob’’ and im-
mediately called a meeting of ‘‘notable’’ Conservatives.68 Convinced that
the turmoil surrounding Gaitán’s death was a prelude to a ‘‘Communist’’
takeover, the group agreed not to recognize any self-imposed revolu-
tionary junta that might claim to have deposed Maríano Ospina Pérez
 Medellín and Core Municipalities 

in Bogotá.69 Arango Ferrer did two more things during the morning of
April : he called Eduardo Berrío González in the town of Santa Rosa and
reappointed him departmental secretary, and he called up Conservative
reserve officers from the southern town of Sonsón to protect Medellín.70
The governor’s first actions had special significance. Eduardo Berrío
González had been instrumental in the appointment of Conservative loy-
alists as mayors throughout Antioquia during his recent stint as depart-
mental secretary. He had also played a seminal role in establishing the
legal basis for the organization of armed rural civilian patrols during the
same period (–). The mobilization of Sonsón’s reserve officers,
moreover, was significant because the town’s inhabitants were known for
their doctrinaire Catholicism and entrenched Conservatism. Arango Fer-
rer thus called upon partisan loyalists to defend Medellín and the gover-
nor’s palace, rather than the departmental police, army, or national police
which contained Liberal agents and non-Antioqueños. Under Arango
Ferrer’s administration, the rabid xenophobia that became a distinctive
feature of Antioquia’s administrative policies throughout the period of
la Violencia, and which contributed in no small fashion to the intensifi-
cation of local resentment against the regional government, received its
first boost.
The eight months of Arango Ferrer’s term as Antioquia’s governor
(April –December , ) were spent eradicating supposed centers of
insurrection in the region. In July , he empowered the customs reve-
nue agents to foment the creation of permanent, auxiliary, civilian police
forces at the municipal level.71 This action made it evident why removing
Liberal Captain Julian Uribe Gaviria from the customs board had been
imperative and gave teeth to the paper tiger created by Governor Bernal
in  when he fought for the legal right of civilians to bear arms and
organize rural patrol groups. The auxiliary civilian police were to be paid
either through additional funds voted by local town councils or, wher-
ever the town council refused to cooperate, by customs monies and the
tax on the regional liquor monopoly.72 The right to allocate funds for the
creation of what amounted to a paramilitary organization that circum-
vented the veto power of legitimately elected municipal councils made
customs officers into one of the most feared and powerful of the regional
government’s forces and marked an important watershed in the evolu-
tion of partisan conflict in Antioquia.
Along with the national police, customs revenue agents were among
 Blood and Fire

the public functionaries against whom the greatest number of complaints

were registered.73 When it became apparent how dangerous the autono-
mous armed groups bankrolled by custom agents could be to even Con-
servative interests, regional attempts (by visitadores administrativos, in
particular) to curb them, were met with defiance and threats of violence.74
A year after Gaitán’s assassination, Conservatives in the southwestern
town of Jardín complained of contrachusmas from the department of
Caldas who had arrived sometime after Gaitán’s assassination and began
‘‘sowing terror’’ in the region.75 Contrachusma forces indiscriminately
harassed both Liberals and Conservatives who repudiated their violent
methods. They also routinely subverted the lines of authority that bound
the regional and local government together. Regional detectives posted
to oversee matters in Jardín, for instance, reported that when they ap-
pealed to the customs agents to help them capture the armed men from
Caldas and protect the townspeople, the revenue agents rebuffed them:
‘‘We asked the administrator of the departmental customs revenues . . .
to help us conduct a sweep against [the armed men from Caldas],’’ the
local detectives complained to their superior and the governor, ‘‘but the
administrator himself told us that when he instructed one of the revenue
agents to help us, the agent replied that he’d rather be fired because he
wasn’t against those people, and we shouldn’t think he was going to help
us out.’’ When the regional detectives threatened the customs agents with
dismissal, they smugly retorted ‘‘that they were connected with certain
political elements who would back them up and come to their aid if [the
detectives] were to proceed against them.’’ 76
The dissemination of violence and the role of specific sectors within
the regional state bureaucracy in promoting it, particularly in the after-
math of Gaitán’s death, further fractured an already disunited regional
Conservative party. While the majority of Antioquia’s Conservatives
might agree about the need to recapture offices and power from the oppo-
sition, no agreement existed among Conservatives regarding the means
used to advance these objectives. When the ‘‘reconquista’’ began to have
wider reverberations—affecting Conservatives as well as Liberals, com-
promising the realization of regional public works projects, threaten-
ing production, and spurring latent socioeconomic conflicts unrelated to
partisan issues—dissension within the Conservative party became severe.
Alarmed by the high-handedness of the individual to whom he had en-
trusted his native region’s political fortunes, President Ospina Pérez dis-
 Medellín and Core Municipalities 

missed Governor Arango Ferrer in late  and named a conciliatory

Conservative in his place.

An Interlude in the Escalation of Violence

Violence was never constant in Antioquia between  and , not
even during the tenure of especially extremist political appointees or
after the demise of such an important opposition leader as Jorge Eliécer
Gaitán. Partisan unrest was cyclical and concentrated, most marked dur-
ing periods of electoral competition or in the immediate aftermath of
a devastating event such as the murder of Gaitán. In fact, reports of
violence waged by public employees against the Liberal opposition in
Antioquia receded during the last months of . Then in early  a
long-simmering feud in Medellín’s Liberal-dominated concejo once more
brought partisan dissension to the forefront of regional concerns.
Council members were locked in a battle over the city’s unresolved
financial difficulties and the possibility that these might lead to bank-
ruptcy. The newly appointed governor, Fernando Gómez Martínez
(December –July ) attempted to reason with recalcitrant Con-
servative council members and entreated them to negotiate with the gai-
tanista opposition who controlled several seats in the concejo. He insisted
on calling together members of the Liberal and Conservative parties
and enjoined them to cooperate to govern the city and resolve its finan-
cial problems without resort to violence.77 The stubborn refusal of some
laureanistas to have any dealings with those they pejoratively dismissed
as ‘‘nueve abrileños,’’ however, moved the governor to issue a manifesto
exhorting Antioquia’s citizens to respect differences of political opinion.78
The struggle between the governor and members of his own party as-
sumed the character of a public debate regarding the meaning of politics
and the appropriate way they should be conducted in Colombia. Dur-
ing a radio interview with the newly appointed secretary of education
in Antioquia, Gómez Martínez seized the opportunity to launch an edu-
cational campaign to teach Colombians how to participate peacefully in
politics. In Colombia, Gómez Martínez lamented,

The period preceding the election is full of violent agitation, of insults

against the opponent. . . . This mistaken understanding of what poli-
tics and parties are about is exactly what I would like to see changed
 Blood and Fire

in the minds of Antioquia’s children, [and this can be done] through

the good offices of the secretary of education . . . . Politics is the art of
governing not in the name of a party but in the interests of the gen-
eral citizenry. . . . I would like to impress upon children’s very soul the
notion that it is savage to fight and hate one another over politics.79

The indiscriminate use of violent rhetoric and incessant wrangling over

jobs and patronage distribution in the concejo also drew the ire of the
regional executive committee of the National Federation of Merchants
(). After the heated council dispute between laureanistas and
Liberals in February , Medellín’s  chapter wrote an open
letter in which it reminded the governor of the region’s historic commit-
ment to technocracy, merit-based hiring, and economic growth. The fed-
eration insisted it was entirely ‘‘removed from partisan political issues’’
and acted ‘‘solely in its capacity as a social force interested in the progress
of Medellín.’’ But it also pointedly reminded the governor that as the
‘‘representative of the commercial sector’’ it ‘‘was the main contributor
to municipal revenues’’ and thus entitled to express an opinion regard-
ing ‘‘the lamentable state of bankruptcy in which the municipality finds
itself.’’ 80
The city’s fiscal situation had deteriorated since once-marginal, Con-
servative, middle-sector politicians had begun to compete with Liberal
bosses over control of the city’s patronage hiring and public works bud-
get. The unprecedented struggle over public revenue had given rise, la-
mented , ‘‘to the most insane policy ever seen in the city’s his-
tory. The [regional] administration is in a state of constant flux: mayors
and executives of public entities are removed because they fail to satisfy
the electoral interests of one or another party, or are obliged to resign
because they fail to meet with the cooperation necessary to develop an
urgent administrative task.’’ 81 The concejo had become little more than
a boxing ring in which members of each party sparred to monopolize
appointments in the public sector and control of the city’s -million-
peso budget and several thousand employees.  complained that
deficits were remedied by overtaxing the city’s citizens and businesses
in an ‘‘illegal and antitechnical’’ manner; employees and workers were
no longer paid on time; urgent ‘‘works of progress’’ were ‘‘paralyzed’’;
and the city had stopped investing in the modernization of working-class
neighborhoods. The merchant’s association asked unions and businesses
 Medellín and Core Municipalities 

to join it in an act of solidarity to demand that the regional government

‘‘place the public interest above partisan interests.’’ In desperation, the as-
sociation also called upon its affiliates to refuse to pay their taxes until the
regional government assumed responsibility for mismanaging the city’s
affairs.
The unfortunate recipient of this missive was the only governor of the
four Conservatives to serve in that office since  who had explicitly
not engaged in partisan hiring. Indeed, Gómez Martínez had had to with-
stand the excoriation of hard-liners within his party who berated him for
refusing to dismiss faculty members at the University of Antioquia simply
because they were Liberal. He had also refused to give in to Conserva-
tive criticism regarding his appointment of members of the opposition
who he considered to be the best qualified as contractors and supervis-
ing engineers on the region’s public works.82 But ’s point was
well taken. For decades Antioquia had taken pride in not participating
in ‘‘Bogotá’s political train’’ and had considered itself a national leader in
‘‘administrative honesty and true political liberty.’’ 83 The region’s native
son, Mariano Ospina Rodríguez had coined the phrase ‘‘to govern is to
manage [gobernar es administrar] while Tulio Ospina Pérez and Alejan-
dro López (a Conservative and Liberal, respectively), two of the region’s
best-known statesmen, had consistently argued that public service posi-
tions would be more efficiently run if they were not part and parcel of
political patronage.84 All three believed that the nation’s ministerial posi-
tions should be filled with ‘‘engineers who are men of ministerial quality
who can inspire confidence in the public’’ rather than professional poli-
ticians.85 It was this tradition of suprapartisan technocratic management
that  mourned had been sacrificed in the name of electoral
interests.
The governor’s speech on the need for political civility and ’s
letter warning that continued partisan conflict held back regional de-
velopment prompted a derisive response from middle-sector politicians
such as those connected with the newspaper La Defensa. These politicians
made it clear that they were indifferent to comments that attempted to
shame them into negotiating with their adversaries.86 When several days
later the governor wrote an editorial exhorting politicians to limit the
use of hate speech, former regional secretary Eduardo Berrío González
retorted by denouncing the existence of a Liberal Communist conspiracy
against the nation.87 In response, the governor warned fellow Conserva-
 Blood and Fire

tives not to manipulate the fear of Communist revolution to block or

interfere with Liberal campaign meetings. He recommended Conserva-
tives stay home to avoid possible clashes, and he prohibited the public use
of ‘‘abajos’’ (down with) and ‘‘mueras’’ (death to) against the opposition.
Gómez Martínez’s attempts to reduce the escalation of partisan antago-
nism could not compete, however, with Eduardo Berrío’s red-baiting. In
such a climate, the governor’s insistence that ‘‘Conservatives . . . should
respect and accept the free exercise of the right [of free speech] by Lib-
erals’’ fell on deaf ears, as did his warning to municipal mayors to be
strictly impartial in the upcoming June elections.88
The governor’s explicit denunciation of violence as a legitimate politi-
cal tool and his refusal to countenance the idea that any means were
legitimate in the pursuit of electoral victories marked a turning point in
regional politics and in the development of la Violencia.89 Officeholders
and politicians, who were the standard-bearers of a hegemonic approach
to politics, increasingly eclipsed Conservatives who defended and pro-
moted the rights of the opposition to take part in political office and de-
bate. Politicians such as Eduardo Berrío González embraced a no-holds-
barred approach to campaigning, one that relied extensively on appeals
to past sacrifices and perceived wounds received at the hands of the oppo-
sition. He and other extremists whipped up popular Conservative sen-
timent to discredit and upstage moderates. In the lexicon of the day,
Gómez Martínez and moderates like him were dismissed as ‘‘notables’’
and ‘‘oligarchs,’’ men whose aloof approach to politics deliberately ex-
cluded all but the members of the regional economic elite from real politi-
cal participation and decision-making. These accusations were largely
valid. Individuals such as Gómez Martínez did represent a paternalistic,
top-down approach to politics, one that typically assumed that the popu-
lar classes were in need of civilizing and elite leaders who could govern
for them. However, it was a poor bargain for regional voters to have to
choose between populists who did not shy away from deploying physical
coercion to achieve their political goals and an elite who deplored vio-
lence but could not bring itself to expand the parameters of the political
arena to include others.
By April the break between the governor and his party directorate and,
in a broader sense, between a regional bipartisan political tradition and
the new politics of violent hegemony was complete. On April , Antio-
quia’s Conservative directorate openly disobeyed the governor’s call for
 Medellín and Core Municipalities 

a dialogue between Conservatives and gaitanistas in the still-simmering

municipal council dispute. In a last ditch effort to restore a modicum of
civility to the conduct of regional politics, the governor turned once more
to the editorial pages of El Colombiano, this time to denounce recent at-
tacks by the contrachusma against Liberal directorates and voters. ‘‘Our
party,’’ he warned, ‘‘does not accept services of that sort. . . . We are a col-
lectivity of ideas not ferocious instincts, and we operate with reason or
by reason. . . . Consequently, conservatism not only does not endorse at-
titudes of that nature, but rather publicly rejects and condemns them.’’ 90
The governor’s party, however, was now in the hands of extremists for
whom electoral enforcers who terrorized the countryside and who dis-
missed anyone who disagreed or stood in their way as ‘‘traitors’’ were
legitimate political allies. Fernando Gómez Martínez’s dismay that vio-
lence had become an accepted strategy for members of his party was poi-
gnantly apparent in one of his last attempts to repudiate its use for elec-
toral purposes: ‘‘We the directors of the parties unreservedly condemn
the use of [violence] and ask all party members to also condemn it, what-
ever camp from which it may emerge, since it would be unfair to repudiate
its use when one’s own party is the victim while stimulating its exercise
against the adversary.’’ 91
The governor’s plea for a return to civility bore little fruit. Elections
perceived to be critical to the future of the Conservative party were just
around the corner; the first for municipal council, the regional assembly,
and congress in June, and then in November the most important elec-
toral contest of all, that for president. Having failed to consolidate their
power in local and regional representative bodies during the  elec-
tions, laureanistas perceived the need to win the  elections with even
greater urgency. The governor’s public commitment to nonviolence and
moderation was thus interpreted by middle-sector politicians as the prin-
cipal impediment to the satisfaction of their ambitions. In such a context,
the actions that won Fernando Gómez Martínez a modicum of Liberal
trust could only guarantee him the enmity of ambitious and ideologically
extremist members of his own party. José Corréa, Antioquia’s Conserva-
tive party whip, in fact publicly derided Gómez Martínez as ‘‘a small intel-
lectual and mamatoco’’ and accused him of governing ‘‘in fear of Emilio
Jaramillo [the editor of the Liberal El Diario].’’ 92
As the pressure to ensure a Conservative victory grew, the active par-
ticipation in partisan attacks by official forces such as policemen and
 Blood and Fire

mayors became more frequent. Of twenty-three reports of violence or

abuse sent to the governor’s office between January  and June ,
eleven specifically attributed violence to the actions of a mayor. Seven re-
ports also mentioned that the police took part in or refused to intercede
on behalf of a victim being harassed by members of the opposition party.
Indeed, as Eduardo Berrío González made clear to Governor Fernando
Gómez Martínez in , mayors who simply belonged to the party but
who failed to use their position to aggressively promote the party’s elec-
toral interests were little better than ‘‘partisan functionar[ies], traitors
[entreguistas], and as such a social menace.’’ 93 Berrío recommended re-
placing such ‘‘elements’’ to avoid electoral defeat. The actions of ‘‘loyal’’
mayors like those endorsed by Berrío, however, prompted an escala-
tion of local complaints regarding abuses of authority. Fights broke out
in numerous municipalities over who should be allowed to register to
vote; Liberals complained that their right to suffrage was inadequately
protected, and local inhabitants in general protested the lack of reliable
police agents or army troops able to respond to outbreaks of public dis-
order in Antioquia.94
The police were the most frequent accomplices of mayoral abuse, and
unionized workers were usually the main target. In April the police were
accused of mistreating Liberals in the towns of Cisneros and Itagüí where
numerous railroad and industrial workers were based.95 Itagüí’s mayor,
party whip, and police engaged in what the local Liberal directorate called
‘‘cannibal-like excesses’’ against the town’s Liberals. This incident was fol-
lowed the next month by partisan-motivated skirmishes in which several
Liberals were wounded or killed.96 Violence in the largely working-class
town escalated so dramatically that the minister of government himself
interceded to demand that the governor appoint a military mayor in the
civilian mayor’s place.97
Indeed, the initiation of the  electoral season revived tensions be-
tween workers and the regional authorities that had seemingly dissipated
in the months following Gaitán’s assassination. In January, the police
mounted a train carrying railroad union officers on their way to deliver a
report to the union’s annual meeting in Puerto Berrío and pushed these
passengers off. The union promptly accused the government of engaging
in ‘‘a new crime against the right to unionize.’’ 98 In March the Frontino
Gold Mines Union in Segovia complained that the mayor appointed by
the governor was ‘‘an ANGLOPHILE’’ who defended the company in-
 Medellín and Core Municipalities 

stead of being ‘‘a safeguard for the people [pueblo].’’ 99 On April , road
workers in the same town struck to protest the fact that they had long
gone unpaid and that local merchants were refusing to extend them any
more credit. Rather than negotiate, the governor sent a squad of police,
exacerbating the strikers’ defiance. The road workers threatened to take
violent action and to commandeer government warehouses if in two
weeks they were not paid their salaries that were four months overdue.100
On April , when matters between the governor and the road workers
reached a stalemate, the minister of public works in Bogotá expressed his
concern that if the dispute with militant road workers in Antioquia were
not resolved, it would give rise to ‘‘complications with tremendous reper-
cussions in the country’s other labor camps.’’ 101 To make matters worse,
shortly after this, the mayor of Medellín, Julio Arias Roldán, dismissed
 municipal public works employees with between five and twenty
years of seniority after first inquiring as to their political affiliation.102
In a resolution issued during their eighth annual union meeting in
May , the Federation of Antioqueño Workers (), a  af-
filiate, publicly denounced the state’s systematic use of violence against
state workers. The union complained that public road workers in San
Andres, Ituango, and Segovia ‘‘have ignominiously been harassed by both
the civilian and eccleasiastic authorities’’ without the government lifting
a finger in the workers’ defense. Indeed, mayors appointed by the re-
gional authorities in San Andrés and Ituango led the attack on workers,
prompting  to warn both its affiliates and regional workers not af-
filiated with a union ‘‘not to allow themselves to be provoked or tricked by
imposters and humbugs.’’ In the same resolution,  also accused
Segovia’s mayor and the labor inspector, who were appointed by the gov-
ernor, of using their positions to abuse workers at the Frontino Gold
Mines Company.103 Jobs as much as votes appear to have been at stake in
the growing conflict between workers and the regional authorities. This
became clear from a complaint made by the Conservative committee in
Yarumal (near Ituango and San Andrés) in which members accused the
road workers’ supervisor of threatening to dismiss Conservative workers
if they voted in the upcoming elections.104
Despite abundant evidence that the source of conflict in most Antio-
queño municipalities was the result of public employee and departmental
or national police abuse, the regional government insisted local conflicts
were the work of labor agitators and excessive intemperance. In late April
 Blood and Fire

 the national government passed a general prohibition of all union

and party-related demonstrations regardless of the participants’ affilia-
tion. It prohibited as well the sale of liquor in the period immediately
prior to the elections.105 Attempts by the regional Liberal party to hold
meetings with their followers without violating the central government’s
prohibition of public political demonstrations, however, only served to
infuriate the local Conservatives and to further incite them into acts of
violence. When Liberal leader Captain Julián Uribe Gaviria held a pri-
vate conference with his followers in Andes that had the explicit prior
approval of the governor, for instance, the mayor and police chief broke it
up violently.106 On the same day, the municipal Conservative committee
in Bolívar telegraphed the governor to complain that Liberals from vari-
ous southwestern towns had gathered en masse in Bolombolo to await
the train bearing the Liberal leader. Campaign stops, like those which had
taken place since time immemorial during election season in Colombia,
were twisted by local extremists into conspiracies in which the ultimate
aim of meetings held by the opposition was to ‘‘take over the southwest
starting with Bolívar.’’ The police and mayor then justified their attack
against Liberals as an act of self-defense intended to restrain a ‘‘violent
Liberal uprising’’ in the making.107
Even when the police did not actively intervene to abuse Liberal voters,
they often refused to restrain public employees who did. Liberals in Yaru-
mal complained that the town secretary (the mayor’s right-hand man)
had insulted another resident from a tavern, pulled out a gun and chal-
lenged the victim, yelling ‘‘down with those who accused the mayor,
down with the son-of-a-bitch Reds, up with the Conservative party. . . .
I’ve only got this gun and six bullets, but I’ll pump them into you. We’re
in control, down with the Liberal party.’’ The police, privy to the entire
incident, stood by idly.108
Indeed, while many citizens attributed the escalation of violence to
the absence of official forces, these forces were so often the perpetrators
of violence between  and  that one wonders why anyone both-
ered to suggest that the presence of the authorities could have been much
help. And yet citizens repeatedly requested that the regional government
send forces to defend them. Liberals in San Roque and Caracolí in eastern
Antioquia complained, for instance, that the police, Conservative civil-
ians, and public employees had made them live ‘‘terrifying hours . . . en-
 Medellín and Core Municipalities 

dangering our lives, our homes, and our families,’’ and then added that
they ‘‘lacked any authorities.’’ 109 In a broader sense, of course, citizen
complaints that the state failed to protect them and that its forces were in-
adequately deployed throughout the region were intended as an explicit
indictment of the regional state’s policies. Antioqueños were expressing a
profound sense of outrage that the regional state had violated its citizens’
trust. By reminding Antioquia’s regional authorities of their obligation
to defend citizens’ rights regardless of partisan affiliation, citizens (many
of them Liberal) gave evidence that the use of violence by public employ-
ees for partisan purposes had transgressed the region’s political traditions
and expectations.
Persistent citizen complaints of official abuse that were addressed to
the governor also underscored the existence of two competing mandates
within Antioquia’s regional administration. While Gómez Martínez de-
fended the right of the opposition to take part in elections and hold
offices, his subordinates followed other orders. Widespread insubordina-
tion within the regional government was apparent in a complaint filed
with the governor’s secretary protesting a local mayor’s use of police and
contrachusma forces to intimidate Liberal voters during the elections of
June . The petitioner described the terror unleashed in the once-
Liberal town of Olaya where laureanista Conservatives were in the process
of forcibly conservatizing the town’s electorate. When confronted with
complaints of partisan abuse and the threat that these actions would be
brought to the attention of the governor, the mayor ‘‘mumbled under
his breath that the upcoming elections would be won by the Conserva-
tive party by ‘FIRE AND BLOOD’ and that he didn’t fear the governor of
Antioquia [Fernando Gómez Martínez] or his secretary of government
so long as [a certain individual] held the office of assistant secretary of
government and [a certain individual] were in the administrative visi-
tor’s office, since the latter had expressly appointed him to the munici-
pality to ensure the party’s triumph.’’ The petitioner pointed out that this
attitude directly contradicted the governor’s public declaration ‘‘of not
tolerating the presence of politically belligerent individuals in the office
of municipal mayor.’’ 110 Indeed, when a bipartisan committee of Liberal
and Conservative leaders in Jardín filed a similar complaint, they also em-
phasized the dichotomy between the governor’s stated opposition to the
participation of public employees in partisan affairs and the local reality.
 Blood and Fire

In their town, for instance, those deploying the police and contrachusma
against members of the opposition simply laughed when asked by a bi-
partisan commission to put an end to such abuses and scoffed that they
had the full support of members of the regional government. They defi-
antly declared ‘‘that they are Conservatives and that they are connected
to certain political elements who support them and would take up their
defense were we to proceed against them.’’ 111
No monolithic party structure existed to ensure coordination between
the various levels of party organization nor was the state sufficiently inte-
grated to enable the governor to demand the loyalty of his subordinates.
Power was a complicated thing negotiated at seemingly obscure levels
of authority within the regional administration. Conservative loyalists
who controlled the day-to-day workings of the regional bureaucracy in
what appeared to be unimportant secondary posts knew that they could
defy the governor because they enjoyed the protection and endorsement
of political bosses such as Eduardo Berrío González (the party whip) or
Manuel Chavarriaga (the superintendent of customs agents). Thus, by
May  when Gómez Martínez’s policy of equitably distributing politi-
cal posts among members of the opposition and Conservatives (other-
wise known as cruce)—in keeping with Mariano Ospina Pérez’s national
policy—was attacked by local extremists, neither the governor nor the
president were able to respond effectively. Extremists consistently com-
plained that cruce enabled members of the opposition to hold politi-
cal offices without having to commit themselves to any of the governing
party’s policies and also that the policy deprived local Conservatives of
positions they regarded as their political due for having brought the party
to power. To circumvent cruce, hard-liners had simply gone behind the
governor’s back and ignored or undermined the authority of Liberals
who were appointed by the governor to local office.
Despite consistent disobedience from his subordinates in the regional
administration, Gómez Martínez’s efforts to defend the rights of all
Antioqueños to take part in politics do appear to have made some
small difference. Liberals who had held political office during Gómez
Martínez’s administration wrote the governor to reassure him that he had
offered them ‘‘complete support [plenas garantías].’’ Only the breakup
of the ‘‘national union’’ accord between Liberals and Conservatives in
May  had forced them to resign from the ‘‘government your lordship
 Medellín and Core Municipalities 

so ably directs.’’ 112 The governor’s commitment to the politics of civility

also prompted prominent Liberals to intercede on the behalf of particular
towns to request a military mayor even though they knew that every mili-
tary appointment to a civilian office represented one less patronage post
in the Conservatives’ hands. Ricardo Moreno assured Gómez Martínez
that ‘‘I have always admired you from my Liberal trench. Today as I read
your public address I admire you more as a government leader,’’ and he
insisted that it was precisely because he believed ‘‘in the sincerity of your
words’’ that he dared request a change of mayor for the overwhelmingly
Liberal town of Remedios.113
But on May , , assurances of loyalty and support for a bipartisan
elite political tradition such as that rendered by Ricardo Moreno to Fer-
nando Gómez Martínez became moot. Liberal governors resigned and
the ministries of war, government, and justice were assigned to military
officers. In a final, desperate gesture toward bipartisan accommodation,
Governor Gómez Martínez offered those Liberal mayors in Antioquia
who wished to do so the possibility of remaining in office, but it was al-
ready too late.114 Divisions between the governor and his subordinates
and between different sectors of the regional bureaucracy had reached
critical proportions. In such circumstances no Liberal in his right mind
was willing to risk his life or political future to support the governor.
Chaos reigned within the regional bureaucracy. Rentas officers and
soldiers in Puerto Berrío knocked down the doors of the state-run liquor
monopoly, looted the stocks, wounded employees, and came to blows
with local state monopoly and tax officers [Resguardo agents].115 In Cara-
colí on the eastern railroad line and in Conservative San Rafael to the
southeast, Liberals requested that the municipal police be confined to
their barracks on election day because they distrusted their ability to re-
main neutral or guarantee public order. Local Liberals demanded instead
the presence of departmental police agents who were not the minions
of local Conservative politicians.116 In Titiribí, in contrast, the mayor re-
quested that the military be left to keep order in the town until after the
elections, because the municipal police were all Liberal and were likely
to celebrate their electoral victory by leading violent attacks against local
Conservatives.117
By late June it seemed as if any remaining will to restrain the escala-
tion of conflict in Antioquia had dissolved. Both regional parties were
 Blood and Fire

internally divided and the chain of command between the governor and
his subordinates had broken down. A free-for-all, played out mainly at
the municipal level, ensued.

The Outcome of Electoral Violence

Three days after the June  elections, the regional newspaper, El Colom-
biano, publicly accused laureanistas of having waged Antioquia’s elec-
toral campaign against Governor Fernando Gómez Martínez and the
daily.118 What the newspaper’s editors meant was that moderate Conser-
vatives and the region’s bipartisan tradition were the real losers in the
June elections. Indeed, the irony of the June elections was that while the
pronounced use of official intimidation irrevocably split regional Con-
servatives, repression had once more failed to change the partisan com-
position of most Antioqueño towns.119 Liberal municipalities continued
to retain their majorities and few competitive towns tilted in the Conser-
vatives’ favor.
Of the forty-four towns in which  percent or more of the votes cast
in the legislative assembly elections of  had been Liberal, twenty-four
continued to be predominantly Liberal in , nine still boasted a nar-
row Liberal majority ( to  percent of total votes), and eleven shifted
to a Conservative majority. Of seventeen historically competitive towns
( to  percent of votes being either Liberal or Conservative), seven
shifted over to a Conservative majority in the  elections. While Con-
servatives were able to win a majority in the regional assembly (twenty
to fifteen) and in the house of representatives (ten to seven), Medellín’s
town council remained solidly in Liberal hands (with a majority of nine
to six).120
Thus, several towns that had always delivered a Liberal majority con-
tinued to do so despite the use of official force against Liberal voters. In
some towns Liberals actually increased the number of votes cast for their
party: nine towns increased their percentage of Liberal votes between 
and  percent between the  and  Assembly elections. Liberals
also won control of municipal councils in nearly all of the towns where
they had historically constituted a significant majority of the electorate
(see map ).121 In municipalities with a strong presence of public road
workers and miners, moreover, a significant number of votes continued
to be cast for candidates identified as gaitanistas (see table ).
 Medellín and Core Municipalities 

Map . Sites of Liberal power. (Source: Instituto Geográfico Augustín Codazzi and
Colombia, Departamento Administrativo Nacional de Estadística)

Like the results of the  elections, the towns in which electoral
change as a result of the threat or use of violence was most pronounced
tended to be historically competitive towns or towns where Liberals out-
numbered Conservatives but could not be said to constitute an over-
whelming majority (see table ). There were also towns where an unchar-
acteristic surge in Conservative voters narrowed but failed to eliminate
Liberal majorities between  and  (see table ).
Several possible factors account for the reduced ability of these towns
to withstand Conservative pressure when contrasted with the success-
ful resistance of other towns in which organized labor had an impor-
tant presence, such as those of the northeast, Magdalena Medio, west,
and Urabá. Workers in centrally located towns tended to be industrial
 Blood and Fire

Table 3. Percent total concejo votes for gaitanistas, ,

concejo elections, June , 
Percentage
Region Municipio gaitanista

Occidente Dabeiba 
Urabá Turbo 

Bajo Cauca Zaragoza

Occidente San Jerónimo

Nordeste Segovia

Central Ebejico 
Bajo Cauca Cáceres 

Occidente Frontino 
Oriente Retiro 
Urabá Chigorodó 

(Source: Colombia, , Anuario Estadístico de Antioquia, Años , , , Apén-
dice /, ‘‘Estadística Electoral’’)

workers affiliated with Catholic trade unions. Moreover, unlike the towns
in which the weight of organized labor was most strongly felt (Magda-
lena Medio, Urabá, the Bajo Cauca, and so on) Envigado, Itagüí, and
Amagá had a significant number of Conservative residents. It generally
proved impossible to conservatize towns such as those in eastern or west-
ern Antioquia where local Conservatives were few in number and where
the surrounding municipalities boasted Liberal majorities.
The existence of ‘‘islands of liberalism’’ was acknowledged by regional
laureanistas and essentially left alone. The laureanistas perceived the en-
tire area north of the western town of Cañasgordas, the eastern half of
Antioquia, the Bajo Cauca, and the Magdalena Medio (that is, the periph-
ery) to be impervious to Conservative proselytizing, no matter how in-
timidating and severe.122 Given these limitations and the fact that no
consensus existed within the regional party regarding the use of force
to achieve electoral victories, laureanistas focused their electoral efforts
in  as they had in : on towns situated near monolithically or
overwhelmingly Conservative municipalities. The strategy of mobiliz-
ing Conservatives from one town to attack Liberals in another helps ex-
plain why individuals who were interviewed about their memory of the
emergence of violence associated it with an invasion led by outsider, or
forastero, Conservatives.123 Such outsider efforts could, of course, only
 Medellín and Core Municipalities 

Table 4. Competitive and Liberal towns that were conservatized,

Assembly elections, –
% Liberal % decline of Liberals
Electoral type Municipio Region      – 

Competitive Anorí Nordeste   

Competitive Caramanta Sur  


Competitive Valparaiso Sur   
Competitive Andes Suroeste   
Competitive Bolívar Suroeste 
 
Liberal Tarso Suroeste 
 
Liberal Pueblorrico Suroeste   

Liberal Campamento Occidente   

Liberal Olaya Occidente 


Liberal Yolombó Nordeste   
Liberal Concepción Oriente 



Liberal San Roque Nordeste 



Liberal Bello Central  

Liberal Montebello Sur 
 
Liberal Barbosa Central   

(Source: Colombia, , Anuario Estadístico de Antioquia, Años , , , Apén-
dice /, ‘‘Estadística Electoral’’)

succeed where they resonated and found local support. Conversely, in

solidly Conservative towns where Liberal voters had historically been a
small minority, Liberals simply abstained from voting in concejo elections
without any violence necessarily occurring.124
The results of the June  elections suggest that towns susceptible
to conservatization had already shifted their dominant affiliation from
Liberal to Conservative between  and , that is, during Governor
José María Bernal’s administration. Very little if anything was thus ac-
complished by increasing the level of intimidation or force used to ex-
tend these electoral gains in . Indeed, perhaps surprisingly, the use
of violence as a tactic of political conversion proved rather ineffective in
Antioquia, perhaps because to be successful local conservatization cam-
paigns required considerable and consistent support from the regional
administration. When hard-liners such as José María Bernal and Eduardo
Berrío González, who endorsed the use of partisan aggression to achieve
electoral victories, were in office as governors, local conservatizing efforts
could succeed. But when the regional administration was internally di-
 Blood and Fire

Table 5. Liberal towns that withstood conservatization,

Assembly elections, –
Percentage Liberal Percentage decline of Liberals
Municipio      – 

Betania   −


Antioquia Vieja  
−
Cisneros   −
Angelópolis   −
Amagá   −
Envigado  
−
Itagüí 

−

(Source: Colombia, , Anuario Estadístico de Antioquia, Años , , , Apén-
dice /, ‘‘Estadística Electoral’’)

vided or the governor actively opposed such efforts, forcible conserva-

tization largely failed. In this sense laureanista accusations leveled against
Fernando Gómez Martínez were ironically accurate, his repudiation of
violence probably did contribute to his party’s inability to forcibly extend
its electoral control in Antioquia.
Violence, in other words, was not inevitable and could not succeed
when it was embraced or pursued by either a handful of local leaders
or the regional government alone. This impression is particularly re-
inforced by the frequency with which local citizens of both parties attrib-
uted the evolution of violence into all-out partisan warfare to the point
at which public employees and policemen appointed by the regional
government intersected with the presence of local extremists. Thus, the
towns where electoral violence had the greatest effect were precisely those
where laureanistas enjoyed at least some local support and where struc-
tures such as party committees, the Catholic Church, and economic inte-
gration were strongest. Hence, the majority of the towns conservatized
between  and  were located in the coffee-producing south and
southwest and in the industrialized center (see map ).125 The success
of official violence was therefore shaped by the presence of strong link-
ages between local society, the regional government, national institu-
tions, and the existence of positions susceptible to patronage control.
Paradoxically, however, these very same towns were also those where
it was easiest to put a stop to electoral or state-endorsed partisan violence
when it threatened economic interests or deviated away from strictly par-
 Medellín and Core Municipalities 

Map . Conservatization in . (Source: Instituto Geográfico Augustín Codazzi

and Colombia, Departamento Administrativo Nacional de Estadística)

tisan concerns. Since the success of electoral violence hinged on local and
regional coordination and the active promotion of violence by public
employees, the absence of structural linkages meant that there were no
mechanisms through which to channel intimidation nor any structural
basis on which to build a strategy of violence. If there were no mayor or
police, who could act as the regional state’s vanguard? And if there were
no patronage posts to distribute, how could one construct a clientelist
electoral machine?
The importance of linkages through which to channel violence as well
as mediate conflict also partially explains why laureanistas avoided wag-
ing electoral battles outside centrally settled areas. Although the over-
whelmingly Liberal affiliation of towns on the periphery was a natural
 Blood and Fire

deterrent to conservatizing efforts, the absence of municipal Conserva-

tive party committees or a nearby area of solid Conservative support and
reliable means of communication were more significant impediments.
It proved logistically impossible to wage a conservatizing campaign or
exert physical intimidation through the presence of police officers or the
contrachusma where institutional supports were lacking.
It is also striking that the dominant partisan affiliation of a particu-
lar town was not an accurate indicator of the likely incidence of vio-
lence or of its success. A Liberal town such as Caucasia, where militant
road workers and miners were an important presence and where one
might have expected the government or the ambitious right-wing sec-
tor of the Conservative party to use partisan force to deal a deadly blow
to  affiliates, nearly quadrupled its Liberal electorate. A similar phe-
nomenon occurred in Puerto Berrío where Liberals not only doubled
their votes but also voted in significant numbers for gaitanista candidates
in the local municipal council election. Ironically, while identification
as members of a militant labor organization made particular individu-
als the likeliest targets for Conservative violence, this same characteristic
proved the most effective deterrent to Conservative efforts to change the
electoral composition and autonomy of peripheral towns with a Liberal
majority.

Electoral Violence after June 

The discouraging outcome of the June  elections taught Conserva-

tive hard-liners in Antioquia a very important lesson. Laureanistas now
knew that intimidation and the appointment of sympathetic public em-
ployees as local mayors were strategies insufficient to radically alter the
partisan composition of most municipalities in the province. This did
not lead them to question the efficacy of the use of coercion for elec-
toral purposes. Instead, they concluded—as in fact the electoral results
also showed—that without a coordinated regional policy disseminated
through an internally unified regional bureaucracy manned by laurea-
nista sympathizers, efforts to conservatize Antioquia would not succeed.
Hard-liners therefore lobbied to increase their presence in the regional
Conservative directorate and to remove from office any Conservatives
(including the governor) who might impede or disagree with their politi-
cal objectives or tactics. While these strategies were being implemented,
laureanistas also promoted and bankrolled the creation of more extra-
 Medellín and Core Municipalities 

official forces such as the contrachusma who had first emerged in re-
sponse to the insufficiency of Conservative police officers to put down
Liberal unrest after Gaitán’s assassination.
The failure to win a mandate in the June elections spurred an intensifi-
cation of Conservative-led violence throughout Antioquia. Indeed, wide-
spread acts of violence only become commonplace in Antioquia in the
second half of , as hard-liners adopted an increasingly all or noth-
ing (a sangre y fuego) attitude toward the pursuit of political power.126
The first casualties of stepped-up violence in the immediate postelectoral
period were the Liberal inhabitants of towns such as Andes, Pueblorrico,
Betania, and San Carlos, who were forced to migrate by contrachusma
forces.127 Armed confrontation between Liberals and Conservatives in the
pay of competing landlords or political brokers began to emerge during
this period as well. Two such gangs paralyzed downtown Bolívar for sev-
eral hours during a barroom brawl in July.128 The next day a Conservative
citizen was killed, ostensibly in retaliation for the assassination of a Lib-
eral in June.129 By August conditions in Bolívar had deteriorated to the
point where the minister of government pointedly asked Antioquia’s gov-
ernor whether the appointment of a military mayor to the town might not
be in order.130 The governor ignored the minister’s request, hell-bent on
achieving Conservative domination of the southwestern town whatever
the cost.
Cantinas emerged as the most frequent sites at which exchanges of
insults and physical confrontations between members of the two parties
took place. Townspeople and their drunken mayor shot a man in a cantina
in Santa Barbara on the same day as the confrontation in Bolívar. A
day later the mayor of Santuario in eastern Antioquia reported that two
drunken peasants had shouted ‘‘viva the Conservative party’’ at the elec-
toral registrar and other men who were also drunk. Santuario’s Conser-
vative mayor insisted to the governor that a fight had been averted by the
timely intercession of the police, but several witnesses swore—in what
became a commonplace occurrence—that the police had stood by and
refused to intervene as the participants came to blows.131
The escalation of municipal violence prompted the national and re-
gional governments to belatedly create bipartisan, Pro Paz, or pro-peace,
committees. Liberal and Conservative representatives of major commer-
cial, industrial, financial, and agricultural associations made up the mem-
bership of these. Members of the regional elite, for instance, dominated
Antioquia’s peace committee.132 The viability of bipartisan compromise
 Blood and Fire

or mediation was undercut, however, by the removal from office of the

region’s most vocal advocate of bipartisan cooperation. In July, Fernando
Gómez Martínez was relieved of his post as governor as part of a strategy
to remove from office not only members of the opposition but anyone
critical of any aspect of the regime or of laureanistas prior to the Novem-
ber elections.
As any possibility of retaining a middle ground faded, towns where
there had previously been no indication of unrest began to complain of
sudden outbreaks of violence between July and October of . As was
true of violence between  and the first half of , moreover, nearly
all the complaints of persecution, extortion, or physical abuse forwarded
to either the governor or the minister of government in the second half
of  involved public employees such as mayors, police inspectors, and
policemen. The appointment of mayors with criminal records or who
were named in cases of homicide and assault that were as-yet untried be-
came common. Medellín’s third circuit judge, for example, was forced to
request that the governor remove the mayors of La Ceja and La Estrella
(where governor Eduardo Berrío González had an important following)
because of outstanding indictments against them for homicide and as-
sault and battery.133 Accused mayors often retained their posts or were
merely transferred elsewhere. Thus, in some cases, mayors who had been
sent previously to pacify a town, and who had earned a reputation for ter-
rorizing the community and committing crimes, would be reappointed
when local matters once more required an iron hand. Such was the un-
fortunate case of Puerto Berrío where the president of the town council
denounced the recent appointment of a mayor he referred to as ‘‘a scourge
in the memory of the citizens of this municipality.’’ 134 The continued pres-
ence of such appointees in local offices, even when their criminal pasts be-
came widely known, destroyed what little public confidence might have
existed regarding the propriety or protective guarantees offered to citi-
zens by the state’s representatives.
The months leading up to the presidential election of November 
also proved unusually tense as Conservative extremists redoubled their
attacks against organized labor. On October , Antioquia’s Railroad
Union denounced the murder of Providencia’s railroad station-master
and the wounding of a railroad employee at the hands of a drunken police
inspector in nearby San José between Yolombó and Maceo in eastern
Antioquia. On the same day at the opposite end of Antioquia in the south-
 Medellín and Core Municipalities 

western village of Bolombolo in Venecia, the station-master and munici-

pal police—yelling vivas to their party—attacked railroad workers enjoy-
ing lunch. The union insisted that the spate of attacks was not haphazard
nor simply the product of spontaneous partisan outbursts. Rather, it
blamed the state and its representatives (either in the form of policemen
or public employees), and their hostility toward organized labor for the
rising tide of violence against workers.
The regional authorities apparently hoped that by empowering Con-
servative civilians to lead attacks against public-sector workers such inci-
dents would be dismissed as the product of unruly elements over whom
the regional government wielded no direct authority, thereby distanc-
ing the state from the appearance of being an instigator of violence. But
union officials saw through such a ploy. ‘‘Those attacks’’ they insisted,
‘‘are not all the product of private citizens in an exalted state of parti-
san fervor, the majority are led, in collusion with individuals of low in-
stincts, by the very public employees and agents charged with protect-
ing the tranquility, honor, life, and goods of the union’s membership.’’ 135
Police hostility toward ‘‘the railroad personnel,’’ union representatives
insisted, ‘‘had been noticeable for some time’’ although the union felt it
had recently gotten worse, forcing workers to choose between ‘‘on the
one hand . . . a natural instinct of self preservation and, on the other, the
interests of the Antioqueño economy.’’ The ‘‘frankly hostile’’ attitude as-
sumed by the police against the railroad workers convinced the union
that the police were concerned by their responsibility to ‘‘protect private
factories,’’ not the lives or rights of workers.136
Two months after the railroad union’s complaint of police abuse
against railroad workers, the military personnel stationed in Venecia re-
ported that a public road worker had been stabbed to death by civilians
while having a drink with his girlfriend at a local cantina. The incident
had taken place directly in front of the mayor’s office and ‘‘with police-
men less than twenty feet away and only two or three minutes after having
conducted a weapons search.’’ But the routine search for arms, the mili-
tary officer suggested, was unlikely to deter assassins for when it came to
searches of ‘‘certain individuals,’’ these were done ‘‘in an entirely staged
fashion.’’ Indeed, in this particular instance ‘‘the two assassins having
committed their crime left to hide their arms in the cantina of a Con-
servative boss . . . suspected of instigating the [contra]chusma to elimi-
nate the Liberals in this town.’’ A few days later neither the mayor nor
 Blood and Fire

the cantina owner bothered to hide their links to the contrachusma. ‘‘All
the contrachusma and part of those drunken Conservative men armed
with a peinilla [small machete] and revolver [and] boasting of their power
and official support’’ sallied forth into the town and terrorized with im-
punity. Attempts to stop the informally organized paramilitary forces
were largely futile, the army officer lamented, because the police colluded
with the leaders and ‘‘because the honorable mayor personally informs
the contrachusma when it is safe to arm themselves.’’ 137

The Presidential Election

Reports of municipal violence received in the governor’s office before

October  had almost all involved violence waged against Liberals,
but in October the governor’s office received reports of Liberal violence
directed against members of the ruling party. These incidents did not in-
volve Liberal guerrilla forces such as those that would soon organize in
peripheral towns, but they did represent the first instances of Liberal re-
sistance in Antioquia. Liberal retaliation against Conservative public em-
ployees and citizens was first reported in La Estrella, Montebello, Amagá,
and Concordia, all towns that were affected by state-initiated electoral
violence during the previous three years: a Conservative mayor was
wounded in La Estrella, two Conservatives were wounded in Montebello,
while Amagá and Concordia in the southwest reported a Conservative
killed in each.138 Meanwhile, a statue of the Virgin of Fátima, the patron of
reactionary Conservatives and a symbol of their anti-Communist agenda
arrived in Antioquia on a plane from Girardot, Cundinamarca. Towns-
people throughout the southwest dated the beginning of la Violencia to
the intrusion of processions honoring the Virgin of Fátima during the
month of October.
The first inkling that unchecked partisan unrest might have economic
repercussions emerged in October as well. To protect against rumored
Liberal assault, the president of the  suggested that soldiers
be deployed in southwestern areas where a heavy coffee harvest was
expected.139 Coffee growers were alarmed by reports that government
agents such as a police sergeant by the name of Bedoya were forcing
prominent hacienda owners in Fredonia—including the president of the
regional Liberal party—to ‘‘abandon their haciendas . . . in the midst
of the coffee harvest.’’ 140 Growers were especially perturbed by reports
 Medellín and Core Municipalities 

that Governor Berrío refused to intercede on behalf of the threatened ha-

cienda owners or to send soldiers, even after the minister of government
in Bogotá had ordered him to do so. Repeated complaints had reached
the minister’s office that violence led by government agents was threat-
ening the coffee harvest and cattle and molasses ( panela) production in
the southwest and was forcing the involuntary migration of many inhabi-
tants. Like the members of the Federation of Coffee Growers, the minister
suggested that the governor deploy soldiers to patrol the area between
Andes and Fredonia to protect farmers and round up ‘‘troublemakers
[maleantes]’’ operating in the zone. He also insisted, however, that ‘‘to
pacify in harmony’’ it was necessary that any policy adopted to protect
coffee municipios be determined through bipartisan dialogue and co-
operation.141 The insistence on the use of negotiation and bipartisan par-
ticipation in the resolution of conflict were features that distinguished the
state’s approach to violence in core municipalities from that employed
in peripheral towns where repression was the more common official re-
sponse to unrest.
The fate of the southwestern coffee towns where Conservatives had
concentrated all their efforts to achieve electoral victories in  and
, and where the deployment of reactionary mayors, partisan police-
men, and armed Conservative civilians had been greatest, for instance,
became a matter of obsessive preoccupation to the central government. In
part this concern was prompted by a desire to protect Antioquia’s coffee
sector, the nation’s second largest exporter of Colombia’s primary source
of foreign exchange. But it also partly reflected the president and the min-
ister of government’s growing concern regarding the possible use of vio-
lence against members of the elite, regardless of their political affiliation,
and disagreements between the national the regional government over
how best to maintain public order. This became apparent in the remon-
strative and impatient telegrams with which Minister of Government Luis
Andrade and the president himself bombarded Eduardo Berrío González
in the aftermath of the forced exile of Liberal leader Julián Uribe Gaviria
from his properties.
When Governor Eduardo Berrío refused to collaborate with the oppo-
sition to put a stop to the disruption of the coffee harvest, the minister
was compelled to write the governor a second, more strongly worded
warning. He reminded Berrío of the serious impact violence could have
on Antioquia’s cattle and commercial agriculture and impressed upon
 Blood and Fire

the governor the damage such incidents inflicted on Colombia’s image

abroad. The individuals responsible for the flight of families and the
threat to the coffee harvest should be pursued, Luis Andrade insisted,
‘‘without any considerations . . . [and] without regard to partisan labels.’’
In case there should be any doubt in the governor’s mind about the
meaning of the minister’s words, the minister reiterated that he explicitly
meant that ‘‘assassins, [those engaged in] armed assault and incendiaries
may not feel protected [by claiming to operate] under the banners of
the traditional parties.’’ 142 Again, he repeated, the solution would have
to be a bipartisan one. When still the governor made no move to punish
the government forces involved in perpetuating violence in the south-
west, the minister admonished the governor once more, ‘‘we reiterate the
[national] government’s preoccupation that the criminal doings com-
mitted in the territory under your jurisdiction be severely and rapidly
investigated so as to determine the delinquents responsible [for these
acts] including possible government agents who, exceeding their man-
date, may prove to be implicated.’’ Punishment of those found guilty of
promoting violence, the minister added, should take place ‘‘without par-
tisan, social, or economic discrimination.’’ 143
Eduardo Berrío González must have known that President Ospina
Pérez and his minister were in not in a position to demand the gover-
nor’s resignation and that power had shifted definitively to the laurea-
nista camp, for from this moment on the governor persistently defied
the central government’s demands and never moved to restrain the local
shock troops crucial to his and other extremists’ electoral ambitions. In
coffee-producing towns where the Federation of Coffee Growers played
an enormously influential role, the governor’s indifference to partisan
violence had less severe repercussions; the private producer association
could and did appeal to the national government and the military to
step in and shunt aside the governor’s minions when the actions of these
threatened the town’s economy or carefully negotiated bipartisan rela-
tions. But in towns where a powerful mediating organization such as
 did not exist, migration or armed resistance proved the only
viable responses to intensified state-endorsed violence. When it became
apparent that not even the president himself could stop the governor or
his henchmen from promoting violence, prominent Liberal businessmen
in towns such as Yarumal and Rionegro, for instance, simply closed up
their businesses and left or good.144
 Medellín and Core Municipalities 

The escalation of partisan conflict between October and November

of  had widespread repercussions. National Liberal party leader,
Alfonso López Pumarejo, ordered fellow Liberals to abstain from par-
ticipating in the presidential elections while Liberal leader Darío Ec-
handía begged President Ospina Pérez to postpone elections and form
a bipartisan government.145 The offer of Liberal support came too late
however. Ospina had already concluded that Laureano Gómez’s sway
over Conservatives and the party was too great to resist, and the presi-
dent became increasingly isolated and coercively minded. Purged of its
moderate members, Antioquia’s newly reorganized Conservative direc-
torate categorically rejected any collaboration with the Liberal party and
declared itself fully in support of Laureano Gómez.146 By October 
any negotiations between Liberals and Conservatives were moot; Liberal
leader Carlos Lleras Restrepo broke off talks, declared an insurrection,
and ordered Liberal abstention across the board. In a powerful sym-
bolic act, moreover, he forbade relations between Liberals and Conser-
vatives.147
The national government responded to these actions and to Liberal
threats of retaliation by stepping up the presence of the army throughout
Colombia, especially in areas perceived to be economically and politi-
cally strategic, such as the coffee zone in Antioquia, the industrial hub
around Medellín, and the region’s ports. By November , Mariano Ospina
Pérez put an end to democratic government and declared the nation to
be in a state of siege. Decree  closed the congress, the regional as-
semblies, and all of Colombia’s town councils.148 The response from in-
vestors and capitalists to these measures was immediate: stocks rose on
the eleventh and cattle prices shot up nationwide as the government’s
prohibition of work slowdowns or solidarity strikes helped spur investor
confidence in the propitious prospects of dictatorial power for capitalist
development.149 Coupled with July’s executive decree nationalizing the
police, the interruption of democratic forums ensured that anyone dis-
satisfied by the increasingly coercive nature of internal rule would have
no nonviolent means through which to register their discontent.150

La Violencia Enters Its Second Phase

Shortly after President Ospina’s declaration of a state of siege, Laureano

Gómez ran and won the presidential election in an unopposed, but re-
 Blood and Fire

markably violent contest in late November. Troops were in such great

demand to maintain public order in towns such as Santa Barbara, La
Estrella, Ebéjico, Yarumal, Salgar, and Betulia that the regional govern-
ment was unable to respond to the overwhelming number of municipal
requests for government forces.151 Towns where Liberal delegates were
present reported that the police, with the tacit support and knowledge of
the mayor, mercilessly harassed them. In some towns, moreover, Liberals
were forced to vote for Laureano Gómez on pain of injury, prompting
many citizens to flee to the surrounding hills to wait out the electoral
period in the hope of escaping persecution by the opposition.152 For in-
stance, when Liberals in towns such as Tarso (which had been forcibly
conservatized between  and ) asked the mayor to protect them
from contrachusmas who stripped them of their cédulas, the mayor re-
fused and coldly suggested the victims leave town instead.153
Laureano Gómez’s election marked other shifts in the development of
regional violence as well. An armed Liberal guerrilla movement within
Antioquia emerged in response to the escalation of official violence.
Armed Liberal groups burned, looted, and led insurrections in Ebéjico,
Betulia, and San Juan de Urabá to protest Gómez’s election.154 The inhabi-
tants of San Juan de Urabá shot the police inspector and other members
of the armed forces, then sought cover in the hills when Conservative au-
thorities detonated a bomb in San Juan de Urabá to punish their refusal
to acknowledge Gómez’s victory or accept the imposition of regional ap-
pointees. The parish priest, acting as the town’s intermediary, urged the
regional government to deploy soldiers to reestablish public order. But
he warned the governor not to send the police, for it was precisely their
deployment during past electoral contests and their repressive acts that
had catalyzed the organization of local resistance in the first place.155
The outbreak of rebellion throughout Antioquia’s periphery in the
wake of Laureano Gómez’s electoral victory marked the end of episodic
conflict in the region and the beginning of prolonged insurrection. Pub-
lic sector workers—the primary targets of official harassment between
 and —increasingly took up arms against the state or colluded
with and protected armed Liberal groups operating in their geographic
vicinity. Railroad workers on the trunk road Tulio Ospina-Anzá, for in-
stance, expressed their repudiation of state policies by ‘‘blocking the
search for Liberal bandits’’ and denying ‘‘the presence of the bandits
to successive police teams deployed to different spots in that sector.’’ 156
 Medellín and Core Municipalities 

Had the army—as the Pacific Railroad Union demanded in November

—been sent to patrol the Anzá and Urrao area and to defend railroad
interests, perhaps railroad personnel might have felt less willing to cover
for the newly created Liberal guerrilla groups.157 Instead, state-employed
workers flaunted their reputation for violent unruliness by threatening to
stop work on important public works projects unless the state removed
its repressive forces.
State-directed partisan violence involving paramilitary groups and
the state’s own employees provoked a full-blown struggle between local,
regional, and national authorities over the monopoly of force and the
jurisdictional rights of specific government sectors in Antioquia. These
sources of contention shaped in determinant ways the escalation and
outcome of conflict in the region from  to . Indeed, the use
of the police to harass public-sector workers became so egregious that
it prompted Colonel Carlos Bejarano, the general director in charge of
the national police in Bogotá, to complain directly to Governor Ed-
uardo Berrío González in January . Colonel Bejarano argued that co-
optation of the police by local party officials and bureaucrats and the use
of the police in matters other than the maintenance of public order were
contributing to the escalation of regional unrest.158 The minister of gov-
ernment seconded the police director’s concerns, warning the governor
that the president was worried about the same issue. The minister spe-
cifically alluded to the persistent police and contrachusma persecution of
road workers and employees in the area of Bolombolo, Venecia, and the
regional and local authorities’ seeming indifference to such abuses. The
minister reminded the governor of the president’s interest in seeing the
road (connecting Bolombolo to the coffee municipalities and Medellín)
completed and urged the governor to speak to his local subordinates—
the mayor and police chief in Bolombolo—so that they would cease to
harass Liberal road workers and their Liberal engineer supervisors.159
The commander of the military’s Fourth Brigade in Medellín, Colonel
Villamil, also warned the governor against using the contrachusma in
public order matters. He pointed out the danger of allowing local mayors
to expand the size and operation of the ‘‘civilian police’’ (contrachusma)
in areas where public sector workers were present and where there were
already sufficient numbers of police and soldiers to guarantee the mainte-
nance of public order.160 The minister of government had already warned
Governor Berrío that his persistent, partisan use of the police did not
 Blood and Fire

enjoy the support of the central government. Yet, less than a day after
receiving the minister’s most recent rebuke, Berrío was busy conspiring
with the governor of the Chocó, Guillermo Valencia Ibañez, about an en-
voy of  police agents and arms from Antioquia to be used to put local
Liberals in their place in the Pacific lowlands department.161 Berrio’s insis-
tence on deploying the police as partisan shock troops finally prompted
the commander of the Fourth Brigade to write directly to the minister
of government to complain about the Antioqueño governor’s behavior.
Colonel Eduardo Villamil informed the minister that he had contacted
Eduardo Berrío several times regarding persistent civilian ‘‘complaints
against the abusive behavior of the police,’’ but each time his complaints
had been ignored. The governor, the army commander insisted, retained
in their posts insubordinate police officers such as Major Arturo Velás-
quez who was responsible for firing upon soldiers and being the com-
manding officer of ‘‘the agents responsible for abuses’’ against citizens in
Bolombolo. The inhabitants of Bolombolo had been ‘‘forced to migrate’’
when it was the duty of the regional government to protect them and see
to it that ‘‘they should return.’’ 162
Liberal citizens stopped bothering to complain about the police to the
regional government shortly before the November presidential elections,
and gradually so did regional Conservatives. The latter concluded that it
was pointless to request that the governor do something about the police
or the contrachusma so long as the man in charge of the region’s for-
tunes was Eduardo Berrío González or someone of his ilk. Instead, well-
off Conservatives who witnessed the exercise of police and contrachusma
abuses directed their complaints to the president, perhaps driven by the
vain hope that Ospina Pérez still represented some modicum of civility.
The crisis of legitimacy prompted by public employees who used vio-
lence to pursue partisan objectives was rendered poignantly evident in a
letter written to the president by a Conservative Medellín businessman
in . The writer registered the growing sense of alienation and horror
experienced by members of the regional elite who suddenly found them-
selves living in a police state. The businessman, unaccustomed to being
a target of the police brutality that was common to lower-class Colom-
bian life, was appalled by the police’s sudden sense of self-importance
and seemingly unlimited power. ‘‘Excellent Sir,’’ his missive to the presi-
dent began,
 Medellín and Core Municipalities 

In my character as a Colombian, a Conservative, and a professional,

I cannot but appeal to the First Magistrate to ardently plead for your
intervention on behalf of fellow citizens who are being made the unfor-
tunate victims of a merciless persecution before the complacent gaze,
not of the highest authorities, but of certain police agents who . . .
allow to occur before their very eyes outrages that daily increase the
sediment of bitterness in those who with or without motive, are vic-
tims of the irresponsibility and cruelty of an infuriated and aggressive
rabble.163

On a weekend drive back from his estate on the outskirts of Medellín,

this businessman recounted how he had witnessed ‘‘fifteen completely
inebriated individuals’’ who detained cars along the road into the city
while shouting ‘‘vivas to the Conservative party.’’ Meanwhile, ‘‘the police
agent demanded [the writer’s] driver’s license and Conservative party
identity card.’’ While ‘‘the agent examined these documents, the occu-
pants of the vehicle were forced to suffer the filthy and shameful vocabu-
lary of extremists [the contrachusma].’’ Incredibly, ‘‘they did not limit
themselves to assaulting us with their words, their machetes tested their
hardiness against the car’s windows, bumpers, and doors, the sparks pro-
duced by their bullets contrasted with the gunmen’s metallic pallor, and
all of this, I repeat, was done before the complacent gaze of three or four
policemen.’’ ‘‘Providence,’’ the author said, ‘‘was kind to us’’ since he hap-
pened to be carrying his party identity card. The occupants of the house
where he had been stopped along the road, however, proved less fortu-
nate. To the sound of shouts of ‘‘long live the Pope and down with the
Reds,’’ the men who had stopped smashing his car once assured he was a
Conservative, had begun to break down the dwelling’s door. He had still,
he assured the president, not recovered from the shock or shame of this
dreadful experience.
Indeed, it was clear from the author’s ensuing remarks that his en-
counter with the contrachusma and the police had marked a turning
point in his moral and political development. ‘‘My ideals have been trans-
formed overnight; the dream of a great Colombia, which always made me
love her more, seems close to becoming a chimera, because it seems that
peace and harmony are exotic plants in our midst.’’ He went on to assure
the president that he did not blame him personally for what was happen-
ing in Colombia. He felt current conditions had deeper historical roots
 Blood and Fire

and were not ‘‘born of the current administration, the previous regime
sowed winds and now we have the unhappy task of harvesting tempests,
all Colombians are victims of a state of uncontrollable events.’’ His point
in writing to the president had been to bear witness, to express the desire
of ‘‘the man on the street . . . who aspires to live in peace with God and
his fellow men . . . to live respecting and being respected by others.’’
A cautious man, the author stored his letter in the drawer of his office
desk, thinking ruefully that it was hardly worth bothering the president
with a complaint of events which ‘‘could easily be controlled by the au-
thorities of this department.’’ But in the interim in his position as an ex-
ecutive at the Banco Industrial Colombiano he had the opportunity to
speak to ‘‘people of all kinds, from different towns and all the political
parties.’’ They had confided that ‘‘la Violencia has once again become in
Antioquia the primary source of unrest.’’ He listed all the towns where
violence was present—Caldas, Itaguí, Envigado, Andes, Concordia, La
Ceja, Amagá, Bolívar, Bolombolo, and Titiribí, and he noted that ‘‘the
lives and goods of ’’ Conservatives who dared to complain to the authori-
ties or to the departmental party committee ‘‘are implacably persecuted.’’
This had created a problem of political refugees made up ‘‘not of Lib-
erals, but of Conservatives pursued by local political bosses,’’ and this
problem, coupled with ‘‘the rainy season and the high cost of living,’’ was
making life in Medellín increasingly difficult. So much injustice and the
seeming absence of any regional recourse for redress had inspired him
to send his missive after all. He begged the president to ‘‘enforce your
authority’’ so that the regional authorities would order town mayors to
stop promoting ‘‘treason [manzanillaje]’’ and so that they would punish
those who wished to create further problems ‘‘using violent means.’’ 164
Once Laureano Gómez won the presidency, however, any possibility
of a negotiated settlement to violence or even of a show of civil au-
thority like that pled for by the Medellín banker who wrote in dismay
to Mariano Ospina Pérez disappeared. Disagreements between moder-
ates and laureanistas in Antioquia’s Conservative party developed into an
open rift after Laureano Gómez’s election. In the corridors of regional
power and in private salons, talk centered on the impending shake-up
of appointees in the regional and municipal bureaucracies as Laureano
Gómez’s presidential inauguration drew closer and on the proliferation
and influence of armed privately organized conservative forces in the
countryside. It became evident, notwithstanding the insistence of Fer-
 Medellín and Core Municipalities 

nando Gómez Martínez and other El Colombiano columnists, that ‘‘har-

mony’’ did not and could not reign between Antioquia’s ospinista and
laureanista camps.165
For Antioqueño laureanistas such as José Mejía y Mejía and Beli-
sario Betancur, who were journalists and idealists far removed from the
bloody realities of rural partisan conflict, Laureano Gómez was a prin-
cipled leader motivated by ideological convictions not materialism. They
could explain away the proliferation of groups of poor, armed extremists
as transient anomalies or as unauthorized fringe groups who were unrep-
resentative of laureanismo’s political objectives. Other laureanista sym-
pathizers such as Dionisio Arango Ferrer and Eduardo Berrío González,
in contrast, reconciled the existence of partisan armed groups as a nec-
essary short-term measure that would disappear once society had been
‘‘cleansed’’ of Communists and radical Liberals. Indeed, they often dis-
missed reports that armed Conservatives were organizing throughout
the countryside to eliminate the Liberal opposition as baseless rumors
spread by the opposition to destabilize the Conservative government.
To middle-sector Conservatives, Ospina and his bourgeois, Antioqueño
supporters represented a different but far more sinister threat than did
the contrachusma. The president and his supporters embodied the aloof,
elite politician guided by technocratic considerations and pragmatism
rather than political ideals or partisan ideology. Laureanistas such as
Mejía y Mejía, Vallejo Alvarez, and Betancur detected a profound indif-
ference to the plight and concerns of the ‘‘little person’’ under the ospi-
nistas’ studied neutrality and pacifism. For these Conservatives, ospinistas
practiced a different kind of violence, the violence of exclusion, margin-
ality, and condescension.
Thus, the August  appointment as governor of Braulio Henao
Mejía, a reticent former dean of the University of Antioquia’s medical
college and an in-law of one of the most powerful Conservative elite
clans, came as a considerable shock to Laureano Gómez’s Antioqueño
followers.166 Some loyalists assuaged their own initial sense of betrayal
by insisting that Gómez’s decision had been politically calculated to ap-
pease Antioquia’s bourgeoisie, the party’s wealthiest and strategically
most important group of national supporters. Conservatives who had
viewed Gómez with distaste but who were apolitical used the appoint-
ment of a bourgeois governor as confirmation that Gómez’s Falangist
rhetoric had always been part of an elaborate facade and that once in
 Blood and Fire

power the president would naturally seek to assuage powerful capitalists

and privilege economic development. Many Liberals, however, and even-
tually the Conservative president’s own local loyalists concluded (rightly)
that Gómez could well afford to make the seemingly magnanimous ges-
ture of appointing a non-laureanista to office in Antioquia, because he
knew that throughout the regional bureaucracy and in the regional Con-
servative directorate, where it counted, his minions were in total control.
Whatever Gómez’s real motives, it became quickly apparent that Gover-
nor Henao Mejía did not have the president’s ear and that his appoint-
ment was at least partly intended as a rebuke to a region and an elite that
had always proved notoriously independent and undisciplined.
When they recuperated from their initial sense of disappointment, re-
gional laureanistas realized they could bypass their newly named but dis-
pensable bourgeois governor. They celebrated their sainted chief ’s long-
awaited rise to power and took their Conservative crusade toward the last
remaining unconverted partisan territory of Antioquia’s periphery. Thus
began the next and far bloodier phase of la Violencia.

Conclusions

By  it was apparent that Antioquia, like Colombia, was a house di-
vided. There was no one who could ‘‘enforce the state’s authority’’: not
the president, not the governor, not municipal leaders, not disgruntled
party members. Real power was organized behind closed doors by politi-
cal bosses and their local cronies and in the backroom discussions of cer-
tain regional appointees. The widespread expression of violence made
manifest not the ‘‘breakdown of the State’’ as Paul Oquist suggested, but
its morally weak and organizationally dispersed nature.167 In such a con-
text, the state could not exercise a ‘‘monopoly of force’’ nor could it fulfill
its role as the defender of the well-being and rights of the citizenry as
a whole. Instead, the very forces that should have represented the prin-
ciple of order were nothing more than one among a competing array of
armed groups, all of whom ultimately answered to private and particular
interests and not to the interests of the public.
The use of the police to pursue partisan objectives until they ulti-
mately grew into a force that not even the president could control repre-
sented not a departure but merely the logical fruition of a series of poorly
thought-out civilian policies regarding Colombia’s public order forces.
 Medellín and Core Municipalities 

Fear underpinned these civilian policies. Unsure of their legitimacy or

strength, Liberal and Conservative party leaders had historically proven
unwilling to create viable forces of public order for fear that these might
challenge or usurp civilian authority. Such public order forces conse-
quently fulfilled the state’s repressive functions but were never allowed
to grow into sufficiently coherent entities bound by a code of ethics or
professional identity. Poor pay, lack of discipline, and the subordination
of public order to the interests of private parties and the shifting winds
of political influence ensured that the armed forces would never compete
for moral or physical parity with civilian rulers. This approach may have
maintained intact the civilian leaders’ monopoly of power in the short
term, but it also guaranteed in the long term the impossibility of estab-
lishing a relationship of respect or trust between the armed forces and
their civilian rulers. The police recognized themselves as the pawns of
those who paid them—‘‘la rosca’’ as one police agent put it—rather than
the neutral guarantors of the general public’s well-being.168
As the state found itself under attack from multiple sources during la
Violencia, it relied more and more upon the repressive force of agencies
such as the police to remain in power. This situation increased the au-
tonomy of the police, while the state’s authority over them lessened in
proportion to its dependence and growing lack of legitimacy. To make
matters worse, when faced with an escalation in violence and its own in-
ability to control or direct the police, the regional government opted to
further fracture the already splintered principle of a monopoly of force.
It empowered paramilitary groups to do the work for which it could no
longer rely upon the police—because they were unable—or the army—
because it was unwilling—to do. Ironically, the regional state created and
armed the very groups that would later most seriously challenge its au-
thority and control over the regional territory.
Machuca, Segovia. October 1998. This man’s
wife was among the sixty-six casualties of
an oil pipeline explosion caused by an Ejercito
de Liberación Nacional (ELN) attack.
2. Bajo Cauca, Magdalena Medio, and the Northeast

The geographic focus of violence and its character and organization fun-
damentally shifted after the presidential election of November . The
areas of Antioquia most affected by violence by early  were located
in the northwest and west (Urabá, Dabeiba, Cañasgordas, Frontino),
extreme southwest (Urrao, Betulia, Salgar), northeast (Amalfi, Reme-
dios, Zaragoza) and lower Cauca Valley (Caucasia, Cáceres), and Middle
Magdalena Valley (Puerto Berrío, Puerto Nare, Puerto Triunfo). Parti-
san disputes over appointments and patronage such as those evident in
centrally settled areas before  also characterized the expression of
conflict between Liberals and Conservatives in peripheral areas, but on
the periphery these conflicts evolved into armed confrontation between
organized groups. Liberal guerrilla groups developed and operated only
in peripheral towns, and it was in these towns and not the core munici-
palities of traditional settlement where the greatest number of regional
casualties and forcible displacements occurred from  to . Partisan
violence in peripheral areas intersected with preexisting ethnic, cultural,
or economic tensions, moreover, to produce a far more complex and
multifaceted struggle than that of the exclusively electoral and patronage-
based conflicts that were characteristic of violence in centrally settled
towns during the first three years of la Violencia.
All of the peripheral towns experiencing violence after  shared
certain traits. They were overwhelmingly Liberal—many had supported
the dissident leader Gaitán and his movement in local, regional, and
national elections—and nearly all were areas of recent intense coloniza-
tion efforts and extractive production. But the evolution of violence in the
periphery also differed from area to area. Paramilitary organizations that
were financed and tacitly or overtly endorsed by the regional and central
governments emerged as the primary form of official public order main-
tenance in some towns, while in other towns paramilitary organizations
were fleeting or nonexistent. In some peripheral municipalities, pecu-
liarities of development, identity, and collective resistance determined
that the army or the police contributed most to the intensification of con-
 Blood and Fire

flict. In still other towns, violence began as a partisan struggle over power
between Liberal guerrilla groups and the Conservative authorities and
remained essentially so until ; while in other places partisan disputes
gave way to incipient social demands that transcended partisan differ-
ences or to social and economic banditry and generalized criminality. In
short, the story of la Violencia in Antioquia lies in the details of local
history, and it is the attempt to explore and underscore these differences
and similarities and to deduce the political and economic implications of
la Violencia’s trajectory that shapes the geographically specific narratives
that follow.

Geography and Violence:

The Bajo Cauca, Magdalena Medio, and Northeast

Three parallel chains of the Andes traverse Colombia from north to south
and come together in a rumpled mass at the border of Ecuador in the
macizo central. Two of these mountain chains—the western and cen-
tral cordilleras—run through Antioquia. Coffee towns perch along steep
mountain slopes in the southwestern part of the department, while in the
core—to the immediate north, east, and south of Medellín—the depart-
ment is characterized by a series of hills and valleys where the climate
varies from tierra templada (,–, feet above sea level) to pockets
of tierra fria (,–, feet). At the edges of the department, how-
ever, the mountain terrain drops dramatically, giving way in the north-
west, far east, and northeast to tropical lowlands (,–, feet above
sea level) where the land is far less broken and the climate is hot.1 To the
east and northeast of Antioquia’s Aburra Valley and Medellín lie major
grasslands, rich lodes of ore, and a powerful river—the Magdalena—that
constitutes the artery that connects Antioquia to the rest of Colombia.
The towns cradled by the Cauca, Magdalena, Porce, and Nechí rivers and
their numerous small tributaries are largely made up of flat, tropical low-
lands characterized by extensive ranching, and gold and petroleum ex-
traction (see map ). The air is humid and heavy, awash in the muffled
drone of cicadas and other insects. Gentle hills alternate with deep for-
ests and flat plains. In the northeast, craterlike forms create islands of
arid devastation amid the lush vegetation; these craters are the result of
centuries of relentless gold mining. To the southeast and extreme east,
grasslands extend to the horizon and drop off into the wide expanse of the
 Bajo Cauca, Magdalena Medio, Northeast 

Map . Eastern Antioquia. (Source: Instituto Geográfico Augustín Codazzi)

Magdalena River. Already in the s the rivers in this region ran a tur-
bid color, contaminated by silt, mercury, and human and animal waste.
Human populations in the northeast, Bajo Cauca, and Magdalena
River regions have historically been transient and sparse. In , the
largest towns—Yolombó, Puerto Berrío, Amalfi, and Remedios—ranged
in population from , to , inhabitants. In contrast, municipali-
ties such as Caucasia, Cáceres, Zaragoza, Segovia, and Maceo boasted few
inhabitants—even though their physical size was considerable—scarcely
those needed to herd cattle or man scattered mining operations. Al-
though the northeast, Bajo Cauca, and Magdalena regions covered a third
of Antioquia’s territory (, square kilometers), they were home to
only  percent of the department’s inhabitants. Because of the links be-
tween guerrilla groups operating over this territory, the ethnic and racial
 Blood and Fire

composition of the local population, and the logistical peculiarities of

local settlement and production, I have decided to treat these three dis-
tinct administrative units as a single geographical area for the purposes
of this study. The mines and haciendas established in these areas and
affected by violence often straddled more than one municipality, and
the workers who labored for them and who became la Violencia’s pri-
mary victims circulated seasonally from the lower Cauca River Valley
through the northeastern mining towns and the port of Puerto Berrío
in search of permanent work and lands to colonize. Violence followed
the circuits traced by such seasonal migration and consequently rendered
largely insignificant the administrative boundaries that treat them as self-
contained or discrete entities.2

The Beginning of Conflict

The first clashes between local inhabitants and the regional authorities in
the Bajo Cauca, Magdalena Medio, and northeast (hereafter collectively
referred to as eastern Antioquia) initially arose as they did in other parts
of Antioquia because of disputes regarding the partisan affiliation of pub-
lic employees and the right to appoint and control them. Shortly after
the Conservative party took national power, Antioquia’s regional party
directorate wrote the governor to insist that the continued presence of
Liberals as mayors and customs agents in towns such as Puerto Berrío,
Remedios, Caucasia, Cáceres, and Amalfi constituted ‘‘a real danger to the
government.’’ 3 The Conservative directorate demanded that these Lib-
erals be dismissed and replaced with Conservative loyalists. However, the
fear that such actions would spark labor disturbances and general pro-
tests in areas of strategic economic importance, where the government
had few supporters and only a weak institutional presence, kept even ex-
tremist governors from heeding the directorate’s demands during most
of the period from  to mid-.
But when revolutionary ‘‘juntas’’ seized control of towns such as
Puerto Berrío and labor uprisings broke out in many of the area’s min-
ing camps following Jorge Eliécer Gaitán’s assassination in April , the
regional authorities were forced to reconsider their policy of restraint.
Local disturbances in the area proved so severe and difficult to put down
that Medellín’s Conservative authorities became convinced that there
were plans afoot to overthrow the regional government. To reassure him-
 Bajo Cauca, Magdalena Medio, Northeast 

self, Governor Dionisio Arango Ferrer appointed a visitador administra-

tivo to conduct a formal survey of the Magdalena Medio, Bajo Cauca,
and northeast. The visitador was charged with measuring the amount
of local Conservative support in the area and the population’s attitude
toward the regional government. After closely surveying Segovia, Reme-
dios, Amalfi, and Zaragoza, the government’s agent concluded that only a
few of the inhabitants of El Bagre (a mining settlement in Zaragoza) could
be considered ‘‘friends of the government’’ and that none of the towns he
visited possessed the counterrevolutionary guidelines issued by the re-
gional government to prevent ‘‘popular uprisings.’’ Local Conservatives
were so removed from access to the regional government and the party
directorate that they were even unaware that a recent decree enabled them
to create Conservative civilian police forces should they suspect the local
municipal police of weakening loyalty or need help putting down future
insurrections.4 The visitador gave his regional superiors little hope that
Conservatives or the regional authorities would soon overcome decades
of official alienation and neglect in the area. In most far eastern towns, the
general absence of infrastructure and the weakness of the regional state’s
presence, he ruefully reported, guaranteed that these towns operated in
a semiautonomous fashion as even radical changes in the region’s public
order policy were never received or could essentially be ignored.
Many of the towns in eastern Antioquia were what might be called
‘‘company towns.’’ These were places where a large percentage of the eco-
nomically active population was employed in the same activity (mining
or oil extraction) and by a single (usually foreign) employer in an in-
dustry considered crucial to the economic interests of both the regional
economy and the central state. The town of Segovia was a case in point.
In ,  percent of the town’s seven thousand inhabitants were esti-
mated to work for the Frontino Gold Mines Company, and the majority
were members of the Frontino Gold Mines Union affiliated with the .
The town was also Antioquia’s principal producer of gold in .5 When
local union members took a stand or mobilized to protest the activi-
ties of the state or its public officials, they were supported by the united
sentiment of nearly all the town’s inhabitants. The same tended to be
true of miners in Zaragoza and oil workers in Remedios. This gave the
inhabitants of towns like Segovia, Zaragoza, and Remedios a leverag-
ing power vis-à-vis both the regional and central authorities that was
not easily reproduced in other peripheral or predominantly Liberal areas
 Blood and Fire

of Antioquia. Additionally, the power to effectively protest government

policies was reinforced in the mining zone by the local presence of pub-
lic sector workers such as road construction crews and railroad workers
who, like the miners, were also affiliated with the  and shared many
of the miners’ concerns. At crucial moments, miners and public sec-
tor workers joined forces and mounted a coordinated offensive against
both the state and their employers. In early , for instance, Segovia’s
miners organized to protest the government’s recent restriction of labor
rights.6 Two weeks later miners employed by Zaragoza’s Pato Gold Min-
ing Company allied with miners in Segovia and also announced a strike,
in this case to protest the company’s noncompliance with the collective-
bargaining agreement signed by workers and management in the after-
math of Gaitán’s assassination. Railroad workers on the Antioquia rail-
road and the departmental public employees union followed suit. The
three groups jointly submitted petitions demanding a raise in salaries
and improved social benefits.7 Faced with a possible work stoppage by
miners, public employees, and railroad workers in an area of the depart-
ment where it enjoyed only a tenuous presence, the regional government
had little recourse but to negotiate eastern workers’ demands.
Coordinated union and political activity also enabled the inhabitants
of eastern mining towns to deter attempts to intensify partisan intimi-
dation during the campaign period preceding the June elections of .
Even when they failed to foil official measures absolutely, organized
labor nonetheless effectively forced the governor to rescind the partisan
policies implemented by local appointees. For instance, two weeks be-
fore the first anniversary of the death of the Liberal leader Jorge Eliécer
Gaitán, gaitanistas in Caucasia complained directly to the governor that
the mayor had taken away their copies of the musical record ‘‘They
Killed Gaitán.’’ The mayor had also prohibited local jukeboxes from play-
ing the song. This seemingly insignificant act was construed by workers
as political censorship and as a violation of their right to free speech.
Workers threatened to retaliate violently if the governor did not restrain
the mayor. The Conservative visitador sent to report on and arbitrate
the dispute suggested that the municipal government was ‘‘infiltrated’’
by gaitanista sympathizers and that the town clerk ( personero) and sev-
eral other local public officials were also followers of the fallen leader.
The complete absence of any local Conservative support forced the visi-
tador to conclude that it would be prudent, given the local population’s
 Bajo Cauca, Magdalena Medio, Northeast 

militancy, to insist that the mayor return the seized records and lift the
prohibition against publicly playing the song.8
Although the visitador’s solution to the conflict in Caucasia was con-
ciliatory on the surface, he had ulterior motives for suggesting that the
government capitulate in its dealings with the townspeople. The visitador
used his general report to drive home his perception of the inherently un-
ruly and culturally inferior nature of Caucasia’s inhabitants and to justify
the future implementation of discriminatory policies against them. He
made much, for instance, of the ‘‘costeño’’ (Caribbean coastal) element
he found dominated the area and spared no effort to warn the regional
authorities of the imminent danger that this Afro-Caribbean popula-
tion posed to Antioquia’s general state of public order. The area’s towns-
people, ‘‘especially those in Nechí and Colorado,’’ he lamented, ‘‘were
people accustomed to living without God or the Law. The number of
marriages may be counted on the fingers of a single hand, the others
live in public and scandalous concubinage for that is what is common
there. Since no one respects an oath, crimes remain unpunished, besides
which it is impossible to advance an investigation because it is for the
most part, a cosmopolitan town.’’ Locals were promiscuous, possessed
no moral sense of right and wrong, and were bent on rising up against the
regional government. ‘‘Authority is not respected,’’ the visitador insisted,
‘‘the agents [of the government] are attacked, they are wounded in the
back, their arms are stolen, and there is made against them, in sum, an
open war if they attempt to fulfill their duty.’’ 9 The term ‘‘cosmopolitan’’
was used as a code word to describe towns perceived to be dominated
by the ‘‘other,’’ where the values and patterns of organization and belief
associated with the ideal of Antioqueñidad held little sway. Despite his
diatribe, however, the visitador was sufficiently realistic to concede that
it was beyond the power of the regional government to fundamentally
alter the ‘‘nature’’ of such areas overnight. As an interim solution, he sug-
gested that the offending mayor be switched with the mayor of nearby
Cáceres where no tensions had been reported between the Conservative
appointee and the local Liberal population.10
Local inhabitants in eastern towns further defied the regional authori-
ties by using commemorations of Gaitán’s death to rally public opinion
against the Conservative government and its campaign of partisan in-
timidation. Some towns insisted on flying red flags at half-mast, others
led public protests, and still others commandeered loudspeakers to pub-
 Blood and Fire

licly denounce past government abuses.11 In Zaragoza, moreover, Liberals

applied the same discriminatory and intimidating policies used against
them to marginalize local Conservatives. El Bagre’s Conservative mi-
nority complained that they felt too intimidated by Liberal public em-
ployees to attend political meetings or to vote and that the Liberal mayor,
mayor’s secretary, and municipal police had conspired to dismiss the few
Conservative workers employed by the Pato Consolidated Mining Com-
pany.12 Complaints of partisan discrimination from Yolombó also came
from Conservative and not Liberal citizens. Conservatives insisted that
the town’s Liberal voting officials (registradores electorales) denied poor,
rural Conservatives, who took time off from their agricultural duties to
register to vote, the possibility of doing so.13 Caucasia, however, remained
the town where Liberal threats against the regional authorities’ control
were most severe. There, the town clerk, public school director, and presi-
dent and vice president of the town council publicly incited the citizens
to attack the departmental police troops stationed in the town. They then
made it clear to the regional authorities that Conservative officials who
dared to meddle in local matters would pay a heavy price.14

The Aftermath of the June Elections and the Evolution of Violence

The outcome of the June  elections vindicated the importance of

local Liberal resistance in eastern Antioquia. The number of Conser-
vatives elected to municipal town council positions in towns such as
Remedios, Caucasia, Zaragoza, Segovia, and Puerto Berrío dwindled
to no more than one or two. Before the elections, Conservatives had
avoided direct attempts to remove Liberals from office. Instead, they had
worked to counter Liberal influence by dismissing public sector workers
and appointing Conservatives as police inspectors, work inspectors, and
mayors. The regional authorities had also been relatively circumspect
about deploying Conservative policemen or sponsoring armed civilian
forces to harass eastern towns, especially mining and port towns. This
meant that the area’s municipal police forces were still overwhelmingly
Liberal and that Liberals still dominated local municipal councils and the
majority of public offices even as late as mid-. However, when the
number of Liberal voters in the June elections actually increased, the re-
gional authorities decided that local defiance had gone too far and that
it was time to replace all Liberal public employees, not just those em-
 Bajo Cauca, Magdalena Medio, Northeast 

ployed in public works. The regional authorities also began to deploy

additional police troops to eastern Antioqueño towns and to encourage
the organization of Conservative contrachusma forces to assault or even
kill workers.
At first, regional Conservative violence was directed against workers
employed by the state, not the Liberal population as a whole. But given
the nature of employment patterns in the region, violence against orga-
nized labor inevitably spilled over to affect even those inhabitants un-
connected to public patronage positions. In the aftermath of the June
elections, the unprecedented presence of departmental police agents in
eastern Antioquia and, in particular, their broader mandate to intervene
in public order matters exacerbated tensions between Conservatives and
Liberals and between the foreign-owned mining companies, the police,
and organized workers. In Segovia in August , a row broke out in
the barrio de tolerancia (red light district) between an off-duty police-
man and an English mining company manager. The manager accused the
policeman and his cronies of impeding the extraction of gold from com-
pany mines and insulted the officer by calling him an ‘‘ass-kisser’’ and a
‘‘flunky’’ (lambón). The policeman responded by pointing his gun at the
manager. Brought up on charges of assault by the company’s executives,
the policeman defended his actions to the regional government, although
he admitted that ‘‘I have a lot of enemies in the municipality of Segovia as
a result of having had to proceed against them in the course of exercising
my powers as an authority.’’ The officer insisted the company was only
persecuting him because he was a proven nationalist who had refused to
allow the company’s British managers to smuggle gold out of Colombia
to avoid paying taxes. The mining company, Officer O’Brien argued, re-
garded the police as a private security force whose primary loyalty was
to the company rather than to the nation. This was reflected in manage-
ment’s ‘‘lack of support for the commanders and guards who have lent
their services in this area.’’ 15
What the dispute between Police Officer O’Brien and the Frontino
Gold Mines Company’s management actually appeared to be about was
something else. In Colombia, the mining companies and foreign-owned
companies in general (United Fruit, Tropical Oil, and so on) had long
been accustomed to having regional governments underwrite the cost
of security at their respective work camps. In return, preferential hiring
for the governing party’s sympathizers was arranged through the vigi-
 Blood and Fire

lantes de seguridad (watchmen or security officers), inspectores de tra-

bajo (labor inspectors), and médicos oficiales (doctors appointed and paid
locally who doubled as the companies’ security forces and health care
providers). Since the mining companies had done much of their hiring
during the previous sixteen years of Liberal rule and since most of the
workers drawn to the camps were migrants from the Caribbean coast or
the departments of Santander and Norte de Santander, the workers who
filtered through this clientelist system in eastern Antioquia were almost
exclusively Liberal and non-Antioqueño.16 The advent of Conservative
rule, however, suddenly put pressure on the companies to hire Antio-
queños and Conservatives. One way in which this pressure was exerted
was through the assignment of policemen such as Officer O’Brien as pub-
lic order forces and security guards at the camps. The overtly partisan and
repressive agenda of many such officers, however, elicited considerable
hostility among the Liberal majority who felt victimized by the regional
authorities’ representatives of order. Policemen were accused of insulting
and arbitrarily restricting workers’ physical mobility and of perpetuating
the notion that workers were continually disorderly or revolutionary in
order to justify official repression. Foreign managers, for their part, were
ambivalent about the presence of regionally appointed police agents in
the camp, despite their role in curbing labor unrest. The companies par-
ticularly worried about the effect that worker hostility toward the con-
servatized departmental police might have on production and labor and
management relations. Newly appointed policemen who were the benefi-
ciaries of extremist patrons, moreover, felt entitled to give free expression
to nationalist, antiforeign, and anti-Protestant sentiments that annoyed
the British and Canadian managers who were typically in charge of min-
ing operations in the region.
If Officer O’Brien perceived the foreign companies’ efforts to limit
the departmental police’s authority in the camps as a transgression of
national sovereignty, workers were no less ready to invoke the issues of
sovereignty and national identity when discussing or justifying their ac-
tions and grievances to the regional state. Miners accused the regional
government of colluding with foreign companies in ways that funda-
mentally violated their rights as citizens of the nation. The regional gov-
ernment allowed the companies to limit worker mobility (restricting
their and other civilians’ access to the camps) and to use Colombian
police troops to restrain and abuse Colombian workers on Colombian
 Bajo Cauca, Magdalena Medio, Northeast 

soil. These issues predated the Conservative rise to national and regional
power but became more urgent and immediate once partisan politics be-
came a more central aspect of the police’s agenda. A year earlier, miners
affiliated with the Pato Consolidated Gold Dredging Company in Zara-
goza had made a point of protesting ‘‘the acute campaign of persecution
currently advanced by the secretary of security [vigilancia] and the health
inspector’s office [inspectoria de sanidad]’’ in the Pato mining camp.
Union officials charged the inspectors with having ‘‘decre[ed] disrespect-
ful and authoritarian measures . . . against the company’s workers.’’ The
measures included foisting decisions on the company’s workers using
‘‘threats and coercive attitudes’’ and submitting workers to ‘‘unjustified
searches and imprisonment.’’ 17 The language in which the miners’ union
couched its complaints drew on a discourse of citizenship that empha-
sized universal rights and obligations. Workers appealed to the state first
and foremost as citizens and only secondarily as individuals whose rights
were protected and regulated by sector-specific limitations (such as a
labor code). Indeed, workers portrayed the foreign-owned company’s
mistreatment of them as an action that, beyond violating the labor code,
‘‘conflict[ed] with the democratic tradition of our homeland [patria],’’
and they demanded that the regional government ‘‘impede the con-
tinued assaults against Colombian citizens who live and travel through
this piece of national territory.’’ 18 In their invocation of citizenship as a
right conferred by birth on Colombian territory (not regional birth or
identity), workers contested the regional government’s notion of citi-
zenship as a privilege predicated on the satisfaction of a series of social
norms and codes of conduct.19 By accusing the government of allying
with the company against workers, moreover, labor turned the tables on
the government, implicitly suggesting that it was the government’s re-
gional insularity, not the militant workers, who compromised Colom-
bian sovereignty.

The Emergence of Armed Liberal Violence

The unprecedented intervention of newly appointed Conservative offi-

cers against mining camp workers and public sector workers, and the
abuse unleashed on the local Liberal population by national policemen
and Conservative public employees after June , prompted the emer-
gence of an armed Liberal reaction in eastern Antioquia. At first such
 Blood and Fire

groups were not based in Antioquia but rather recruited and organized
in the department of Santander. After the assassination of Gaitán, they
initially made their presence known by sporadically crossing the Magda-
lena River and assaulting Antioqueño properties and government offices.
These armed Liberal bands were made up of individuals—many of them
followers of Gaitán—who fled police and civilian Conservative abuse
in Boyacá and Santander. They settled in the hills opposite Antioqueño
port towns such as Puerto Triunfo, Puerto Boyacá, Puerto Nare, Puerto
Perales, and Remedios where few state authorities operated or where they
were badly armed and weak.
The first reported victim of this externally-based guerrilla violence
was the port town of Puerto Perales in the municipality of San Luis
where individuals from the department of Santander (Santandereanos)
were blamed for destroying  houses and causing damages estimated
at , pesos in September .20 Santander-based guerrillas next
struck two months later in Zambito (located on the Santander side of the
Magdalena River between Puerto Nare and Puerto Berrío but a settle-
ment informally under Antioquia’s jurisdiction). The Antioqueño ha-
cendados (estate owners) and colonos (settlers) who dominated the area
complained that they and their properties were the frequent targets of
‘‘escaped criminals [prófugos]’’ who capitalized on the absence of any
government representative to infest the area.21
After these first two assaults, a six-month lull followed in which fur-
ther unrest was absent, but nonetheless the threat of imminent guerrilla
invasion was invoked by terrified Conservative propertyholders to extort
the deployment of police troops and government protection. Local prop-
erty owners insisted that the guerrillas were interlopers with no visible
local following, yet earlier Conservative complaints suggested something
different. During the elections of June , for instance, Puerto Nare’s
Liberal municipal police searched Conservative party headquarters, ar-
rested the local Conservative committee president, and tore down Con-
servative campaign posters. Members of the committee insisted that the
municipal police were gaitanistas who had taken part in the violent up-
heavals that took place in the wake of Gaitán’s assassination and who
now sympathized and colluded with the guerrillas from Santander.22 A
month later the town’s Society of Public Improvements, which was made
up of the area’s wealthiest and most prominent citizens, complained to
the governor that the local police inspector and his secretary were in ca-
 Bajo Cauca, Magdalena Medio, Northeast 

hoots with ‘‘bandits’’ (that is, Liberal guerrillas). They spared no effort
to persuade the regional authorities that such conduct was ‘‘clearly rep-
rehensible and injurious to the tranquility and interests of local inhabi-
tants’’ and that the local authorities (who like the municipal police were
still Liberal) were ‘‘totally untrustworthy.’’ 23 Members of the Society of
Public Improvements pointed to the cases of a bandido who wounded
a businessman and was allowed to go free only ten days later; the mu-
nicipal police agent who levied a -peso fee for returning a citizen’s
stolen property; and the police inspector’s practice of charging citizens
for freeing them from arbitrary imprisonment. By September, Puerto
Nare’s Conservative president was requesting reinforcement troops from
the governor to stem what he called the ‘‘Communist threat’’ represented
by both the town’s public employees and the guerrillas camped on the
shores opposite the port.24
A year later the only officially reported guerrilla-related deaths in east-
ern Antioquia had occurred in two Magdalena River port towns: nine
deaths in Puerto Perales in the municipality of San Luis and three in
Puerto Triunfo in the municipality of Cocorná.25 Conservative complaints
of the threat posed by armed Liberals were thus not without substance,
although there was little evidence of collusion between local Liberals and
the guerrillas from Santander. What is crucial about these spatially cir-
cumscribed guerrilla attacks, however, is that they enabled both local
Conservative extremists and the regional government to justify a wide-
spread repressive campaign against all of eastern Antioquia’s Liberal in-
habitants. This campaign extended to areas where no violence had been
reported or where violence was neither the result of Liberal guerrilla at-
tacks nor of local Liberal ‘‘insubordination.’’
In September , in keeping with the newly established policy of re-
arranging appointments in towns where the government enjoyed little or
no support, Antioquia’s governor finally adopted the regional Conserva-
tive directorate’s suggestion that only ‘‘doctrinaire’’ Conservatives be ap-
pointed as mayors in historically Liberal towns such as Puerto Berrío. But
the governor’s appointee immediately prompted the Liberal president
of the municipal council to accuse the mayor of provoking local unrest
where there had previously been none. The new mayor had proceeded
to usurp the authority of Puerto Berrío’s municipal police (who enjoyed
the support of the majority of the town’s inhabitants) and had replaced
them with Conservative departmental policemen. Next, the mayor and
 Blood and Fire

the departmental police worked to dismiss all of the town’s Liberal mu-
nicipal employees, shut down the taverns (a gross violation in a town of
hard-drinking miners, stevedores, and sailors!), and prohibit the con-
sumption of alcohol. The town had already had a taste of the mayor’s
repressive measures, moreover, for he appears to have briefly served in
the immediate aftermath of Gaitán’s assassination.26
The actions of Puerto Berrío’s Conservative mayor were part of a
broader regional Conservative strategy to marginalize the authority of
Liberal municipal police forces in eastern towns where the police were
perceived to be closely allied to sectors of organized labor in control
of municipal council seats. Conservatives in both Puerto Berrío and
Puerto Nare had persistently complained to the regional government
that local police forces were overwhelmingly Liberal and ‘‘revolutionary’’
and urged the government to replace or circumvent their power. Indeed,
a week after Puerto Berrío filed a complaint against the Conservative
mayor, the regional authorities announced new rules to determine the
selection of policemen assigned to patrol the mining camps of Providen-
cia, Bagre, Pato, and Santa Margarita in the towns of San Roque, Zara-
goza, and Amalfi. Forty agents were to be chosen from areas other than
those in which they were to provide service to ‘‘avoid the grave inconve-
nience posed by police recruitment from among the local inhabitants.’’ 27
The choice of men, however, was left in the hands of the regional national
police commander and not the companies. This shift in the criteria for
selection of policemen in the mining camps was matched by an increased
deployment of Conservative national policemen to areas where public
sector and unionized workers were concentrated. The number of reports
of violent clashes between workers and policemen in eastern Antioquia
rose accordingly.
In October the railroad workers’ union complained that the trains and
their personnel were the constant targets of abusive police activities all
along the line between Medellín and Puerto Berrío. National policemen
were aided, union officials added, by ‘‘individuals of low instincts,’’ that is,
by Conservative contrachusmas who migrated up from nearby southeast-
ern towns or were recruited from among Conservatives in towns such as
Maceo and San Roque, which bordered the rail line.28 Violence involving
railroad workers and police agents broke out again in November when
well-armed policemen disguised as ‘‘bandoleros’’ (bandits) boarded the
train operating between Medellín and Puerto Berrío. Policemen with ma-
 Bajo Cauca, Magdalena Medio, Northeast 

chetes and guns also attacked Liberal workers in San José de Providencia
(one of the mining camps to which national police agents had recently
been assigned).29
The sudden increase in police violence directed against union mem-
bers sparked widespread hostility in towns such as Puerto Berrío. Indeed,
a few months after their arrival, national police agents were forced to
seek refuge in the mayor’s office for fear of the threat of violent retalia-
tion by the army and civilians.30 When the national government decreed
a state of siege in November , for instance, local inhabitants in the
railroad town of Cisneros organized a demonstration in which they killed
one policeman and wounded another.31 The local army commander im-
plicitly condoned the attacks on the police when he attributed them to
popular frustration with the ‘‘mayor, judge, police, and customs agents.’’
These officials, he insisted, had ‘‘used their official positions’’ to lead ‘‘a
wave of violence in which they were accused of [committing] all kinds of
crimes’’ against Cisneros’s civilian population. Abuses were so severe that
civilians had to flee to the ‘‘[military] barracks in search of protection.’’ 32
Clashes between inhabitants and the police and between the police and
public employees also occurred in the nearby railroad town of San José
de Providencia in the municipality of San Roque and in the gold mining
town of Zaragoza.33
Thus, in early , when oil workers at Shell Oil’s Casabe camp heard
rumors that the governor planned to replace the local police inspector,
they correctly surmised that this meant the introduction of a repressive
policy directed against them that would be similar to the one already
under way against railroad workers in the area around Puerto Berrío and
Cisneros and against miners in the northeast. Their union mobilized to
plead with the governor to keep the current inspector in his post, arguing
that the camp had been ‘‘calm’’ since his appointment.34 The governor
ignored the union’s request, however, prompting Shell Oil’s legal repre-
sentative to file a complaint on behalf of the company less than six months
later protesting the conduct of national policemen and the police inspec-
tor assigned to oversee public order in the Casabe camp. The lawyer as-
serted that the inspector had ‘‘introduced a climate of discord among
the personnel’’ which he ominously warned, ‘‘may give rise to a serious
incident.’’ Shell’s legal counsel was quick to reassure the governor that it
was not the company’s intention ‘‘to interfere in any way in the actions
of the departmental authorities in the Casabe camp.’’ The company’s ac-
 Blood and Fire

tions were prompted ‘‘by the fear of future incidents which could have a
very negative effect [on the company] and which could also create serious
problems for the governor’s office.’’ 35 A veiled warning, but a warning
nonetheless to a governor whom the Shell Oil management knew was a
self-avowed nationalist who was critical of the foreign company’s power
and a political extremist who was the major force behind the deployment
of partisan public employees and policemen in the area. The message ap-
pears to have struck home because, two weeks after Shell’s complaint,
local Conservatives wrote to Governor Eduardo Berrío González beg-
ging him to revoke his decision to remove the national police from the
Casabe camp.36
The repressive presence of national police troops and the aggres-
sive imposition of Conservative officials in eastern Antioquia eventually
prompted a large-scale Liberal guerrilla attack against the regional gov-
ernment and its representatives. On August , , guerrillas made up of
men from both Antioquia and Bolívar converged on the hamlet of Gua-
rumo in Cáceres and on the town of Caucasia in what appears to have
been a coordinated operation that included guerrilla attacks in Urrao
and Urabá (western Antioquia) on the same day.37 The guerrillas burned
buildings and killed local Conservatives in Guarumo before continuing
on to the town of Caucasia. The town’s telegraphist, who had managed to
escape to Magangué in the department of Bolívar just prior to the guer-
rillas’ arrival, urgently asked the governor to airlift army troops to the
town to respond to the presence of three hundred armed men.38 During
the three-hour siege guerrillas sacked and looted the local stores. They
also attacked the customs offices, mayor’s office, local treasury and civil
registry office, and killed the captain of the coast guard, the civil registry
officer, two municipal policemen, a local merchant, and a national police
agent.39 It took the authorities several hours to reestablish control, but
they claimed to have captured one hundred guerrillas and recaptured the
majority of the police arms stolen by the guerrillas during the attack.40
It was no coincidence that the attacks on Guarumo and Caucasia oc-
curred just before and immediately after Conservative Laureano Gómez’s
inauguration as Colombia’s new president on August . What was sur-
prising was how unprepared the regional government was to confront
the possibility of an armed Liberal attack when it occurred. The regional
government had been circulating rumors of an armed plot and had jus-
tified police repression during the previous months supposedly because
 Bajo Cauca, Magdalena Medio, Northeast 

such an event was imminent. Indeed, less than a month before the attack
in the Bajo Cauca, the governor had received reports from three differ-
ent sources warning him of possible insurrectionary activity on the San
Jorge Road linking Antioquia to the Caribbean coast through Caucasia.
The police inspector was the first to inform the governor that the authori-
ties had seized several powerful bombs that had been stolen by criminals
from the public road works in operation between Caucasia, Antioquia,
and Montelibano, Bolívar. He warned the governor that the San Jorge
Road was largely unpatrolled and that few workers were present because
of constant ‘‘threats against the public order.’’ 41 Two days later the gov-
ernor of Bolívar confirmed the inspector’s report when he complained
that Antioquia exercised no control over the San Jorge Road and accused
the road crew of helping local ‘‘leaders of the revolt.’’ Bolívar’s gover-
nor begged his counterpart in Antioquia not to send the road workers
any more dynamite because the workers were only interested in ‘‘per-
turbing the public order.’’ 42 Finally, on August , Major Arturo González
Arcila, commander of the Division Colombia Police, informed the di-
rector of the national police in Bogotá that the guerrillas were led by an
ex-army sergeant and former policeman (not an unusual feature of guer-
rilla commanders) named Ortíz. The guerrillas, the police major insisted,
had publicly announced their intention to target the police (but not the
army), and he confirmed that the attack was timed to coincide with an
uprising by guerrillas in the Llanos that was designed to prevent Laureano
Gómez from taking office.43
These reports should have prompted Antioquia’s governor to re-
inforce the area troops against an essentially foregone assault, but
throughout Antioquia the number of anticipated incidents protesting
Laureano Gómez’s rise to power proved too great for the meager public
order forces at the regional government’s disposal. Indeed, the attacks
on El Guarumo and Caucasia were but the first of many armed Liberal
actions. The attack exposed the nearly complete absence of regional au-
thorities in strategically important portions of Antioquia, undermining
any claim that the government might have made regarding its legitimacy
and effective control over public order affairs. It soon became evident
that while the regional government possessed sufficient men to harass
and abuse the local opposition on an intermittent basis, it did not pos-
sess the necessary forces to confront the consequences of such a policy.
In any case, it was easier for the incidence of public unrest in the Bajo
 Blood and Fire

Cauca, to scapegoat the local road crew, agent provocateurs from the de-
partment of Bolívar, and the area’s supposed history of rebellion than to
consider the local effect of the state’s recently appointed repressive forces.
It was not fortuitous that the main objects of guerrilla attack were police-
men and government offices, yet the local Conservative party committee
blamed violence on the local chief engineer in charge of hiring the road
crew. The engineer had precipitated violence by being ‘‘condescending
to paisas (i.e., to Antioqueños)’’ and failing to ‘‘give priority in hiring
decisions to local Conservatives.’’ 44 Regional and partisan identity were
conflated and the problem of partisan conflict was represented as a prob-
lem of cultural as well as political differences. If the public workers on the
San Jorge Road had only been replaced with Conservative Antioqueños,
the committee insinuated, armed assault could have been avoided.
Caucasia was calm for a brief, three-month period after the devastat-
ing attack in August, and no Liberal guerrilla activity occurred elsewhere
in the northeast during that time either. But in November the national
finance administrator informed the governor of a telegram he had re-
ceived from a customs agent in the town who warned of rumors regard-
ing a second guerrilla attack.45 Again, the regional government did noth-
ing to prepare the town, and, again, despite advance warning the town
was caught off-guard by guerrillas penetrating from the department of
Bolívar. Like the earlier attack on Caucasia in August, the one in Decem-
ber also appears to have been part of a coordinated assault on several
Antioqueño fronts. On the same day that guerrillas struck in Caucasia,
attacks occurred in Urabá and Urrao in western Antioquia and in Puerto
Perales south of Puerto Berrío on the shores of the Magdalena River.46
The guerrillas cut the telegraph line that linked Caucasia to Medellín and
prevented any direct communication between the town and the regional
capital for three months.47 While the government failed to send troops
to confront the guerrillas during the assault, Caucasia’s electoral regis-
trar reported that less than a month later the national police were infor-
mally conducting aplanchamientos against Liberal civilians in retaliation
for the guerrillas’ activities.48 When the registrar publicly denounced the
presence of three Conservative contrachusma members who aided the
police in their assaults, the police attempted to assassinate the registrar
and intimidated and threatened the mayor for jailing them.49 Meanwhile
detectives in the pay of the regional security department (Departmento
de Seguridad) insisted—as the Conservative directorate three months
 Bajo Cauca, Magdalena Medio, Northeast 

earlier had—that the enemy in Caucasia was ‘‘within.’’ The departmen-

tal detectives accused the road workers of being guerrillas and the con-
tractors in charge of the road (all of whom were Liberal) of lending the
guerrillas revolvers, trucks, and dynamite to attack the Conservative gov-
ernment and its local representatives.50
Public road workers, miners, and armed Liberal groups in the Bajo
Cauca and northeastern regions did collude with one another, but it
was only after a long period of government-directed dismissals, harass-
ment, and police abuse against railroad and state road workers that they
mobilized in openly insubordinate behavior. A letter written by a Con-
servative hacienda owner who lived in the hamlet of Colorado in Cau-
casia during the entire period of la Violencia corroborates the impres-
sion that local violence was not an organic development but the result
of repeated provocations by the regional authorities. His account, which
covers events in the region from  through early , provides a dev-
astating chronicle of the role played by policemen and contrachusmas in
the spread of violence and of the reluctance with which local Liberals and
Conservatives greeted the emergence of armed Liberal resistance.
The author of the letter referred to himself simply as ‘‘Arturo.’’ 51 He
considered himself unusual among regional men of his class and gen-
eration for he had refused to ‘‘climb aboard the bourgeois train like a
good Antioqueño.’’ Instead, he had buried himself in the Bajo Cauca to
work the land amid ‘‘the sometimes monotonous but often soothing sea
of green.’’ Arturo confided to ‘‘José,’’ the intended recipient of his let-
ter, that he and his neighbors in the Bajo Cauca and the department of
Bolívar were waiting tensely to ‘‘fall into the hands of that den of assas-
sins known as the police.’’ He and other locals had already been forced
to ‘‘witness the worst crimes imaginable, the very memory of which pro-
duces nausea.’’ These experiences had prompted him to ‘‘write a book
which is currently buried for fear of the uniformed killers.’’ Although he
feared the Conservative police, Arturo was no apologist for the Liberal
guerrillas, who, he believed, had vowed to ‘‘fight until the last of them
was alive and had vowed to kill every policeman or Conservative they
happened upon.’’
To indicate this thirst for vengeance that motivated the guerrillas who
operated in the area, Arturo cited the example of one group of twenty-
two adult men and an eleven-year-old boy who were led by a young
woman known as La Cucaracha (the cockroach). This guerrilla group had
 Blood and Fire

vowed ‘‘not to shoot soldiers except as a last resort.’’ When they ‘‘capture
[soldiers], they release them if they are alive, and if they should happen
to have killed them in combat, they leave their bodies undisturbed with-
out even removing their guns.’’ But, just as Major Arturo Velasquez had
earlier warned his superior in Bogotá, the guerrillas offered the police no
quarter. La Cucaracha and her men had repeatedly eluded the authorities
despite the ‘‘two hundred policemen and soldiers’’ sent to capture them.
Instead, the guerrillas proved adept at ambushing the police. They de-
capitated policemen in an elaborate and macabre ritual and then ‘‘played
soccer’’ with ‘‘the heads of the police . . . as if this were part of the pro-
gram.’’ La Cucaracha would dictate sentences of mutilation that corre-
sponded to the aggression and loss of family members that she herself
had suffered at the hands of the police and Conservative civilians: ‘‘ ‘Since
they killed my father,’ she says, and they land a tremendous machete
blow to his neck. ‘Since they killed my brother,’ another blow is struck
against the throat. Then the head hangs precariously suspended by the
skin on either side of the neck. ‘Since they killed my mother,’ she adds,
and the last blow is dealt to the abdomen. In addition to this treatment,
when policemen are the victims, they are sliced into strips.’’ Arturo ac-
knowledged that the armed Liberal bands that roamed the region killing
Conservative civilians on a daily basis were heartless and bloodthirsty.
This he repudiated and abhorred. But he also felt that while the actions
of the guerrillas were execrable, they were also understandable. While
some guerrillas were undoubtedly ‘‘innate assassins or fanatical criminals
for a cause’’ many were also individuals crazed by the violence wrought
upon them. As he put it, ‘‘they are avengers of abuses that weigh them
down. . . . All are starved for vengeance against the government’s people
who earlier assassinated their fathers, their sisters, and their brothers.’’
Arturo also noted that despite the region’s overwhelmingly Liberal
population, the guerrillas initially enjoyed little support from the local
inhabitants. Indeed, the earliest response of Liberal and Conservative
peoples alike to the appearance of the guerrillas had been to mobilize ‘‘in
an open crusade against the rabble [chusmeros] in sincere and disinter-
ested support of the authorities.’’ The local citizenry had been willing to
risk their lives to eliminate the Liberal armed bands, ‘‘even the very Lib-
erals who constitute an overwhelming majority of the local population
(and who had nothing to fear from the guerrillas) in order to see peace
restored to the region.’’ And peace, the author dryly noted, ‘‘arrived with
 Bajo Cauca, Magdalena Medio, Northeast 

the police.’’ Forty policemen from the department of Bolívar, under the
command of police Lieutenant Muñoz, ‘‘made their triumphal entrance
through the region of Sapo, burning the settlers’ homesteads, shamelessly
stealing the money from their pockets and the rice from their barns. No
one to the present day has been able to know with certainty how many
men died imprisoned by the flames of their own homes; any conjecture
would be dangerous and audacious.’’ The police ‘‘attacked, looted, and
burned Villa Uribe’’—a predominantly Liberal settlement that bordered
the predominantly Conservative settlement of Regencia that had earlier
been devastated by the Liberal ‘‘guerrillas [chusma] . . . after first assassi-
nating many of its inhabitants.’’ One unfortunate peasant had mistaken
the police for guerrillas and identified himself as a Liberal. Lieutenant
Muñoz ‘‘had his agents hang the peasant from a tree, they poked his eyes
out with the point of a bayonet, and then Muñoz ordered him killed. The
corpse of this poor Conservative was left there for the terrible crime of
having said he was a Liberal.’’ In another instance, the local police inspec-
tor stood idly by while a worker for Arturo’s cousin was publicly ‘‘mar-
tyred’’ through the streets of the town for two days. And the guerrillas
left him deaf ‘‘after cutting off his ears and making him eat them.’’ The
trauma also left him permanently speechless.
The police engaged in outright theft, cattle-rustling, and extortion,
and their behavior enabled local individuals to justify killing Liberal
neighbors in the name of the Conservative party when the real motive,
Arturo implied, was mere greed, jealousy, or long-standing family feuds.
But Arturo reserved his worst accusation for last, including at one point
the shooting of a public tow truck driver and his assistant for the mere
enjoyment of it and then the delivery of the bodies in a casket to the local
mayor with a note attached that alleged the victims had been ‘‘killed in
an encounter with the guerrillas.’’ As if all these police activities were not
bad enough, the police were not even capable of doing what they had
supposedly been sent to the region to do: defeat and eradicate the Lib-
eral ‘‘chusma’’ (Liberal guerrillas). Instead, Arturo insisted, ‘‘the police
flee the chusma, they refuse to confront them, they are scared of them.
[The police] only appear to pursue them to sow terror, finishing off what
the guerrillas leave behind.’’ And he concluded, presciently, as it would
turn out:

There can be no peace while elements of the government compete in

criminality with the bandits and exceed them in delinquency [ratería].
 Blood and Fire

There can be no peace while men, honest rural workers, withhold their
trust from a government that has completely forfeited its legitimacy.
The government has committed errors of such a magnitude that we
are on the verge of a revolution, a revolution of hatred and vengeance
because in each wounded heart lies the latent germ of a future avenger.
May God prove me wrong, but in this area, in the fields and mountains,
one can feel the truth pulsing with the force of immanent events. . . .
If things continue as they are, we shall have no recourse but to seek
the support of the guerrillas.

One can only speculate what impact this letter had when it was captured
by the police in March , handed over to the Minister of War, José
María Bernal (Antioquia’s former governor), and then forwarded to Gov-
ernor Braulio Henao Mejía in Medellín. Its content, however, could not
have come as a surprise to the regional authorities. Antioqueño Con-
servative criticism of the behavior of extremist public employees, fellow
Conservative civilians, and the police was not new. On many separate
occasions citizens had indicated that the most frequent target of these
forces—contrary to the justification given by the regional authorities
for their existence—were not Liberal guerrillas but rather the unarmed
civilian population.
Among the more than thirty-three reports of contrachusma activity
and sixty-one reports of Liberal guerrilla violence filed with the gover-
nor’s office or with his secretary between  and —the vast majority
of which were reports of violence in  and —in the Bajo Cauca,
Magdalena Medio, and northeastern regions, there is not a single report
in which the Conservative civilian police attacked Liberal guerrillas or
contrachusma forces and Liberal guerrillas had any kind of intentional
armed encounter with each other. Of the sixty-one reports involving Lib-
eral guerrilla attacks or encounters between them and the government’s
forces, moreover, there was not a single instance in which the police took
the offensive and sought out guerrillas in their camps before a guerrilla
attack occurred. Encounters between the police and guerrillas typically
only took place when the latter assaulted a town in which the police hap-
pened to be stationed or when guerrillas led ambushes of policemen to
steal their arms, uniforms, and supplies. In the few cases where the police
did go after the guerrillas it was only after the guerrillas had already at-
tacked and left behind civilian casualties. Unable to locate their objec-
tive, the final police report invariably concluded that the much sought
 Bajo Cauca, Magdalena Medio, Northeast 

after guerrilla forces had melted into the surrounding hills and been lost
from sight.
The police and contrachusma violence that was directed against the
civilian population, combined with these forces’ evident failure to protect
civilians from guerrilla attack, embittered many inhabitants in eastern
Antioquia. Among them were moderate Conservatives like Arturo who
refused to condone the violent behavior of their partisan brethren and
who opposed the use of violence against innocent members of the oppo-
sition. They warned the regional government that it was unwise to push
local inhabitants too far. The government, they insisted, was in no posi-
tion to actually restrain violence should local Liberals decide they had
endured enough abuse and opt to support or join the incipient, armed
Liberal bands that were increasingly active in the region. Juan de Dios
Arango, a Conservative engineer in charge of Puerto Berrío’s electric
plant and a local landowner, for instance, denounced fellow Conserva-
tives for falsely accusing the Liberal opposition of subversive activities
solely to legitimize taking jobs away from them. Arango insisted that the
local Liberal mayor had never abused Conservatives as some extremists
insisted and that his only ‘‘crime’’ was to focus on advancing the develop-
ment and growth of an area where ‘‘progress has been put off for twenty
years.’’ The mayor was the first appointee to pay serious attention to fin-
ishing much needed sewerage, electricity, and aqueduct projects in the
town. Laureanistas, bent on punishing moderate Conservatives who re-
fused to collude in the abuse of the opposition were, in Arango’s estima-
tion, the real sources of local unrest. In his own case, fellow Conservatives
had attempted to force ‘‘the undersigned to resign as the head of the elec-
tric company’’ for defending the mayor and for attributing Puerto Berrío’s
‘‘backwardness’’ to Conservative-led ‘‘intrigues around the distribution
of public jobs.’’ Arango, however, had resisted attempts to force him to
resign insisting that ‘‘under no circumstances can we operate subject to
the intrigues of these politicians.’’ He also reminded the governor that
while there might be no harm in constantly fiddling with local political
appointments to public office in other towns where local needs were less
pressing, doing so in a town as strategically important as Puerto Berrío
would have disastrous consequences.52 The local Conservative party com-
mittee surprisingly seconded Arango and also pleaded with the governor
to restrain local party extremists and to block their attempts to force the
removal of the town’s Liberal mayor.53
 Blood and Fire

Local inhabitants, including many Conservatives, moreover, had been

protesting the partisan activities of fellow Conservatives in eastern Antio-
quia since . The former customs administrator and collection agent
of the national finance office in Zaragoza, for instance, complained to
the governor that the local mayor had removed him from office and de-
nounced him to the local Conservative committee as ‘‘a bad Conserva-
tive’’ for refusing to arbitrarily dismiss Liberal public employees.54 Two
months later in January  Caucasia’s municipal registrar narrowly es-
caped an attempt on his life after he refused to condone the arbitrary
treatment of members of the opposition by national policemen and their
contrachusma aides.55 Later that same month, Puerto Berrío’s Conser-
vative committee denounced the mayor and national police agents for
abusing the town’s citizenry when they were drunk (which, the commit-
tee assured the governor, was often because both the mayor and police
suffered from ‘‘alcoholic intemperance’’). The mayor and the police had
assassinated an unarmed citizen and caused frequent public scandals.
This behavior, the committee reminded the governor, ‘‘reflects badly on
the Conservative government.’’ 56
The determinants of local violence in eastern Antioquia were hence far
more complex than any innate, unavoidable differences between mono-
lithic groups of Liberals and Conservatives—the traditional explanation
given for la Violencia—might suggest. In many areas there were no innate
conflicts, these were created and fed by the state’s own agents who capi-
talized on a few disgruntled, local adherents or otherwise imported them
from nearby areas to fan and exploit the flames of partisan difference. It is
also apparent from the numerous complaints filed by moderate Conser-
vative public employees and elected officials in peripheral areas that no
consensus existed among Antioquia’s Conservatives regarding the pro-
priety of using violence for political purposes. In some cases this may have
stemmed less from a sense of ideological sympathy with local members
of the opposition and more from fears of the economic impact of partisan
unrest. But whatever the motivations for the absence of overwhelming
local Conservative support for the activities of the regional authorities,
what is evident is the need to question a generalized or ‘‘inherently’’ vio-
lent concept of partisan affiliation in Colombia as an explanation for la
Violencia.
For a long time Governor Braulio Henao Mejía was loath to respond
to or believe Conservative reports suggesting that the principal catalyst
 Bajo Cauca, Magdalena Medio, Northeast 

of local violence in eastern Antioquia was the unruly presence and seem-
ingly unlimited arbitrary authority exercised by extremist public officials,
contrachusmas, and national police troops. He also ignored suggestions
that rivalries between competing Conservative factions played a deter-
minant role in the rise of police and contrachusma influence. But events
in western Antioquia, where violence had already grown quite severe and
where comparable complaints against extremists had been received by
his office, gradually convinced him that there was some truth to both the
rumors of police inadequacy and to accusations of arbitrary actions filed
against Conservative contrachusmas in areas where violence was present.
The governor first voiced his growing doubts about the exact nature and
scale of the conflict being played out in Antioquia’s peripheral zones in
a detailed, confidential letter he sent on March , , to his close friend
and fellow Antioqueño, Gonzalo Restrepo Jaramillo, the minister of for-
eign relations in Bogotá.57 In it the governor confessed that he no longer
knew whom to believe nor had he any idea how many men were really
up in arms, whether they were ‘‘thousands’’ or simply ‘‘whether their ex-
traordinary mobility makes it appear as if there were a great number of
them.’’ He also noted that violence only recently attributable to conflicts
organized along clear partisan lines no longer seemed so easily classified.
‘‘More serious than the initial twisted political inspiration or the gang
[pandilla] or guerrilla actions directed against the government and the
nation’s institutions—which is what we had been witnessing until just
a couple of months ago,’’ he wrote with evident concern, ‘‘is what we
are currently witnessing, that is, groups of true vandals who are solely
dedicated to pillage as their main motivation and who are now made up
of members of both parties.’’ Henao Mejía acknowledged knowing that
there were now at least two different kinds of armed groups in Antioquia.
Those whose sole purpose was to ‘‘steal and, while they are at it, assas-
sinate, burn, and rape’’ and for whom partisan concerns seemed to play
a secondary role, and others who fulfilled a variety of functions (‘‘there
are those taking part in several different tasks,’’ that is, both partisan and
economic). Liberals and Conservatives were involved in both types of
armed bands.
The governor’s letter to the minister of foreign relations was mainly
motivated by worries regarding the presence of guerrilla groups on the
stretch of road from Dabeiba to Turbo in the northwestern region of
Urabá. But disappointment in the ability of the police to meet the chal-
 Blood and Fire

lenge of Liberal guerrillas and escalating economic banditry elsewhere

in Antioquia influenced his change in attitude toward the issue of public
order throughout the department. Preoccupied by the ineffectuality of
the police, Governor Henao Mejía insisted to the minister of foreign rela-
tions that only the complete militarization of regions severely affected by
violence (such as eastern Antioquia) could begin to meet the problem of
unrest in the department. Only absolute control by the army would do,
the governor concluded, because, as he confessed to Restrepo Jaramillo,
‘‘our valiant police force—whether because of a lack of manpower or sup-
plies—has been lamentably defeated twice in the last month on occasions
in which it could have performed splendidly.’’ The central government,
moreover, appeared to have abandoned the region to deal with escalating
violence entirely on its own. Although in secret meetings the president
had promised the governor needed supplies, arms, radios, and airplanes,
Antioquia had still not received any aid. Without the requested materiel,
the governor warned, he could hardly be expected to stem ‘‘Liberalism’s
increasingly subversive attitude.’’ As the central government delayed, the
conflict in Antioquia intensified. Every day areas of recent colonization
and intensive agricultural production were ‘‘razed, harvests are stolen or
destroyed and workers are assassinated or they flee in the hope of saving
their lives.’’ Henao Mejía was convinced that he could win the war against
subversion if only the central government would send him adequate sup-
port with which to confront it. As such, he concluded his missive on a
bellicose note and demanded, ‘‘more judges, more arms, more soldiers,
but above all much more diligence and cooperation [from the central
government].’’
Braulio Henao Mejía never received the additional troops and sup-
plies he insisted were necessary to reestablish public order in Antioquia.
He was reduced during his entire term in office (August –July )
to writing a stream of memorandums to the central government, begging
for aid and for additional monies with which to meet the increased ex-
pense of public order maintenance in the region. Meanwhile laureanistas
and other extremist Conservatives publicly upbraided him and portrayed
him as an ineffective regional leader who was too timid to impose au-
thoritarian measures to end Liberal subversion. In response to local pub-
lic order incidents, moreover, Henao Mejía continued to deploy on an ad
hoc basis the very policemen whose efficacy he had questioned in his mis-
sive to Bogotá. The governor also came to tolerate, whether through in-
 Bajo Cauca, Magdalena Medio, Northeast 

ertia or commitment, local Conservative efforts to arm volunteers and to

ignore or minimize disturbing complaints from concerned party mem-
bers regarding the long-term implications of continuing to promote ar-
bitrary and violent policies against the opposition.
The predictions made by moderate Conservatives that the escalation
of government repression would invite intense armed Liberal retaliation
were ultimately fulfilled when several Liberal armed groups went on an
extended rampage against haciendas, mines, and towns that began in
February  and lasted through July. By October, guerrilla violence was
matched by equivalent levels of contrachusma violence, and rural areas
in eastern Antioquia, where public-order matters had not been of press-
ing concern to the government before , suddenly emerged as some
of the departmental sites most severely affected by la Violencia.

The Intensification of Guerrilla Conflict

On June , , a month before a military coup brought down the gov-
ernment of Conservative president Laureano Gómez, a young woman
was captured in Puerto Berrío and brought to Medellín to testify about
her life among the Liberal guerrillas who operated in eastern Antioquia.58
Her name was Angela Rosa and she was twenty-four years old, single, of
illegitimate birth, and part of the wave of seasonal laborers who floated
from hacienda to hacienda throughout eastern Antioquia looking for
work. Angela Rosa had been born in the northeastern mining town of
Amalfi, where declining production and the growth of large estates had
pushed hundreds of settlers out in search of livelihoods from the late
s through the early s. She had been living in another mining
town to the south, Maceo, when a local landowner turned her over to the
police and accused her of being a spy for the men up in arms against the
Conservative government. After her arrest, the chief of the department of
criminal investigation in Medellín asked Angela Rosa to recount every-
thing she knew about ‘‘where that group of bandoleros that operated in
your vicinity lived, the number of men up in arms, their names, how they
dress, how they obtain food, their leaders, what connections they have
with nearby towns and their activities.’’
The verbatim transcript of her testimony provides a rich account of
the ordinary lives and hardships of migrant settlers in the Magdalena
Medio region of Antioquia and of the effect of violence on their way of
 Blood and Fire

life. The events Angela Rosa recounted in her judicial testimony took
place between mid- and May —the year in which Liberal guer-
rillas were most active in eastern Antioquia and the year in which the
greatest number of deaths, expropriations, and general criminal activity
were reported for the area.
Angela Rosa began her narrative by disabusing the criminal inves-
tigator of his impression that she had ‘‘lived’’ with the guerrillas: ‘‘we
didn’t exactly live with them, but we did live near them and that’s how
we knew everything they did.’’ Proximity turned to intimacy when some-
time in  Angela Rosa and her lover, Alfonso, were informed by the
proprietor of the estate where they had been working as tenants that they
would have to ‘‘get out.’’ Angela Rosa and her companion moved to her
stepfather’s farm where they stayed for a month. ‘‘They were rounding
up Liberals in those days supposedly to kill them,’’ and so Alfonso and
five other young men ‘‘fled to the hills.’’ Angela Rosa and her stepsis-
ter Teresa were left behind. After wandering around for six days with-
out sighting the guerrillas, Alfonso returned to the women’s side and
picked up another friend, Manuel, with whom he planned to renew his
search for the ‘‘Liberal chusma.’’ Before they set out, however, a contra-
chusma group came down from the hills and there was a ‘‘shoot-out [se
echaron candela].’’ Wounded in the skirmish, Alfonso fled into the hills
with Jesús (Teresa’s lover) and Pedro to renew his search for the guerrillas
while the women sought refuge on another farm ( finca). Some days later,
the men rejoined the women as they were about to move on to a third
farm and begged the women to accompany them into the monte (hills).
Prompted by rumors that the Conservative contrachusmas ‘‘were killing
women, children, men, and everything in their path,’’ the group traveled
and lodged at four different farms in the space of less than a week. The
group finally arrived at a farm called ‘‘Nuevo Mundo,’’ where Angela Rosa
noted that ‘‘we had lived before’’ as sharecroppers or tenants.59
A week after the group’s arrival at Nuevo Mundo, members of a Lib-
eral band headed by a man nicknamed ‘‘Pielroja’’ (red skin) arrived and
insisted the men join their guerrilla group because ‘‘they’d be killed there
all by themselves.’’ 60 The problem once more became what to do with the
women. Alfonso and Jesús insisted that the women go back to their step-
father’s farm, but the women refused because ‘‘we were scared to go back
on our own.’’ They proposed continuing with the men and then stopping
at a farm located at a league’s distance from Pielroja’s camp on the near
 Bajo Cauca, Magdalena Medio, Northeast 

shore of the Rio San Bartolo.61 Alfonso was still suffering from the wounds
from the skirmish with the contrachusma, so Pielroja sent Alfonso back
to be nursed by the women on the farm, but agreed to send for Alfonso
when ‘‘Alfonso was better and they [the guerrillas] had a mission to per-
form.’’ Pielroja’s men kept careful tabs on the recovery of their recruit
during the two months of Alfonso’s convalescence. When his arm healed,
‘‘he joined them on missions and then came back home, and he did that
for another month.’’ During the month in which Alfonso accompanied
the guerrillas on their ‘‘missions,’’ the band attacked one rural neighbor-
hood in the municipality of Yalí and killed five people. They then moved
on to Ité where they had an encounter with government forces in which
they killed another six men. In July , their month-long spree culmi-
nated in an aborted attempt to blow up the railroad bridge at Monos.62
Alfonso abandoned the guerrillas after that and sought work on the
hacienda where his female companions had found employment. He and
the proprietor drew up a six-month contract in which ‘‘half the farm was
negotiated and [Alfonso] began to sow the fields and fix up the animal
pens and beds.’’ At the end of six months, news reached the group that
Pielroja’s band had recently killed eighteen people in ‘‘Las Partidas’’—
among them a man, his four sons, a son-in-law, and others whom Angela
Rosa knew. Around the same time, two other Liberal guerrilla groups ar-
rived at the farm, one led by Capitán Corneta and the other led by a guer-
rilla leader named Santander. Santander was originally (as his name sug-
gests) based on the other side of the Magdalena River, but he had linked
up with the Antioquia-based band led by Captain Corneta and become
Corneta’s second in command. These two groups, who numbered around
 men, were about to attack ‘‘El Coco,’’ a corregimiento near ‘‘El Tigre’’
in the municipality of Amalfi.63 Once they had successfully burned half
the settlement and killed ‘‘a lot of people,’’ they moved on to nearby ‘‘La
Susana’’ where they burned numerous homes and killed another twenty-
five people, among them a woman whom Angela Rosa also knew.64 After
resting three days at Alfonso’s farm, Captain Corneta’s and Santander’s
men moved on to join forces with Pielroja’s band.
The attack on El Coco and La Susana finally pushed the army into
action; they came looking for the guerrillas, surrounded them, and forced
them to flee north toward the mining town of Segovia.65 Government
forces killed three guerrillas from a detachment of seven sent to search
for salt on one of the farms where Angela Rosa and Alfonso had sought
 Blood and Fire

refuge, but four others escaped after killing several miners, among them
the father-in-law of the farm’s owner. Terrified that the army would kill
them, Angela Rosa, her stepsister, the concubine of one of Pielroja’s men,
and Alfonso fled to a safe house owned by another guerrilla where they
stayed for a week while the farm owner built his wife and mother-in-law
a hut in the hills where they could hide. Alfonso and Angela Rosa joined
the farmer and his family in their hideaway, while Teresa set up house with
another of Pielroja’s men nicknamed ‘‘Relámpago’’ (Lightning). Three
weeks later, Alfonso and Angela Rosa had ‘‘a falling out [un pereque]’’
with the farmer and moved onto a farm owned by ‘‘Lalo,’’ one of the guer-
rillas’ local suppliers. They had to abandon that farm, too, when Pielroja’s
men commandeered it a week later.
Resting in a hut hidden in some scrub on the farm ‘‘Pescadero,’’ the
group heard that forty of Santander’s men had arrived at Lalo’s farm
and were looking for ‘‘Amanda, Luis (Amanda’s partner), Alfonso, and
[Angela Rosa].’’ They demanded that the men ‘‘join up’’ and that the
women become ‘‘their [the guerrillas’] women.’’ Hearing this, the women
and their companions fled but failed to reach a safe farm before dusk.
Ten of Santander’s men caught up with them and announced: ‘‘OK men,
what’s the deal here? Either you’re Conservatives [godos] or maybe we’re
the Conservatives, but you keep fleeing from us, so we’ve come for these
women, these are our orders. We’ll take them with your consent or with-
out it.’’ When the women began to cry, the guerrillas told them they
would leave them alone if they stopped crying, but later that night, the
guerrillas gang-raped them. ‘‘At night they grabbed us and did with us
whatever they liked, all of them used us, and since Alfonso and Luis
grabbed us back, they got mad because we’d been taken away from them
[the guerrillas], and one of them who they call ‘Lucero’ took his gun from
his holster and took aim.’’ Angela Rosa then concluded that ‘‘everything
calmed down and we went to bed.’’
Santander’s men then left to link up with Pielroja’s band, while Alfonso
and Luis, apparently after agreeing to rejoin the guerrillas and deliver the
women later, set out with Angela Rosa and Amanda in the direction of
‘‘El Presidio.’’ There, Luis hoped to sell his only worldly belongings, his
shotgun and his poncho (ruana), in the hope of cobbling together enough
money to enable Amanda to escape to the port town of Puerto Berrío.
Instead, the men were accosted by two of Pielroja’s band and forced to
lead the guerrillas to the women’s hiding place. On the road, Alfonso es-
 Bajo Cauca, Magdalena Medio, Northeast 

caped and ran to warn the women to flee. At one of the several farms at
which the women stopped to ask for refuge, a servant fed and agreed to
lodge them but was discovered by the proprietor and his overseer who
turned the women over to the Maceo authorities and accused them of
being ‘‘spies for the chusma.’’ Angela Rosa and Amanda were eventually
sent to Medellín, where the authorities took down their testimony and
pressed them for details regarding the whereabouts and strength of the
guerrillas.
Six guerrilla bands operated in eastern Antioquia. Pielroja’s had
seventy men and two women and was based near ‘‘La Susana’’ in Maceo
between Puerto Berrío and Yolombó. Santander and Corneta’s band had
 members and had its base in ‘‘San Vasconio’’ near the mining camps
of Remedios and Segovia. Vicente Mejía (alias ‘‘El Dormido’’) had eigh-
teen men and three women and was based along the railroad tracks be-
tween Maceo and Puerto Berrío. A fourth group, led by ‘‘El Chicote,’’ was
based in the department of Santander, but Angela Rosa purported not to
know of it. Two other bands, led by Rafael Rangel and Trino García, were
based on the other side of the Magdalena River near Santander’s camp
but sometimes crossed the river to operate in Antioquia. Angela Rosa
had had no contact with them and could supply no further information
(see map ).
To survive, Pielroja, Corneta, Santander, and Vicente Mejía ‘‘went out
all the time to rustle cattle.’’ 66 Pielroja’s group kept the branded cattle for
their own consumption and sold unbranded cattle to an absentee land-
lord from Maceo who resided in Puerto Berrío. Salt, of crucial signifi-
cance to the guerrillas’ survival and a commodity whose trade was strictly
controlled by the authorities, was mainly obtained by stealing what was
left out in pastures for cattle. Knowing the dire need for salt by many
guerrillas, the authorities frequently poisoned it. To avoid this, guerril-
las only stole salt from troughs in pastures where the cattle had just been
let out to graze. When I interviewed the former guerrilla leader, Captain
Corneta, he said the salt was dyed with methylene blue, but that the guer-
rillas were so desperate that they washed it off and ate it anyway.67 The
guerrillas obtained other day-to-day necessities such as grains, lard, mo-
lasses loaves, and clothes by stealing from nearby farms. Salt, saltpeter,
and cigarettes, however, had to be brought in from Puerto Berrío. Several
local farm owners and merchants residing in Puerto Berrío colluded to
supply these items to the guerrillas. When questioned about the degree
 Blood and Fire

Map . Guerrilla operations and movement in Eastern Antioquia. (Source: In-
stituto Geográfico Augustín Codazzi; Archivo Privado del Señor Gobernador de
Antioquia, –; Archivo de la Secretaría de Gobierno de Antioquia, –
)

to which inhabitants other than those in Maceo provided logistical sup-

port for Pielroja’s men, Angela Rosa responded that ‘‘I haven’t heard it
said that any of the other nearby towns help them out.’’ She was unable
(or unwilling) to say whether locals acted as ‘‘spies’’ or informants who
relayed information regarding government operations to the guerrillas.
Angela Rosa was able to reel off the aliases of numerous participants
in Pielroja’s band. But the men under Santander and Corneta, she in-
sisted, were ‘‘all strangers to me.’’ 68 Whereas Pielroja’s men appear to
have been largely local and to have mainly confined their activities to
the Maceo/Puerto Berrío area, Corneta and Santander’s group was highly
mobile and of largely nonlocal origin. Angela Rosa testified that ‘‘[Cor-
 Bajo Cauca, Magdalena Medio, Northeast 

neta’s] band moves around a lot, they say it’s been sighted in the depart-
ment of Bolívar in the Cienagas de Barbacoas as well as Ité, and that lately
they’ve been seen around Segovia and Remedios in the ‘Mata’ canyon.’’
Corneta and Santander’s group was also considerably better armed than
Pielroja and his men.69 Except for a few shotguns and revolvers, the ma-
jority of Pielroja’s band possessed no guns and was armed only with the
peinillas, or small machetes, commonly carried by agricultural laborers
throughout Colombia. A shortage of weaponry had in fact compelled the
group to scale back their activities to such a degree that ‘‘they hardly go
out anymore.’’ In contrast, Corneta and Santander possessed seventeen
rifles—fifteen captured from the police and two marked as army-issue—
several carbines and revolvers, many shotguns, swords (yataganes), and
peinillas. Santander and Corneta’s men also had stolen police uniforms.70
The majority of the bands’ participants, however, donned the typical garb
of the paisano (khaki pants, shirt, ruana, and straw or felt hat) that en-
abled them to blend in with local inhabitants and to wander unnoticed
throughout the countryside.
Angela Rosa ended her long account of her experience with the guer-
rillas with the fate of the group of eight men who had initially set out
in search of the guerrillas less than a year earlier. Alfonso and Luis had
joined Vicente Mejía’s band; three other men had met their deaths at the
hands of the government’s forces; the whereabouts of two others were
unknown, and Angela Rosa’s stepsister’s lover had joined Santander.

The Context of Angela Rosa’s Narrative

It is a striking feature of Angela Rosa’s testimony that la Violencia appears

not to have touched her or her neighbors’ lives until sometime in mid-
. Indeed, her narrative—spanning an approximately eleven-month
period between June  and May —encompasses almost exactly the
duration of consistent armed conflict in most of eastern Antioquia. Un-
like either Urabá or Urrao in western Antioquia, where partisan violence
gave rise to a well-organized, locally-based Liberal guerrilla response by
late  and where a state of nearly continuous armed conflict existed
from then until June , most civilians and hacienda workers in east-
ern Antioquia largely escaped the effects of intense violence until .
Although there were guerrilla attacks against Antioqueño port settle-
ments and the mining areas around Caucasia since , they had all
 Blood and Fire

been led by guerrilla groups based outside Antioquia proper (from the
nearby departments of Bolívar and Santander) and appear to have en-
joyed little local Liberal participation or support. Government forces, al-
though quite brutal, had also encountered difficulties in applying a wide-
spread policy of harassment throughout eastern Antioquia. This was due
in part to the vastness of the terrain and the lack of infrastructure, par-
ticularly around the mining regions of the northeast and Bajo Cauca, but
was also due to the relative absence of local Conservative support for
such efforts. Despite the use of official violence against organized state
workers and miners since , moreover, organized workers had largely
succeeded in deflecting or resisting government attempts to eliminate
them or usurp their control over local political offices and power.
The turning point in the spread of violence occurred in  when at
least one major guerrilla group (under the leadership of the Antioqueño-
born Captain Corneta) based in Antioquia proper emerged, and when
Governor Braulio Henao Mejía endorsed an incipient expansion of the
role and number of paramilitary Conservative forces operating in east-
ern Antioquia. When Dionisio Arango Ferrer, a longtime supporter of
armed Conservative contrachusma groups assumed control of the gover-
nor’s office in July , moreover, contrachusmas were made primarily
responsible for the maintenance of public order in the department. The
combination of these two developments shifted the focus of violence
away from the harassment and dismissal of workers and political office-
holders toward a far bloodier and more generalized violence against rural
inhabitants and hacienda personnel.
In a body count of civilian and guerrilla (‘‘bandolero’’) deaths regis-
tered by the regional armed forces in , eastern Antioquia reported
two civilian deaths (in Amalfi and Cáceres) out of a departmental total of
sixteen casualties. In , eastern Antioquia reported twelve dead civil-
ians out of a regional total of forty-nine (concentrated in two towns,
Puerto Triunfo/Cocorná and Puerto Perales/San Luis). It was not until
 and especially , however, that the number of casualties in the
area rose to alarming proportions and constituted a significant percent-
age of the violent deaths occurring throughout Antioquia. Amalfi, Puerto
Berrío, Remedios, and San Luis accounted for  percent of the region’s
total of  civilian deaths and  percent of the region’s total of  ‘‘ban-
dolero’’ deaths in  (see appendix A.). But, by  eastern towns ac-
counted for more than half the departmental total of , civilian deaths
 Bajo Cauca, Magdalena Medio, Northeast 

and  percent of the total number of the region’s ‘‘bandolero’’ casualties

(,). This is especially striking as these eastern towns held less than
 percent of the department’s total population. The largest number of
deaths in  and , moreover, was registered in Puerto Berrío and
its surrounding area, precisely the zone encompassed by Angela Rosa’s
account (see appendix A.).
The governor first received reports that a locally-based, armed Lib-
eral group was operating in eastern Antioquia when spies complained
that an hacienda owner in Yolombó, who was also a Liberal regional as-
sembly representative, had sent his mayordomo to deliver provisions to
Liberal guerrillas who were camped out in the hills nearby.71 Five days
later, eight guerrillas appeared on an hacienda in the Alicante region be-
tween Puerto Berrío and Yolombó to inquire about the political affiliation
of the hacienda’s workers. When it turned out that they were all Liberal,
the workers’ lives were spared, but the hacienda was robbed. By late Feb-
ruary the guerrillas were reported to have ‘‘taken up positions’’ around
the hacienda ‘‘La Gallinera’’ in the vicinity of Yalí.72 Meanwhile, in the
hamlet of Nechí in Caucasia, guerrillas crossed the border from the de-
partment of Bolívar and attacked the Conservative caserio ‘‘Regencia’’
(located five leagues from the border with the department of Bolívar), de-
stroyed forty-three houses, left many dead and wounded, and raped most
of the settlement’s married women. Meeting with no official opposition,
the guerrillas escaped via Ayapel, Bolívar, without losing any men.73 A
massive migration of the area’s inhabitants ensued, prompting the gover-
nor to beg Caucasia’s notary ‘‘not to formalize during these days of terror,
any land deeds, so as to avoid the abuses audacious individuals wish to
commit with simple, terrified people.’’ 74 No further guerrilla attacks were
reported in eastern Antioquia for about a month.
In April, however, the guerrilla activity flared up once more and lasted
until July. Armed Liberals struck in Puerto Nare in early April killing one
woman and leaving another eight inhabitants missing.75 Two weeks later
a guerrilla force of between  and  men marched down from the rail-
road line near Puerto Berrío and attacked an hacienda in the settlement
of Santa Rita in the southeastern municipality of San Luis. The guerrillas
left fourteen dead while seventy-five terrified families fled to seek refuge
in the hills near the village.76 The attack on Santa Rita sparked a panic
among Conservatives in other parts of San Luis such as Samaná, where
rumors of an impending guerrilla incursion forced dispersed rural in-
 Blood and Fire

habitants to converge upon the town proper. In their hurry to abandon

their properties before the guerrillas arrived, the inhabitants left ‘‘troughs
filled with corn, hundreds of cattle on different farms, [and] thousands
of wandering pigs’’ while they themselves were ‘‘without housing, food,
or work.’’ 77 They joined together to send a petition to the governor re-
questing the establishment of an army post in the settlement to woo
back the rural residents who had left ‘‘the lands in Samaná uninhabited
for fear of the chusma’’ just at the onset of the planting season. Only a
week after Samaná’s residents begged the government to send soldiers
to protect them, however, an armed Liberal band of forty men, led by
the Santander-based guerrilla leader Trino García, crossed the Magda-
lena River into Puerto Nare and once more spread terror in Antioquia’s
port towns and the surrounding rural areas.78
While Trino and his men attacked the southeast, two other guerrilla
leaders, Captain Corneta and Sergeant Santander, were being chased
down the mountains by the army in Caucasia. In the absence of any public
authorities who could do so, Caucasia’s telegraphist took it upon him-
self to provide the regional government with a public-order report from
Tarazá, where he had escaped to after the guerrillas cut the telegraph line
in Caucasia proper. He informed the governor that all the stores in the
town were closed, that the town’s inhabitants had fled, and that the priest
and all the nuns from the local girls’ school had escaped across the fields
to Puerto Antioquia in the company of the town’s public officials. Three
days later, national police troops fought with the guerrillas when they
encountered them attempting to set fire to the government-owned gaso-
line reserves. The national police assassinated eleven people.79 Nearby
Cáceres, moreover, resembled a ghost town. All of its farms had been
abandoned as inhabitants fled in the wake of a guerrilla attack in which
fields were razed and houses burned.80 Corneta and his men eventually
escaped via the Cauca River and a month later attacked mining camps in
El Bagre, Remedios, and Segovia. Just prior to the attack, local inhabi-
tants warned the governor that there were no policemen, no soldiers, and
no arms with which to defend the town.81 Having robbed the gold mines,
Corneta’s men proceeded toward Puerto Berrío, where his men and those
of Pielroja and Vicente Mejía wreaked havoc on the numerous haciendas
concentrated in the area between Maceo, Amalfi, Yolombó, and Puerto
Berrío.
On May , thirty to forty guerrillas attacked a farm in Virginias. It
belonged to a Conservative engineer in charge of Puerto Berrío’s elec-
 Bajo Cauca, Magdalena Medio, Northeast 

trical plant—the same engineer who had earlier denounced Conserva-

tive extremists and warned the governor not to allow the police and
contrachusmas to harass local Liberals lest they rise up in arms against
the state. Although some of the farm workers managed to escape (many
were Liberal), others were captured and killed by the guerrillas.82 Guer-
rillas next burned an hacienda in La Susana belonging to a prominent
Medellín industrialist family on June , then assaulted an hacienda in a
place called Alicante in Puerto Berrío and stole all its cattle.83 Four days
later guerrillas attacked again, this time at an hacienda belonging to the
same family in Murillo, Puerto Berrío.84 Within days, other haciendas
became the focus of guerrilla attacks in Remedios and all along the rail-
road line between Cristalina and Puerto Berrío.85 In a single attack on one
farm, twenty-five workers were killed, while in nearby Yalí on June ,
a group of ninety guerrillas assassinated the Liberal hacienda owner,
Raul Isaza Sierra, because he employed thirteen Conservative workers
and denied it when asked about their political affiliation.86 Guerrilla at-
tacks on haciendas in Maceo and Puerto Berrío continued through July
until finally the flood of guerrilla violence ended with attacks on Shell
Oil’s Casabe camp and against the Pato Consolidated Mining Company’s
camp in El Bagre where guerrillas stole gold dust, dynamite, tents, and
machines.87
The speed and intensity of guerrilla assault on eastern Antioquia
shocked the regional government. Although Conservatives residing in
the port towns along the Magdalena River in the southeast and in min-
ing hamlets in Caucasia had been the occasional targets of Liberal attack
since , such assaults had been sporadic and led by guerrillas based
outside Antioquia proper. The typical targets of these early guerrilla at-
tacks, moreover, had not been hacienda workers but rather the offices and
agents of the state. In contrast, the majority of guerrilla attacks occurring
between February and July of , October through November of the
same year, and the first four months of , were led by forces based in
Antioquia and conducted as part of a systematic and intensive strategy
of human eradication. The principal objects of these later attacks were
the mining camps and haciendas that produced the bulk of Antioquia’s
gold, cattle, and commercial agricultural products. Threats to the region’s
economy coupled with the fact that attacks were now led by forces based
in Antioquia—who were perceived to be likelier to garner the support
of local hacienda owners and inhabitants—prompted the regional gov-
ernment to finally treat the issue of public order in the area with the at-
 Blood and Fire

tention it had until then only lavished on questions of public order in

western Antioquia. The central problem facing the regional government
in early , however, was how to meet public-order challenges emerg-
ing simultaneously on several fronts when it lacked sufficient troops and
supplies.
On April , , Governor Henao Mejía received a telegram from
Bogotá’s secretary of the ministry of government encouraging the Antio-
queño leader to respond to growing guerrilla activity by arming Con-
servative civilians. To give added force to his message, the secretary for-
warded the complaints of numerous Conservatives from towns scattered
throughout Antioquia who expressed their discontent with being left at
the mercy of armed Liberal bands. He gave Puerto Berrío as a prime ex-
ample of a town that would benefit from the organization of ‘‘personal
defense groups.’’ 88 By the time Luis Ignacio Andrade’s telegram reached
the governor, however, Henao Mejía had already held a meeting with the
members of Antioquia’s Conservative directorate. Together they agreed
to create and arm contrachusmas in towns such as San Luis and Remedios
in eastern Antioquia and Caicedo and Titiribí in the southwest. In theory
these contrachusma groups were intended to complement, not supplant,
official forces when they were too few to defend particular areas against
Liberal attack. However, in reality local sponsors seized the opportunity
to form paramilitary groups to act independently or in lieu of official
troops regardless of whether sufficient official forces existed or not.89
The inhabitants of the municipality of San Luis, for instance, bluntly
informed the governor that the national police were ‘‘cowards’’ and that
‘‘the only solution . . . is to arm all the area’s Conservatives so that they
can defend themselves, and to give rifles to all the local members of the
army reserve who are older than twenty years.’’ 90 Requests for arms soon
poured into the governor’s office. On each telegram demanding weapons,
the governor reassured impatient petitioners: ‘‘we have sent arms,’’ and
‘‘we have distributed more than one thousand revolvers. We’re doing this
every day.’’ In a half-hearted attempt to maintain at least the semblance of
some kind of control, he also added that he needed ‘‘at least the names’’
of the reserve volunteers before they could be issued uniforms, food, and
weapons. He also warned Conservatives in San Luis not to repeat what
had happened in the western town of Ituango where Conservatives had
also volunteered to form a group of civilian police and then quickly dis-
appeared once arms had been distributed.91
 Bajo Cauca, Magdalena Medio, Northeast 

This was not the first attempt to organize Conservative civilian police
troops in eastern Antioquia. Extremist Conservatives and selected re-
gional authorities had first mobilized them in anticipation of the senate
elections of September . Most of these efforts had been concentrated,
however, in towns with a modicum of Conservative support and not di-
rectly in the mining towns of the northeast or in the Bajo Cauca where
Conservative loyalists were few. There was simply not enough local an-
tagonism between members of the two parties in these areas, or the Lib-
eral opposition was too large for Conservatives to seriously contemplate
forming paramilitary units to attack Liberal neighbors in these towns.
The strength of organized labor in mining towns such as Remedios, Zara-
goza, Segovia, and Caucasia, moreover, deterred the emergence of the
kind contrachusma-led electoral harassment that had occurred in Liberal
towns throughout the southwest between  and . The regional
government had also encountered little support from Conservative land-
owners in eastern Antioquia for their paramilitary project. Local land-
owners proved uninterested in promoting a partisan agenda for extrem-
ist regional authorities if such a strategy meant the possible disruption
of tenuous labor markets and valuable production. This left extremist
regional Conservatives with only one good option: to import or encour-
age individuals with no vested economic stake or local bipartisan ties in
eastern Antioquia’s Liberal-dominated areas to migrate there in order to
promote violence in return for material and political rewards.
Thus, it was not until the events of  that the possibility of de-
ploying armed Conservative civilians to usurp power from local Liberal
majorities in eastern Antioquia became a reality. The creation or deploy-
ment of Conservative civilian forces was ultimately most successful in
towns where Liberal railroad and public sector workers had been the vic-
tims of official abuse since the death of Gaitán in  and one or two
towns where Conservatives were a strong presence. Conservatives in San
Roque, for instance, colluded with Conservatives in Maceo and Caracolí
who were commissioned by Maceo’s parish priest to force Liberals to vote
in the  senate elections. Together they obliged Liberals to shout vivas
to the Conservative party and to parade in the streets on election day
‘‘bearing the blue flag [the Conservative party’s color] before them to save
their skins [para poder salvar el pellejo].’’ Liberals tolerated such humilia-
tion, one man recounted, because they feared the implicit threat of being
taken for ‘‘a ride’’ on the train that ran from San Roque to Puerto Berrío,
 Blood and Fire

where they would be ‘‘decapitated at the Monos railroad station,’’ which

was, not surprisingly, one of the targets of guerrilla attack.92 Other local
citizens confirmed that contrachusma reinforcements from Maceo and
Caracolí were sent to San Roque to ‘‘preach violence and finish off the
Liberals.’’ 93
Contrachusmas recruited and organized in Maceo and Caracolí even-
tually expanded their activities into towns such as San Carlos, San Luis,
and Cocorná in the southeast. In San Carlos a man named Emilio Espi-
nosa, referred to as a ‘‘stranger [ forastero]’’ by local inhabitants, was said
to have arrived mounted on a white steed, waving a blue flag, and leading
‘‘a mob of peasants which he’d collected’’ along the way. With the support
of San Carlos’s parish priest, Espinosa and his mob recruited unemployed
local agricultural workers and ‘‘began hacking down Liberal houses and
businesses.’’ 94 The possibility of effecting contrachusma violence against
local Liberals with no apparent connection to the Liberal guerrillas, how-
ever, was contingent on the support of local Conservatives. When the
local Conservative mayor of San Carlos interceded to defend a local Lib-
eral town council member and party leader from becoming the contra-
chusma’s victim, the guerrillas ‘‘put away their machetes and left.’’ 95 But
in places where the contrachusma encountered the support of local offi-
cials, they prospered and eventually challenged or usurped the control of
less extremist fellow party members. Such forces gradually grew in such
a manner as to contemplate penetrating into nearby areas where Con-
servative support was weaker and where Liberal workers were abundant.
Contrachusmas from Yolombó, for instance, spread northward into Yalí
and Amalfi, while those from Caracolí mobilized toward Alejandría and
Puerto Berrío.96
The mobilization of a landless, largely unemployed or underemployed
peasant ‘‘army’’ brought to the surface the very real issues of agricultural
crisis, land concentration, and declining employment that underlaid
many of the tensions between the predominantly agricultural southeast,
disappointed colonists in the northeast, and workers in the public and
foreign-owned mining sectors. Violence was waged through officially-
sanctioned paramilitary units (the contrachusma) that became legitimate
channels of expression for long-simmering rivalries and feuds not di-
rectly related to partisan differences. The contrachusma were formed
and led by ambitious local politicians and priests or by recently ap-
pointed public employees such as mayors and police chiefs who were
 Bajo Cauca, Magdalena Medio, Northeast 

brought into local areas from elsewhere in the department. The use of
force against Liberal citizens and uncooperative Conservatives was jus-
tified by invoking the defense of religion or preemptive strikes against
possible ‘‘Communist’’ subversion. The real purpose, as colluding parish
priests and local leaders made abundantly clear when called upon to jus-
tify their decision to arm poor civilians before disapproving regional and
national critics, was to take away jobs, solve rural unemployment, and re-
store ‘‘order.’’ In other words, the contrachusma became a crucial means
of putting right both political and economic wrongs.
Building on these earlier efforts, a consolidated, armed Conservative
response to the Liberal guerrilla’s presence began to take shape in the port
towns on the Antioqueño side of the Magdalena River shortly after the
first guerrilla attacks against haciendas in April . The explicit mission
of these public order reinforcements was to repel and eradicate Liberal
guerrillas and to inhibit the migration of peasants and settlers (colonos)
crucial to the area’s hacienda production.97 But contrachusma forces were
very quickly viewed by significant sectors of the regional population—
including many prominent Conservative property holders and a number
of Conservative public employees—as a remedy worse than the problem
they were meant to alleviate. The contrachusma’s violent and arbitrary
actions prompted even Conservatives to suggest the adoption of alter-
native solutions to the problem of public order that did not involve the
formation of armed civilian bands. Liberal and Conservative citizens in
Cáceres, for instance, jointly signed and mailed the governor a petition
in which they openly declared their support for the governor’s Pro Paz
Initiative, which was announced by both party directorates on the day
the governor secretly sent out arms to Conservative civilians, and asked
the regional government to promote a colonization project to repopu-
late the areas devastated by guerrilla violence. They insisted, moreover,
that rather than arming more people, the regional state would be better
off making its presence felt in the area by appointing legitimate authori-
ties and making a commitment to investing in the area’s economic de-
velopment.98 This would seem to suggest that locals were fully aware that
at least one of the attractions of participating in armed groups was the
hunger for jobs and land.
The inhabitants of Cáceres were not the only citizens to suggest a link
between the absence of state authority, economic incentives, and the inci-
dence of violence in the Bajo Cauca and the Magdalena Medio. A promi-
 Blood and Fire

nent Conservative family who owned numerous haciendas in the area be-
tween Remedios and Puerto Berrío, for instance, joined a Liberal member
of the Sociedad de Agricultores Colombianos and a Liberal member of
Antioquia’s legislative assembly, who were also landowners in the North-
east and Magdalena Medio, in a unique private sector offer to Colonel
Luis Abadía, the commander of Antioquia’s Fourth Brigade in May .99
These landowners offered land to any soldier willing to commit himself
to protect the area’s haciendas for five years. The signatories were willing
to alienate a portion of their own lands to the soldiers and to supplement
these with concessions obtained from the reserve of public lands held in
trust by the nation. They also offered to pay for barracks, a landing strip,
and supplies in order to attract a permanent presence of the state’s officers
to the area.100 Both in this letter and in another that these men wrote in
conjunction with others in June, the landowners stressed eastern Antio-
quia’s strategic economic importance. They also implied that the state’s
failure to make itself felt and respected in eastern Antioquia had contrib-
uted to the region’s becoming a major site of violence and an easy target
for attack.
In their June letter, petitioners impressed upon the governor ‘‘the ex-
tremely serious situation in which the cattle haciendas situated on the
west side of the Magdalena River have been left, especially those located
in the area between Puerto Nare and the shores of the rivers San Bar-
tolomé and Ité, encompassed by the municipal boundaries of Puerto
Berrío and Remedios.’’ Landowners also pointedly reminded the gover-
nor of the area’s importance as a supplier of agricultural products and
cattle to Medellín and the industrial satellite towns that surrounded it:
‘‘your lordship [Su Señoria] cannot ignore that the haciendas located in
the above-mentioned area hold more than sixty thousand head of cattle,
not to mention the fact that they constitute the pantry of our department.
The farms in that region supply  percent of the cattle for Medellín’s
cattle fairs. Besides, they also supply a large percentage of the agricultural
products consumed in this city [Medellín], and in many other Antio-
queño towns.’’ The petitioners complained that public insecurity in the
area had forced them to limit their purchase of livestock, while the threat
of cattle-rustling had seriously compromised their ability to transport
to markets the cattle ready for slaughter. Landowners were unequivocal
about the effect of a reduction in hacienda productivity. They warned the
governor that prices for meat and agricultural commodities would rise
 Bajo Cauca, Magdalena Medio, Northeast 

‘‘exorbitantly’’ to the detriment of consumers faced with already unstable

and escalating price cycles and would ‘‘produce serious setbacks in the
collective economy.’’ Violence, moreover, was forcing workers to migrate
away from haciendas—leading to a scarcity of labor and a concurrent
rise in operating costs and demand as former rural producers became
urban consumers. Concerned landowners also suggested a number of
ways of protecting private property in the region and of avoiding ‘‘immi-
nent socioeconomic upheaval resulting from the multitudinous exodus
of rural workers.’’ They requested an increased presence of the military
and the police, particularly those already familiar with the area (that is,
locals); the establishment of a body of armed personnel within certain ha-
ciendas to restore the confidence of workers who feared being killed; per-
manent foot patrols throughout the region; and boats and other means of
transportation that might aid in the pursuit and capture of guerrillas.101
What is striking about the landowners’ appeal for a greater military
presence in the area and their offer to ‘‘cooperate with the Armed Forces
with all the means at our disposal’’ is that they were members of both
parties. The threat posed to private property was attributed by land-
owners to the presence of ‘‘pillage and . . . banditry’’ which had trans-
formed the region ‘‘into the most terrible site of vandalism and ruin,’’
not to violence motivated by partisan differences. Given the highly politi-
cized discourse employed by the regional authorities when referring to
the problem of violence, it is notable that the economic motivation be-
hind armed mobilization was made explicit by cattlemen with holdings in
the Remedios/Puerto Berrío area, while partisan issues were rarely men-
tioned. This may have been due, as other evidence would seem to suggest,
to the fact that landowners were aware that violence that endangered the
region’s economic interests was committed by Conservatives as well as
Liberals, and they were thus careful not to alienate the regional authori-
ties by directly accusing them of fomenting violence. But it also seems to
suggest the persistence of a regional, elite bipartisan attitude that privi-
leged economic development and social calm above distinctly partisan
objectives.
The strategies advanced by local hacendados—most of whom were
absentee landlords with industrial and commercial interests in Medellín
—to end violence suggest that they and the regional government were
motivated by quite different concerns in the issue of public disorder. In
eastern Antioquia, hacendados affiliated with both parties were equally
 Blood and Fire

dependent on a predominantly Liberal workforce and were aware that

it was unrealistic and dangerous to attempt to alter its composition by
force. Landowners also feared the possible long-term consequences of
delegating the maintenance of public order to armed groups of poor indi-
viduals in an area characterized since the s by struggles over land and
resources, even when such individuals shared the landowner’s partisan
affiliation. The fate of private property in one particular corregimiento
during la Violencia, ‘‘El Tigre’’ in Amalfi, was well known to all the land-
owners in eastern Antioquia and vividly illustrates the material basis of
landowner fears during this period.

El Tigre

In December  the owners of the Hacienda El Tigre, located on the

border between Amalfi and Yolombó in a settlement also known as ‘‘El
Tigre,’’ complained to Antioquia’s secretario de gobierno that colonos (a
term that can mean either ‘‘settlers’’ or ‘‘squatters’’) were invading their
lands. To protect their property, they demanded that they be allowed to
form private, armed defense groups like those they claimed were orga-
nized and paid by the foreign-owned mining companies that operated
nearby.102 Although El Tigre’s proprietors did not specifically make refer-
ence to it, only six months earlier Governor José María Bernal and his sec-
retary, Eduardo Berrío, had lobbied hard to defeat a measure that would
have prohibited the sale of arms and ammunition to civilians. The cen-
tral government had proposed the measure as part of an effort to limit
the escalation of partisan conflict.103 But the governor and his secretary
argued that such a law discriminated against farmers who relied upon
shotguns and other firearms to hunt and defend their harvests and ani-
mals from thieves. The right to bear arms in self-defense, not surprisingly,
quickly became the basis for a government policy of distributing arms to
Conservative civilians between  and  and greatly increased the
number of hired guns employed by local political bosses and landowners
in areas where colono disputes were most common.
Two months after El Tigre’s proprietors demanded the right to form
armed defense groups, Amalfi’s town clerk informed the government that
the dispute between landowners and settlers over control of the Hacienda
El Tigre had been going on for a very long time and that the landowners
had already ‘‘used violent means to . . . establish gangs of armed men in
order to dislodge the settlers.’’ The private land dispute had become a
 Bajo Cauca, Magdalena Medio, Northeast 

public order matter and the clerk requested the immediate presence of
six departmental police agents to keep the peace.104 The police intervened
on behalf of the landowners, but colono attempts to gain recognition of
their rights continued. In June  eleven peasants made the trip to the
governor’s office in Medellín to complain that over the course of sev-
eral years they had ‘‘peacefully’’ made improvements [mejoras] consisting
of ‘‘houses, gardens, plantings . . . covering a large extension along the
mountain ‘La Gallinera’ ’’ on the contested lands.105 Yet a man named Luis
Restrepo, who claimed he held legal title to the lands they had settled,
had expelled, illegally imprisoned, and fined them. Settlers did not dis-
pute the legitimacy of Restrepo’s titles (although their validity had yet to
be determined), but they insisted on being reimbursed for the improve-
ments they’d made before being forced to move on. When the conflict was
still not resolved in August, settlers once more protested that Restrepo
was abusing them. He had ordered the police to surround them and had
threatened to take away their goods and harvests.106
The economic dispute between settlers and landowners in El Tigre
eventually intersected with the emergence of partisan violence, and the
police and private security forces originally employed to defend the ha-
cienda became the nucleus of a contrachusma.107 The village in which the
hacienda was situated, moreover, became the repeated site of assaults and
invasions while squatters and rural peons alike bore the indiscriminate
brunt of both contrachusma and Liberal guerrilla violence. Conserva-
tive inhabitants reported that the contrachusma attacked farm workers in
general, even Conservative ones, ‘‘beating [them] up, demanding money,
arms, and terrorizing workers as happened to a member of this locality’s
Conservative committee.’’ 108 Meanwhile, Liberal guerrillas sacked the
customs office, stole money and the majority of the liquor stocks, burned
down the telegraph office and archives, and destroyed half of the vil-
lage’s houses. The guerrillas then invaded the church and hacked the Vir-
gin of Fátima statue to pieces.109 The two partisan forces concentrated
their fury upon the civilian population but never engaged in combat with
each other, even though the goal of eliminating the other was what os-
tensibly served to justify their mutual existence. By February of ,
most of the inhabitants of El Tigre had fled, and violence had become so
severe that the Caja de Crédito Agrario y Minero (Agrarian and Mining
Credit Agency) complained to the governor that even financially desper-
ate ‘‘working peoples categorically refuse to go into that region.’’ 110
El Tigre’s situation was widely known to the property holders and
 Blood and Fire

working peoples of the northeast, and its history illustrates one facet of
the complex situation raised by partisan violence in areas where there
were preexisting conflicts over land, labor, or resources. What had begun
as a struggle over land between settlers and landowners had opened the
way for armed groups (initially sponsored by the landowner to dislodge
the discontented settlers who challenged his authority) to gradually as-
sume control of the hacienda and act on their own. ‘‘Self-defense’’ groups
certainly dislodged settlers, but they also invited Liberal guerrilla attacks
on the area and ultimately created a situation of violence so severe that the
original landowners could neither reoccupy their lands nor attract labor
to work them. For local landowners the lesson of El Tigre was clear. Em-
powering armed groups, particularly those who could justify their violent
activities by representing themselves as allies of the regional government
in an area characterized by struggles over property and labor, was simply
too risky a strategy for landowners, even Conservative ones, to adopt.
Logic argued against the creation of informally organized, armed
groups regardless of their official objectives. The majority of the area’s
settlers and workers were Liberals and many were non-Antioqueño mi-
grants. Many of the area’s landowners, however, were Conservative ab-
sentee landlords whose residence was in distant Medellín. For years before
the outbreak of violence, absentee landlords had a system for ensuring
a stable labor force and the continued productivity of their lands: they
had hired Liberal mayordomos (estate managers) and administrators and
allowed them to determine the political affiliation of the majority of ha-
cienda workers. Many of the workers were transient or seasonal laborers
or tenants with precarious status, who, like Alfonso in Angela Rosa’s ac-
count, drifted from one limited contract to another to work someone
else’s land. For the landowners, the advantage of such a system was that
it never enabled workers to establish a claim on lands whose status (pri-
vately owned or in part usurped from public lands) was often murky.
Several prominent industrialists and merchants from Medellín with large
properties in the Puerto Berrío/Remedios/Yolombó zone had already
had to fight the claims of poor colonos and expelled thousands of them
from the land in the s and s.111
Paradoxically, the emergence of Liberal guerrillas did not threaten ha-
cienda labor relations or private property in eastern Antioquia as did
armed Conservative bands operating in conjunction with the police. This
was not because Liberal guerrillas were less violent than the Conservative
 Bajo Cauca, Magdalena Medio, Northeast 

contrachusma; the list of thefts, assaults, and assassinations perpetrated

by Liberal armed groups on the area’s haciendas between  and 
leaves little doubt as to the guerrillas’ brutality. But the guerrillas were
primarily interested in supporting themselves economically in order to
finance their rebellion against the Conservative government; they were
not motivated by land hunger or unemployment. Most of the members
of the guerrilla groups operating in the area, moreover, were not native to
the department and had not been poor farmers. They were former pub-
lic workers, miners, and cowboys, not sedentary agriculturalists.112 What
the guerrillas had in common with the majority of workers and mayor-
domos in eastern Antioquia was a shared political affiliation. Thus, while
guerrillas might rob an hacienda or work out a deal with mayordomos
to take a cut of the hacienda’s production, if the workers were Liberal,
the guerrillas tended to leave them and the land alone. Local landowners
also appear not to have believed the arguments of the armed forces and
regional government that local Liberals were inherently rebellious or in-
distinguishable from the guerrillas. Indeed, the very small size of most
guerrilla bands operating in the area and the overrepresentation of non-
Antioqueños in them suggest that while locals may certainly have looked
to the guerrillas to defend them from Conservative violence, few flocked
to join such bands. For instance, Angela Rosa’s companion, Alfonso, and
his friends only felt compelled to seek out and join a guerrilla group after
Liberal hacienda workers became the target of contrachusma attacks in-
vading from other areas in mid-. The threat of unemployment and
sudden, forced displacement by a landowner seems to have acted as a
stronger catalyst to guerrilla membership than the threat of personal in-
jury inflicted by roving bands of the Conservative opposition or any sen-
timent of partisan loyalty.
The peculiar structure of the local economy thus intersected with
political pressures to create a seasonal labor market in which theft, attacks
against the state, and general violence alternated with more traditional
employment in mining, clearing pasture land, and agricultural labor.
Young men with no other viable means of support and little other means
of defense easily shifted in and out of armed bands and from violence to
seasonal labor on nearby estates. Guerrilla attacks increased when sea-
sonal work on local estates dried up and declined when the young men
whom they recruited found employment clearing forests, sowing fields,
breeding cattle, or mining. The structure and operation of the guerrillas
 Blood and Fire

therefore adapted to and reproduced the area’s seasonal cycles of produc-

tion and employment without unduly threatening or rearranging these
activities.
Collusion between contrachusmas and local authorities, in contrast,
transformed indiscriminate robbery into the prelude for more compli-
cated and permanent patterns of extortion and, ultimately, land usurpa-
tion. Unlike the Liberal guerrillas, the contrachusmas threatened not only
hacienda production, they also affected the stability of the area’s labor
market since they operated on the assumption that all Liberals (includ-
ing hacienda workers) were guerrillas or guerrilla sympathizers and that
this justified their abuse and elimination. The contrachusma and their
police allies also justified a wide range of activities, such as expropriation
of cattle and agricultural production and the appropriation of lands, in
the name of eradicating subversion and defending the interests of values
such as religion, order, and democracy. In doing so, the contrachusma
threatened landowners’ interests by accusing them, as the Liberal guer-
rillas could not, of betraying the government by hiring Liberal workers
and of being complicit with guerrilla forces.
In January  a Conservative proprietor in San Roque (where contra-
chusmas were particularly active) complained to the governor’s secretary
that he had returned to his farm after a two-month absence only to find
that it had been overtaken by ten armed men who beat him up and ac-
cused him of leading a guerrilla group of forty men against the govern-
ment.113 Another Conservative complained that his land had been taken
over by a former administrator with the help of the ‘‘police quartered in
this parish’’ and that they had barred his own and his workers’ entry to
his property.114 Incidents such as these probably influenced the decision
of a Conservative owner of several haciendas in Remedios, whose prop-
erties had repeatedly been the target of Liberal guerrilla attack, to refuse
the local contrachusma leader’s offer to set up an armed group within
his haciendas’ limits. The hacienda owner informed the governor that
he rejected the offer of contrachusma protection ‘‘spurred by my belief
that this [the monopoly of force] is the sole prerogative of the authori-
ties.’’ 115 But he must also have worried that once inside the hacienda there
was nothing to keep an incident like El Tigre from happening or noth-
ing to prevent the contrachusmas from eventually usurping control of his
properties.
Conservatives who disagreed with the tactics employed by their ex-
tremist party members even began to turn the extremists’ justification
 Bajo Cauca, Magdalena Medio, Northeast 

of the need for Conservative militancy on its head. In addition to re-

pudiating the use of violence against innocent local Liberals, moderate
Conservative citizens in Puerto Triunfo, for instance, accused the contra-
chusmas and their civilian supporters of attracting rather than impeding
Liberal guerrilla attacks. Conservatives singled out the local police in-
spector as the principal instigator of ‘‘the assaults which our party mem-
bers have been conducting against peaceful settlers of that land.’’ The
inspector’s actions worried moderate Conservatives because they might
goad ‘‘the guerrillas that operate in the mountains of Boyacá to organize
and attack the port with . . . unfavorable repercussions for the Conser-
vative party and the government.’’ The fact that Conservatives were in
power and constituted a local majority, moderates insisted, was a poor
excuse for provoking the opposition. The Liberal’s economic importance
weighed heavily with Conservative moderates, many of whom were land-
owners. They described local Liberals as ‘‘Liberal peoples who . . . are
workers who haven’t been mixed up in anything’’ and who had the high-
est rate of productivity in the entire Magdalena Medio region. Common
sense thus prompted moderate Conservatives to urge the regional gov-
ernment not to harass local Liberals in places where ‘‘the government
lacks adequate police forces to deal with so many disturbances.’’ But their
complaints were also motivated by a deeper concern regarding the appro-
priate functions of the state. The state should ‘‘respect the lives of those
citizens in order to maintain the region’s tranquility’’ and not promote
and perpetrate violence.116
The armed forces and extremist Conservatives, however, rejected the
attempt by local moderates to persuade the governor to stop deploying
or tolerating official attacks against rural Liberal workers. They suggested
that local landowners had enabled their workers and mayordomos to use
their privileged position within the haciendas to rob and despoil them
at will, and that they were therefore criminal accessories of guerrilla vio-
lence. Puerto Berrío’s army commander made this point explicitly in re-
lation to an hacienda in Murillo just north of the town. The hacienda had
been attacked by masked men who bound the watchman and took his
shotgun ‘‘after all the workers docilely allowed themselves to be tied up.’’
The guerrillas knew exactly where the mayordomo kept a second gun,
removed it, and ‘‘told the mayordomo to take out his family and all their
things because they were going to set fire to the house.’’ The workers were
then let go unharmed so that they could ‘‘watch their employer’s house
go up in smoke.’’ Only one person was injured, ‘‘an individual affiliated
 Blood and Fire

with the Conservative party’’ whom the attackers had sought ‘‘to kill.’’ 117
The owners, the authorities argued, had created the very conditions that
now endangered the economic viability of their properties. ‘‘The majority
of Conservative and Liberal hacendados who do not support the guer-
rillas, dare not go to their haciendas for fear of being killed. In order to
protect their properties from attack they surround themselves with ad-
ministrators, mayordomos, and workers who are completely disaffected
from the government. . . . Liberal hacienda owners are even more prone
to do this.’’ 118 Indeed, the army commander asserted, ‘‘the bandits are
located inside the very haciendas. . . . This makes it extremely difficult
to control the situation when there is no sincere desire on the part of
the hacienda owners to cooperate with the government in the reestab-
lishment of normalcy.’’ The only possible means of eradicating Liberal
banditry, the authorities concluded, was to import ‘‘new personnel . . .
healthy [sano] workers who unlike the rest of this region aren’t full of
revolutionary ideas.’’ In the military’s estimation, this was the only way of
‘‘cleansing [sanear] the region’’ and pacifying it.119 One way of pacifying
was to encourage and deploy contrachusmas from the southeast where
there was not only a greater Conservative presence but also a large num-
ber of landless poor with few if any employment opportunities.
The formation of the contrachusmas or ‘‘an army of armed peasants
[campesinos],’’ as the parish priest of Yolombó explicitly suggested doing
in , had long been advocated by local priests and municipal Con-
servative committees as an efficient solution.120 The formation of armed
Conservative bands solved the problem of growing enclosure around
once public lands in the southeast and accompanying growth in unem-
ployment.121 The contrachusmas provided jobs for needy Conservatives
either by employing them as civilian policemen or by using them to strip
Liberals of state patronage jobs and employment on haciendas. The cre-
ation of the contrachusma, moreover, was perceived to solve other ‘‘prob-
lems’’ as well. Civilian volunteers with a stake in the region were thought
to be more committed antagonists of the guerrillas, and they consti-
tuted an inexpensive and effective means of supplementing what every-
one agreed were insufficient government troops in the area. The contra-
chusma were also perceived to exert a rehabilitative moral and ethnic
presence on the very populations the police in Puerto Berrío suggested
were full of ‘‘revolutionary’’ ideas. Contrachusma volunteers were drawn
from among ‘‘white’’ migrants from core municipalities who had recently
 Bajo Cauca, Magdalena Medio, Northeast 

settled in the southeast, and, as such, they became the ethnic shock troops
of extremist Conservative interests who attributed local violence to the
presence of unruly blacks and non-Antioqueños.122
Efforts to persuade the regional government to modify its indiscrimi-
nately repressive policies against Liberals in eastern Antioquia were thus
drowned out by the arguments of the armed forces and extremist Con-
servatives who approached the issue of public order as a moral crusade.
The repeated defeat of the army and the police at the hands of guerrillas
in Urabá, western Antioquia, the Bajo Cauca, and the Magdalena Medio
in , moreover, convinced extremists in the regional government of
the necessity of adopting more severe—not more lenient—measures to
restore public order.

The State’s Response to Guerrilla Violence

The role of the military in the containment of civilian conflict was the
first aspect of public order policy to undergo a drastic shift in the Magda-
lena Medio region in mid-. By June, the commander of the troops
stationed in Puerto Berrío issued a memorandum to hacienda owners
and farm administrators in which he announced a series of policies in-
tended to control the hiring and movement of hacienda workers in the
region. The armed forces once more accused Conservative landlords of
knowingly hiring Liberal mayordomos and having them ‘‘hire [engan-
char]’’ Liberal workers. The army insisted these were ‘‘potential bandits’’
who robbed and burned haciendas. So, the military announced that, be-
ginning in June, ‘‘hacienda proprietors or their administrators will have
to put together a registry of hacienda employees’’ in which the names of
all workers, their job descriptions, dates of employment, town of origin,
national identity number, and reservist status had to be noted.123
When a worker left his job or was fired, whoever was in charge of the
hacienda was obliged to ‘‘immediately inform the local barracks com-
mander regarding the worker’s date of departure and intended desti-
nation.’’ The hacienda’s registry of workers had to be kept in ink, and
owners were obliged to submit it to the local military base for inspec-
tion at the beginning of every month. Hacienda employment was also
made contingent on the agreement of workers to collaborate with the
authorities in the identification and capture of guerrillas. Permits (salvo-
conductos) which enabled individuals to obtain work were to be issued
 Blood and Fire

only to those who agreed to help the military and police and only after the
‘‘well-known honorability’’ and ‘‘absence of a criminal record’’ of these
individuals had been vouchsafed by employers. Hacendados were for-
bidden to hire ‘‘unfamiliar personnel’’ or those who couldn’t provide the
information demanded by the salvoconducto process. Anyone found in
the region ‘‘without his appropriate pass,’’ moreover, was to be ‘‘consid-
ered a bandit suspect and will be detained by the army and the police,’’
and any worker who lost his pass ‘‘will not have a new one issued to him
but will be forced to abandon the region.’’ 124
The effect of these measures on the region’s workers was felt only a
week after the military issued its memorandum. On July , police raided
‘‘suspicious’’ haciendas in the Alicante and La Florida parishes, where
they stole cattle and ten cargas ( kilograms) of corn and beans and
assassinated thirty workers. On July , Puerto Berrío’s police arrived in
Maceo and its haciendas and killed another forty campesinos they sus-
pected of being guerrillas.125 Since employment had been made contin-
gent on collaboration with the authorities, moreover, workers increas-
ingly acted as informants in order to remain alive and keep their jobs.
Liberals who oversaw Conservative-owned haciendas in Maceo, for in-
stance, reported sighting Pielroja’s guerrillas and informed the army that
these guerrillas had called in one Liberal worker and charged him with
asking the town’s Liberal leaders to send supplies. The guerrilla leader
had let the workers know that ‘‘he thought it very strange that given that
Maceo’s Liberals knew he and his men were hanging around in those
mountains, that no one had offered to help them out with anything.’’ 126
Other Liberals reported that shipments of bullets, shoes, and clothes des-
tined for guerrilla use were being transported between Maceo and an
hacienda by the name of ‘‘Playa Rica’’ in Remedios.127
The military’s aggressive campaign to gain the upper hand over pub-
lic order matters in eastern Antioquia was somewhat undermined by
the reappointment of Dionisio Arango Ferrer as Antioquia’s governor
in late July . Arango Ferrer had last been named governor the day
after Gaitán’s assassination when the central government had deemed it
necessary to use force to stem outbreaks of Liberal subversion in the de-
partment. During his earlier administration, the governor’s preference
for privately organized, Conservative civilian forces to maintain public
order and his violent tactics had elicited an outcry from moderate Antio-
queño Conservatives, and he had left office only eight months after his
 Bajo Cauca, Magdalena Medio, Northeast 

appointment. His reappointment in  reflected the central govern-

ment’s desperation and the extent to which it feared public order had
deteriorated in the nation’s most important department. Arango Fer-
rer was no friend of the army.128 He suspected that sectors of the armed
forces were sympathetic to the Liberal opposition or, at the very least, did
not endorse the Conservative government’s instigation of partisan war,
and he feared army hostility might eventually instigate a coup. For this
reason, Arango Ferrer was reluctant to increase the power of the mili-
tary over public order matters. The governor put his trust instead in the
police and armed Conservative contrachusma forces, despite overwhelm-
ing evidence of their brutality and ineffectiveness. It was these two forces
then which he chose to bolster and reinforce in the campaign to restore
public order in Antioquia during his tenure from July , , until a
military coup against Laureano Gómez also toppled him on June , .
Shortly after assuming office in August, Arango Ferrer ordered the
mayors of Puerto Berrío, Remedios, Maceo, San Luis, Caracolí, and Yalí
to contact the presidents of their local Conservative committees. The
presidents were asked to send a special delegate to a meeting at the gov-
ernor’s office in Medellín to discuss issues of public order. The delegates
were to bring detailed reports of the state of public order in their areas
and the names (and national identity card numbers) of potential Conser-
vative volunteers.129 Arms and supplies were distributed to each of these
volunteers. Two weeks later, the armed forces issued a confidential mes-
sage to all its commanders in Antioquia in which, as a way of regaining
the institution’s prestige and the public’s faith in the army as an effective
combatant against guerrilla violence, it announced a policy of no quar-
ter in dealing with Liberal guerrillas. Puerto Berrío, they admitted pri-
vately, was one of the places where they deemed their counterinsurgency
strategy to have been most deficient and, aside from Urrao and Dabeiba
in western Antioquia, the municipio where it was most important to stop
violence in the department.130
The combination of these two measures—the strengthening of contra-
chusma forces and the decision to adopt a policy of no quarter against the
guerrillas—soon provoked a wave of terror throughout eastern Antio-
quia. Less than a month after distributing arms and condoning the cre-
ation of contrachusmas in Remedios, Maceo, and Yalí, officials and citi-
zens reported widespread assassinations and indiscriminate thefts led by
contrachusma forces against the area’s citizenry. In late September the
 Blood and Fire

mayor of Remedios warned the governor in a coded message that the

town’s Liberals were fleeing because of a rumor that the Conservative
contrachusmas were about to arrive.131 Meanwhile, the severity of contra-
chusma violence in Maceo provoked Liberal and Conservative citizens
to band together to ambush and assassinate the ‘‘contrachusma leader
Evelio Carmona’’ while he was unarmed and his men were confined to
their barracks.132 In Yalí and Vegachí, Yolombó, moreover, the mayor re-
ported accompanying the visitador administrativo to investigate reports
of contrachusma violence against both Liberals and Conservatives. They
found that men under the leadership of Emeterio Castro had assassinated
ten peasants, five of whom were Conservatives; looted and robbed Con-
servatives and Liberals alike; and raped local women belonging to both
parties. The visitador concluded that the contrachusmas conducted ‘‘gen-
eral extortions’’ and that they operated ‘‘without any control,’’ and he
recommended that they ‘‘should be totally exterminated’’ because their
behavior only discredited the government. Instead of arming unsuper-
vised and informal paramilitary groups, the visitador suggested that the
governor would be better off taking up the local hacendados’ offer to pay
‘‘lodging and . . . the expenses associated with supporting and arming the
police’’ and appointing a police inspector to oversee the region.133
Conservative repudiation of partisan violence was even more pro-
nounced in towns where Liberals were an overwhelming majority. A
meeting held by moderate Conservatives in Remedios was violently inter-
rupted when the local contrachusma leader burst into the party’s head-
quarters and announced that ‘‘he had a list of Liberals to be eliminated
approved by [Antioquia’s] Conservative directorate.’’ Dismayed, mod-
erate Conservatives immediately telegraphed the governor to ask if this
was true. They warned him that if it was true, then the governor was
being complicit with ‘‘abuses against peaceful citizens whose only crime
is to belong to the opposition party.’’ They angrily insisted that ‘‘a good
 percent of all the many tragedies which this town has suffered’’ were
not the product of guerrilla violence but of the contrachusmas who have
forced ‘‘families of known honorability belonging to both parties’’ to have
to ‘‘abandon the city.’’ A month later these same moderate voices reiter-
ated their concerns regarding the activities of the contrachusma and its
use against ‘‘peaceful Liberal citizens, helpless people who are engaged in
working in the countryside.’’ The Conservative mayor complained that
abuses were conducted by ‘‘a group of Conservatives led by Remedios’s
 Bajo Cauca, Magdalena Medio, Northeast 

treasurer’’ who had organized ‘‘groups that they call contrachusma whose
purpose is to go out into the countryside and kill Liberals.’’ The mayor
justified his resignation by saying that the contrachusma wanted him ‘‘to
take part in all kinds of vengeances; things I don’t want to accept.’’ It
was one thing, the mayor insisted, to go after ‘‘those who lend their sup-
port to groups up in arms’’ and ‘‘to pursue and punish those whom it
can be proven are responsible [for guerrilla violence],’’ but another thing
entirely to assume that simply because they belonged to the opposition
that all Liberals were bandits. The local priest and Conservative commit-
tee members agreed with the mayor, but the governor did not, and he
accepted the mayor’s resignation with considerable alacrity.134
The police inspector of Puerto Nare similarly denounced the police
and their contrachusma aides and accused the governor of violating the
public’s trust by presupposing that ‘‘the mission of lower-ranking em-
ployees is to assassinate ‘traitors’ [manzanillos] in order to pacify the de-
partment.’’ If this were the governor’s expectation, then Puerto Nare’s
inspector would resign because ‘‘my condition as an honorable man who
has a soul entrusted to God will not permit me to stain my name and
reputation.’’ The police inspector added that the oft-invoked justification
for the state’s indiscriminately repressive tactics in the area, the so-called
Liberal ‘‘chusma,’’ moreover, was ‘‘nothing more than people abused by
the police who fled to the hills to seek refuge and revenge.’’ 135 The very
forces whose reprehensible behavior the police inspector condemned—
‘‘people with a bad conscience who call themselves Conservatives, but of
the kind that Doctor José María Bernal classifies as rustlers who want to
assume ownership of other people’s properties,’’ as the inspector’s dis-
traught son put it to the departmental secretary eight months later—con-
spired to remove the police inspector from his post.136 Colorado’s police
inspector in the municipality of Caucasia also made it clear that interests
that transcended partisan objectives were at the heart of many disputes
between the contrachusmas and their opponents. For instance, he attrib-
uted accusations leveled against him by local hard-liners—accusations
similar to those leveled against Puerto Nare’s police inspector—to his re-
fusal to be an ‘‘instrument of vengeance’’ against local Liberals who were
‘‘rich and apolitical.’’ 137
Many Conservative observers residing in the northeast and Magda-
lena Medio regions were quite simply appalled by police and contra-
chusma brutality. Their publicly expressed outrage corroborated and ex-
 Blood and Fire

panded upon the critiques offered by the public employees of Puerto

Nare and Puerto Triunfo. The issue at stake was not the legitimate use of
force to defend against Liberal guerrilla assaults. Even local Liberals ad-
mitted that the guerrillas were often as bloodthirsty and sadistic as their
Conservative opponents. Rather, what bothered those who denounced
contrachusma and police actions was the indiscriminate use of violence
against civilians simply because they belonged to the opposition. The
state was charged with protecting and not harming citizens, these crit-
ics argued, and the official endorsement of paramilitary forces violated
the contract binding citizens and the state. It was one thing, as Arturo in
the Bajo Cauca had argued, when ‘‘the bandits commit horrible crimes,
have always committed them, and will continue to commit them,’’ for
they were outlaws ostracized and excluded from civil society. But it was
another thing entirely when ‘‘the others, the policemen, are exonerated,
are paid and glorified so that they may pervert their mission of peace for
that of shameless theft and criminality.’’ 138 The same yardstick could not
be used to measure the behavior of those sworn to uphold the law and
those up in arms. Local citizens reminded the regional authorities that a
state that modeled its behavior on the comportment of the very sectors
it was meant to discipline and that justified criminal behavior by arguing
that the means justified the ends risked becoming illegitimate in the eyes
of its citizenry.
The regional government’s strategy of promoting contrachusma
forces to combat the guerrillas had three major effects on the situation of
public order in eastern Antioquia. First, it brought to the surface irrec-
oncilable differences within the ranks of the regional Conservative party
regarding questions of public order maintenance and attitudes toward the
opposition. These differences ultimately contributed to catalyzing local
Conservative support behind a military coup in June .139 Second, the
use of contrachusma forces provoked a violent reaction among the Lib-
eral guerrilla groups operating in the area without significantly lessen-
ing the impunity with which they acted. And, finally, the proliferation of
paramilitary forces condemned the region to becoming the site of per-
sistent conflict long after Liberal guerrillas had given up their arms and
ceased to pose a challenge to the regional authorities or the state.
The months between October  and June , when the deploy-
ment of Conservative contrachusma forces in eastern Antioquia esca-
lated, were characterized by brutal acts of vengeance and retaliation be-
 Bajo Cauca, Magdalena Medio, Northeast 

tween the Liberal guerrillas, the contrachusma, and the state’s official
forces (especially the police). The focal point of struggle continued to be
the region’s haciendas and the civilians who worked on them. First, one
group would sweep through the rural areas of towns such as Remedios,
Amalfi, and Puerto Berrío raping, killing, stealing, and burning and then
the forces of the opposition would follow and eliminate what little was
left or would indiscriminantly go after the opposition in nearby commu-
nities to exact revenge. El Tigre in Amalfi, Santa Isabel in Remedios, the
railroad stations of Virginias, Sabaletas, Cristalina, and San José de Nus in
Maceo and Puerto Berrío became sites of repeated martyrdom, assaulted
and looted by opposing armed bands until neither people nor goods were
left. Those who survived migrated, swelling the number of starving refu-
gees in towns such as Puerto Berrío. By December  the parish priest
calculated that Puerto Berrío had at least five hundred unemployed heads
of household, an alarming rate of infant deaths due to malnutrition, and
an innumerable number of poor, ‘‘all the result of banditry.’’ 140
The distribution of arms by the government, moreover, increased
the number of predators in the region, but the carnage and destruction
caused by the guerrillas and their ability to elude the authorities did not
diminish. Indeed, all that changed was that now instead of one active
force that stole, raped, and killed, there were two. These forces rarely
if ever faced off against each other in direct combat. Instead, they con-
ducted an intricate dance of evasion, circling and feinting but never cross-
ing, preying mercilessly on the populations unfortunate enough to lie in
their path. Civilian casualties in eastern Antioquia amounted to  per-
cent of all civilians killed in Antioquia ( of a department total of ) as
a result of violence between January and May of , but eastern Antio-
quia accounted for only  percent of guerrilla deaths in the department
during the same period.141 The presence of police troops in the area grew,
but so did the incidence of violence. There were sixty national police-
men stationed in Puerto Berrío and thirty-two in Zaragoza in January
 when other violent municipios in Antioquia had fewer than twenty-
five agents each.142 In February there were  policemen in Maceo while
Urrao in western Antioquia had only  agents; and, in March,  police-
men were stationed in Remedios when the most violent municipios in the
southwest (Betulia, Salgar, and Urrao) had  agents each.143 Any justifi-
cation that the contrachusma were necessary because government forces
were absent or insufficient was clearly risible, for more troops were de-
 Blood and Fire

ployed in eastern Antioquia than anywhere else in the region in . In

any case, soon even the pretense of combat faded, overshadowed by the
lucrative opportunities to usurp and occupy the lands and goods that
desperate, frightened people had left behind. By the beginning of  it
was clear that the primary sources of instability in the region were the
government’s own forces.

The ‘‘End’’ of Violence in Eastern Antioquia

Ideologically motivated conflict in eastern Antioquia gave way in  to

a war in which the only meaningful objectives of armed conflict were ma-
terial goods and access to employment. Already by  it was difficult to
state with any degree of precision just who made up the various ‘‘sides’’
of the conflict. Generalized competition for goods gave rise to situations
like that reported by two hundred colono families from Puerto Triunfo
who had fled to La Dorada to save their ‘‘seriously threatened lives.’’ They
left behind their ‘‘homes and plantings, as well as the animals . . . used
for agricultural labor.’’ In a petition to the governor begging for protec-
tion, they delivered a painful list of those assassinated in their town, all
of them poor settlers and hacienda workers. The colonos insisted that
the killings and robberies that had forced them to flee ‘‘were prepared in
Puerto Triunfo by people who were perfectly well known to all’’ and that
this could be corroborated by their ‘‘Conservative neighbors . . . who out
of good will and Christian spirit had impeded other attacks.’’ The assas-
sins ‘‘seek out every possible means to befriend and obtain the passive
obedience of the police,’’ but partisan concerns appeared not to be the
central motivation for their violent acts. This led the refugees to conclude
that the problem was not a partisan one, but rather the result of ‘‘social
breakdown.’’ ‘‘Neither the undersigned, nor the instigators and perpetra-
tors of the depraved acts we have recounted,’’ they insisted, ‘‘belong to a
single political party.’’ 144
The refugees suggested that an implicit bargain between themselves
and the government had been broken by the events of la Violencia, and,
in doing so, they expressed a sentiment common among many of east-
ern Antioquia’s inhabitants. The life of the majority of local inhabitants
was dominated by the hardship of agricultural labor and by the modest
expectation that years of ‘‘physical effort to forge wealth and homeland
[patria] in inhospitable regions’’ would guarantee them but the barest
promise of survival. It seemed relatively little to ask that in return for
 Bajo Cauca, Magdalena Medio, Northeast 

‘‘years of arduous effort bereft of the smallest comforts . . . and [consid-

ering] the payment of our taxes and obligatory contributions’’ that the
government should at the very least ‘‘safeguard our lives and tranquility.’’
That the very government entrusted with protecting them promoted vio-
lence by empowering individuals such as the police and contrachusmas
who ‘‘for reasons we cannot fathom obey orders issued by irresponsible
persons who are only interested in damaging the government, the re-
gion, and private citizens’’ bewildered the former inhabitants of Puerto
Triunfo and deepened the sense of isolation and alienation to which east-
ern Antioquia’s inhabitants had long been condemned.
Indeed, as the number of denunciations of contrachusma violence
grew to alarming proportions, and as the number of these filed by Conser-
vatives increased, it became evident that the supporters and detractors of
a paramilitary response to the issue of public order were divided by more
than partisan differences. What had begun as partisan conflict gradually
crystallized along opposing lines of economic interest. Those supporting
the formation of contrachusma forces and those who defended them even
when their fellow party members denounced them as assassins, tended to
be select members of the local clergy accustomed to acting as local politi-
cal brokers, extremist public officials who gained authority and prestige
through their association with armed men over whom they exerted a
certain degree of control, and poor Conservative individuals who often
migrated from economically depressed areas and volunteered to form
contrachusma groups in the hopes of material as well as political rewards.
Those who most vocally opposed the contrachusmas, in contrast, were
either wealthy property holders, locally-born Conservative public offi-
cials linked by interest and friendship with prominent Liberals, or poor
Conservatives suspected of disloyalty by their party members because
they had long coexisted with and married into the local Liberal majority
and refused to collaborate in eliminating their neighbors and kin. Unlike
the mixed feelings elicited by the contrachusma, Liberals and Conser-
vatives differed little in their view of the Liberal guerrillas. Landowners
affiliated with both parties repudiated the guerrillas’ often arbitrary and
vindictive acts and attributed many of the area’s economic difficulties
to their constant cattle-rustling and thefts. Popular Liberal support for
the guerrillas was more widespread but seems often to have been largely
shaped by circumstance and necessity rather than from any deep sense
of ideological communion or identification.
By June  complaints of indiscriminate violence waged against Lib-
 Blood and Fire

erals and Conservatives alike, and shaped largely by economic and not
partisan competition, were echoed by inhabitants from all walks of life in
eastern Antioquia.145 The main perpetrators of violence were uniformly
described as the very forces charged with controlling public order in the
region and protecting its inhabitants’ lives. Insult was added to injury,
moreover, when the inhabitants were expected to maintain paramilitary
groups through forced contributions of  pesos or more in towns such
as Yalí.146 The exaction of forced contributions from locals and the direct
support provided by the regional authorities enabled well-armed, largely
independent bands of roving assassins, legitimized by official authority,
to become entrenched in eastern Antioquia. These bands continued to
operate long after Liberal guerrilla groups such as those manned by Capi-
tán Corneta accepted the military’s amnesty in July .147 By August of
, Antioquia’s Liberal party directorate warned the military governor
that not even the military regime was safe from the contrachusma. The
contrachusma were said to be intent on ‘‘toppling’’ the regime with the
support of the bishop of Santa Rosa, Miguel Angel Builes, and former
governor, Dionisio Arango Ferrer.148
Much to the despair of regional moderates, Antioquia emerged as a
national leader in the organization and distribution of arms to paramili-
tary forces and became a source of reference for other governors and
departments wishing to follow its example (see appendix B.). The gov-
ernor of Huila, for instance, enjoined Dionisio Arango Ferrer in April
 to share with his regional government ‘‘by what means the governor
managed the purchase of revolvers and ammunition to arm Conserva-
tives in order to counteract bandit actions and maintain public order.’’
The ‘‘state-of-the-art’’ arsenal of recently purchased Winchester repeat-
ing rifles and Hornet .-caliber rifles that was deployed by Antioquia’s
contrachusma groups had become the source of considerable admiration
by Conservative authorities anxious to replicate Antioquia’s success in
their own regions.149

Conclusion

The creation and consolidation of a paramilitary response to problems

of public order was perhaps the single most important outcome of the
struggle that began as a conflict motivated by partisan differences in east-
ern Antioquia. It shaped the emergence of endemic violence around the
 Bajo Cauca, Magdalena Medio, Northeast 

area’s haciendas and mines, and in contrast to what scholars studying the
effect of violence on partisan identification in other Colombian regions
have suggested, la Violencia appears to have undermined, not cemented
local loyalty to Colombia’s traditional parties.150 People in towns such
as San Roque, San Luis, Maceo, and Cocorná, for instance, turned away
from the mainstream of the Conservative and Liberal parties to swell the
ranks of dissident political movements that challenged both the tradi-
tional two-party system and the legitimacy of the regional government in
the decades following la Violencia.151 Incidents such as land invasions and
the forcible expropriation of property, moreover, became commonplace
in eastern Antioquia wherever Conservative paramilitary forces and gov-
ernment agents had been intensively deployed during la Violencia. Towns
such as San Roque, Puerto Nare, Nús, Puerto Triunfo, Amalfi, Puerto
Perales, Yolombó, Remedios, and Cisneros all reported such incidents in
the months following the military coup. Yet the military government had
by then already passed a decree to collect the arms that had been distrib-
uted to Conservatives under the previous civilian government.152
Liberal guerrillas who surrendered and gave up their arms when the
military came to power found it impossible to return to civilian life or to
their traditional labors as cowboys, public works employees, or miners.
The continued presence of the contrachusma ensured that they were mer-
cilessly harassed and forced to migrate. Pushed off lands and barred
from working for local mining concerns, several former guerrillas even-
tually left for the Llanos and joined the incipient leftist guerrilla groups
organizing there.153 Former guerrillas who remained in eastern Antio-
quia complained that the armed Conservative civilian forces even forbade
them from expressing open support for the government of General Rojas
Pinilla.154 The seeds of future armed rebellion that eventually came to
characterize the daily lives of eastern Antioquia’s inhabitants in the s
and s may be traced to the forcible migrations, expropriations, and
indiscriminate abuse exercised by the groups to which the state alienated
its ‘‘monopoly of force’’ in the s.
Peque, July 2001.
Townspeople carry the
days-old, mutilated
body of a local leader
in the aftermath of
a paramilitary attack.
3. Urabá and Western Antioquia

Urabá and certain areas of northwestern Antioquia had long been viewed
as valuable but undomesticated regions. Luxuriant, virgin nature mag-
nified and reinforced perceptions of unlimited economic promise and
political danger. Urabá was still largely a jungle in the s, a densely
vegetated, insalubrious lowland area wedged between the Pacific low-
land department of Chocó to the west and Córdoba (then part of the
department of Bolívar) to the east. Cattle ranges and wild pineapple
groves dotted the landscape as did forests of valuable hardwoods and
coconut trees. On its shark-infested shores were numerous coves where
contraband trade flourished. Despite encompassing a significant portion
of Antioquia’s physical area (, square kilometers), in  Urabá
boasted only four municipalities: Turbo, Chigorodó, Murindó, and Pava-
randocito. A mere , individuals, dispersed in widely scattered settle-
ments, constituted Urabá’s official population in the census of .1
Settlements that have since become municipalities such as Mutatá, Apar-
tadó, Necoclí, San Pedro de Urabá, and Arboletes were but small settle-
ments [caserios] or corregimientos and were subordinated to the jurisdic-
tional authority of either Turbo, Chigorodó, or Pavarandocito (see map
). Between one oasis of human concentration and another, enormous,
often unexplored distances prevailed.
Western Antioquia, in contrast, included fourteen municipalities,
covered a physical area half the size of Urabá (, square kilometers)
and boasted a population nearly eight times greater (,).2 Most of
western Antioquia was broken and steep, crisscrossed by streams and
rivers laced with gold, with land too rocky for most kinds of farming.
Large sections of it (Frontino, for instance), were still largely covered in
unexplored primary forest. With the exception of Cañasgordas where ex-
tensive coffee plantations existed, the municipalities of western Antio-
quia mainly produced cattle, sugar, gold, and commercial agricultural
commodities such as cotton (in Dabeiba) or cacao (in Sabanalarga).
Much of this production was concentrated in large estates (especially in
towns such as Frontino and Ituango) situated at considerable distance
 Blood and Fire

Map . Urabá and Western Antioquia. (Source: Instituto Geográfico Augustín
Codazzi)

from one another and from each town’s urban center, making these areas
difficult to patrol or monitor. The absence of paved roads or established
mule paths meant that most travel required an expert knowledge of hid-
den jungle routes or mountain passes or involved navigation. Canoes and
dugouts were the principal means of transportation in and out of the
labyrinthine channels, rivers, and streams sunk deep in the jungle and
known only to experienced guides and longtime local inhabitants.
Antioquia had lobbied hard to gain control of Urabá since the early
nineteenth century, but it was not until  after the ignominious loss of
Panama to the United States in  that Urabá officially came under the
province’s jurisdiction.3 Urabá gave Antioquia access to the Caribbean
Sea, a stake in the lucrative trade between Panama, Central America, the
 Urabá and Western Antioquia 

United States, and Colombia and a region rich in forest products, public
lands for colonization, and the possibility of extensive agricultural plan-
tations. When Conservative president Pedro Nel Ospina signed a contract
in  to begin construction of a road (the Carretera al Mar) linking
Medellín to Turbo with a series of interlocking trunk roads, Antioquia’s
dream of harnessing Urabá’s resources finally seemed close to becoming
a reality. Communication between the center of Antioquia and Urabá
continued to be nearly nonexistent, however, well into the s, despite
the initiation of road construction and government-sponsored coloniza-
tion efforts. Consequently, although Antioquia claimed to control Urabá,
on a day-to-day basis most public order matters concerning Urabá were
resolved in the departments of Bolívar or Chocó and not Antioquia. In
addition, almost all of the region’s production was directed north toward
Cartagena or Panama, not Antioquia proper. Antioquia could not even
be said to have colonized Urabá, for most of the area’s settlers were drawn
from Chocó, Bolívar, or the Caribbean coast.4 Matters were somewhat
less dramatic for western Antioquia, parts of which (Santa Fé de Antio-
quia, Buriticá, and San Jerónimo) had been linked to Antioquia since the
colonial period because of their importance as mining and commercial
centers. Still, the physical location of northwestern municipalities such
as Frontino, Dabeiba, Peque, and Ituango (at considerable distance from
Medellín and the seat of regional power) and the sparseness of human
settlement relative to their vast physical size made these areas isolated
and only unevenly integrated into the rest of Antioquia.
Violence altered the administrative arrangements that separated
Urabá from western Antioquia, or at least exposed the inconsistencies
inherent in this division. For the purposes of this book, I have opted to
give priority to the parameters of subregions as they were redefined by
violence rather than to the official administrative divisions established
by the state (see map  on page ). Indeed, at the time of la Violencia
the regional government itself ignored its own administrative jurisdic-
tions and, for public order purposes, considered towns such as Ituango
to be part of western Antioquia, and Dabeiba to be a continuation of
Urabá. There were good reasons for doing so. The guerrilla groups that
operated in these areas were often linked to each other. Moreover, the
Carretera al Mar emerged as a central locus of violence binding Urabá
and western Antioquia into a coherent public order area. And, finally, the
movement of goods, people, arms, and official forces during the period
 Blood and Fire

of la Violencia created a broad circuit of exchange from Urabá through

western Antioquia as far as Cañasgordas and back. For instance, cattle-
rustling, which constituted a significant aspect of the violence in this re-
gion, found one of its central outlets through Ituango and into the de-
partment of Bolívar (today the department of Córdoba), thus tying a
nominally northern municipality into the western/Urabá complex. For
these reasons I will consider the question of violence in western Antio-
quia and Urabá together.

The Early Years of Violence

The strategic implications of an absence of infrastructure and decades

of state indifference in Urabá and western Antioquia became painfully
evident in the days following Gaitán’s assassination. Buriticá went up in
flames while armed gaitanista sympathizers who were connected to the
mayor by blood and friendship occupied his office and threatened the
municipal telegrapher’s life when he dared to inform the regional au-
thorities that the entire town was in a state of open rebellion.5 Indeed,
weeks after regional authorities had successfully put down unrest in other
parts of Antioquia, Turbo and Peque remained up in arms.6 Gaitán’s death
brought to the surface a complex combination of partisan rage and deeply
held local resentment against a regional government that for decades
had ignored local needs, ruling through imposition without consulta-
tion or negotiation. Just four months after the uprisings affecting these
towns, the minister of hygiene noted with alarm that these and other
towns in western Antioquia and Urabá were among those where sani-
tary and health services were poorest or nonexistent and where even the
obligatory official doctor normally appointed by the state was absent.7
These towns were also characterized by communication and transporta-
tion infrastructure that was so antiquated, modest, or inadequate as to
create serious difficulties for both the region’s economic development
and the state’s ability to respond effectively to episodes of disorder. When
guerrillas cut the telegraph line that constituted Turbo’s only form of
communication with Medellín in late , for instance, more than a
month passed before the regional authorities were in a position to re-
establish radio contact with the area.8 During that time contraband arms
flowed freely from Central America and Panama to supply emergent in-
 Urabá and Western Antioquia 

surgents along the Carretera al Mar while the area’s few regionally ap-
pointed representatives were powerless to impede it.9
Like most of the rest of the department, however, western Antioquia
and Urabá were relatively quiet for almost a year after Gaitán’s assassina-
tion. But in May  during the heated pre-electoral season latent par-
tisan tensions resurfaced. Liberals and Conservatives accused each other
of engaging in acts of intimidation and harassment. Turbo’s Conserva-
tive minority insisted that the town’s Liberal majority had mercilessly at-
tacked the town’s Conservative mayor and requested the appointment of
a military mayor to provide protection during the upcoming June elec-
tions.10 In contrast, Turbo’s Liberal voters and the presidential delegates
sent to oversee local elections insisted that military troops were needed to
protect not Conservatives but Liberals from intimidation and fraud.11 As
in other Antioqueño areas where Liberals constituted an overwhelming
majority, however, Conservative intimidation had little effect on electoral
returns in either the west or Urabá. Liberals handily won the June 
elections while gaitanistas in Turbo, Dabeiba, Peque, and Frontino won
majorities in local town council elections.
Had the regional government accepted these Liberal victories and
understood that further pressure was unlikely to increase the number of
government supporters in Urabá and western Antioquia, it is possible
that violence such as that which became prevalent by late  might have
been averted. Instead, after a brief respite, the regional government re-
newed its campaign to intimidate and publicly humiliate members of the
opposition in preparation for the November presidential elections. An
unrepentant laureanista politician of the era confessed that he and other
Conservatives corralled local Liberals into the central plaza in towns such
as Cañasgordas and stripped them of their cédulas. The laureanistas then
smacked Liberals with the back of a machete to further ensure that local
returns for the Conservative candidate could be boosted without impedi-
ment.12 Such tactics often produced the desired outcome of inflating the
number of Conservative votes, although they fooled no one as to their
fraudulent nature. In Dabeiba, a town where the total number of Conser-
vative votes had never numbered more than  before , for instance,
, votes were cast for Laureano Gómez.
The use of public order forces and public employees for partisan pur-
poses during the  elections deepened divisions between local inhabi-
 Blood and Fire

tants in the west and between Liberals in these areas and the regional
authorities. Partisan differences overlapped with ethnic antagonisms in a
volatile combination that even affected different branches of the govern-
ment’s armed forces sent to maintain order in the region. The intersec-
tion of partisan and ethnic divisions among the different branches of the
government’s armed forces impeded from the very start the possibility of
developing a coherent public order policy in Urabá and western Antio-
quia. Army troops deployed to maintain public order in Urabá in 
and  (and later in the s), for instance, were generally drawn from
Cartagena’s Second Brigade, not the army based in Medellín. The long
distances involved in deploying troops to Urabá from Medellín made the
use of Antioqueño forces costly and inefficient as they had to be flown
in. Soldiers from the department of Bolívar, unlike many of the soldiers
recruited in Antioquia, tended to be black and to share the dominant
partisan affiliation of the majority of Urabá’s inhabitants; that is, they
were Liberals. Thus, in contrast to the hostility with which southwest-
ern inhabitants greeted public order forces brought into Antioquia from
other Colombian regions, Urabeños embraced soldiers from the depart-
ment of Bolívar with greater sympathy and support than they gave to
regional government forces staffed by Antioqueños. Urabá’s Conserva-
tive minority, most of whom were migrants and settlers from Antioquia
rather than Bolívar or Chocó, in contrast, suspected the troops brought
from Cartagena of collusion with the guerrillas and doubted their will-
ingness to defend Conservative lives. These perceptions fed rumors that
the army did not support the Conservative government. This was such
a commonly held belief in western Antioquia and Urabá by late 
that the commander of the Fourth Brigade in Medellín, Colonel Eduardo
Villamil, felt compelled to officially warn Governor Eduardo Berrío that
a subversive campaign was under way to try to turn the army against the
government.13
In contrast to the soldiers from the Liberal-dominated, black Carib-
bean coast sent to patrol Urabá, most national policemen deployed to
Urabá and western Antioquia were drawn from traditional, heavily in-
digenous departments such as Boyacá, Cundinamarca, and Huila where
the Conservative party had historically been strong. Conservatives in
western Antioquia and Urabá tended to prefer the police to the army be-
cause they perceived the former as more sympathetic to their partisan
interests, although at times the indigenous identity of many policemen
 Urabá and Western Antioquia 

also gave rise to frictions among race-conscious Antioqueño Conserva-

tive colonists. The majority Liberal population in the area, in turn, de-
spised the national police for both their ethnic origin and their perceived
partisan sympathies. To complicate matters further, Urabá’s municipal
policemen tended to be overwhelmingly Liberal and locally born and op-
posed to the national police who were nonlocal and Conservative. Cus-
toms officers, who were regionally appointed and among the few govern-
ment representatives present throughout western Antioquia and Urabá,
in turn, were resented by local Liberals both because the customs ad-
ministration had been conservatized after  and because smuggling
constituted a primary and fiercely defended local survival strategy. Long-
term cultural, ethnic, and regional differences (whether perceived or real)
thus intersected with partisan differences to create the sense of an align-
ment between specific sectors of the government’s forces and the popu-
lations over whom they were sent to exercise authority.
The departmental government at first attempted to dismiss as local
fantasy the perceptions of partisanship among the armed forces and de-
nied the influence of ethnic and interregional conflicts in the formulation
or implementation of public order policies in Antioquia. The regional
authorities’ own actions, however, belied such denials. When Turbo’s citi-
zens led an armed attack against regionally appointed customs agents
and injured one of them in November , authorities in Medellín did
not send in national policemen, their only available public order troops
at the time, because they anticipated the local citizenry would react vio-
lently.14 In early  when Governor Berrío entertained bids from Smith
and Wesson for the purchase of two thousand of their . ‘‘Specials’’ to
improve the region’s defenses, moreover, he privileged the distribution of
arms to forces whose cultural as well as partisan proclivities were likely to
garner local Conservative support. When Conservative committee mem-
bers in the western municipality of Dabeiba questioned the partisan and
ethnic loyalties of soldiers (blacks from Bolívar) and policemen (indige-
nous recruits from Cundinamarca and Boyacá) sent to maintain public
order in the town, the governor liberally distributed arms to locally con-
stituted, volunteer contrachusma groups and culturally acceptable police
forces instead. These were groups whose regional and partisan creden-
tials were beyond reproach.15 Indeed, the governor generally opted to
arm Antioqueño-born civilians and police rather than official forces in
which non-Antioqueños played an important role, regardless of the tac-
 Blood and Fire

tical needs of these forces.16 It sent the recently negotiated Smith and
Wessons to outfit the reorganized and conservatized customs and border
officers (resguardo de rentas), trustworthy policemen, and Conservative
civilians in select municipalities. None of the new revolvers were distrib-
uted to the army.17
Once lines were drawn between Liberals and the army on one side
and Conservatives, locally formed contrachusmas, and national police-
men, on the other (at least in the minds of many of the local inhabitants
and the regional authorities), violence escalated dramatically in western
Antioquia. In towns such as Caicedo, police agents brazenly destroyed
the pictures of Liberal council members and heroes that hung in the
town council office and replaced them with posters of Conservative Lau-
reano Gómez. Anyone who attempted to impede the police was arrested
and jailed. Policemen then colluded with civilian Conservative forces to
paper over the entrance of Liberal stores with the Conservative presi-
dent’s image, warning disgruntled Liberal shopkeepers that they would
have to pay a five-peso fine every time the poster was damaged.18 When
asked to investigate and condemn such actions, Jorge Salazar, the police
officer in charge of Antioquia’s department of criminal investigation, ex-
onerated the national policemen and civilian Conservatives responsible
for these violations. The police chief insisted that the army was only issu-
ing permits to bear arms to Liberals in western Antioquia, and that Con-
servatives therefore had a right to defend themselves from the opposition
by any means possible, including arming themselves and employing the
police for partisan purposes.19

The Organization of Armed Liberal Resistance

Guerrilla groups in Urabá and western Antioquia emerged shortly after

the June elections of . Their initial objective was to harass the Con-
servative government and topple it. They quickly found both consider-
able local support and a strategically ideal place from which to mount
concerted attacks against the state, its representatives (public employees,
police, army, customs officers, and so on), and the area’s small minority
of Conservative citizens. Although many guerrilla bands varying in size
from a dozen men to several hundred operated in the area between Turbo
in Urabá and Frontino in western Antioquia, several groups stand out
both for the range of their activities and their ability to effectively at-
 Urabá and Western Antioquia 

Map . Guerrilla operations and movement in Urabá and Western Antio-
quia (Source: Instituto Geográfico Augustín Codazzi; Archivo Privado del Señor
Gobernador de Antioquia, , vol. , ‘‘Asociación, instigación para delinquir.’’
Ministerio de Justicia, Juzgado  de Instrucción Criminal, Oficio #,  Feb.
)

tack and elude the authorities (see map ). One of these groups was led
by Sergio David and had its base in the valley between the Sinú, Sucio,
and San Jorge rivers in the department of Bolívar (in an area that is now
part of the department of Córdoba) immediately north of Antadó and
Ituango and east of Urabá.20 Satellite camps led by members of the David
clan along with members of the Arias, Cartagena, Duarte, Velásquez,
Higuita, Montoya, Pino, Romero, Serna, Torres, Tuberquia, and Usaga
families in turn spread the reach of the guerrillas into Chigorodó, Mutatá,
and along much of the Carretera al Mar. An offshoot of the David-led
group, under the immediate command of Captain Patricio Usaga, more-
 Blood and Fire

over, established control over the hamlet known as Caucheras near Mu-
tatá where important rubber plantations were located.21 A second guer-
rilla group numbering approximately one hundred men operated in El
Carmen, Chocó, across the border from the town of Bolívar, under the
leadership of Ramón Elías Calle, while a third guerrilla base was located
in the municipality of Frontino in a settlement known as ‘‘La Blanquita’’
within Murrí. In the base in La Blanquita, several dozen guerrillas under
the leadership of Aparicio Escobár manned a cattle-rustling operation
that extended as far as Peque, Juntas de Uramita, Frontino proper, and
Pabón (in Urrao).22 Other smaller groups or satellites of larger guerrilla
organizations also operated in the area between Sabanalarga and Barba-
coas (Captain Vidal Torres), in Cañasgordas (Salomon Marín), Antadó
(made up of former penal colony prisoners), and Vijagual/Apartadó (led
by an individual named Sandón).
The most important of the guerrilla camps operating in western
Antioquia, at least from the perspective of the regional government and
the armed forces, however, was that located in Camparusia (today the
municipality of Armenia) in the town of Dabeiba. The military com-
mander in charge of overseeing the maintenance of public order in Fron-
tino deemed Camparusia the ‘‘best organized guerrilla camp in Antio-
quia and the one with the greatest number of men prepared to enter
into combat with any one of the government’s forces.’’ 23 Here, Arturo
Rodríguez Osorio and Aníbal Pineda Torres (the latter a native of the
Urama/Uramita area in Dabeiba) built a nearly impenetrable headquar-
ters where several hundred men and many displaced Liberal families
sought refuge. Although under a separate command, the guerrillas based
in Camparusia were nominally linked to the guerrilla organization led
by Captain Franco in Urrao (in the upper southwest).
Unlike the single men who operated as Liberal guerrillas in eastern
Antioquia and who rarely drew their recruits from among the local popu-
lation, guerrillas based in the areas of Dabeiba and Chigorodó and along
the Sinú River in Bolívar participated in armed resistance as members of
locally based family clans, although these clans often straddled the bor-
der between Antioquia and Bolívar. In a list of men and women sought
by the government as suspected guerrillas,  men and women sharing
common surnames or singled out as members of distinct family groups
appear to be the principal source of disorder in the area.24 The guerrillas
or ‘‘chusma’’ (rabble), ‘‘bandoleros’’ (bandits), and ‘‘maleantes’’ (hooli-
 Urabá and Western Antioquia 

gans) as the government liked to refer to them, initially began their activi-
ties as defensive organizations. But their relative isolation, easy access to
contraband arms, and the general weakness of the regional government
and its forces quickly enabled the guerrillas to take the initiative against
the government rather than merely defend local Liberals from it. Within
two years of their initial appearance, a number of these armed bands ex-
panded their activities to primarily serve the political and personal inter-
ests of powerful individuals in the region. There gradually emerged a
flourishing informal market of stolen articles and animals (mules, horses,
cattle) of which the guerrillas formed an integral part.
In Dabeiba, for instance, guerrillas took part in what one observer
referred to as ‘‘all kinds of business deals’’ and received money and tac-
tical assistance from seven well-known merchants and Liberal political
leaders. The interaction and intersection between local civilians and the
guerrillas was so great that one of these patrons boasted publicly that ‘‘he
could stop the bandit rabble from killing any more people’’ with a single
order. Pronouncements such as these led the regional authorities to con-
clude that civilians and guerrillas were largely indistinguishable and that
the civilians were also ‘‘the heads of guerrilla bands.’’ 25
In return for merchandise, food, money, and logistical support,
the guerrillas guaranteed their supporters and patrons the ability to
travel freely through strategically contested areas. One of the guerril-
las’ Dabeiba supporters, for instance, had rented fourteen estates for a
pittance from Conservative property owners who ‘‘were forced to aban-
don them’’ precisely because they feared being killed by the guerrillas.
The guerrillas accompanied their patron whenever he needed to oversee
production ‘‘throughout the rural areas and saved him from having to
worry about losing his life.’’ For other sponsors the guerrillas acted as en-
forcers and hit men that eliminated or forced to flee unwanted squatters,
workers, or rivals. One landowner contracted the ‘‘bandit chief Patricio
Usaga so that he would come in the night and kill one of his peons, Justo
Giraldo, in his very own home.’’ The guerrillas also acted as private secu-
rity forces or bodyguards for local landowners. Several landowners ‘‘fre-
quently go out to their farms located at some distance from the county
seat [and] have been observed now several times in the company of eight
or more bandit riflemen [escopeteros].’’ 26 Even Liberal political leaders
in Medellín who were sympathetic to the guerrillas’ goal of fighting to
restore their party to power, and who occasionally offered the guerril-
 Blood and Fire

las advice and instructions, admitted privately that in some cases par-
tisan objectives among the guerrillas in western Antioquia rapidly gave
way to more base monetary objectives. Western guerrilla leaders such
as Salomon Marín were considered by even supportive Liberal political
leaders in Medellín to do ‘‘little more than rob people of their money.’’ 27

The Beginning of Liberal Guerrilla Activity

After months of reported arms smuggling from Panama through Urabá

and parts of the Chocó, Liberal guerrillas in western Antioquia led their
first major attack against Murrí, in the extreme northwestern corner of
the municipality of Frontino in June . They occupied the settlement
and stole cattle worth sixty thousand pesos.28 The regional government
had been warned in advance that the attack was imminent—just as the
governor was warned some three weeks before a similar attack took place
in Caucasia in the Bajo Cauca in August.29 It made no difference. The re-
gional authorities were simply in no position to respond to unrest on the
department’s borders. Years of neglect and failure to invest in infrastruc-
ture or the well-being of such areas could not be overcome at a moment’s
notice. The attack on Murrí spread to other areas of western Antioquia,
spurring local Conservative officials to report that a guerrilla army of
five hundred men was on the verge of taking control of the region.30 The
areas bordering the departments of Chocó to the west and Bolívar to the
northwest were places where little other than cattle was produced, but
they were crisscrossed by concealed paths and hideaways that made them
ideal refuges for men up in arms.
Unable to actually respond to the presence of Liberal guerrillas with
permanent forces of its own, the regional government attempted to starve
the guerrillas by blocking their supply routes and means of survival. But
like so many other policies embraced on different occasions by authorities
far removed from the day-to-day realities of the areas they nominally gov-
erned, food rationing penalized the local civilian population but did little
to deter the guerrillas who operated in their midst. Local landowners
sympathetic to the regional government were among the first to insist that
food rationing was an ineffectual means of dealing with armed insurrec-
tion in the region: ‘‘There is a scarcity of chocolate, lard, molasses loaves,
and other foodstuffs indispensable to the lives of honest workers in all
that area, but to think that the bandits will retreat for lack of food is an
 Urabá and Western Antioquia 

error, since there are immense amounts of available corn, . . . cattle, plan-
tains, yucas, etc., sufficient for the bandits to live on for many years.’’ 31
Indeed, while the guerrillas supplied themselves through theft and illicit
exchanges with smugglers, investors in the Chocó complained that food
shortages had raised prices for basic goods in the area to dangerous levels.
This created labor problems in the area’s mines and in turn had negative
repercussions on Chocó’s economy and state of public order.32
It was clear, moreover, that even in instances where merchants, land-
owners, and noncombatants did not necessarily support the idea of
taking up arms against the government, there often was little incentive to
deny the guerrillas material support, especially if there existed the pos-
sibility of considerable profit. The second sergeant inspector stationed
in Uramita, who had been accused of collaborating with local Liberals
and guerrillas but adamantly denied this to his superiors, gave a persua-
sive explanation of why attempts to eliminate the guerrillas with poli-
cies such as food rationing were doomed to failure. Uramita’s merchants,
the inspector insisted, had at first abided by the government’s decree
forbidding the sale of food or merchandise to those suspected of being
guerrillas or guerrilla supporters. But the guerrillas had simply shifted
their business to Urama, Juntas de Uramita, and Peque where merchants
were perfectly willing to sell them goods instead. As the inspector rightly
concluded, ‘‘the rabble-rousers [chusmeros] can always find provisions in
other areas,’’ especially since there were no government forces present to
enforce the prohibition against sales to the guerrillas. And the inspector
continued, ‘‘in this corregimiento we could say that all the merchants are
purveyors of goods to the bandits since all of them sell their goods with-
out asking questions. All they care about is making money.’’ 33 The state’s
inability to enforce its own policies and laws, in other words, destroyed
any incentive for even its most fervent supporters to abide by them. But
it also seems true that for many locals partisan differences were simply
not compelling enough to stand in the way of doing business with the
opposition.
Shortly after the attack on Murrí, three hundred men armed with
shotguns, revolvers, and machetes attacked a troop of soldiers in Tucura
(today Córdoba) several miles north of Playones in Dabeiba on July .
The guerrillas wounded four soldiers, committed a series of murders
and robberies, and then fled south to attack Playones proper.34 Attacks
occurred simultaneously in Urrao and La Camara in the municipality
 Blood and Fire

of Salgar, where guerrillas forced Conservative peasants and workers

to leave their haciendas.35 These events—especially the occupation of
valuable estates—obliged the regional government to acknowledge the
severity of the threat to state control in western Antioquia and Urabá
and prompted it to deploy soldiers and additional policemen to the area.
On July , Colonel Eduardo Villamil, the commander of the Fourth Bri-
gade in Medellín, ordered the sub-lieutenant and commander of the local
barracks in Dabeiba to permanently patrol the road between Mutatá,
Dabeiba, and Chigorodó with two cars. He also sent two lower-ranking
officers and thirty national policemen from the Antioquia division as re-
inforcements. Anticipating that the deployment of police would be re-
sisted by both some soldiers and local Liberals, Villamil admonished his
subordinates to form mixed police and army patrols and to ensure that
the police agents be treated with ‘‘respect and kindness by the [army]
troops.’’ 36 In the further hope of dispelling rumors that the army was co-
operating with the guerrillas, moreover, the colonel insisted that his men
penetrate deep into the jungle in pursuit of the bandits and that these
bandits be offered no mercy.
A few days later, government officials in Dabeiba were able to report
their first minor victory against the Liberal guerrillas when they cap-
tured six and detained nine other people suspected of providing supplies
from the settlement of ‘‘la Montañita,’’ located just outside the town.37
In other parts of the west, however, in a pattern that would become
commonplace, frustrated public officers who were unable to locate the
guerrillas wreaked vengeance on the local civilian population instead.
In Caicedo, for instance, members of the national police and Conserva-
tive volunteers attacked ten inhabitants with machetes and guns in Gua-
sabra, and justified the attack by arguing that this was one of the many
corregimientos suspected of giving aid to the guerrillas.38 Excesses such
as these prompted bickering and mutual accusations between police and
army officials. Only a day after the report by Caicedo’s military com-
mander of police abuses against civilians, Colonel Villamil complained to
Antioquia’s Governor Berrío that Jorge Salazar, the police chief in charge
of Antioquia’s department of criminal investigation, had overstepped
his powers and violated several national decrees. Salazar in turn ac-
cused Villamil of being disloyal and a secret Liberal sympathizer.39 Mean-
while as the police and army exchanged barbs, towns such as Ituango
and Caucasia on the border with Bolívar worried aloud about the ab-
 Urabá and Western Antioquia 

sence of any government officers to defend them from imminent guerrilla

attack.40
The expected attack came on August  when guerrillas simulta-
neously struck settlements in Urabá north of Turbo, Dabeiba, Caucasia
in the Bajo Cauca, and Urrao. News of the assault first reached the gover-
nor’s ears not through an official channel, but via the secretary of the re-
gional Conservative party directorate who informed him that numerous
Conservatives had been assassinated in Urabá and that Dabeiba’s parish
priest had been murdered by Liberal guerrillas. Later that same day, the
minister of government, Domingo Saresty, forwarded a list to the gov-
ernor of all those killed in the town. The minister complained that the
local mayor and his secretary and the soldiers stationed in the region by
the Fourth Brigade had physically impeded national police troops under
the command of Major Arturo Velásquez from sending out a group of
armed civilian volunteers and policemen to track down the guerrillas,
thus ‘‘allowing the assassins to go free.’’ 41 These events prompted the min-
ister to rebuke Governor Berrío and to make patent the minister’s own
lack of faith in the army as a reliable public order force. The minister
openly sided with Dabeiba’s Conservatives when they declared that ‘‘the
army Lieutenant Quintero is a social menace as far as we are concerned,
for he has become one with the Liberals who are his fellow party mem-
bers.’’ 42 From this point on, Dabeiba became the site of constant conflict
and a major center of contrachusma organization. Dabeiba formed a kind
of gateway between Urabá proper and western Antioquia, the last pos-
sible geographic staging ground of Conservative paramilitary strength, as
it was impossible to rustle up enough Conservatives in Urabá. The town
consequently became a major battlefield and suffered Antioquia’s great-
est number of officially registered deaths as a result of violence between
 and . In three and a half years, Dabeiba lost  inhabitants or
 percent of the town’s  population of ,.43
As bad as matters were in Dabeiba, the situation of public order was
far worse in El Carmelo and Turbo in Urabá. A week went by before
the regional government even acknowledged that the guerrilla invasion
had taken place. By August , when government forces had still not ar-
rived to take stock of the situation, local observers tentatively reported
a death toll of more than fifty Conservatives. When troops finally made
it to Turbo on August , eleven days after the initial attack, ‘‘one thou-
sand’’ armed men from the department of Bolívar were reported to have
 Blood and Fire

attacked El Carmelo once more.44 The guerrillas looted the ‘‘principal

stores,’’ assaulted the customs offices, and destroyed the police inspec-
tor’s office. Two policemen and several customs officers were wounded
and one policeman was killed. The number of civilian deaths was difficult
to calculate because the bodies of the dead had been thrown into the river
where they were torn apart by sharks. The police inspector reporting on
the attack pleaded desperately for the governor to send ‘‘a strong, perma-
nent contingent of the army’’ to counter the area’s civilian authorities, all
of whom ‘‘are enemies of the current government’’ who would take ‘‘any
opportunity to attack the regional authorities.’’ 45
The only troops available for deployment to Urabá, however, were
members of Cartagena’s marine infantry. Turbo’s Conservatives thought
of costeños as lazy and inept and were convinced they ‘‘would arrive, get
drunk, and do nothing.’’ Whether the infantry did ‘‘nothing’’ is not clear,
but it is clear that many of its men disagreed with Antioquia’s and the
Conservative government’s actions against Urabá’s Liberal citizenry. This
was abundantly corroborated by a letter from one of the infantry’s men
who was captured several months later by the national police. In it an
officer gave testimony of the deep sense of camaraderie that bound the
coastal forces to Urabá’s population. He expressed a sense of shame be-
cause ‘‘I and my comrades make up part of the public administration,
that is, of this bloodied government, [even though] we did not come to
commit assaults against those victims who are only persecuted because
they are Liberals. We came to establish safeguards.’’ Posted for duty in
the Port of San Juan de Urabá, the officer made evident the weight of
racial and ethnic considerations in public order matters: ‘‘The port’s in-
habitants are dark-skinned [morenos], but they are noble and respect-
ful of the authorities. We were well received in this settlement, all the
settlers loved us because they were able to understand that we were there
to guarantee the safety of their property, their lives, [and] their honor
and not intent upon assaulting, robbing, or persecuting them, unlike the
preceding blue [Conservative] authorities who hid behind their military
uniforms to assault their houses, rape their daughters, and murder their
fathers. . . .’’ 46 Sentiments such as these on the part of forces sent from
the Caribbean coast prompted Turbo’s Conservatives to remind the gov-
ernor of the area’s strategic location between the department of Bolívar
and Panama and to request ‘‘thousands of soldiers’’ commanded by high-
ranking officers whose political loyalty was beyond question.47 But it was
 Urabá and Western Antioquia 

not until August  that Medellín’s Fourth Brigade was able to spare addi-
tional men to send to Urabá. On that date Colonel Villamil sent six planes
full of army personnel for public order maintenance to Turbo.48
In the very midst of the repeated guerrilla attacks that took place in
numerous peripheral towns along the region’s northern borders in Au-
gust , Antioquia’s governor’s office changed hands. Eduardo Berrío
González, who preferred arming Conservative civilians as volunteer
policemen and deploying national policemen to maintain public order
and bring recalcitrant Liberal areas to heel—regardless of the effect
of such policies on local populations—was replaced by Braulio Henao
Mejía. In contrast to Berrío, newly appointed Governor Henao Mejía was
a retiring man who maintained cordial relations with the military and
was wary of, if not explicitly opposed to, arming Conservative civilians
or forming contrachusmas. The new governor was also supportive of the
police but was sensitive to deploying them in places where they were
hated so much as to catalyze popular rebellion. While former governor
Berrío blamed the spread of Liberal violence on the lack of cooperation
provided by the commander of the Fourth Brigade whose removal he bel-
ligerently reminded the president, ‘‘I have been begging the national gov-
ernment to undertake for more than six months,’’ Governor Henao Mejía
attributed the ineffectuality of the region’s public order strategy to the
sheer insufficiency of government troops, not the disorderly or partisan
affinities of army officers.49 A scant two days after taking office on Au-
gust , Governor Henao informed the president that Antioquia simply
did not have enough men to suppress the numerous sites of ‘‘sedition’’ in
the department. The local army battalion had even been forced to bor-
row troops from Pereira and Manizales in the department of Caldas to
the south.50 Moreover, far from suspecting the army’s loyalty as Berrío
or Minister Domingo Saresty had, Governor Henao entrusted the diffi-
cult and potentially sensitive task of conducting an in-depth analysis of
the reasons for widespread unrest in Urabá to a military officer, Colonel
Luis Abadía, who was given broad powers to deal as he saw best with the
situation of public order in the northwest.

The State of Public Order in Urabá

If what the regional government hoped for was a glimmer of local iden-
tification or sympathy between Urabá’s inhabitants and Antioquia’s re-
 Blood and Fire

gional government, the colonel’s findings were disheartening. Colonel

Abadía’s report covered five crucial areas in Urabá: San Juan de Urabá,
Arboletes and Damaquiel on the Caribbean coast near the border with
the department of Bolívar, Necoclí near the Gulf of Urabá, and Turbo,
the northernmost point of the Carretera al Mar (the road to the sea, con-
necting Medellín with Turbo). The army officer first noted to what degree
economic interest and class played a role in the perception of guerrilla
threat in the region. The region’s wealthiest citizens, he noted with ill-
disguised irony, were the most likely to complain of the threat of a guer-
rilla attack. Large property owners ‘‘sensed the presence of bandits every-
where.’’ One individual who had ‘‘properties and is a rich man’’ insisted
that the chusma were stationed near Turbo, while another insisted they
‘‘were expected to attack on the same day in San Juan and Arboletes’’ (an
extremely far distance away).51
Colonel Abadía was convinced that nervous landowners were mainly
moved to request the immediate deployment of military troops by con-
cern for their large capital investments and estates. Local haciendas were
so large, the officer noted, that they swallowed up all the land between
certain towns or villages. This forced travelers to cross the haciendas
in order to move between settlements.52 Indeed, between  and 
twenty-four concessions of public lands were made in Turbo in which
the average lot amounted to , hectares (approximately , acres),
while the average size of forest concessions in Chigorodó between 
and  was , hectares (, acres).53 The colonel made it clear
that fears of economic extortion, rather than partisan-inspired conflict,
spurred the numerous rumors of impending disorder in Urabá. Land-
owners belonging to both parties, for instance, repeatedly alluded to
having been ‘‘threatened by Hincapie’s band,’’ which wandered from
hacienda to hacienda soliciting ammunition and supplies or extorting
mayordomos (estate managers) and workers.
Turning his attention next to the town of Damaquiel, Colonel Aba-
día noted that one hundred armed civilians had turned up at the port
when news of the arrival of government troops spread throughout the
town. Thinking that the police had been sent to monitor public order in
the area, Damaquiel’s citizenry mobilized to forcibly turn away the gov-
ernor’s agents.54 It was only when they were reassured that the governor
had sent the army and not the police that people dispersed and put away
their arms. The townspeople’s attitude convinced the colonel that many
 Urabá and Western Antioquia 

of Urabá’s inhabitants felt alienated and suspicious of Antioquia and that

it would be foolhardy for the regional government to insist on sending
national policemen to patrol the area. The colonel’s observations about
the inhabitants and conditions present in the settlement of Damaquiel
also reveal the degree to which hierarchies of cultural difference imbued
and shaped official assessments of local conditions and violence. Dama-
quiel was described as ‘‘painfully abandoned. . . . There is no one who
may act as the representative of authority. . . . There is the greatest indo-
lence. . . . The men don’t work because of the state of abandonment [and]
the hostile and suspicious climate. . . . I found all of them stretched out
on their hammocks, the grimace of a defeated race tormented by laziness
reflected in their faces.’’ The same was true of San Juan de Urabá, a town
in ruins with ‘‘multitudes of abandoned houses . . . in the most horrify-
ing state of filth.’’ 55 Moreover, those who had not fled the town after the
guerrilla attack in August hid from the authorities.56
Finally, the colonel concluded his inquiry by explaining why, purely
from the perspective of self-interest, it behooved the regional authori-
ties to take immediate action to remedy Urabá’s current state of pub-
lic disorder: ‘‘[Urabá] represents for the national economy one of the
most promising sources of future development; and if it attaches itself
to Antioquia, the economic features represented by such faraway cor-
ners of the department will be of critical and of incalculable value in
the future interests of Antioquia.’’ 57 Colonel Abadía’s report exposed the
colonial fantasies of extraction and wealth that had historically under-
pinned the regional government’s attitude toward Urabá. He suggested
that Urabá’s integration into Antioquia and the department’s control of
the peninsula would remain tenuous so long as the area was economically
oriented toward the Caribbean rather than Antioquia. The area’s rice,
lard, corn, and plantains (produced in ‘‘large quantities’’) were shipped
to Cartagena and other ports on the Caribbean, while bananas and coco-
nuts had historically been shipped to Panama rather than used to ‘‘supply
the scarce markets of Antioquia.’’ 58 In other words, as long as few of
Urabá’s ‘‘economic features’’ were oriented toward Antioquia, the army
officer seemed to suggest, Antioquia’s dream of political control would
remain unfulfilled. There was no official motorboat to patrol the region.
Air service was provided haphazardly at the whim of Avianca. Commu-
nication from Urabá was accomplished by couriers who traveled by foot
or canoe because there were no radio or telegraph lines to link the region
 Blood and Fire

to Medellín, and the only government authorities present were customs

agents and a sporadic mayor or police inspector.59 From the colonel’s per-
spective, then, the problem of violence in Urabá was at least as much a
result of cultural, political, and economic differences as it was of parti-
san ones.
The only hope for Urabá’s future integration into Antioquia and iden-
tification with its values and interests was the completion of the road link-
ing Medellín to Urabá. The colonel was convinced that only the Carretera
al Mar could begin to solve Urabá’s problems and introduce ‘‘progress
and economic resurgence.’’ More importantly, the road would also guar-
antee the assertion of Antioquia’s political control and cultural values.
The army officer argued insistently for the ‘‘need to extend to these areas
the moral and civic aspects of the Antioqueño people [pueblo], which will
ensure the redemption of many men who in these distant regions live a
primitive, morally and mentally lax life.’’ However, before any change in
Urabá’s attitude could be expected to occur, Colonel Abadía warned his
regional superiors and the department’s civilian authorities that the de-
partment would have to prove that ‘‘all of Antioquia is truly attentive to
them.’’ 60 This was unlikely to occur in the immediate future. Indeed, but
a month after the colonel concluded his extended inquiry into the public
order situation in Urabá, the region was once more awash in rumors of
arms smuggling and imminent guerrilla activity. Guerrillas, moreover,
had established a stranglehold over the area stretching from Turbo to
Puerto Abaldía in the Gulf of Urabá.61
The problem of open insurrection in Urabá ceased to be a strictly re-
gional matter of public order concern by late . By November, inci-
dents of extensive arms smuggling, attacks on private properties, and
assassinations had spilled across Antioquia’s borders into Chocó and
Bolívar and prompted the governors and public order forces of these
two departments to express their concern regarding Antioquia’s seem-
ing inability to police its own frontiers and internal affairs. The trouble
was that Urabá was not and had never been a strictly Antioqueño area;
its settlers, trade, investments, police forces, and resources easily bled
across multiple administrative and physical borders. It had until very re-
cently been little more than a vast commons of sparsely populated jungle
where the state, the law, and its representatives—whether regional or
national—were rarely felt. A small army of customs officers, policemen,
coast guards, and soldiers with modern forms of communication and
 Urabá and Western Antioquia 

transportation would have been needed to submit Urabá’s many hidden

coves, rivers, and dense forests to the effective surveillance of the regional
government. In , however, Antioquia did not even have a motorboat
with which to patrol Urabá’s coast, much less a central radio station for
Turbo, six boat stations, watchposts along the region’s roads and rivers,
or two planes, items that in December the governor of Cartagena warned
Braulio Henao Mejía were absolutely necessary to inhibit the massive
contraband of arms through the area and to impede the spread of guer-
rilla forces.62
Indeed, joint efforts by the departmental police forces of Bolívar,
Antioquia, and Magdalena to control the trade in illicit arms proved
futile. This was not because the smugglers were aided by ‘‘Communist
ideologies emanating from Russia’’ as the police liked to insist, but be-
cause the government’s forces were poorly equipped and lacked almost
any logistical support for their endeavor. In contrast, most of the com-
mercial boats owned by merchants based in Cartagena who plied their
goods up and down the Arquía River and the banana boats traveling back
and forth along the Atrato River were armed with machine guns and
rifles and had two-way radios.63 The government, in contrast, lacked cus-
toms officials to stop and search these boats, leaving them free to trade
in weapons intended to supply guerrillas employed or hidden along the
Carretera al Mar. Reports of just such illicit traffic in Sautatá on the Su-
cio River—where Abuchar Brothers, a Cartagena-based merchant house,
had the region’s largest sawmill and logging operation—were frequent.64
In several cases the regional government received detailed inventories of
exactly what and how many arms were passing through Urabá and west-
ern Antioquia, the routes through which these were shipped, and who
was trading, but did absolutely nothing in response.65
The combined effect of the flow of arms to guerrillas and the state’s
poor performance in western Antioquia and Urabá made December 
the most turbulent month in the region since the multiple incidents of
guerrilla attack in August. Armed disturbances occurred in Turbo on
December , while Conservatives in Cañasgordas complained that more
than two hundred farms had been invaded by guerrillas and that the
two hundred soldiers sent to patrol the area by Medellín’s military com-
mander refused to enter the invaded areas to reclaim them.66 By mid-
month the Caja de Crédito Agrario, Industrial y Minero reported that
guerrillas were also occupying valuable agricultural areas and leading
 Blood and Fire

invasions of public lands reserved by the regional government for rub-

ber development in Villa Arteaga and Pavarandocito.67 The installations
and workers on the experimental rubber plantation in Villa Arteaga—
which was jointly operated by a U.S.-Colombian consortium—had been
attacked several times in the previous eight months and sustained losses
of sixty thousand dollars. Economic loss was accompanied, moreover, by
fears of worker starvation and revolt as the regional government’s food-
rationing policy took its toll on the civilian population.68 Although they
had already sunk . million into the rubber plantation and were on the
verge of producing their first crop, the investors were considering pulling
out of the project because government protection was so unreliable and
sporadic. Terrified lest such threats become reality, the Caja de Crédito
Agrario offered to pay out of its own pocket the expenses of fifty soldiers
to be stationed in the area and also offered to fund, in conjunction with
the Red Cross, medical and humanitarian aid and services for the area’s
settlers (colonos).69 By December , verbal threats and boletas (extortion
notes) were circulating along the Carretera al Mar and as far south as
Dabeiba. The regional government’s own officers admitted that it was im-
possible to control the ‘‘bandits’’ or to lead a surprise attack against them
because everyone in the town—whether from conviction of simply out
of fear—informed the guerrillas of government operations before they
occurred.70

The Militarization of Urabá

The severity of unrest in western Antioquia and Urabá and the repeated
defeat and ineffectual performance of official forces prompted regional
authorities to once more commission government agents and members
of the armed forces to analyze the situation of public order in Urabá.
A specially charged visitador (visitador encargado) provided the first of
these commissioned reports in January . The visitador concluded that
the towns most affected by violence within the larger western region were
Dabeiba, Frontino, Cañasgordas, and Peque, and he attributed the state
of constant disorder to the government’s inability to impede collusion
between the Liberal civilian population in these towns and the men up
in arms.71 A detailed report issued around the same time by a prominent
Conservative officer in Caicedo seemed to confirm the visitador’s conclu-
sions. But, the Caicedo official added that moderate Conservatives who
 Urabá and Western Antioquia 

refused to break off relations with the opposition in areas where their

party was weakly represented, the state was largely absent, and the oppo-
sition dominated local affairs also posed an obstacle to the establishment
of regional Conservative control over the area.72
Caicedo had a Liberal majority of  voters to  Conservative
voters, a situation that seriously compromised the influence of the Con-
servatives on local affairs. In addition, the Conservative mayor con-
tributed to the local Conservatives’ continued marginality by being an
‘‘enemy of conflict and a friend of the peaceful life.’’ The mayor was on
awkward terms with the newly installed and openly reactionary local
Conservative committee and the local priest. He was also ‘‘economically
dependent on Liberals since whenever he buys goods for cash or on credit
he does so in the stores and grain dealerships belonging to Liberals, since
these are the best. This creates, at the very least, a material obstacle to
[conservatizing] administrative affairs.’’ 73 Caicedo’s analyst touched on
a sensitive but important problem for extremist regional authorities in
his report: the regional government enjoyed too little credibility even
among its own sympathizers to risk breaking off ties with the opposition
or engaging in openly defiant acts that might trigger additional guerrilla
attacks. Who would come to their defense were they to do so? Ill-paid,
poorly trained, and scarce public order forces, nearly as much at odds
with each other as they were with the individuals they were sent to com-
bat? Party officials who rarely ventured beyond the confines of Medellín?
While extremist sectors of the regional government insistently pro-
voked and harassed the opposition in distant areas, they rarely consid-
ered the cost of such a policy to local citizens or took into account the
government’s inability to effectively respond to local Liberal reactions to
such harassment. Regional authorities also failed to recognize the degree
to which the interests of their party’s local membership were intertwined
with those of the opposition, whether because of kinship, business, or
other long-term associations. For many local inhabitants the demands
of extremist sectors within the national or regional Conservative party
were unrealistic or simply had little bearing on local affairs. Competing
loyalties produced a contradictory response among local Conservatives.
Unable to win the support of moderates, local Conservative extremists
increasingly preferred paramilitary or unofficial solutions rather than the
deployment of either the army or the police to confront guerrilla-led at-
tacks. Moderate Conservatives, meanwhile, distanced themselves from
 Blood and Fire

their party and the extremists in it, limiting themselves to unsuccessfully

pressing the government to avoid promoting partisan conflict or, in some
cases, opting to surreptitiously collude with members of the opposition
against regional government officers.
Conflict labeled as strictly partisan simply failed to address the com-
plexities of local strife or the possible existence of long-standing relation-
ships that cut across party lines. Indeed, the bewildered tone in which the
governor of Antioquia reported on the composition of the armed bands
operating in western Antioquia to the minister of foreign relations in
Bogotá made apparent an official tendency to reduce violence to a series
of binary oppositions, and this was one of the central obstacles to de-
vising an effective public order policy in western Antioquia. In response
to the president’s theory that violence in Urabá and western Antioquia
was organized and led by members of Medellín’s Liberal directorate and
that the guerrillas were all former gaitanistas, Antioquia’s governor was
forced to acknowledge that no such clear-cut conclusion could be drawn
about the participants in the area’s unrest.74 Instead, the governor rue-
fully admitted that he had received numerous reports that violence was
increasingly motivated by economic rather than purely partisan factors
and that among the participants in it were numerous Conservatives—
acting in their own armed bands or in groups that included both Liberals
and Conservatives.75
The difficulty of identifying exactly who was up in arms and why was
a point repeatedly alluded to by a number of the analysts charged by the
regional government with diagnosing the sources of unrest in the west.
Detectives employed by the office of internal security, for instance, in-
sisted that there were no separate guerrilla forces per se, but rather that
road workers and guerrillas were one and the same (exactly as military
analysts suspected was true of the Bajo Cauca region).76 Guerrilla activity
in the name of a party and union militancy was conflated. This opinion
was seconded by a public order report filed by the military in February
in which road workers were accused of providing the guerrillas with in-
formation, food, and clandestine transportation of arms through Urabá.
It proved difficult to separate the interests of organized labor from par-
tisan interests because the Conservative authorities had made it a point
to conflate workers, Communists, and Liberals, and because state em-
ployees were necessarily the products of a partisan patronage system and
therefore tended to belong to whatever party happened to be in power
 Urabá and Western Antioquia 

when they were hired. New road personnel sent to replace those sus-
pected of collaborating with the guerrillas, for instance, were killed by
old road workers (hired during an earlier period of Liberal rule) lest the
former collaborate with or betray their work mates to the army.77 Road
workers employed on the Carretera al Mar clearly used threats such as
those directed at potential Conservative personnel appointees as a means
of protecting themselves from both dismissal and from partisan discrimi-
nation and harassment.
The national forest inspector also suggested that partisan conflict
thinly disguised the struggles between competing economic interests in
the region. The inspector reported that he had gone on a trip to Chi-
gorodó to survey wood concessions leased to several prominent Antio-
queño Liberals who owned an important logging and sawmill operation
in the area. In the course of his inquiry he found ‘‘that local opinion is
unanimous in asserting that the company has fomented, given aid to, and
sustained the reigning state of insecurity in the area in order to monopo-
lize control of the forest products which abound in the region.’’ 78 Indeed,
the governor had already received earlier reports linking the Surambay
sawmill with the contraband of arms through Urabá.79 It is difficult to
know, however, whether the inspector’s accusations were motivated by
the desire to benefit a group of equally prominent and well-connected
former Conservative officeholders who had also formed a consortium to
exploit a forest concession for wood products in Urabá or by the genu-
ine protests against privatization expressed by locals.80 The privatization
of extraction from what had long been construed as public lands was
clearly an issue of considerable conflict and concern among local resi-
dents and had been for several decades.81 Either way, more was at stake
in the armed struggles taking place in Urabá than the question of which
party was to control Colombia’s fortunes. Reports that linked violence
to the emergence of an important, informal market for the sale of stolen
wood and cattle in which members of both parties were said to take part
further reinforce the impression that in some cases so-called partisan vio-
lence masked conflicts more appropriately construed as economic and
personal.82
Analysts sent to report on western Antioquia and Urabá all agreed
that the existence of illicit economic networks in which members of both
parties colluded and benefited was dependent upon a continued state
of public disorder. Analysts also believed that local collaboration with
 Blood and Fire

the guerrillas and the reluctance of some Conservatives to participate

in the wholesale oppression of the opposition undermined the govern-
ment’s ability to impose order in the area. But the principal determinant
of continued unrest in western Antioquia was thought to lie elsewhere.
The government’s informants concluded that the central impediment to
the defeat of armed insurgents was the absence of a collective consen-
sus regarding the legitimacy of using force to impose partisan hegemony.
This lack of consensus had created an intractable problem: the persis-
tent refusal of members of the government’s party and the armed forces
to cooperate with one another or to take part in what was perceived as
partisan-motivated violence.
In late January, Turbo’s civilian mayor complained that the soldiers
stationed to defend the town refused to cooperate with the Conserva-
tive civilian authorities or the police to search for and combat the ‘‘ban-
doleros.’’ 83 Two weeks later the governor deployed twenty-nine national
policemen to provide backup in Turbo, but Colonel Luis Abadía (the
newly appointed commander of Medellín’s Fourth Brigade) notified the
governor that the police had refused to be put under the command of
the army (despite the governor’s orders). Indeed, the police had pub-
licly declared they would not cooperate with the army under any cir-
cumstances. The regional head of the national police in Antioquia, Major
González, was forced to fly out on a military plane to deal with his men.84
The conflict between soldiers and policemen was temporarily resolved
by entrusting them with separate jurisdictions. The police were assigned
to defend Turbo and the Vijagual region, while the military was granted
jurisdiction over everything else.85
Ethnic and partisan tensions between different branches and divisions
of the armed forces such as those suggested by Major González contrib-
uted to the absence of a united official front against armed insurgence.
In January, the major informed the governor of the need to relieve the
troops from Cauca and Huila who were stationed in the area because they
‘‘adversely affect [the] behavior of Antioquia’s troops and are totally un-
disciplined.’’ 86 The scion of a Conservative Medellín industrialist family
who owned a large hacienda in Chigorodó seconded the major’s obser-
vation. The Conservative hacienda owner warned the governor that so
long as responsibility for maintaining public order in Urabá lay in the
hands of national policemen who were ‘‘cundiboyacense’’ (that is, of in-
digenous descent) little hope existed of winning local civilian trust or
 Urabá and Western Antioquia 

eradicating the guerrillas.87 In a confidential letter to Laureano Gómez

in early , Governor Henao, moreover, attributed the guerrillas’ re-
peated defeat of government forces in western Antioquia to the ‘‘lamen-
table human material of many of the guardians of order’’ and implied
that this was partly a result of their ethnic and regional identity (that is,
they were non-Antioqueños and of indigenous descent).88 Local Conser-
vatives, in contrast, blamed the ‘‘Liberal army from the Caribbean coast
[costeños]’’ sent to patrol the Carretera al Mar for an increase in guer-
rilla directed assassinations, robberies, and extortion and accused these
of openly supporting the insurgents by selling them saltpeter and arms.89
Disputes over the relationship between ethnic identity and the ability
to maintain public order were in the majority of cases but an expression
of Antioqueño xenophobia and an excuse for what were really conflicts
about jurisdiction and prerogative. The regional authorities resented the
intrusion of forces appointed from outside Antioquia in the maintenance
of public order because they could not so easily subordinate them to a
dominant regional ideology or value system. But paradoxically, just as
the regional authorities regarded such forces as a threat to their power,
localities also resented the intrusion of regionally appointed forces who
were perceived to be alien to or not so easily subordinated by local machi-
nations. In late January the Conservative town of Anzá reported that two
hundred guerrillas had attacked one of the town’s corregimientos, which
was protected only by six departmental police agents, a single municipal
policeman, the police inspector, and three customs officers. The town had
been left defenseless because racism had prompted it to refuse to allow the
deployment of soldiers from the Caribbean coast in the area.90 National
policemen from Boyacá and Cundinamarca stationed in Dabeiba, more-
over, were targeted by guerrillas operating along the Carretera al Mar
who knew that black costeño soldiers (from Bolívar) stationed in garri-
sons in Chigorodó and Mutatá would make no effort to aid or rescue the
‘‘cundiboyacense’’ policemen.91
In the wake of events in early  and after having considered the
various reports analyzing the sources of continued and escalating vio-
lence in Urabá and western Antioquia, Governor Henao Mejía took a
series of steps to assert regional authority over an area he and the central
government feared was rapidly escaping the government’s grasp.92 First,
the governor set up a meeting between Colonel Abadía of the Fourth
Brigade, Major González of the department of security, the chief of the
 Blood and Fire

regional office of the national roads department, Antioquia’s secretary

of public works, and Gregorio Mejía, the Liberal contractor in charge
of hiring workers on the Carretera al Mar. These men agreed to station
one hundred members of the army corps of engineers on the Carretera
al Mar where ‘‘bandits . . . harass workers and private citizens, inter-
rupt transportation, rob settlers, and maintain a climate of terror.’’ 93 The
regional government also decided to once more rearrange the deploy-
ment of troops in the region. Cañasgordas, Peque, Ituango, Dabeiba, and
Frontino (western towns) were put under the control of the police while
Mutatá, Caucheras, Chigorodó, and Riogrande (towns in Urabá) were
assigned to the military.
Regional authorities justified their decision to assign policemen and
soldiers to different geographic areas by arguing that the police were
posted to areas where there were a greater number of inhabitants and
where incidents of violence were more related to ‘‘police issues,’’ while
the army was sent to patrol areas involving ‘‘technical’’ matters (such as
the construction of the road).94 In reality, however, the reasons for assign-
ing police and army to different public order sites and to which specific
areas followed other concerns. Both the central government and Antio-
quia’s regional authorities worried that worker disturbances might im-
peril the completion of the road. Indeed, in February the governor had
confessed that he feared a ‘‘massive revolt by workers.’’ 95 It was the fear
of alienating the workers (almost all of whom were Liberal) and jeopar-
dizing Antioquia’s control and access to important natural resources and
trade that ultimately determined the assignment of the army to patrol this
zone. In this sense, road workers in Urabá effectively capitalized on their
reputation as militants who would stop at nothing to impede dismissal
and harassment to force the hand of the regional government much as
miners in eastern towns such as Segovia were able to do. In contrast to the
considerations shaping the governor’s approach to public order matters
in worker-dominated areas of eastern Antioquia, however, in western
Antioquia the regional authorities were initially more concerned about
alienating Conservative supporters who distrusted the army but sup-
ported the police. Western Antioqueño towns were also areas where in
early  the regional government was convinced that the motivating
force of violence was more narrowly partisan, not economically moti-
vated, spurred by cultural differences, or involving organized workers as
it appeared to in Urabá.
 Urabá and Western Antioquia 

The regional government’s policy changes initially worsened the situa-

tion of public order in Urabá. In the first weekend after the placement
of army troops along the Carretera al Mar between Mutatá and Dabeiba,
a series of guerrilla attacks took place which prompted the governor to
report to an alarmed president that ‘‘I fear the rebellion is spreading in
a dangerous fashion.’’ 96 Indeed, guerrilla activity escalated so dramati-
cally throughout the zone during the last days of February that the visi-
tador administrativo in the area recommended bombing Rio Sucio and
the guerrilla camps established in Mutatacito and Chadó as the only way
of stopping the guerrillas’ advance.97 By March , the governor admitted
that government forces had been unable to defeat armed men present in
Turbo, Riogrande, Micuró, Vijagual, San Juan de Urabá, and Necoclí. He
limited himself to ordering scarce army troops to the area to ‘‘maintain
bases to conduct the necessary sweeps to defend the local citizenry . . .
until such time as the governor’s office may be in a position to build up
its defenses in those areas.’’ 98 But the regional state never managed to
increase its ability to protect those areas, and their continued vulnera-
bility simply invited further guerrilla aggression. On March , guerrillas
took advantage of the fact that army troops deployed in El Carmelo had
abandoned the settlement twenty days earlier. The guerrillas attacked the
police inspector’s office, killed fifty Conservatives, and forced surviving
inhabitants to flee in terror.99 Five days later a group of fifty guerrillas
struck again on the outskirts of Turbo. Meanwhile, local Conservatives
complained that the soldiers posted to defend the town were drunk and
shouted vivas to the Liberal party and vulgar insults at the town’s mayor
and local Conservatives.100
By the end of March the government had lost the battle against guer-
rilla forces in Urabá. The tax collector in Chocó reported that guerrillas
controlled Titumate and Tanela and were about to sweep settlements in
the Acandí district, an area through which much of the contraband be-
tween Antioquia and Panama flowed.101 Meanwhile, the military officer
in charge of overseeing public order in the area between Chigorodó and
Cañasgordas reported that despite the presence of military garrisons in
Mutatá and Chigorodó and the permanent assignment of soldiers on the
rubber plantations in Caucheras, three active guerrilla camps existed be-
tween Caucheras and Chigorodó.102 The army’s inability to destroy the
camps was partly due, the officer noted, to the inhabitants’ general refusal
to cooperate with the army and to their political affiliation as Liberals. But
 Blood and Fire

he also noted a widespread perception among wealthy hacendados in the

area (who did not support the guerrillas) that the army would be unable
to defend them if they were to turn in the guerrillas. This, too, contributed
to the government’s inability to effectively contain unrest. Overwhelmed
by the seeming intractability of armed insurrection in Urabá and under
considerable pressure from both Liberal and Conservative landowners
in the area, Governor Henao Mejía informed his superiors in Bogotá on
March , , that he had done as much as could reasonably be expected
of him and that the only solution to Urabá’s violence was to hand over
control of the area to the military.103

The Locus of Violence Shifts Westward

Once the governor opted to put Urabá completely under military control,
the principal theater of armed confrontation shifted away from Urabá
toward western Antioquia, particularly to the towns of Peque, Frontino,
Cañasgordas, and Dabeiba. Guerrilla attacks in western Antioquia after
March , in contrast to those reported for Urabá in  and early
, tended to avoid targeting either government offices or public order
personnel to concentrate instead on attacking private property and civil-
ians. In an ironic twist—given the governor’s supposition that western
violence was more partisan than economic—cattle-rustling, extortion of
hacienda owners and administrators, forced sales of land, and theft be-
came increasingly common expressions of violence. For instance, in late
March four hundred head of cattle and one hundred thirty mules and
horses were stolen from Mora Hermanos y Cia.’s hacienda, ‘‘Argelia,’’ in
Urama and taken to Camparusia, where Liberal guerrilla leader Arturo
Rodríguez had his headquarters. The hacienda’s administrator was able
to identify the thieves by name, and he insisted that they were well known
to everyone in the region.104 A little over a month later, guerrillas led by
Aparicio Escobár of Frontino stole another ninety-five head of cattle, and
in June another local hacienda and much of its livestock was burned to
the ground by ten or so guerrillas. The result was fifty thousand pesos of
damage.105
Indeed, cattlemen from both parties in Frontino increasingly com-
plained of rustling and of the movement of stolen cattle to Murrí and
nearby municipalities.106 By August it became apparent that established
circuits for the distribution and sale of stolen cattle and property ex-
 Urabá and Western Antioquia 

ceeded any scale of rustling guerrillas might be engaged in for sheer sur-
vival. Mora Hermanos, which had been robbed of several hundred head
of cattle and horses in March, once more reported attacks against its
properties, this time ‘‘Argelia’’ and ‘‘El Palermo.’’ 107 The administrator
again knew the thieves and insisted that the identities of those buying
the stolen cattle were known as well. The cattle thieves were concentrated
in Peque, Juntas de Uramita (Cañasgordas), Antesalas (Ituango), and all
along the Sinú River in the department of Bolívar on the border with
Ituango, while cattle stolen in the corregimiento of Tabacál in Buriticá
were routed through Cañasgordas and Uramita.108
In September  the Caja de Crédito Agrario, Industrial y Minero
announced that more than seven thousand head of cattle had been stolen
in western Antioquia over a period of several months. The institution
warned that if the situation continued, within seven or eight months rus-
tling ‘‘would absolutely destroy the economy of this important region.’’ 109
To counteract the extensive losses to the Caja’s reserves from forfeited
loans, the credit agency opted to restrict loans in areas affected by vio-
lence, especially those intended to finance the purchase of cattle. The
lending organization also requested authorization from its main office in
Bogotá to refuse all requests for credit while the situation of public order
in the west remained severe. The Caja argued that it was pointless to ex-
tend further credit for cattle purchases when the guerrillas had become
so bold in the face of the regional government’s inability to protect the
region as to attack haciendas only six or seven kilometers from the urban
core of the area’s towns.110
The impact of violence on the property and economic interests of
western Antioquia’s inhabitants, as well as the rising toll of civilian deaths
wreaked by guerrilla attacks and the government’s inability to effectively
impede violence, increasingly prompted some local Conservatives to ap-
peal to the regional government for arms. The mayor and disgruntled
Conservatives in Ituango had already offered to organize a group of Con-
servative contrachusmas to combat the group of six hundred ‘‘bandole-
ros’’ that were stockpiling everything they could steal in the area. The
mayor insisted that an armed volunteer force of local citizens was the only
effective means of eradicating guerrillas in the area, because the police
that had been sent to defend the predominantly Conservative town were
‘‘useless.’’ 111 Conservatives in nearby Dabeiba on Ituango’s western bor-
der had already threatened to create a contrachusma four months earlier
 Blood and Fire

in retaliation for the guerrilla attack that had left their parish priest dead.
There could be no harm in arming Conservative civilians, Dabeiba’s Con-
servatives insisted, since ‘‘no matter how armed we are, . . . we are armed
only to defend the public order.’’ 112
As was true of contrachusma forces that came to operate in other parts
of Antioquia such as the Magdalena Medio and Bajo Cauca, those in west-
ern Antioquia elicited considerable concern and criticism from members
of the government’s own party. Some months after Dabeiba’s offer to
support a local contrachusma, the bishop of Santa Fé de Antioquia, Luis
Andrade Valderrama, felt sufficiently alarmed about the course of events
in Dabeiba and nearby Cañasgordas to send one of the area’s priests for a
private audience with the governor in which the former relayed ‘‘several
very serious observations . . . regarding the situation of public order in
that region.’’ 113 The bishop also sent the parish priest of Giraldo to meet
with the governor. Giraldo’s priest had publicly declared that the only
local violence was that created by a perpetually drunken police force. The
priest accused the police of being ‘‘the sort of evildoers who so deeply
dishonor the prestige of the government and who are really only Com-
munists in disguise who are out to undermine the honor of the govern-
ment.’’ 114 The priest’s accusations prompted Bishop Andrade to insist to
the governor that those promoting violence should be stopped ‘‘regard-
less of who they may be,’’ a posture which won the bishop accolades from
Liberal leader Alberto Jaramillo Sánchez.115 During a dinner in Medellín
at the Nutibara Hotel in July , Jaramillo Sánchez declared that Bishop
Andrade was ‘‘the only prelate in Colombia who had raised his voice to
plead for the restoration of peace and tranquillity’’ and who had spo-
ken out against the ‘‘civil guards’’ (that is, the contrachusma) created by
Dionisio Arango Ferrer in  and the ‘‘aplanchadores’’ sponsored by
Eduardo Berrío González soon after.116 Months earlier, in fact, both Jara-
millo Sánchez and Captain Agustín Salcedo, the military attaché of the
Fourth Brigade, had sounded the alarm against contrachusma activities
in western Antioquia and called for the total disarmament and dissolu-
tion of ‘‘civic police forces’’ in the area traversed by the Carretera al Mar.
The two men insisted that if soldiers were assigned to control public order
in areas affected by violence they would personally guarantee workers’
‘‘respect and obedience to the army and the return of workers to their
tasks, because they were fully confident that army officers and soldiers
would guarantee the security and tranquility of the road workers.’’ 117 To
 Urabá and Western Antioquia 

extremist Conservatives such pronouncements served only to confirm

long-held suspicions that the army and the Liberal party were working
together.
Despite the protests of selected armed forces personnel, the bishop
of Santa Fé de Antioquia, numerous parish priests, and Liberal leaders
such as Alberto Jaramillo Sánchez, the number of towns opting to cre-
ate contrachusmas in western Antioquia grew rather than diminished by
May . Peque, squeezed between Dabeiba, Cañasgordas, and Ituango,
and fearful of an impending attack led by guerrilla leader Aníbal Pineda
and three hundred men from the Llano de Urarco, offered to organize vol-
unteers to defend the town in exchange for arms from the regional gov-
ernment. But the few available policemen in the area, Conservative vol-
unteers pointed out, possessed only twenty bullets between them. Faith
in the government’s forces had reached such a low ebb, moreover, that
the parish priest, from the pulpit during Sunday Mass, had exhorted his
parishioners to abandon the town, and he fled two days later with several
local families.118 Meanwhile, many Conservative hacienda owners fled
south from Pineda’s stronghold in the Llanos de Urarco to seek refuge in
Anzá, while Peque’s Conservatives sought refuge across the Cauca River
in the town of Sabanalarga.119 Partially in response to the wave of refu-
gees seeking asylum and to the presence of guerrillas led by Captain Vidal
Torres and Pablo Emilio López who were camped some fourteen kilo-
meters from the corregimiento of Barbacoas, Sabanalarga’s priest urged
the regional government to empower him to arm volunteers and pay
them ‘‘a daily wage as police auxiliaries’’ so as to ‘‘recuperate lands’’ and
pursue the guerrillas.120 By June , it appears that the governor had heeded
the priest’s call, for the priest reported that a group of armed civilian
police had just succeeded in killing the Liberal guerrilla leader Vidal
Torres.121
In June the regional government and Medellín’s Fourth Brigade re-
organized and increased the distribution of soldiers throughout Urabá,
along Antioquia’s border with Chocó, and in western Antioquia to reflect
the intensification of guerrilla and contrachusma activity in the area. The
number of soldiers present in the region rose dramatically to three hun-
dred eighty from the mere ninety-two that had previously patrolled the
area between Turbo and Frontino.122 Two weeks later, a message from the
armed forces declared Urabá to be the third most violent area in Colom-
bia (following only the Llanos and southern Tolima) and noted that west-
 Blood and Fire

ern Antioquia was rapidly becoming as great a matter of public order

concern to the central authorities as Urabá.123
Several incidents of public unrest fueled the armed forces’ increas-
ing preoccupation with western Antioquia. Guerrillas had led an attack
against Peque in which they desecrated the local church, hung children
alive by meat hooks, and flayed a woman to death. When thirty-five
national policemen were sent out to search and capture the guerrillas,
however, four ran away and two others were wounded, while the guer-
rillas lost no men at all.124 A wiretap conducted by the chief of the secu-
rity department recorded the reaction of the army officers stationed in
nearby Dabeiba whose responsibility it was to provide backup troops for
the national policemen under attack in Peque. The transcript of their con-
versation leaves little doubt that they considered the police one of the
central impediments to the effective defense of public order in the re-
gion. On hearing that four of the policemen sent to pursue the guerrillas
had fled and that the others had stopped their operations simply because
some of their number were wounded, Major Peña Sánchez complained
to his fellow officer, Major Márquez, that ‘‘the police are useless. . . . The
moment someone takes a shot at them from the hills, they run.’’ When
Major Márquez lamented that he had no other men to send but thirty
national policemen as backups, Peña Sánchez grumbled, ‘‘man, what a
bitch [que vaina carajo]. . . . What I’ll do is take fifteen soldiers away from
the posts I control and I’ll take charge of the matter.’’ To which Márquez
slyly suggested, ‘‘Better yet, at every post you come to, take the police
away from them so they’ll [the guerrillas] be screwed, because [police] are
what those bastards have been wanting.’’ 125 The thirty policemen, just as
Major Márquez predicted they would, attracted another guerrilla attack
three days later.126
As terrified refugees fled Peque toward Sabanalarga, the priest there
once more called upon the regional authorities and local Conservatives
to arm and organize contrachusma units to track down the guerrillas
and avenge Conservative deaths.127 But the regional government took
other steps in response to the wave of guerrilla activity assaulting Urabá
and western Antioquia. In August the government decided to impose a
system of salvoconductos, or permits, that limited local mobility within
the area covered by the Carretera al Mar. Any person wanting to enter
or leave the Carretera al Mar’s jurisdiction (that is, between Turbo and
Cañasgordas and through Dabeiba and Frontino) could only do so with
 Urabá and Western Antioquia 

a government-issued pass. Those found not to have a salvoconducto were

to be expelled from the area. The new mandate also applied to workers
on the rubber plantations in Chigorodó and Caucheras.128 Just as it would
do with hacienda workers in eastern Antioquia in mid-, the regional
authorities effectively criminalized western Antioquia’s workers while
making them a captive labor force as well.
The intensification of violence in western Antioquia also prompted
the private sector to take steps to mitigate the effects of unrest on local
investments. In September, the Federation of Coffee Growers hammered
out an agreement with the governor to forestall the possible loss of Antio-
quia’s leading export commodity. Any producer who feared that his crop
might be endangered by the threat of violence could immediately request
troops from the regional government to collect the harvest.129 There was
good reason to suspect that violent incidents might interrupt the har-
vest, for senate elections in which Conservatives were the only candi-
dates were to be held at the end of the month. Elections usually meant
the widespread distribution of liquor and use of intimidation to force the
opposition to vote. Frontino, Buriticá, Dabeiba, and Juntas de Uramita in
Cañasgordas all reported electoral disturbances that coincided with the
onset of the coffee harvest and all appealed to the government for troops
to protect them from attack. In Uramita such threats emanated from the
government’s own forces as contrachusmas allied with members of the
national police ‘‘sowed panic’’ in the rural working population.130
Growing fear of the police and their contrachusma aides gripped not
only Liberals but also moderate Conservatives who worked with or em-
ployed Liberals and who did not support the creation of paramilitary or
vigilante groups who might use partisanship to further specific economic
interests. The extremes to which such groups could go became apparent
when police and civilian Conservatives lined up eleven Liberal peasants
employed by the prominent Martínez Villa family in Frontino and shot
them dead because they held jobs that might have been filled by Con-
servative peasants.131 This incident and another a month later signaled a
shift from violence intended to achieve a Conservative monopoly over
offices and votes or to limit Liberal guerrilla activities to one in which
economic considerations played a determinant role. A group of coffee
producers in Cañasgordas informed the governor in November  that
eighty ‘‘rural police’’ from Tabacál, Buriticá, were dedicated to ‘‘conduct-
ing all kinds of abuses such as homicides, major thefts of cattle, hogs,
 Blood and Fire

coffee, corn, [and] beans.’’ Things had gotten so bad, the coffee growers
complained, that ‘‘they had been forced to abandon the region.’’ 132 The
growers reported that the rural police had stolen twenty thousand pesos
of harvested goods from one store and then burned it down, stolen an
additional forty thousand pesos in goods from local farms, and killed
many women and children in the corregimiento of ‘‘El Naranjo.’’ Sixty
rogue policemen and their officers, moreover, owned ‘‘a store, a cantina,
and a butcher shop where they sell the cattle, coffee, [and] corn they’ve
stolen from us’’ in Tabacál. Seriously alarmed by the possible repercus-
sions of such developments, the regional chapter of  insisted
that the regional government conduct a rigorous investigation to ascer-
tain whether or not such accusations had merit. The Federation of Coffee
Growers warned that if they did, measures would have to be taken to
protect ‘‘an economic force of incomparable value to the nation.’’
In late February , a worried parish priest and members of the mu-
nicipal conservative committee in Caicedo also reported that the national
police stationed in the town and their civilian volunteer aides were not
pursuing the guerrillas. Instead, the contrachusma and the police were
terrorizing the countryside ‘‘abusing women [and] stealing animals and
money from innocent peasants.’’ 133 Several western towns nonetheless
continued to enthusiastically organize additional Conservative paramili-
tary units. Caicedo’s Conservatives, for instance, were split and, despite
earlier complaints of police and contrachusma terror, Caicedo was among
those towns reminding Governor Henao Mejía in early April  that
he had promised to outfit volunteers with official arms to supplement
the insufficient number of available troops to combat the guerrillas.134
The governor sent nearby Frontino and Cañasgordas more than twenty
guns each.135
Because they enabled certain local factions to wreak revenge on the
opposition with impunity and to use the threat of terror to obtain real
material advantages, contrachusma forces were popular despite the evi-
dent risks involved in arming untrained local extremists. But locals also
preferred unofficial armed groups, because they perceived the govern-
ment’s official forces as inefficient and corrupt, a perception that the gov-
ernment’s own policies and attitudes did little to counteract. The parish
priest of Ituango, one of the earliest municipalities to call for arming Con-
servative civilians, for instance, wrote the governor five months after the
creation of these to denounce atrocities committed by the national police
 Urabá and Western Antioquia 

assigned to defend the town. The policemen, the priest angrily insisted,
spent their time ‘‘drinking up the government’s liquor stocks, killing de-
fenseless and peaceful people, [and] . . . terrorizing the residents who
discern in the police an enemy more terrible than the very bandoleros.’’
While Liberal guerrillas took away local people’s lands, the police lived off
the town’s meager public budget and did nothing to stop them. This state
of affairs led the priest to muse aloud, ‘‘I ask myself if these peasants—
who are left unprotected by the government and are subject to extortion
by a police force that only obeys depraved instincts—would defend the
government should it become necessary.’’ 136
The police’s failure to actively pursue the guerrillas may have been
the result of having not been paid, being poorly armed, and knowing
that there were no backup forces. But these reasons cannot excuse the re-
peated reports of police brutality against the civilian population in the
areas they were sent to patrol. Complaints of corruption among the gov-
ernment’s forces, moreover, were too common to dismiss as exceptional.
Uramita reported that national policemen had confiscated seven horses,
saddlebags filled with goods, and three cargas (loads) of coffee (about 
kilograms) worth four hundred pesos in early May and colluded with a
local civilian who acted as their fence to hide and divide the spoils. The
incident came to light when one of the officers sold the items in ques-
tion but cheated his subordinates out of their share. A vicious, public
struggle ensued. Those denouncing the policemen’s behavior suggested
that incidents such as these were commonplace.137 The possibilities for
profit appear to have been so significant that they attracted the attention
and participation of policemen from nearby Frontino. They were said
to migrate to Uramita in search of ‘‘alternative forms of economic sur-
vival’’ as a way of overcoming the scarcity of food and supplies and lack of
pay they suffered in the neighboring municipality to which they’d been
assigned.138
Collusion between members of the government’s official forces and
unscrupulous civilians and between civilians and the guerrillas or contra-
chusmas reached alarming proportions by June . The military warned
the governor of Antioquia that it was imperative to protect rural folk as
they transported their goods to market ‘‘to free them from the extor-
tion they are subjected to by unscrupulous merchants who capitalize on
the region’s abnormal state of affairs to advance their own interests.’’ 139
Meanwhile, local merchants spread rumors that guerrilla attacks were
 Blood and Fire

about to occur or that the guerrillas were close by in order to terrify

farmers into selling their goods for less than their market value and at
great profit to themselves.
At first, the illicit activities carried out by the police and local contra-
chusma forces were matched by those perpetrated by Liberal guerrillas
in the region. Cattle owners detected and reported the development of
networks of merchants and even fellow ranchers who dealt in cattle and
goods stolen by the guerrillas in early . Evidence of the importance
of these newly formed, informal markets and of the indiscriminate theft
of goods without regard to partisan affiliation, however, became more
frequent in the latter half of . In September, for instance, members
of Toledo’s Conservative committee complained to the minister of war
that a hundred guerrillas armed with shotguns and machetes had recently
stolen three hundred head of cattle and thirty beasts of burden from land-
owners ‘‘without regard to whether the owners are Conservatives or Lib-
erals.’’ 140 Both Liberal and Conservative stockmen complained as well
that known cattle rustlers haunted the area around Murrí and Rioverde
in Frontino.141
The possibilities for profit created by a permanent state of dis-
order, moreover, affected even the livelihoods of supposedly partisan-
motivated guerrilla leaders. In early February , Arturo Rodríguez
Osorio, Dabeiba’s guerrilla chief, was overheard responding disdainfully
to an offer made to him by officers of Medellín’s Liberal directorate of
one hundred fifty pesos to start a new guerrilla group in the department
of Santander. Rodríguez rejected the directorate’s offer, boasting that he
could easily earn four to five hundred pesos a day leading an indepen-
dent, armed band.142 Indeed, throughout the first half of  reports
of guerrilla attacks in which the sole objective appeared to be the theft
of cattle and goods abounded. Rapes, torture, and abuse also increas-
ingly accompanied these robberies.143 By March the excesses committed
by Liberal guerrilla leader Aníbal Pineda in Dabeiba were so egregious
that they prompted disgusted members of his own band to kill him.144
Three months later, for similar reasons, Patricio Usaga met the same fate
at the hands of his followers.145 In April, the Caja de Crédito Agrario,
noted with alarm that guerrilla groups in Urabá had advanced from rob-
bery to land occupations (see appendix B.). Guerrillas took over pri-
vately held lands and occupied them against the owners’ wishes in Urama
(Dabeiba), Urarco (Buriticá), Carepa (Mutatá), Carauta, Murrí, and Pla-
 Urabá and Western Antioquia 

tanales (Frontino).146 Meanwhile, the townspeople of Cañasgordas re-

ported in July that guerrillas based in Camparusia, Dabeiba, had sowed
the entire surrounding area with beans and corn and were anticipating a
large harvest in August. If no effort were made to put a stop to this, local
townspeople warned, the guerrillas would be able to buy arms and food
and remain active in the region indefinitely.147
In addition to the rise in theft, by mid- the Magdalena region,
Urrao, Urabá, and western Antioquia were caught in the grip of an un-
precedented wave of homicides. Statistics kept by the armed forces in
Antioquia of civilian and guerrilla deaths estimated that , individuals
lost their lives as a result of violence in , more than double the num-
ber of casualties reported in  () or  ().148 Western Antio-
quia accounted for  percent of the officially recorded, violence-related
deaths in the department. Indeed, although the number of deaths in west-
ern Antioquia was incredibly high what was more horrifying was that
the town of Dabeiba accounted for more than  percent of Antioquia’s
officially reported casualties in  (). This made Dabeiba the sec-
ond most violent Antioqueño town that year after Puerto Berrío in east-
ern Antioquia. Other western towns such as Cañasgordas accounted for
 percent of the departmental total of deaths and Anzá for  percent in
the same year. One-fifth of Antioquia’s Violencia casualties in  took
place in just three western municipalities (see appendix A. and A.).
The military responded to the rise in violence-related homicides by
stepping up its counterinsurgency tactics. But the escalation of military
activity, coupled with the proliferation of contrachusma groups, worked
to dramatically increase rather than curb violence throughout the region.
Events in a single western municipality illustrate this process vividly.
Abriaquí had registered no deaths before , but in that year it became
the destination point for numerous Conservative refugees from Caicedo
and Anzá who were fleeing Liberal guerrilla violence.149 The refugees
began to organize contrachusma forces to combat the guerrillas and to
lead forays into the areas where Liberal groups had usurped their prop-
erties. Four months after the mass exodus to Abriaquí, the town’s mayor
reported in distress that ‘‘some Conservatives want me to support the
killing of peaceful, honest Liberals who have never attempted to rise up
against the authorities. They circulate false rumors in which they insist
that your lordship [the governor] has ordered the indiscriminate mas-
sacre of Liberals and that mayors who don’t comply with that order will
 Blood and Fire

suffer the same fate.’’ 150 ‘‘If it was true’’ that the governor had ordered
the killing of Liberals, the mayor announced, then he would resign. ‘‘If
mayors cannot offer equal safeguards to all citizens . . . then I beg your
lordship to replace me in the post in which I have served loyally during
three and a half years without ever violating the rights of others.’’ 151 As it
turned out, the newly reappointed governor, Dionisio Arango Ferrer, and
his secretary had encouraged mayors throughout the west to organize
volunteer Conservative forces and to meet with the regional authorities
for the distribution of arms. By the end of , forty-four individuals in
Abriaquí had become the victims of violence, twenty-nine of these casual-
ties were supposed bandoleros tracked down and killed by newly created
paramilitary forces made up of outsider refugees.152
Among the many problems incurred by the encouragement of para-
military forces to fulfill what should have been the sole responsibility of
the state and its legally constituted armed forces was the government’s
failure to supervise. The governor and his secretary, Alfredo Cock, ex-
horted mayors and Conservative committee members charged with arm-
ing civilian volunteers to keep close tabs on the use and location of the
arms distributed to contrachusmas.153 But in most cases, once arms left
the governor’s office, he ceased to have any real influence over the indi-
viduals employing them in the government’s name. The majority of arms
ended up in the hands of economically and politically influential local
caciques or public functionaries who failed to report on their status or use.
The lack of control exerted by the regional government over its local
followers led to a situation in which all manner of excesses could be justi-
fied as the fulfillment of official orders. On October , for instance, Con-
servative volunteers in Toledo and members of the national police sta-
tioned in the town joined together to organize a group of fifty-five armed
men. They were to link up with Conservative volunteers and police-
men in nearby towns ostensibly to pursue guerrillas based in Bocas de
Peque and Orobajo in Sabanalarga.154 Nine days later, however, the local
mayor of Toledo reported that instead of pursuing the guerrillas, contra-
chusma forces and the police had been engaged in cattle-rustling and
thefts in both Toledo and Sabanalarga.155 Conservative farm owners in
Uramita, Cañasgordas, meanwhile complained that their cattle had also
been stolen by the group.156 Cañasgordas’ mayor—obliged to commu-
nicate in code lest local sympathizers of the contrachusma find out and
kill him—informed the governor that Conservative contrachusmas were
 Urabá and Western Antioquia 

engaging in widespread cattle-rustling, ‘‘including from Conservatives,’’

and that he urgently needed ten departmental policemen to keep the
contrachusmas in line.157
The illicit activities of the contrachusma operating in Dabeiba reached
such a high level by December that the town’s Conservative mayor rec-
ommended they be disbanded immediately. There was an ‘‘imperative
need to uproot five local Conservatives in order to avoid sending them to
jail for a long time,’’ the mayor insisted, for they had been ‘‘dedicated to
stealing horses and corn, selling them later in the corregimiento of Ura-
mita.’’ 158 It must have seemed a painful irony that contrachusma forces
were selling goods stolen from their fellow party members to merchants
in Uramita who were widely known to be the principal suppliers of the
Liberal guerrillas.
Violence, initially waged in defense of particular party interests or
to protect the lives of party members against the actions of the opposi-
tion, evolved into a free-for-all in western Antioquia. Duty and partisan
loyalty gave way to the pursuit of personal accumulation among armed
members of both parties and even the government’s own forces. Indeed,
recruitment into the national police and contrachusma forces became—
as it also did in eastern Antioquia—a way for the government to chan-
nel in nonthreatening ways poor rural men’s resentment in areas where
lands were concentrated in the hands of a few, and the only available
employment was as poorly paid peons on local haciendas. By directing
violence against members of the opposition of similar social status, local
authorities deflected aggression away from the stark economic and social
inequalities typical of an economy that revolved around cattle ranching
and large sugar estates dominated by hacienda owners who doubled as
political bosses. Partisan competition and repressive state policies pro-
duced a Hobbesian world in western Antioquia in which the opportu-
nity for profit (sometimes but not always mediated by desire for revenge
or partisan hegemony) gradually emerged as the only viable objective of
violence by late .

The Regional Government Responds to the Breakdown of ‘‘Order’’

In January , national policemen stationed in Frontino set out on a

mission to capture guerrillas and the stolen cattle they had hidden on
the outskirts of the town. Several days later the police returned with two
 Blood and Fire

hundred eighty animals they had confiscated from outlying farms with-
out regard to the owners’ affiliation or their possible relationship to the
guerrillas. Forty head of cattle were entrusted to the mayor to hold while
their ‘‘rightful’’ owners were supposedly located and contacted, while
the remaining two hundred forty head of cattle were taken by the police
to an hacienda in Musinga, Frontino. A prominent member of the local
Conservative committee owned the hacienda and he was the individual
with whom ‘‘the deal was to be clinched.’’ The hacendado then resold
the stolen cattle captured by the police to other Conservative hacenda-
dos in the area. When the mayor protested, the local committee members
intervened to have him removed.159
Simultaneously, the Caja de Crédito Agrario, Minero y Industrial com-
plained that the state of disorder in nearby Frontino (caused, it implied,
by the increasingly violent activities of the contrachusma) had made ‘‘it
almost impossible to obtain labor’’ in the area.160 Alarmed by reports
that local Conservatives were at each others’ throats over whether or not
to support armed Conservative volunteers and that some Conservatives
were colluding with Liberals in open defiance of their party’s interests,
Dionisio Arango Ferrer asked Antioquia’s Conservative directorate to
send a visitador especial to Dabeiba, Cañasgordas, and Frontino to report
on the state of the party in the region and the role of armed Conservative
forces in the promotion of local violence.
The visitador’s report on Cañasgordas revealed (unintentionally per-
haps) a sordid, corrupt, divided, and violent society riven by faction-
alism, family feuds, local animosities, personal jealousies, vindictive-
ness, greed, conflicts between haves and have-nots, and struggles over
power.161 The report excused homicides and brutality as ‘‘youthful ex-
cesses’’ and dismissed opponents of partisan extremism such as the priest
or mayor or an elderly patrician political leader as corrupt, sexually licen-
tious, or senile. Nepotism and collusion ran rife and were treated as un-
remarkable aspects of a local political culture in which both the regional
and central government were powerless to confront the complex machi-
nations of cliquish, family-defined roscas (patronage networks). Locals
routinely defied the governor or any government official that attempted
to restrain or condemn their actions. A Conservative who dared to sug-
gest that Liberals were not safe was met with the steely reply that ‘‘the Lib-
erals could hardly meet with better treatment than we have given them,
[I cannot understand] why some people should go around saying that
 Urabá and Western Antioquia 

it’s impossible to live in Cañasgordas, that the persecution here is hor-

rible.’’ As proof of their kindness toward the opposition, local Conser-
vatives noted that ‘‘the bust of [General] Santander is intact even after
the Liberals have persecuted us so many times.’’ Meanwhile the mutilated
bodies of those opposing the contrachusma rotted amid clouds of flies
on outlying farms.
Egregious acts of violence by Conservatives were blamed on an im-
plausible ‘‘Liberal conspiracy.’’ Liberals had supposedly agreed to ‘‘pass
themselves off as Conservatives in order to commit crimes in the name
of the Conservative party.’’ When it proved impossible to get around the
fact that a well-known Conservative had engaged in violence, the visi-
tador found an explanation for that, too. ‘‘Emilio Cifuentes (Milo), is a
Conservative from head to toe and has a reputation for being a terrible
bandit,’’ the visitador acknowledged, but ‘‘What has he done? He always
goes on missions accompanied by the police and other civilian Conserva-
tives. He has attacked areas where a bandit has surely been.’’ Indeed, when
performed by the Conservative contrachusma, violence was excused as
the unfortunate product of economic necessity rather than personal am-
bition. The visitador insisted that ‘‘it cannot be argued that he [Cifuentes]
only robs defenseless people for his own personal enrichment. He is mar-
ried, has three kids, and is very poor.’’ Another Conservative assassin,
Samuel Ruiz, was lauded for being ‘‘principled and well mannered’’ even
though he had robbed the Liberal coffee hacienda of which he was the
administrator. Finally, ‘‘Rapidol,’’ another contrachusma leader, was de-
scribed as a paragon of virtue who ‘‘prohibits his followers from commit-
ting theft, the rape of women, etc.’’ Unfortunately this appears to have
been Rapidol’s death sentence, for disgruntled members of his own band
murdered him.
The visitador found that none of the men participating in Dabeiba’s
or Cañasgordas’s contrachusma was an army reservist even though the
regional government had specified that only reservists could be issued
arms. Indeed, in some cases, members of the contrachusma did not even
possess a national identity card and therefore could not be tracked down
legally for any crime they might commit while supposedly acting in the
government’s name. Contrachusma recruits were uniformly described as
poor, young, married, with many children, and employed as day laborers,
several of them on haciendas owned by local Liberals. Wealthy hacen-
dados and tienda (store) owners who dominated the local economy and
 Blood and Fire

political decision-making in the town stood in counterpoint to the per-

petually underemployed reserve army of rural laborers. The visitador
reported that some landowners and merchants supported the contra-
chusmas, seeing in them a vehicle for the expansion of their own influence
and the opportunity for considerable profit. Others vehemently opposed
the contrachusma, terrified lest a ragtag band of underemployed, resent-
ful, armed poor men eventually challenge their property rights and their
right to determine the fortunes of the party within the municipality. In-
deed, the portrait that emerged from the visitador’s observation of west-
ern Antioquia’s local Conservative party committees and the qualities of
those serving on them is pathetic and devastating.
In Dabeiba, the local pharmacist was the president of the municipal
Conservative committee, but he was opposed by an hacendado originally
from the southwestern town of Fredonia. The hacendado harbored am-
bitions to ‘‘be Dabeiba’s political boss [gamonal] since, according to his
own words, he’s got money and he’s white.’’ The hacendado from Fre-
donia’s opposition to the committee’s president was supported by the
committee’s former secretary, who was the administrator of various local
haciendas belonging ‘‘exclusively to Liberals.’’ The former party secre-
tary was accused of buying and selling goods and cattle with Liberals,
even suspected ‘‘bandits.’’ These two opposed the creation of the contra-
chusma and refused to make a ‘‘voluntary’’ contribution to the local party
committee to underwrite the contrachusma’s activities. This led the visi-
tador to conclude that the contrachusma’s detractors were nothing more
than ‘‘three or four rich guys who aren’t even from Dabeiba, and there’s
no telling which of them is a worse Conservative.’’ The quotas, which
these men refused to pay, were exacted from both public employees and
local party members to support party activities. The wealthy received let-
ters requesting a contribution of five hundred pesos with an accompany-
ing note from the party president ‘‘energetically warning members that
[the committee] was demanding the quota because of the apathy which
they have always shown when it was time to serve the party in this mu-
nicipality.’’ Wealthy members who refused to contribute were threatened
with death, while public employees who did not pay a percentage from
their salaries were fired and replaced with those who would.
Those contributing to the party coffers could demand certificates stat-
ing that they and their dependents, workers, or friends were ‘‘Conserva-
tives in Good Standing,’’ a document which was a prerequisite for ob-
 Urabá and Western Antioquia 

taining employment on local haciendas and for avoiding assassination by

the local contrachusma. When an hacendado refused to pay his assigned
quota, for instance, the committee informed him ‘‘not to ask the commit-
tee for recommendations or any favors since if he [the member] wouldn’t
serve the party then the party wasn’t obliged to do anything for him.’’
Nepotism, moreover, was well inscribed in this system of doing politics.
Favorites and relatives of those in good standing were routinely paid to
act as intermediaries and messengers between the local committee and
the regional government.
In the course of describing the way politics worked in Dabeiba, the
visitador also provided a brief synopsis of its operation in the nearby
western municipality of Sabanalarga. Like public employees in Dabeiba,
those hired in Sabanalarga also had to pay a forced contribution to the
local party committee; this resulted in the creation of a system of bosses
who distributed patronage and subverted any possible neutrality among
those they helped hire. When candidates needed to be fired or replaced,
they were simply accused of ‘‘being turncoats [manzanillos],’’ ‘‘keeping
concubines,’’ or ‘‘drunkenness.’’ In Sabanalarga a pair of ‘‘uneducated’’
brothers filled the offices of mayor and judge. The tax collector was the
brother of the president of the local Conservative committee. The com-
mittee president was a cousin of the mayor and the judge. The local in-
spector de corregimiento was an uncle of both the tax collector and the
committee president; the town clerk was related by marriage to the in-
spector, and the customs agent was a first cousin of the town clerk. The
Conservative president’s two daughters were public school teachers who
had gained their appointments by falsely accusing the two teachers before
them of being party traitors.
If the local party committees in Dabeiba, Cañasgordas, and Sabana-
larga were dens of competing private interests and petty rivalries, the
state of the Conservative party in Frontino, perhaps the most important
and largest western municipality, the governor’s special envoy reported,
was even worse. To begin with, the seventy-year-old medical doctor who
had historically been the president of the local Conservative committee
refused to collaborate with the newly appointed president or any of the
newly elected members. He informed the visitador that he considered
‘‘some of them little better than bandits and the rest idiots.’’ The doctor
was a ‘‘respected’’ man but opposed by a faction of the town’s Conserva-
tives because he continually berated them and publicly declared that he
 Blood and Fire

would like to ‘‘see them all in jail, doing worse sentences than those mer-
ited by the acknowledged bandits [that is, the Liberal guerrillas].’’ The
visitador concluded that such a violent assessment of his party members’
behavior was due to the doctor’s ‘‘nervous state that impedes him from
seeing political reality with seriousness and good judgment.’’ Those ac-
cused of being brutal assassins by the former president were, in the visi-
tador’s estimation, nothing more than ‘‘boys, enthusiastic about politics,
but clean and honest when it comes to respecting other people’s prop-
erty and human life.’’ But despite this vindication of local Conservative
comportment, even the visitador concluded that the majority of those
currently serving on the local party committee were ‘‘incapable of work-
ing in politics’’ and that the only useful member was a local mechanic
although even he ‘‘doesn’t exactly possess excellent abilities.’’
The Conservative party’s problem in Frontino boiled down to a simple
one: the likeliest and most desirable candidates for committee member-
ship were primarily landowners and merchants suspected of colluding,
doing business with, and hiring Liberals. The unreliability of such wealthy
Conservatives had created the opportunity for individuals whom the
visitador essentially characterized as part of a lumpen class—poor, un-
employed or underemployed, and ‘‘resentful’’—to finally enter politics.
These were the shock troops of right-wing ideological movements every-
where. The lumpen allied with arrivistes—individuals with a modicum of
wealth or status who hoped to dislodge those who had historically ruled
local fortunes through a strategy of terror and intimidation. The visita-
dor envisioned the contrachusma not as appointees to the committee,
but as its enforcers. They would constitute the popular support neces-
sary to give the current committee the strength to withstand opposition
from Conservatives such as the former committee president. Depending
exclusively upon armed goons clearly posed problems for a local party
committee, but these seemed less serious obstacles than those posed by
the neutrality or bipartisanship of the wealthy. As an example, the visita-
dor mentioned the names of three brothers who owned several haciendas
in Santafé de Antioquia, Cañasgordas, Frontino, and Dabeiba, and whose
cousins owned Antioquia’s Conservative daily El Colombiano. Using their
influence on Frontino’s Conservative committee, all three had repeat-
edly requested and obtained certificates attesting to the Conservative af-
filiation of the workers on their haciendas even though most of these
 Urabá and Western Antioquia 

were actually Liberal. Workers such as these typically carried two forms
of identification—a Conservative certificate and a Liberal party identity
card—that they showed selectively, depending upon who was asking.
The composition of Frontino’s reconstituted Conservative committee
included a fair share of landowners, but unlike the hacendados deemed
unworthy by the visitador, the landowners supported the formation of
the contrachusma. This may have been due in part because, aside from
‘‘defending’’ the party’s interests, the contrachusma also served the land-
owners’ illicit economic ambitions. Three of the hacendados named to
the committee were in the process of being investigated for purchasing
stolen cattle from policemen and their Conservative civilian aides dur-
ing supposed raids to free the region of Liberal guerrillas. The hacenda-
dos in question had joined forces with the policemen and contrachusma
to recuperate cattle and goods they had earlier lost to the Liberal guer-
rillas. This they appear to have done with great success as less than two
years after their initial losses all of them owned new estates nearer to the
urban core and had bought cafés and stores in town. It was commonly
known in Frontino, the visitador acknowledged, that the new committee
members had become wealthy dealing in cattle and goods ‘‘brought from
places affected and unaffected by violence . . . and even from farms be-
longing to Conservatives, and that they kept the proceeds from the sales
to themselves . . . without being authorized by anyone.’’
The visitador acknowledged that local party members were hope-
lessly divided and that those who most adamantly defended the organiza-
tion of armed civilian Conservative groups were also those most deeply
involved in benefiting economically from la Violencia. They purchased
stolen cattle and goods, resold them, and used the contrachusma as their
own private army to terrorize and eliminate those who disagreed with
them or posed a challenge or obstacle to their economic ambitions. The
visitador made no effort to hide his findings, he freely admitted that those
supplying the Liberal guerrillas were ‘‘weak but entrepreneurial Con-
servatives from Dabeiba, Uramita, and Nutibara’’ and that prominent
committee members were engaged in robberies against their own party
members. Even the police were assessed disdainfully as ‘‘Boyacenses’’ (in-
digenous people from the department of Boyacá) who robbed and ex-
torted for ‘‘personal gain’’ and who ‘‘frequently engaged in annoying
snubs and humiliations directed at Antioqueño members of the police
 Blood and Fire

and the general citizenry, provoking highly inconvenient arguments.’’

Ultimately, however, the visitador was willing to overlook the misbehav-
ior of Conservatives such as the ‘‘ardent boys’’ who, with dubiously ac-
quired wealth, made up the contrachusma, the police, and the hacenda-
dos, because ‘‘at least we can be sure of their political affiliation.’’
All that mattered, the visitador suggested in his report to the governor,
was an unquestioning loyalty to the party regardless of the moral or ethi-
cal nature or real motivations of those acting in its name. The unruly be-
havior of the contrachusma, for example, was not only exonerated by the
visitador, he even recommended that party fervor expressed as violence
could be put to good use by the regional authorities in other areas. He
thus concluded his report by recommending that contrachusma mem-
bers accused of murder and robbery be recruited as policemen and sent
to other Liberal-dominated towns such as Puerto Berrío ‘‘so that they can
have a change of pace. This way they could be removed from Cañasgordas
where they’ve become a nuisance, and their valor could be put to use
under the discipline of the police in specific needy areas.’’ It is unsurpris-
ing that the police enjoyed so little esteem from the civilian population
when the government and its representatives envisioned the institution
as a channel through which lower-class aggression could be safely and
legitimately played out against the populations of marginal areas who,
from the government’s perspective, were colonized by undesirables.

The Denouement of Western Violence:  and Beyond

Two months before General Rojas Pinilla toppled the Conservative gov-
ernment of Laureano Gómez in June , the situation of public order in
Urabá and western Antioquia remained as severe as it had been in .
Cattle-rustling had become a permanent feature of the area’s economy
despite repeated decrees intended to strangle the export of livestock from
the region and repeated forays conducted by the police and army to dis-
lodge the guerrilla camps that they argued were the conduits for stolen
animals in the region.162 Guerrillas, moreover, continued to successfully
lead attacks against civilians, rural workers, and the state’s officers de-
spite the increased number of troops stationed in the area.163 The only
tangible change in the area’s public order situation was the increasingly
entrenched character of the contrachusma. Municipal authorities who
 Urabá and Western Antioquia 

came to regret arming civilians would remember the first months of 
as the point at which the armed men they organized to defend against
Liberal guerrilla forces took on a life of their own.
In a long, coded message that Caicedo’s mayor sent the governor in
April, the mayor lamented that when he had attempted to expel the
local contrachusma by arranging a distant mission, the members of the
group had threatened to return and ‘‘form a new guerrilla group with
the Montoyas.’’ 164 Also, local leaders in one town increasingly deployed
the contrachusma to extend their control over individuals in another.
Those organized by the mayor of Dabeiba, for instance, were sent to
operate in Chigorodó where they assassinated three adults and six chil-
dren in April and then with impunity returned through the hills to their
home base.165
The government and its forces, meanwhile, were still a largely inef-
fective and absent presence in the region. Workers and mule drivers in
Uramita, Cañasgordas, suffered assaults along the main road through the
town because of the absence of military posts in San José, San Benito,
and Aguadas, while moonshine flourished in Chigorodó because all of
the customs officers had been forced out of the zone.166 Disorder that
initially appeared to be a horrifying but temporary product of partisan
conflict in late  had thus become endemic and difficult to catego-
rize by . Conservatives who had initially supported any government
effort intended to stem the tide of Liberal guerrilla thefts and assassina-
tions gradually came to lose faith in the ability of the regional authori-
ties or the armed forces to defeat the guerrillas or maintain public order.
In many instances, the police whom local extremists had insisted were
the only force whose partisan identity made them trustworthy, proved
as disorderly and rapacious as the armed opposition. Some Conserva-
tive landowners even went so far as to urge the regional authorities not
to appoint national policemen to oversee local public order matters, be-
cause the police force ‘‘sweeps away any human element in its path.’’ 167
Private citizens ultimately offered to underwrite the cost of public de-
fense—as landowners in eastern Antioquia had also done—in order to
ensure the minimal protection of their lives and economic interests.168
Other Conservatives opted to arm bands of Conservative civilians to ful-
fill the government’s responsibility to protect private property and lives,
while others withdrew their support from the government and their party
 Blood and Fire

and colluded with members of the opposition in order to defend their

personal interests.
But disillusion also set in among those initially supporting the cre-
ation of contrachusma groups as these increasingly proved to have agen-
das that transcended a subordinated role as strict defenders of partisan
interests. By  local attitudes toward the contrachusma and Liberal
guerrillas mirrored to some degree those present in eastern Antioquia.
Moderate Conservatives and established local landowners or business-
men with no stake in deploying armed members of their party to attack
Liberal workers or the properties of the opposition with whom they had
shared business interests, opposed the contrachusma, often at consider-
able risk to their own lives. Arrivistes and locals resentful of collaboration
between an old guard and the opposition, or resentful of the opposition’s
control of local affairs, seized upon the contrachusma as a medium for
simultaneously advancing their personal fortunes and settling old scores.
Poor Conservative farmers, settlers, and day laborers, who bore the
brunt of guerrilla attacks, supported or participated in the contrachusma
out of self-interest or in self-defense as it became apparent that the gov-
ernment’s forces could provide little protection. Some eventually came
to regret their participation as the indiscriminately vindictive nature of
these forces became increasingly evident. A similar phenomenon oc-
curred among poor Liberals who initially supported the guerrillas but
eventually could see little difference between one criminal organization
and another, both of whom taxed, assaulted, and stole from them in-
discriminately. Disorder, in other words, created both unparalleled op-
portunities for some and tragedy for others. Lands and properties were
cleared of unwanted and troublesome squatters and tenants.169 Some
local merchants packed up and left, others capitalized on the emergence
of new informal markets of stolen goods. The marginally employed or
poor found an alternative source of economic survival as members of pri-
vate armies and political offices and party committees were rearranged
to reflect newly acquired fortunes and recently achieved status.
Of course, all of this occurred at considerable cost. Western Antio-
quia alone accounted for  percent of all the officially registered deaths
caused by violence in the department between  and May of  and
for  percent of all the deaths registered in the six worst months of la
Violencia between August  and January . In real terms, these per-
centages represented a total of , deaths in western Antioquia, nearly
 Urabá and Western Antioquia 

two-thirds of which occurred in just three towns: Dabeiba (), Cañas-

gordas (), and Frontino ().170
The trajectory of violence did vary, however, between Urabá and west-
ern Antioquia, as did its long-term impact to some degree. Urabá, con-
sidered by the armed forces as Colombia’s third most violent area in
, actually registered relatively few deaths between  and . Two
percent (seventy-seven) of the department’s total casualties occurred in
Urabá, while only four deaths were registered in , the year in which
the rest of the department recorded the most deaths from violence.171
It is significant that the bulk of officially registered deaths in Urabá oc-
curred during  (sixty-three) in the months before the regional gov-
ernment relinquished control of the area to the military and pulled out
the national police. For while Urabá accounted for a very low percentage
of the region’s total Violencia deaths, it accounted for a considerable per-
centage of the political prisoners (that is, captured guerrillas) interned
in Medellín’s ‘‘La Ladera’’ prison ( percent) () (see appendix A.).
In contrast, civilian and guerrilla casualties in western Antioquia were
high while the number of guerrillas captured and imprisoned amounted
to only  percent (fifty) of those jailed.172
The explanation for this discrepancy may lie in the entrenched pres-
ence of the contrachusma in western Antioquia and its relative absence
in Urabá. The majority of those held in ‘‘La Ladera’’ from Urabá were
captured after the military took control of Urabá. With the exception of
the portion of the road in Dabeiba controlled by the military (which ac-
counted for the majority of the prisoners from western Antioquia [thirty-
three of fifty]), in contrast, almost no prisoners were taken in western
Antioquia after  when the contrachusma became particularly active
in the area. These facts lead to several conclusions. Contrachusma groups
began to operate extensively in western Antioquia in late  and early
. They received an added impetus and additional arms from the re-
gional government after July . The greatest number of deaths in the
area were registered between August  and January  when these
groups were most active. Urabá, on the other hand, experienced very little
contrachusma activity (except for sporadic attacks led by contrachusma
groups organized in Dabeiba against the section of the Carretera al Mar
that linked Dabeiba to Mutatá) and registered very few deaths after .
This would seem to suggest that violence was directly proportional to the
presence of irregular forces, especially the contrachusma. When the gov-
 Blood and Fire

ernment relinquished control of Urabá to the military (most of whom

were from the department of Bolívar) in March , for instance, vio-
lence in Urabá declined substantially.
In other words, the unprecedented rise in civilian casualties seems di-
rectly related to the presence of paramilitary forces in western Antio-
quia rather than to conflict between the armed forces and guerrillas or
guerrilla-led activities. Indeed, a summary of the town of Peque’s ills,
provided by the parish priest in the immediate aftermath of the military
coup, gives a sense of the intransigent and significant impact of paramili-
tary violence on the region even after the military government had come
to power. Father Blandón Berrío informed Píoquinto Rengifo, Antio-
quia’s new military governor, on June  that approximately eleven thou-
sand inhabitants had lived in Uramita (Cañasgordas) and Peque prior to
la Violencia. By mid-, however, only three thousand exiles inhabited
the town of Peque proper, while several hundred former residents were
dead and another seven thousand inhabitants were holed up ‘‘in the hills,
surrounded by misery, lack of clothing, hunger, and sickness.’’ They were
too terrified to come back to the town for fear of losing their lives, not
to the Liberal guerrillas, but to contrachusma groups still active in the
surrounding area (see appendix B.).173
It is little wonder that the priest of Liborina seized the pulpit to de-
nounce the contrachusma and to declare during a Mass in late May 
that ‘‘the government led by Braulio Henao Mejía had been a disaster,
but that under Dionisio Arango Ferrer it was worse.’’ 174 In a long missive
addressed to Governor Dionisio Arango Ferrer in June, former governor
Fernando Gómez Martinez in fact argued that the main instigators of
violence in western Antioquia were the contrachusma groups endorsed
and armed by the regional government. Gómez Martinez also accused
Arango Ferrer’s administration of reassigning public employees indicted
for criminal behavior instead of firing and jailing them; of illegally detain-
ing suspects and arbitrarily executing them with the aid of the national
police; and of using torture to obtain confessions even when evidence
obtained in such a manner had been declared illegal and inadmissible by
Colombia’s highest court.
The former Antioqueño governor used the excesses of the paramili-
tary forces endorsed by the state in Cañasgordas to illustrate his accusa-
tions. After a unit shot at each other in a public dispute over goods they
had recently stolen from innocent local inhabitants, they forced ‘‘the in-
 Urabá and Western Antioquia 

habitants, especially Liberal women, to don mourning clothes while the

republic’s armed forces acted as the honor guard to the funeral proces-
sion as it wound its way to the cemetery.’’ 175 Shortly after this incident,
the town’s contrachusma erected a bust in Arango Ferrer’s honor that
confirmed the degree to which their existence had been the product of
the governor’s support.
The contrachusma were, in fact, the big winners in western Antio-
quia. They continued to operate long after August  when the govern-
ment officially demanded the return of arms distributed by the former
regional government and after local Liberal guerrilla groups had been
captured or voluntarily taken advantage of the military government’s
amnesty policy.176 Their continued presence made it impossible to heal
the wounds of inhabitants exhausted by a conflict many of them had
neither condoned nor understood. Caicedo’s priest, who prefaced his
criticism of the contrachusma by identifying himself as a lifelong Conser-
vative loyalist, emphatically denounced the persistent effect that armed
civilian Conservative groups had on local life. In July, shortly after the
military took power, he held a reconciliation Mass on the feast of the
Virgin of Mount Carmel to reincorporate Liberals who had stayed away
from the church during the previous four years. His good intentions were
ruined, however, when ‘‘three guys of the very lowest Conservative scum
in this town, . . . three men who were probably once recalzados [Liberal
converts to Conservatism] and implicated in the great robberies con-
ducted by the famous Montoya Giraldos who led the contrachusma,’’
erupted in the churchyard and dragged away several ‘‘peaceful’’ Liberal
congregants and imprisoned them.177 Elsewhere policemen and contra-
chusmas continued to work for private interest groups and individuals
to intimidate workers and deny them their legal rights.178 By August, re-
ports were circulating in Juntas de Uramita and Cestillal, Cañasgordas,
that one hundred contrachusma were in the hills between these two vil-
lages and were organizing ‘‘throughout western Antioquia’’ ‘‘to topple the
government.’’ 179
Although all of the Liberal guerrillas once active in western Antioquia
and Urabá surrendered or had been eliminated by October , homi-
cides, cattle-rustling, and forced displacement of workers, squatters, and
rural property owners grew rather than diminished.180 In western towns
such as Sabanalarga, Cañasgordas, Dabeiba, Caicedo, Uramita, and Fron-
tino, citizens of both parties complained to the military governor that
 Blood and Fire

‘‘four or five layabouts have been dedicated to sowing terror,’’ forcing

people to flee and then seizing their lands.181 The perpetrators of disorder
organized publicly and were well known to all. Lists of those forming the
contrachusma as well as their sponsors in towns such as Cañasgordas,
Uramita, Buriticá, and Frontino were repeatedly sent to the governor and
to various private sector entities such as the Caja de Crédito Agrario, the
regional Federation of Coffee Growers, and the Conservative and Liberal
regional party directorates during the several months following the mili-
tary coup.182 But the very sectors of society that might have been counted
on to put a stop to privately motivated disorder—the local clergy, politi-
cal leaders, prominent businessmen and landowners, police, and public
employees—were precisely those most enmeshed in the perpetuation of
paramilitary violence in the region. The list of future victims in Buriticá,
for instance, was routinely submitted in advance to the parish priest for
approval.183
The incidence of forced land abandonment as a result of contrachusma
terror became so severe in western Antioquia that the Caja de Crédito
Agrario warned the governor that it would have to deny any credit to
these regions until the destabilizing contrachusmas were eradicated.184
Antioquia’s Auditoria General del Departamento (regional comptroller’s
office), moreover, officially declared that the bulk of the cattle stolen in
Uramita, Frontino, and Chigorodó had been taken with the collusion
and participation of the authorities and the government’s forces.185 In-
deed, the local Conservative party members whom the Conservative di-
rectorate’s visitador especial had found guilty in Frontino a year earlier
of colluding with the authorities to resell stolen cattle and force out ten-
ants and farmers, were once more accused in November  of leading
‘‘gangs of assassins and thieves’’ to terrorize the inhabitants of Cañas-
gordas, Dabeiba, Frontino, and Buriticá.186
It is impossible to draw an absolute correlation between paramilitary
violence, land concentration, the growth in property values, and the ex-
pansion of large-scale commercial production of sugar and cattle in west-
ern Antioquia as a result of la Violencia. Some data does suggest, however,
that violence supported by powerful economic agents in selected west-
ern municipalities and parts of Urabá affected by contrachusma violence
had long-term repercussions on local land markets and forms of pro-
duction. Regional catastro, or property tax assessment records, contain
the number of properties assessed in a given year and their total value
 Urabá and Western Antioquia 

in each municipality from  through .187 Dividing total munici-

pal land values by the number of municipal properties and the resulting
value by the number of inhabitants recorded by the respective popula-
tion census, enables us to form a general picture of land trends in the
immediate aftermath of la Violencia (see appendix A. and A.).
While the departmental average for the number of properties per
capita in Antioquia between  and  was ., towns in western
Antioquia and Urabá figured among the fifteen Antioqueño municipali-
ties with the least number of properties per capita. The average number of
properties per capita in Turbo, Sabanalarga, Santa Fé de Antioquia, and
Dabeiba was half or less than half the departmental average. The towns
experiencing the greatest increase in the value of properties per capita in
Antioquia between  and , moreover, were those dedicated to the
production of cattle, commercial agriculture (such as sugar processing,
cacao, rice, rubber, and coconuts) and wood extraction. These included
Chigorodó and Turbo in Urabá, and Buriticá, Anzá, and Ituango in west-
ern Antioquia. Land values in these areas outstripped even the value of
land in industrial towns, the traditional leaders in land values in Antio-
quia through . For the period encompassed by the land tax survey of
 and , moreover, the town with the greatest increase in the aver-
age value of property was Frontino. This was the very municipality where
the contrachusma was most actively mobilized by up-and-coming busi-
nessmen and landowners to promote their economic interests and where
the strongest evidence for the collusion of these forces in land usurpations
and cattle-rustling was collected by the regional authorities.
Indeed, both the absolute and relative growth of property values in
western and Urabá towns was phenomenal between  and  and
again between  and . Chigorodó went from an average value per
property of  pesos in  to , pesos in  and , in .
Dabeiba experienced a similar rise in property values during the same
period, climbing from  pesos in  to , in  and , in .
Frontino whose average property value amounted to  pesos in 
and , pesos in , posted the fourth-highest average property value
in all of Antioquia by  (, pesos). Sabanalarga, a major center
of contrachusma organization, registered the department’s third-highest
average property value in . It had risen from an average of  pesos
in  to , in  to , in .
While the completion of the Carretera al Mar may partially account
 Blood and Fire

for the rise in property values in towns such as Turbo, Chigorodó, and
Dabeiba, it cannot account alone for either the concentration of land
ownership or increase in land values in the area: The road did not run
through either Sabanalarga or Frontino, the western municipalities with
the highest growth in average property values. What all the towns experi-
encing the most dramatic increases in average local property values had
in common was the presence or operation of well-organized paramilitary
forces supported by and deployed in cattle-rustling, theft, worker elimi-
nation, and land usurpation by sectors of the economically powerful.

Conclusion

The military coup that brought an end to Conservative rule failed to bring
peace to western Antioquia. This was at least in part a result of the fact that
beyond the partisan antagonism that had fueled initial unrest in Antio-
quia, la Violencia had laid bare the weakness of the state’s authority and
invalidated any claim to a monopoly of force in Colombia. It was true
that the guerrillas in western Antioquia had access to arms, logistical in-
formation, and supplies that exceeded on occasion those available to the
government’s own forces, but the cause of continuing violence in the re-
gion was not primarily instigated by the Liberal guerrillas. It was the lack
of consensus within the regional and central government and within the
regional Conservative party, coupled with an absence of trust between
the government and its own armed forces, that first doomed the state’s
efforts to contain unrest in the region. Moreover, when the state handed
over responsibility for the maintenance of public order to armed, infor-
mally constituted groups of partisans, it lost legitimacy and credibility
not only among the alienated opposition but also among many of its own
supporters.
Whatever the nature of the disputes initially catalyzing unrest in west-
ern Antioquia and Urabá—and in most cases these disputes were par-
tisan, although inextricably linked to economic and cultural issues as
well—by the official conclusion of la Violencia, Liberal guerrilla contain-
ment and the imposition of partisan hegemony were not the overriding
objectives of public order forces. Indeed, it is hard to escape the impres-
sion that for distinct local sectors what began as partisan violence even-
tually evolved into unprecedented opportunities for personal profit that,
 Urabá and Western Antioquia 

contrary to the stated objectives of groups such as the contrachusma, re-

quired the persistence and not the conclusion of violence to thrive. In
effect, the state and local forces not only promoted violence, they main-
tained it long after the justification of unrest—the existence of organized
armed insurrection led by members of the opposition—had ceased.
Betulia, March 2000. People
are displaced in the aftermath
of repeated battles between
paramilitaries Autodefensas
Unidas de Colombia (AUC)
and members of the Fuerzas
Armadas Revolucionarias de
Colombia (FARC) guerrillas
on the border between Urrao
and Betulia.
4. Urrao and the Southwest

: Why weren’t you killed?

 ,  : Because God was on our side
and the devil was on theirs.

Seeing as how the inhabitants of La Mina and La Guamala have given great
support to bandits [bandoleros] since the War of the Thousand Days, [I], the
undersigned, without consulting with my superiors, but with the best intentions,
ordered the complete disoccupation of these hamlets. . . . There is no reason to
offer safeguards to the enemy in places where the families of honorable Conser-
vatives cannot live.—Major Arturo Velásquez Acosta

The area most associated with la Violencia in Antioquia is the southwest,

specifically Urrao. Its experience has been the benchmark by which re-
gional violence between  and  has historically been measured.
Yet, in many ways the trajectory and character of violence in the south-
west followed a quite different path from what was typical of violence
in eastern Antioquia, the west, and Urabá, the other areas of Antioquia
where violence was severe. Part of the reason why la Violencia and the
southwest have become synonymous may have to do with the fact that
in certain ways southwestern violence appears to conform most exactly
to what became, in hindsight, the idealized version of la Violencia. Con-
servatives opposed Liberals, Liberals took up arms to defend their party,
and when the military toppled the Conservative government of Laureano
Gómez, Liberal guerrillas put down their arms, accepted amnesty, and
la Violencia came to an end. It is a tale of loss and tragedy, but it is also
heroic and familiar: a tale in which the two sides and their objectives were
clearly drawn in a pattern reminiscent of the nineteenth century’s civil
wars. Or, as the wife of Urrao’s former guerrilla leader succinctly put it,
‘‘We knew what we were fighting about then, but what are they fighting
about now?’’ 1
In the southwest, there were no outside migrants, militant union-
ized workers, ethnic hatreds or overt rivalries over resources and land
 Blood and Fire

to muddy the development of la Violencia or to suggest that its ultimate

purpose had been more complex than a straightforward partisan dispute.
A recognizable language existed—the language of traditional partisan
politics—to constitute a familiar frame of reference among Antioqueños
and other Colombians for thinking about violence, regardless of how ex-
treme, as the product of age-old hatreds. When what was fought over
were principles and loyalties intelligible to all, the conflict between people
who had only shortly before been neighbors, friends, business associates,
and kin could be infused with a kind of nobility (and inevitability). This
was particularly the case when the objectives of struggle in the southwest
seemed not to call into question the distribution of power in Colombian
society or its basic economic and social arrangements as it did in eastern
Antioquia, the west, and Urabá.
Violence in the southwest is worthy of study precisely because its tra-
jectory seems to have differed so markedly from the way violence evolved
elsewhere in Antioquia during the same period. The most commonly ac-
cepted version of la Violencia, in other words, was the regional exception,
not the rule. Yet, even in the southwest where violence appears to have re-
produced an uncomplicated version of la Violencia—a conflict waged ex-
clusively around partisan issues—ambiguity as to the exact objectives of
violence also surfaced by , particularly in relation to the distribution
of land and jobs. The presence of a struggle over land in certain south-
western towns, especially those characterized by large estates, suggests
that la Violencia, at least in Antioquia, was a dress rehearsal for the explic-
itly Marxist-inspired land struggles that emerged in the s and that
are still a critical determinant of Colombian violence today. If this was
so even in an area of supposedly traditional settlement, what inhibited
violence in the southwest from becoming the vicious internecine struggle
over material resources and rights that it ultimately became in eastern
and western Antioquia?
Why did violence manage to remain relatively contained in the south-
west? Why was the state able to eventually defeat the guerrillas without
recourse to the paramilitary organizations to which it alienated its au-
thority over public order in other parts of Antioquia? And, finally, why
was Urrao, the southwestern epicenter of la Violencia, able, as other re-
gions affected by violence were not, to heal the vestiges of fratricidal con-
flict and reknit the bonds of collective identity in the aftermath of la Vio-
lencia?
 Urrao and the Southwest 

Redefining the Geographic Parameters of Southwestern Violence

The term ‘‘southwestern violence’’ is a bit of a misnomer. As a map of the

towns affected by violence in the southwest makes clear, violence was not
a generalized phenomenon throughout the southwest (see map ). While
nearly all the towns of the coffee-producing southwest experienced some
electoral conflict between  and  and/or suffered from violence
spearheaded by regional Conservative appointees, only Urrao, Betulia,
Salgar, and, to a lesser degree Concordia and Bolívar experienced vio-
lence generated by conflicts between armed Liberal guerrillas and gov-
ernment forces from  to .2 Of these five towns, four were im-
portant coffee producers, although in Urrao, where the guerrillas were
concentrated, coffee played but a minor role in the local economy. What
linked these five towns into a coherent area of violence was the emergence
of armed Liberal resistance in Urrao. Violence radiated out from Urrao
to affect surrounding municipalities. For these reasons, in the analysis
which follows, the bulk of the narrative will focus on Urrao and only sec-
ondarily on the four southwestern and two western municipalities that
fell within the radius of Urrao’s influence.
While technically Urrao is considered part of the administrative sub-
unit known as suroeste (southwest), it resembles and is linked by inter-
est and location more appropriately with the administrative area known
as occidente (the west). The municipality is situated in a broad val-
ley crisscrossed by clear streams and the Penderisco River. To the west,
dense jungle forms a natural boundary between Urrao and the Pacific
lowland department of Chocó, while to the northwest Urrao abuts the
sparsely settled municipality of Murindó in Urabá. Tropical highlands
(the Paramo de Frontino) separate Urrao from Frontino and the Urabá
peninsula to the northeast. The southwestern coffee belt ends at Urrao’s
southern border. Situated between wilderness and ‘‘civilization,’’ Urrao
formed an intermediate space or gateway between the frontier of tradi-
tional Antioqueño settlement and the less explored territories of Chocó
and Urabá.
With the exception of several hamlets along Urrao’s southern border
where smallholder coffee farms flourished, the town’s economy revolved
around ranching, the cultivation of sugarcane, the manufacture and sale
of moonshine, contraband, guaqueria (the unearthing and sale of items
from Indian burial grounds), and subsistence farming. The more acces-
 Blood and Fire

Map . Southwest Antioquia (Source: Instituto Geográfico Augustín Codazzi;

Archivo Privado del Señor Gobernador de Antioquia, –; Archivo de la
Secretaría de Gobierno de Antioquia, –)

sible land was concentrated in large estates, some of whose twentieth-

century titleholders were descended from colonial owners. A consider-
able amount of land, however, remained as tierras baldias, or public
lands, covered in dense forest and jungle. Nearly four thousand square
kilometers of total land made Urrao Antioquia’s second largest munici-
pality by area (after Turbo in Urabá) with a population of approximately
twenty thousand at the onset of la Violencia. However, this population
was concentrated, and the vast majority of the municipality’s territory
was uninhabited.
Urraeños perceived themselves and were perceived by other Antio-
queños as independent, rebellious, fierce individuals with a reputation for
 Urrao and the Southwest 

licentious behavior and an iconoclastic attitude toward life. Like frontier

towns elsewhere in Latin America, economic survival and social com-
portment in Urrao often revolved around activities at the margin of licit
society.3 The fortunes of many otherwise ‘‘respectable’’ inhabitants were
the product of the opportunities for illicit gain offered by Urrao’s unique
topography and location. Contraband had flowed freely since colonial
times from the Gulf of Darien through the Atrato River in Chocó and
from there into Urrao’s countless hidden paths and streams. Locals had
developed ingenious systems for smuggling goods and arms in and out
of the region, and these systems stood them in good stead during la Vio-
lencia.
Smugglers capitalized on local beliefs in witchcraft and animism by
camouflaging contraband in coffins that were paraded through the jungle
borders of the municipality by men dressed in sheets and accompa-
nied by flickering torches in the dead of night. The fear of encounter-
ing angry spirits effectively deterred both regional customs officials and
local policemen from pursuing such incidents too closely, and this was a
boon to armed insurgents who knew the terrain and the power of local
superstition only too well. The tenacity of local resistance was expressed
as well in a long tradition of moonshine distilleries which no amount of
aggressive customs revenue pursuit had ever managed to dislodge. Re-
gional customs agents still ruefully complained in the s that Urrao’s
illegally produced local cane liquor (tapetusa) was so popular that it con-
sistently forced the regional liquor monopoly to reduce its prices just to
compete.
Nonconformity took social as well as economic forms in Urrao.
Middle-class norms of piety and respectability that formed the basis of
the regional ideal of Antioqueño comportment rarely operated in a con-
sistent fashion there. Many of the town’s residents typically lived in free
union rather than in legally sanctioned marriages, and the rate of illegiti-
mate births was considerably higher than the regional norm.4 The town
abounded in taverns and brothels. These facets contributed to Urrao’s
regional stereotype as a place of loose morals and radical political ten-
dencies. Coffee farmers from nearby towns whose behavior was closely
scrutinized by parish priests and the Legion of Mary perpetuated Urrao’s
fame as a moral-free zone by fleeing to it during holidays to find release
in prostitutes and drink.
In addition to its reputation as a place of ‘‘misrule’’ where the re-
 Blood and Fire

strictions of conservative, smallholder, ultra-Catholic society could be

temporarily shed, Urrao was famous for the independent spirit of its
local inhabitants and their defense of local autonomy.5 The town’s repu-
tation dated back to the colonial period when Urrao became a place of
refuge for free people of color, runaway slaves, and others fleeing the re-
strictions of Santa Fé de Antioquia’s mining and ranching economy to
the northwest. The area’s natural attractions—its lush vegetation, gentle
hills, and sparsely inhabited valleys of well-irrigated land and abundant
woods—offered a respite from colonial scrutiny and demands. Antio-
quia’s archives are filled with lengthy petitions and drawn-out lawsuits
initiated by Urrao’s lowborn, free people of color against abuses exer-
cised by powerful landlords and colonial bureaucrats. Although few such
petitions were successful, a precedent of local resistance and an unwill-
ingness to subordinate local needs to established rules of hierarchy and
power continued to form an important aspect of local lore and self-
representation in Urrao throughout the period of la Violencia.
Urrao was also famous as a center of Liberal fervor in a historically
Conservative department. Liberals had always controlled the majority of
the town’s public offices and wealth. This did not change when the Con-
servative party came to power in , but, like other Liberal towns in
the s, local Liberals were divided. The bulk of the town’s Liberals
voted for the official party candidate, Gabriel Turbay, in the  presi-
dential election, but gaitanistas were in control of the local town coun-
cil in .6 Urrao’s Conservatives, who constituted perhaps  percent
of the town’s eligible electorate, primarily identified with the moderate
politics of Mariano Ospina Pérez, although some support for Laureano
Gómez existed as well. It was not until the assassination of Jorge Eliécer
Gaitán in April , however, that local Liberals had any sense that their
monopoly over local authority might be endangered by the rise to power
at the regional and national levels of the Conservative party.
As news of the assassination spread, Liberals in towns to the south
of Urrao, such as Andes and Venecia, especially those characterized by a
concentration of men employed by the railroad and in public road con-
struction, organized demonstrations to protest the Liberal leader’s death
with the support of Liberal municipal policemen and local public em-
ployees. Governor Dionisio Arango Ferrer responded to these protests
by promoting the formation of Conservative volunteer forces. These were
concentrated in the southwestern municipality of Andes and in Bolívar.
 Urrao and the Southwest 

Arango Ferrer also deployed army troops to guard against possible insur-
rection in Fredonia, Jardín, Andes, Betania, Tarso, Pueblorrico, Salgar,
Bolívar, Betulia, and Urrao, but fears of imminent rebellion proved un-
warranted.7 No incidents of unrest or partisan skirmishes were reported
either in Urrao or in other southwestern municipalities during the year
following Gaitán’s demise.
In February , however, Urrao’s mayor and the Liberal town coun-
cil president wrote Antioquia’s governor (by then moderate Fernando
Gómez Martinez) to request an immediate audience. A variety of admin-
istrative issues were beginning to create tensions among the local citi-
zenry.8 The parish priest, an ardent laureanista named Manuel Zapata,
was accused of conspiring to establish a meat monopoly and of increas-
ingly intruding on the secular prerogatives of the municipal authorities
in an attempt to consolidate his position as a local power broker.9 A
marginalized Conservative schoolteacher was making threatening noises
about the local Liberal monopoly of offices and was clamoring to be ap-
pointed as town clerk. Embroiled in a bitter battle with a divided mu-
nicipal council in Medellín and threatened with the city’s bankruptcy,
Governor Gómez Martinez paid scant attention to Urrao’s complaints.
The governor then unpleasantly surprised the town by appointing a Con-
servative mayor without consulting local officials. The newly appointed
mayor snubbed Urrao’s Liberal officeholders and refused to collaborate
with the local municipal council. It seemed increasingly evident that the
regional government was intent on eclipsing the authority of Urrao’s Lib-
eral majority and its elected representatives.
Administrative tensions involving predominantly Liberal municipal
councils and Conservative appointees to municipal public office emerged
in other southwestern towns around the same time in early . A month
after the events in Urrao, the recently appointed Conservative mayor
of Valparaiso informed the governor’s office that he had uncovered a
dynamite-making operation, ammunition, and arms in a local Liberal
politician’s house, and accused the politician of preparing a local revolt
against the Conservative authorities.10 Two weeks later the mayor and
police chief of Jericó assaulted two Liberal visitadores fiscales (regionally
appointed legal aides) who had been sent to review irregularities in the
municipal account books. Moreover, in Bolívar, where the memory of
Gaitán’s assassination and the Conservative repression in its aftermath
was still fresh, disturbances on the April anniversary of the fallen leader’s
 Blood and Fire

death prompted the governor to deploy ten policemen to preempt the

spread of unrest to nearby municipalities.11
Neither escalating tensions between members of the Liberal and Con-
servative parties nor the usurpation of local authority by Conserva-
tive appointees, however, impeded predominantly Liberal southwestern
towns such as Urrao from returning a resounding victory for Liberal can-
didates in the municipal council elections of June . Seven Liberals—
four of them self-described gaitanistas—and two moderate Conserva-
tives won seats on Urrao’s town council. Temporarily emboldened by
their ability to resist Conservative pressure without resort to violence,
Urrao’s political leaders continued to express their grievances with the re-
gional government through petitions, letter-writing campaigns, and bi-
partisan meetings of local party representatives.
By September , however, peaceful attempts to oppose the re-
gional government’s conservatization campaign had come to naught.
The mayor who had been named by the governor preferred to coordi-
nate his activities with Father Zapata rather than cooperate with Urrao’s
popularly elected officials. Real municipal power began to rest increas-
ingly in the reactionary priest’s hands. He hand-picked victims for public
humiliation and beatings and directed the local police to carry them out.
On September , , Liberals and moderate Conservatives from the
town council called an emergency meeting to discuss the deterioration
of relations between the parties and the priest’s excesses. But the mayor,
whom the council had especially invited, failed to attend.12 Two weeks
later, a procession led by Conservatives from nearby towns and Urrao’s
few laureanista supporters took over the town square. Bearing aloft the
statue of the Virgin of Fatima, a small group of drunken, armed men shot
up the bust of Liberal hero Rafael Uribe Uribe in Urrao’s central plaza.13
Fights escalated in the aftermath of the Virgin’s appearance. Brawls
broke out in local taverns and served as dress rehearsals for the organi-
zation of Liberal armed resistance. Chócolo, who became a prominent
member of Urrao’s Liberal guerrillas, heckled and insulted departmen-
tal police agents and local Conservatives from the safety of the cantina
Cantaclaro in December . ‘‘Go ahead and abuse us you son of bitch
Conservatives [godos], our day is coming soon!’’ the town heard him
shout defiantly, as unsympathetic laureanistas stared. Conservative com-
plaints in the town judicial records dismissed Chócolo as an ‘‘antisocial’’
who ‘‘provokes peaceful citizens . . . inciting them to fight’’ and a ‘‘rabble-
 Urrao and the Southwest 

rouser’’ who flaunted his relationship with a prominent Liberal lawyer

and politician as proof of his impunity and power.14 But Chócolo’s out-
burst served as a warning about the growing unease of the town’s Liberal
majority whose only outlet was through raucous insults issued under the
protective guise of inebriation.15
The presidents of the local Liberal and Conservative parties, in con-
trast, were careful to eschew any incendiary action. They preferred to
respond to violence by sending the governor reports in which they ac-
cused the mayor and the police of encouraging citizens to loot stores
and attack the town’s inhabitants. They sent copies of their complaints
to regional and central authorities, Medellín’s private producer associa-
tions (, , and ), and the city’s principal indus-
trial boards (Coltejer, Fabricato, Litografía de Bédout, and Cervunion).16
Urrao’s party leaders seem to have still been laboring under the impres-
sion that a threat to the local economy would mobilize the region’s elite
against the local spread of violence. But the region’s elite was enmeshed
in its own ineffectual attempt to restrain the regional government from
engaging in partisan activities and proved deaf or too impotent to ad-
dress Urrao’s petition. Instead, those daring to make public the misuse of
authority by Conservative public appointees were met with swift reprisals
by local extremists. One individual’s office and its contents were set afire,
another’s gas station was threatened with arson, a moderate priest was
beaten up, and a prominent moderate Conservative committee member
was forced to leave town or be killed.17 In the days that followed these
events, local laureanistas destroyed or invalidated local Liberals’ national
identity cards, while Father Zapata directed the police to conduct lynch-
ings and aplanchamientos.18
When the central government declared a state of siege in November,
the president of Urrao’s town council, who was a Liberal, and his close
friend and fellow council member, a moderate Conservative, decided to
act independently of their respective party directorates and take the mat-
ter of public order into their own hands. The two men were concerned
that violence waged by the regional authorities and Conservative out-
siders would irreparably alter what had been competitive but not openly
violent relations between local members of the two parties. They also
feared that their own ability to influence events in the town was rapidly
ebbing away as regional appointees and previously marginal elements
within the town insinuated themselves into positions of power. These
 Blood and Fire

men resolved to consult the two institutions they continued to believe

could still influence government policies and the town’s fate: the church
and the military.19 Separately and secretly, so as not to arouse the suspi-
cions of the regionally appointed local authorities, each traveled to meet
the bishop of Santa Fé de Antioquia, the ecclesiastical head of Urrao’s
parish, and the commander of Antioquia’s army base in Medellín. They
begged Bishop Luis Andrade Valderrama to remove Father Zapata and
begged Colonel Eduardo Villamil to send soldiers to regulate the behav-
ior of the police.
The religious and military leaders to whom these councilmen ap-
pealed were not chosen haphazardly. Each had forged a reputation of tol-
erance and rectitude and had publicly expressed disapproval of partisan
violence. For several years prior to la Violencia, Bishop Andrade Valde-
rrama had issued letters warning parish priests under his authority to
abstain from using partisan rhetoric or the pulpit to influence politics.20
In retaliation, extremist Conservatives spread rumors that the bishop
was conspiring in a possible ‘‘general uprising led by Liberals.’’ 21 Colonel
Villamil was similarly distrusted by sectors of the Conservative regional
bureaucracy as his attitude of neutrality was misinterpreted as advocacy
of Liberal interests.22 Both Andrade Valderrama and Villamil agreed to
help Urrao’s elected representatives. The offending priest was eventu-
ally posted elsewhere, and in the interim a less partisan prelate was ap-
pointed to mitigate Father Zapata’s influence in Urrao. The deployment
of soldiers was equally remembered by locals as a significant deterrent
to police abuses. Urraeños insisted that until  when Colonel Villamil
was removed from his post in Medellín, and aerial strafing against local
guerrillas began, the army occupied a neutral, even sympathetic presence
within the town.23 But despite the attempt undertaken by Urrao’s Liberal
and Conservative representatives to mediate incipient partisan conflict in
nonviolent ways, they proved unable to prevent the emergence of Urrao
as a prime site of armed Liberal and Conservative struggle between 
and .

The Organization of Urrao’s Liberal Guerrillas

Local lore states that a man named Juan de Jesus Franco arrived in Pabón,
a southern hamlet in Urrao near the border with Betulia and Salgar, with
nothing more than the clothes on his back, a letter of introduction from
 Urrao and the Southwest 

national Liberal party leader, Carlos Lleras Restrepo, tucked in his shoe,
and the change from the bus fare given to him by Liberal party head-
quarters in Medellín.24 In June  after the Conservative party had been
toppled from power, in a letter to Antioquia’s military governor, Pío-
quinto Rengifo, Franco gave his reasons for taking up arms and his choice
of Urrao as the site for an armed insurrection. A lifelong Liberal who had
taken little overt part in party affairs before la Violencia, Franco claimed
that his political education began when he attended a meeting held at
Medellín’s Liberal party headquarters shortly after Gaitán’s assassination.
During the session, armed Conservatives attacked the Liberal offices, de-
stroyed party propaganda, and vandalized the premises. After the melee,
several Liberals were hauled off to prison. Franco was among them.
This event convinced Urrao’s future guerrilla leader that only arms
could defend his party and restore it to power. He approached the Lib-
eral directorate and asked the members to cover his travel expenses to
Urrao. Dubious, but figuring there was little to lose, the directorate sup-
posedly agreed.25 Franco’s choice of Urrao was shaped by a prior stint
in the army and police; he had once been stationed in the western town
and found its difficult, broken terrain, proximity to the jungles of Chocó,
and staunchly Liberal population advantageous for mounting an insur-
rectionary group. The story may be more apocryphal than true; despite
Franco’s signature at the end of his public apologia for taking up arms, it
was a member of Medellín’s Liberal directorate, not the guerrilla leader
himself who penned the missive to Antioquia’s military governor. Con-
cerned lest Franco ‘‘go down as a mere bandit,’’ the Liberal party officer
had felt compelled to justify the emergence of armed popular rebellion
and to reassure the regional authorities that such groups had been politi-
cally motivated, not merely criminal, and that their ideology had been
Liberal not Communist.26
When Franco arrived in Urrao he found the vestiges of an earlier
armed group organized by a man named Arturo Rodríguez. Rodríguez,
whom Liberal leaders in Medellín later referred to as a glorified bandit
(bandolero), had left about a dozen men with a few shotguns wander-
ing about the hills outside town when he abandoned Urrao and moved
to Dabeiba. Rodríguez ultimately established one of Antioquia’s most
powerful guerrilla camps (Camparusia) in Dabeiba.27 Shortly afterward,
Franco was joined by a strapping, blue-eyed, blond recruit, known affec-
tionately as ‘‘El Mister,’’ who came from the department of Valle and con-
 Blood and Fire

tributed his Liberal fervor and machine gun to the local effort to form a
guerrilla band. As news of Franco’s arrival spread throughout the region,
young Liberals who were persecuted in their towns of origin flocked to
Pabón, the village where the guerrillas established their main headquar-
ters. In addition to men from Urrao, others came from Salgar, Caicedo,
and parts of Betulia. The majority of guerrilla recruits, however, were
local men and boys from Urrao. No men, the guerrillas later fiercely in-
sisted, had joined from Altamira, the Betulia parish dominated by the
reactionary priest, Manuel Vargas, who personally recruited and armed
Conservative civilian volunteers to counter Urrao’s guerrillas.
Franco’s men eventually ranged across most of the western region of
Antioquia (see map ). Pabón’s forces were linked to guerrilla bases in
Camparusia and worked the areas along the Carretera al Mar connecting
Antioquia to Urabá, west along the border with the Chocó (with Ramón
Elías Calle’s men in El Carmen, Chocó), and south and southeast through
Salgar, Bolívar, and Caicedo. There were infrequent reports of guerril-
las sighted as far south as Jardín in the heart of the southwestern coffee
zone, and as far north as Mutatá in Urabá, but the core of Franco’s group
was concentrated in the hills and jungle between the Chocó, Urrao, and
Dabeiba. Their range of operation included Urrao, Caicedo, Betulia, Sal-
gar, Bolívar, Anzá, Santa Fé de Antioquia, Peque, Frontino, Dabeiba, and
various towns in the Chocó such as El Carmen. To great advantage, the
guerrillas established their bases in what government agents later ruefully
concluded was ‘‘a relatively inaccessible region.’’ 28
Official estimates of the size of the guerrilla group varied between
eight hundred and several thousand.29 However, former guerrillas calcu-
lated that a core group of one hundred fifty permanent men, which could
swell to three hundred or more in times of combat, was a more realis-
tic assessment of the number of men up in arms in Pabón.30 The marked
difference between these estimates of the group’s size reflects the govern-
ment’s initial failure to defeat the guerrillas. In order to excuse the inepti-
tude of government forces, particularly the police, it became necessary
to wildly exaggerate the enemy’s number, arms, and range. The guerrillas
also contributed to misperceptions about their number. To improve their
negotiating position with the central state, they claimed to have incorpo-
rated many more recruits than the actual number of men who reported
to the Fourth Brigade seeking official amnesty after the military came to
power in June .
 Urrao and the Southwest 

Map . Guerrilla operations and movement from Urrao (Source: Instituto Geo-
gráfico Augustín Codazzi; Archivo Privado del Señor Gobernador de Antioquia,
–; Archivo de la Secretaría de Gobierno de Antioquia, –)

Initially, all those who wanted to join the guerrillas were accepted. Re-
cruitment became more selective in late  and early  as the mem-
bers of Franco’s band became wary of possible spies and were unwilling
to integrate those prompted to join the guerrillas by a romantic notion
of the outlaw life or who were motivated solely by revenge. Only those
who could produce a letter of introduction from their respective Liberal
party committee—and, some said, who knew the secret handshake—
could hope to be admitted.31 The guerrillas were organized in a series of
camps, each of which was under the leadership of a single man, typically
a member of the guerrilla’s plana mayor, or junta. Each camp had between
twenty and forty recruits who operated over a specific territory. Military
 Blood and Fire

hierarchy defined the guerrilla group’s internal structure of command: at

the top a supreme commander (Captain Franco), next a group of about
twenty lieutenants and sergeants, and, at the bottom, the rank and file.32
Assignments were made on a daily or weekly basis and posted on a camp
blackboard. Officers were armed with rifles or automatic pistols; rank and
file soldiers used shotguns or machetes. Only officers had ‘‘official’’ caps
(kepis) and sported police or army fatigues, the other men were dressed
as paisanos, that is, in common rural civilian dress: ruana, khaki pants,
white shirt, and felt or straw hat.33
Franco’s men built a foundry to repair and manufacture arms, ap-
pointed a number of male cooks who never engaged in combat, and em-
ployed male nurses. Medicine and food were prepared and stored in the
guerrilla camps but were rationed out to individual guerrillas so that they
could take them home to share with their wives and children. The guer-
rillas set up mobile rancherías (groups of huts) a short distance from
headquarters so that women remained in constant contact with their hus-
bands, lovers, and kin up in arms. Supplies, clothes, ammunition, and
smuggled goods were kept in the rancherías, as well as sewing machines
for the manufacture of uniforms. A distinctive feature of Urrao’s guerril-
las—one not found in any of the other guerrilla forces operating in either
Antioquia or the rest of Colombia—was the incorporation of Catholic
ritual into daily guerrilla life. Guerrilla encampments boasted a portable
altar and shrine for the three figures of guerrilla devotion: the Virgin of
Mount Carmel, the miraculous Child of Buga, and the Holy Trinity.34 Be-
fore each battle the guerrillas would gather around the makeshift altar
and say the rosary or recite the trisagio.35 One member of the guerrillas
was always on duty to ensure that the altar was carried away to safety
should government forces ever attack the camp. Many years later mem-
bers of the group insisted that they had lost so few men despite the heavy
militarization of the region because ‘‘God was on our side and the devil
was on theirs.’’ 36 Father Ramirez, who eventually replaced Father Zapata,
visited the camps to say Mass, conduct baptisms and marriages, and con-
sole the guerrillas. (Disapproving Conservatives said the priest drank and
danced with the guerrillas, too.) With the exception of the male nurses
and cooks—employed rather than women to ‘‘avoid possible jealousies’’
among the men—the guerrillas reproduced the coordinates of rural do-
mestic life within the camps.37
The core of the armed organization was made up of clansmen. Seven
 Urrao and the Southwest 

Urrego brothers from Caicedo, who had moved to Urrao with their sis-
ters (one of whom, Graciela, became Franco’s common-law wife) were
joined by families such as the Cañolas and Cartagenas. The list of guer-
rillas incarcerated after la Violencia in ‘‘La Ladera’’ prison in Medellín or
listed in Urrao’s parish registry of the deceased is rife with networks of
uncles and nephews, sons and fathers, brothers and cousins. They joined
up all together or individually as they rotated between guerrilla activity
and scratching out a living as farmers and day laborers ( jornaleros). Re-
cruitment and conflict frequently grew out of long-standing family feuds.
Liberal Urregos, for instance, joined Franco, while their longtime ene-
mies, the Cossios and Montoya Montoyas from Caicedo, made up the
ranks of the police and Conservative contrachusma in nearby towns.38
The intricacies of kin-based loyalty—in which many families included
members affiliated with both parties, especially through marriage—and
dispute meant that Conservatives sometimes joined the guerrillas as well.
The men who formed part of Urrao’s guerrillas were well and widely
known to the local population. A large number of the town’s families had
at least one near or distant relative involved in guerrilla activities. When
testimony was taken from civilians who had had contact with Franco’s
band, they frequently referred to the guerrillas by name, situating them
within established local family networks and neighborhoods. One Urrao
farmer noted, for instance, that when he was stopped by guerrillas who
patrolled the rural areas and who exacted a kind of tax on goods trans-
ported to and from peasant fields, he ‘‘was perfectly familiar with’’ three
of the five men who accosted him.39
Urrao’s womenfolk played a pivotal role in maintaining the link be-
tween civilian and armed resistance in ways that seem not to have been
reproduced to such a degree in any of the other guerrilla occupied areas
within Antioquia. Women served as spies, messengers, providers of food
and clothing, carriers of arms, and informal sentries.40 They penetrated
the police lines when the town was occupied, shifting critical information
across a broad network of rumor and gossip that wrapped the town in a
close-knit relationship of conspiracy and complicity. Like the soldaderas
(camp followers) of the Mexican Revolution, Urrao’s women followed the
guerrilla camps from place to place ensuring that life was organized like a
household, where religious devotion, regular conjugal relations, and ritu-
alized festivities reproduced local customs and formed an integral part of
war.41 In the lulls between armed encounters, women and men held the
 Blood and Fire

threat of death at bay with boisterous drinking and dancing enlivened by

impromptu music strummed on rough-hewn instruments. Women were
so intimately a part of guerrilla organization that government spies sent
to observe the composition of the guerrilla camps around Urrao reported
with astonishment the presence of ‘‘at least fifteen women.’’ 42
The omnipresence of women among Urrao’s guerrilla forces appears
to have ensured the maintenance of social and sexual norms of conduct
dictated by local custom.43 Unlike other areas of Antioquia, for instance,
guerrilla warfare in Urrao appears not to have involved the systematic
use of rape or sexual mutilation. In areas where guerrillas were made up
largely of nonlocals or forasteros and where the battlefront and home-
fronts were divorced from each other, in contrast, rape, sexual abduction,
and mutilation appear to have been common.44 Much of the guerrillas’
popular support was contingent upon avoiding rape or torture in their
treatment of civilian enemies and the local population. Indeed, Urrao’s
guerrillas became known for sparing women and children in villages or
settlements they attacked—often warning them in advance of an attack—
and concentrating their violent activities against male heads of household
and the police.
Women within Urrao who had no direct contact with the guerrillas
and who felt ambivalent about the use of popularly led violence against
a legally elected (if disliked) government, agreed that the guerrillas had
distinguished themselves from many of the government’s troops precisely
because they did not widely engage in rape. The daughter of the local Lib-
eral Party chief, who noted with distaste that the guerrilla leader Franco
was a swarthy, impulsive, hard-drinking man, nonetheless pointed out
that he was able to maintain discipline over his followers. In fact, Franco
imposed the death penalty upon members of his band who took part in
the two acts which he considered violated the norms of legitimate war-
fare: rape and torture.45 Although this draconian measure met with re-
sistance within the guerrilla group and was not consistently obeyed, it
constituted an important way in which the guerrillas could differenti-
ate themselves from the authorities.46 This is not to say that rape did not
occur, but rather that a widespread perception existed—corroborated by
testimony and judicial records—that it was a tactic more frequently em-
ployed by the authorities than the guerrillas. At the very least, rape was
punished harshly by the guerrillas.
Cross-class cooperation and support for the guerrillas in the interests
 Urrao and the Southwest 

of local and partisan defense was another critical component of Urrao’s

experience during la Violencia. That is, while cross-class alliances and col-
lusion certainly characterized relations between members of the Liberal
guerrillas in eastern Antioquia, Urabá, and western Antioquia, in Urrao
the end result of this was not the advancement of the interests of spe-
cific landlords or merchants but the defense of the town and the Liberal
party. A woman of elevated social class within Urrao remembered wait-
ing every night until her husband fell asleep and then crawling across the
floor of their bedroom to the balcony that faced the town’s main street.
Outside, police sentries posted at the doors of prominent Liberal homes
would chat with each other and let slip strategically important informa-
tion about troop deployments, arms, and reinforcements. The woman
would eavesdrop, memorize the information, and pass it on to her sis-
ter who in turn passed it on to members of the guerrilla group. Another
sister, known as a local Mata Hari, made it her duty to flirt with local
policemen in the hopes of extracting valuable information.47
In this manner, the web of complicity was drawn to incorporate nearly
everyone in the town. The Sisters of the Poor, who ran the local hospital
and dispensed medicine, sent the guerrillas penicillin and other drugs on
the sly. Local merchants—before the town was so heavily policed that it
became impossible to smuggle out food or merchandise—sold the guer-
rillas matches, candles, lard, cigarettes, and cloth, while local farmers,
including several Conservatives, sent the guerrillas food and molasses
loaves. Even children had a role to play as they used their routine ubiquity
to drift between the police station, the army headquarters, the school, the
streets, and home, picking up information, observing details, acting as
the town’s ‘‘ears.’’ 48 So pronounced was the degree of everyday conspira-
torial participation in Urrao that the regional authorities came to regard
the townspeople in their entirety as enemies. Nowhere else in Antioquia
did visitadores or army officials who were sent to analyze the situation
of violence report as they did for Urrao that ‘‘the majority of the local
people are untrustworthy since they are all partisans of the bandoleros
who inhabit the hills surrounding the municipality and who they abso-
lutely adore.’’ 49
Even if it were not true that everyone in the town ‘‘adored’’ the guer-
rillas, the very fact that the authorities thought so and assumed that the
town as a whole could not be trusted, encouraged and reinforced a sense
of local identity and collective purpose. This sense of collective involve-
 Blood and Fire

ment enabled local inhabitants to justify having taken up arms against

the government. This local sense that they were invested in a just cause
and that it had been the government that had left them no option but to
take up arms extended beyond Liberal party lines to incorporate Con-
servatives who refused to actively endorse the use of violence for parti-
san purposes against their neighbors. The Conservative in charge of the
local airport—a critical point of contact and access between Medellín,
the cattle-rich towns around Urrao, and strategically placed Urabá—was
removed in  because his lifelong friendship with several prominent
local Liberals made him suspect of complicity with them.50 Conservative
telegraphers who were thought to sympathize with local Liberals suffered
the same fate. For the first two years of armed struggle (–), the
tight interaction between guerrillas and the civilian population made it
nearly impossible for the regional administration and its troops to locate,
penetrate, or attack the guerrilla groups. Even when the regional authori-
ties succeeded in choking off outside support to the guerrillas and re-
duced them to intermittent defensive attacks by , the struggle be-
tween guerrillas and authorities never advanced beyond one of attrition
and stalemate; neither side could win an outright victory.

The Initiation of Guerrilla Conflict

The first incidents of organized guerrilla violence in the southwest took

place not long after the November  presidential election and Chó-
colo’s drunken outburst in the Cantaclaro bar. On December , several
dead and wounded Conservative peasants were found on outlying farms
that had been burned to the ground in the vicinity of Concordia and
Betulia.51 The event set off a battle between the soldiers and the national
police troops who were sent to investigate the crime and pursue the guer-
rillas.52 This blatant show of internal dissension within the government’s
forces encouraged the guerrillas to think that undermining the Conser-
vative government would prove a relatively simple affair. Soon after this
incident, rumors similar to those which had become common in west-
ern Antioquia and Urabá flourished. In towns such as Salgar and Betulia,
rumors alleged that the army was sympathetic to the Liberal rebels and
that the police were the minions of the Conservative government.53
The national police’s extensive use of intimidation and physical vio-
lence (beatings, armed assaults, and arson) during the November presi-
 Urrao and the Southwest 

dential elections was the immediate catalyst for the armed retaliatory
Liberal attacks which began against Conservatives in December .
The behavior of the police and their Conservative civilian aides even
prompted the commander of Medellín’s Fourth Brigade, Colonel Ed-
uardo Villamil, to advise the minister of government in Bogotá—be-
cause his requests to Governor Eduardo Berrío had been ignored—of
the need to remove the departmental police agents stationed in Betu-
lia and Venecia. These policemen were led by Captain Arturo Velásquez,
an officer who made no bones about his Conservative sympathies and
who the military commander noted ‘‘had been [in the area] too long,
making it impossible to pacify the town without replacing the agents re-
sponsible for leading local abuses.’’ With the support of Conservative
civilians, drunken national police agents had on several occasions fired
their guns at the unarmed Liberal citizenry several times until, finally, the
army interceded to disarm them.54
During the first half of , violence around Urrao was limited to
periodic assaults on coffee and cattle haciendas in Urrao proper and the
corregimiento of La Camara in Salgar and did not involve warfare between
armed groups of Liberal and Conservative antagonists or the guerril-
las and the armed forces. The guerrillas forcibly dislodged Conservative
peasants from areas such as La Camara and took over their properties.
Franco’s men also secured control of Puerto Arquía on the Chocó bor-
der and killed the Antioqueño authorities stationed to monitor the entry
of contraband through the port.55 These initial acts of violence met with
no organized government response because the number of police and
army troops stationed in Urrao was insufficient to allow for deployments
into the countryside. Instead, the mayors of Betulia, Salgar, and Con-
cordia gathered together national policemen, one hundred Conserva-
tive civilian volunteers, customs agents from Betulia, and policemen and
Conservative civilians from Urrao and stationed them along the Betulia
and Urrao border.56 These forces were used to keep at bay guerrillas in
Pabón after rumors began to circulate in late June  that they were
planning a major attack.57
By July, however, the expected guerrilla attack had failed to material-
ize. Instead, Pabón’s guerrillas focused on rustling cattle from Conserva-
tive haciendas such as that owned by Francisco Ospina Pérez, the Colom-
bian president’s brother, in Concordia.58 In the meantime, the mayor of
Betulia (where no guerrilla actions had yet occurred) took it upon himself
 Blood and Fire

to appoint and arm additional ‘‘civic’’ policemen, that is contrachusmas.

Colonel Villamil made it clear to Governor Berrío that the mobiliza-
tion of contrachusmas in Betulia constituted an unnecessary provocation
since in his opinion there were more than enough national policemen
and soldiers assigned to the area to oversee public order matters.59 The
guerrillas, in any case, had not yet engaged in any major confrontation,
thus avoiding actions that might force them into a showdown with the
government’s forces. Instead, they continued to force Conservative peas-
ants off lands that were located in areas of difficult access, selectively am-
bushed the police and stole their arms, and penetrated and took control
of important coffee plantations in the settlements of El Planchón, La Mar-
garita, and La Camara in Salgar.60 The guerrillas were biding their time,
building up their store of arms, gaining control of territory, and eliminat-
ing local Conservative men who might be recruited by the contrachusma.
Unlike the sporadic guerrilla attacks that took place during this period
in eastern and western Antioquia, Urrao’s guerrillas targeted members
of the opposition and estates, not government offices or officers for at-
tack. By late July, the local Conservative committee reported that Liberal
guerrillas operating in La Camara numbered two hundred and that they
were well armed as they had attacked ‘‘defenseless’’ Conservatives and
their single police escort with a machine gun.61
By August , Conservatives in Urrao and Salgar were in a panic. They
demanded that the government immediately deploy one hundred fifty
men to defend the area and its economy from guerrilla assassinations of
Conservatives, the usurpation of coffee estates, and cattle-rustling.62 But
despite his earlier assurances of sufficient personnel to oversee the region,
the commander of the Fourth Brigade was unable to provide more than
fifty national policemen to the town because of all his other troops were
already deployed elsewhere in anticipation of violent protests against
Laureano Gómez’s inauguration as president. The guerrillas had astutely
waited until just before the inauguration to strike, knowing that Liberal
protests were likely to explode in numerous places and that the govern-
ment’s forces would be reduced and dispersed. The military commander
was forced to concentrate his few men within the urban perimeter of the
towns likeliest to erupt in violence, leaving the guerrillas free to act unim-
peded in the countryside. The commander could do little otherwise, for
as he pragmatically noted, ‘‘with the existent personnel at my disposal we
can only attend to the public defense needs of the urban center and under
 Urrao and the Southwest 

no circumstances the hamlets or villages where the bandoleros are actually

located.’’ 63 Meanwhile, fear of property attacks similar to those occurring
daily in Salgar and Urrao convinced other southwestern towns that it was
pointless to rely on the regional or central government to supply troops.
They requested and bought arms from the regional government to arm
contrachusmas instead.64
Numerous assassinations of Conservatives occurred in and around
Urrao in early August  at exactly the same moment in which Lib-
eral guerrilla forces in the Bajo Cauca, Urabá, and western Antioquia
went on a similar spree of killing and house burning. Indeed, from this
moment on, every major guerrilla attack that took place in Urrao or its
immediate environs was echoed in similar guerrilla attacks throughout
Antioquia.65 The national police and their contrachusma aides retaliated
against the wave of assassinations by killing ten Liberal peasants in the
hamlet of Guasabra in the western municipality of Santa Fé de Antio-
quia, but they did not attempt to penetrate guerrilla-dominated territory
in Urrao itself.66 Two weeks into August, another fifteen Conservatives
were assassinated in rural Urrao, while more Conservatives were forced
to give up their farms and goods to the guerrillas.67 When the number of
Conservative refugees seeking asylum within Urrao’s town limits swelled
to more than a hundred, local Conservatives impatient with the govern-
ment’s failure to attend to their pleas for help threatened to ‘‘mobilize
two thousand men in the southwest to avenge the blood of our fellow
party members.’’ 68 Then, as suddenly as guerrilla attacks had begun, they
stopped. No further violent incidents were reported during the next three
months, which was the period of the main coffee harvest season.
The guerrilla’s activities during the early days of August set the tempo-
ral pattern and general character of future violence in the southwest. At-
tacks occurred most frequently in July and August before the onset of the
main coffee harvest—a period in which large numbers of men were with-
out jobs or partially unemployed and beginning to be desperate for food
and money. Guerrilla activity was next most intense in February, March,
and April, just before and at the onset of the travesia, during the coffee
crop’s second flush. The second harvest typically employed less than half
the labor force of the main harvest. Guerrilla activities were preceded
by the boleteo (written messages sent to threaten potential victims and
exact a forced contribution); the penetration of isolated or unpatrolled
villages and settlements; the confiscation of land, arms, and food; wide-
 Blood and Fire

spread cattle-rustling; and selective assassination and arson. If the vic-

tims of guerrilla aggression were police agents, they were typically shown
no mercy. Before entering an area and burning it down, guerrillas some-
times advised Conservatives to leave if they were not ‘‘sapos’’ (‘‘toads’’
or informants) for the regional government or contrachusma sympathiz-
ers. Peasants too poor or unable to travel to Medellín to seek refuge were
forced to seek asylum within the nearest town.69
Guerrillas were slower to consolidate their presence and activities in
Urrao than were guerrilla groups elsewhere in Antioquia, but once they
did, violence quickly escalated and spread. In December  guerril-
las resumed their assassinations of Conservative peasants in Urrao and
nearby Betulia and then attacked Urrao’s army barracks and the town
itself in the early hours of the morning of December .70 Nearly two
weeks passed before the government was able to lead a retaliatory at-
tack, but on December  the military strafed and mortared the area of
Guasabra in Betulia where guerrilla forces had pushed out Conservative
peasants and established a base of operations. Unfortunately, as the mili-
tary itself admitted some time later, the deployment of heavy armament
against the guerrillas proved largely ineffective. The logistical implica-
tions of transporting men and arms to guerrilla-occupied areas made the
government’s approach known to everyone in the area long before the
actual initiation of military activity. By the time troops arrived in Betu-
lia, the guerrillas had melted into the surrounding hills and taken their
stolen livestock with them.
Capitalizing on the weakness of the government’s forces and their in-
ability to penetrate into isolated regions of guerrilla support, Franco’s
men aggressively spread and consolidated their presence and control over
a broad territory. The guerrillas occupied the settlements of Orobugo,
Santa Isabel, Pabón, and San José in Urrao and took control of north-
western Betulia during the first half of . By May of that year the guer-
rillas had bases in the neighboring department of Chocó, up through
the region of Urabá, and in the southwest. Reports of killing, arson, and
theft increased accordingly, although the intensity of these varied from
area to area. Betulia complained that the guerrillas recruited young peas-
ant men, engaged in widespread cattle theft, extorted and intimidated
wealthy townspeople, and went on rampages burning the homes of Con-
servative peasants and resettling Liberal peasants or their own guerrilla
recruits in such areas.71 But in Salgar and Concordia, complaints regard-
 Urrao and the Southwest 

ing the presence of the guerrillas focused primarily on their extensive

cattle-rustling and coffee theft rather than on the assassination or dis-
placement of peasants.
The elimination of members of the opposition from particular ham-
lets appears to have obeyed the logic of personal feuds, partisan differ-
ences, and intermunicipal rivalries, and it was in the context of forcibly
displacing individuals from such contested areas that the greatest inci-
dence of brutality was reported. On two separate occasions in ,
Antioquia’s Conservative party representatives prefaced their pleas for
government action, additional judges, and public order personnel with
descriptions of guerrilla barbarity directed against their party members.
Summing up the events of the year, they reported that by late December
twenty Conservatives had been assassinated in Urrao during the preced-
ing months and that the preferred guerrilla tactic was to invade regions
and kill ‘‘everyone,’’ women and children as well as adult men. Guerrillas
were accused of raping and sexually mutilating women in the presence
of their families. The killings were done ‘‘with a truly horrifying degree
of cruelty and butchery.’’ 72
Without denying the veracity of these reports, I found that explicit
references to sexual violence and mutilation were rare in official reports,
judicial testimony, or the complaints regarding violence perpetrated by
Urrao’s guerrilla forces that were forwarded to the governor’s office in
Antioquia.73 In contrast, complaints of extortion, cattle-rustling, and the
forced displacement of people from settlements that were about to be
burned to the ground were the most commonly reported incidents of vio-
lence. Indeed, when more sensationalist accounts were invoked in reports
of public disorder, they tended to constitute a rhetorical strategy, a pre-
lude to far more detailed and lengthy complaints of economic violence.74
The report of twenty dead Conservatives in Urrao, for instance, dedicated
less than a paragraph to mourning the victims of violence or describing
what had happened to them but spent more than two pages detailing the
economic and social effects of tolerating such violence. The guerrilla ac-
tions ‘‘seriously injured the local economy because of the abandonment
of rural areas [and] harvests’’ while ‘‘the considerable number of people
left without employment [who] suffer the rigors of hunger and misery’’
constituted ‘‘a real social problem.’’ ‘‘Banditry,’’ moreover, had begun to
spill over into adjacent municipalities. The complaint’s principal con-
cern was once more the economic effect of violence: ‘‘several incidents
 Blood and Fire

of coffee looting had already occurred, as had orders to hand over sig-
nificant sums of money.’’ 75 Affluent hacienda owners from Urrao, for in-
stance, had begun to relocate to Betulia to seek protection in what had
emerged as a contrachusma stronghold. The corregimiento of Buchadó,
on the Chocó side of the border with Antioquia, Salgar, and Concordia,
and the hamlets along the Carretera al Mar also reported similar incidents
of economically motivated out-migration.76
The few local accounts (typically written by Conservatives) alleging
acts of barbarity or torture that were sent to the regional government
tended to be viewed with some skepticism, perhaps because conflicting
versions so frequently surrounded a single event. Regional authorities
also appear to have suspected that exaggeration and sensationalism were
strategies used by local Conservatives to justify demands for arms and
the right to organize paramilitary organizations that might well exceed or
escape the control of either the party or the regional government.77 Betu-
lia’s request for one hundred fifty reinforcements and the appointment
of a ‘‘trustworthy’’ army officer—a ploy characteristically used to replace
appointed officers who would not conform to local civilian demands—
after a guerrilla attack spilled over into the municipality, for instance,
elicited a restrained response from the commander of the Fourth Brigade.
He advised his second in command in the area to ‘‘safeguard your inde-
pendence, don’t allow yourself to be counseled by civilians who, in order
to be believed by you, should present evidence and not mere rumors.’’ 78
In his instructions to his subordinate, the army commander touched
upon a sensitive issue at the heart of public order matters and violence in
Antioquia. The interests of those affected and taking part in la Violencia
were many and complex, and gossip and rumor created an atmosphere
of confusion and distrust. Death might be swift, but reliable firsthand re-
ports describing the conditions in which it occurred were few. Days or
even weeks might elapse before the regional or central governments re-
ceived news of local affairs, and a similar amount of time might pass be-
fore government agents or troops were deployed to the scene of violence.
Delays and conflicting reports made the identification of the perpetra-
tors of violence nearly impossible. Confusion led the governor of Antio-
quia to complain in private to his friend the minister of foreign affairs in
March  that he no longer knew what violence was about. Liberals and
Conservatives attacked their own party members as well as the opposi-
tion, reports of the size and reach of guerrillas were unreliable and con-
 Urrao and the Southwest 

stantly shifting, and internal divisions within the party made consensus
about public policy impossible.79 Three months later when Urrao’s Con-
servative committee president berated the governor for ‘‘doing nothing
to stem the tide of violence’’ affecting the southwest, Governor Henao
responded in frustration that it was difficult to know exactly what to do
‘‘since our fellow party members in areas affected by violence suggest dif-
ferent strategies not only regarding the deployment of the armed forces
but also with regard to the tactics these should adopt.’’ 80
In a rough draft of a letter to President Laureano Gómez in the same
year, the governor was forced to acknowledge, moreover, that his re-
peated orders that public officials be impartial and respect the opposi-
tion’s civil rights were routinely subverted or ignored.81 For example, in
April  the governor had ordered the engineer in charge of overseeing
public road construction between Urrao, Betulia, Concordia, and Bolom-
bolo to speed up work on the section between Concordia and Bolom-
bolo as a way of facilitating troop deployments to the southwest. Betulia’s
mayor and police chief, however, arrested the workers and threatened to
kill them because they were ‘‘Urraeños-Paboneños’’ (i.e., suspected Lib-
erals and guerrilla sympathizers). When the engineer confronted Betu-
lia’s authorities about disobeying the governor’s orders, the mayor bel-
ligerently responded, ‘‘that I should come and complain in person and
then I’d see that the same thing would happen to me.’’ The mayor then
sneeringly instructed the engineer to tell the governor that ‘‘if the gov-
ernment is interested in finishing the section of road between Bolombolo
and Concordia, they should request personnel from the mayor and he’d
have these ready in half an hour, but that he wasn’t about to accept foras-
teros around there.’’ 82
A city-bred professional had run up against the ugly underbelly of
power at the municipal level. There, petty tyrants like the ‘‘group of police
agents [polizontes] who radiate aggression everywhere’’ ruled with im-
punity and could submit regional appointees like the engineer to the hu-
miliation of having his belongings ransacked ‘‘by a bunch of illiterates.’’
As the engineer (who was a Conservative) sadly concluded, ‘‘you’d have
to have experienced the pain to understand what it felt like.’’ 83 Little won-
der then that governors like Braulio Henao did not know whom to be-
lieve and how many men to deploy in defense of public order. Towns in
the same region could report that the guerrillas ‘‘are not very numerous’’
and six months later that they numbered a thousand.84 Indeed, military
 Blood and Fire

strategists eventually came to suspect that such conflicting rumors were

either a guerrilla strategy to distract and disperse the government’s lim-
ited forces in order to facilitate guerrilla attacks on unpatrolled areas or
they were the ploy of local authorities to further their own private ven-
dettas and interests.

Growing Fears of the Economic Effect of Violence

In August , largely in response to events in Urrao, Governor Braulio

Henao Mejía sent an urgent request for arms to the minister of govern-
ment in Bogotá. Less than two-thirds of the department’s national police
force was armed ( out of ,), and many of its weapons were out-
dated or unserviceable.85 Four months earlier, the governor had notified
the president that the soldiers stationed in Urrao needed six new auto-
matic rifles to meet the quality and number of guerrilla arms.86 An in-
ventory of arms provided by an ex-guerrilla and cross-referenced with
the list of arms submitted by the guerrillas when they sought amnesty in
 confirms that Franco’s men were well provisioned. They possessed
machetes, peinillas,  escuadras (automatic pistols),  revolvers, 
shotguns,  rifles,  machine guns, and  boxes of ammunition.87 It was
also rumored that Liberals obtained contraband arms from the Carib-
bean through Urabá and that the department of Valle had sent Urrao’s
Liberals a machine gun to aid in the area’s defense.
By , the fears of economic decline in the heart of the department’s
coffee region and the anxiety that the guerrillas were better armed than
the armed forces—and also were rapidly gaining ground in the southwest
where the regional government had historically regarded itself as exerting
effective control—spurred the regional government to take the threat of
armed insurrection in the area very seriously. While the number of people
killed was not spectacular, the loss of cattle was. Guerrillas reported with
some glee that they had survived for an entire year on the cattle con-
fiscated from the haciendas of their former enemy, Father Zapata, and
former president Mariano Ospina Pérez (whose family owned land in
Urrao).88 In the same month Urrao’s Conservative committee (which had
been keeping careful records and periodically updating them for the re-
gional authorities) insisted that , head of cattle had been stolen in
Urrao alone,  houses had been burned and,  Conservatives had been
murdered since late .89 Similar calculations prompted the Caja de
 Urrao and the Southwest 

Credito Agrario, Minero y Industrial to insist that the only way to restore
order in the area would be to deploy one thousand men to patrol the area
in and around Urrao. Had such a measure been implemented, it would
have tied up almost all of Antioquia’s available troops.90 Instead, the re-
gional government settled on a two-pronged course of action. Cannon
and machine-gun fire had proved ineffective in stopping the guerrillas,
and the government’s forces had been unable to penetrate the enemy’s
networks of local support.91 The governor decided that the army should
shift tactics and use military planes to bomb guerrilla camps while im-
posing severe restrictions on the sale of all goods and the movement of
cattle in the region.
On September , the army bombed Bocas de Peque to stop a guerrilla
retreat from the area.92 Two weeks later, a joint operation by the air force,
army, and national police used planes to bomb and strafe a guerrilla-held
area near ‘‘La Despensa.’’ Stolen cattle were recovered, but the govern-
ment’s forces found only women and children in the area and no guerril-
las.93 On November , aerial strafing was directed against the guerrillas’
main camp in Caríazul on the Cerro de Pabón.94 This time, the guerril-
las knew in advance about the government plan to bomb them, and they
reacted accordingly. Franco spread his people throughout the targeted
area and used cattle as decoys. The bombs killed eighty head of cattle, but
only three humans. The army claimed a victory even though no guerrillas
were captured, while the guerrillas congratulated themselves on foiling
the government’s attempt to destroy them and their followers.95
The initiation of bombing in  marked a shift in the struggle
between the guerrillas and the government. Any real or purported
secret agreement to avoid confrontation between the guerrillas and army
leaders stationed in Urrao came to an end. Franco was said to have de-
clared that from now on ‘‘they [the guerrillas] were to shoot at any hel-
met [cachucha],’’ and the army expanded its definition of bandits to in-
clude anyone who might be perceived to aid the guerrillas or disagree
with the Conservative government.96 Increasingly, moreover, a division
of labor and tactics began to characterize official responses to guerrilla
violence. The military concentrated its efforts upon eradicating guer-
rilla camps and supply lines, bombing them or conducting pincer move-
ments involving the coordination of troops in several municipalities, but
avoided one-on-one retaliatory sweeps in the towns affected by guerrilla
action.97 The police and civilian Conservative forces, on the other hand,
 Blood and Fire

tended to respond to guerrilla attacks by selectively targeting Liberal-

dominated hamlets and settlements regardless of whether or not these
were areas where guerrillas were present or had recently struck. After
a series of guerrilla assaults on hamlets in Urrao in August , for
instance, police and contrachusmas attacked supposed subversives and
Liberal civilians in towns such as Salgar and Betulia where no guerrilla
actions had taken place.98
During the main coffee harvest of , guerrilla and government mili-
tary activity ceased as it had the year before, but on January , , a
troop of ‘‘six hundred guerrillas’’ resumed their activities by attacking El
Yerbal in Betulia. Two weeks later, twenty-five guerrillas attacked a Con-
servative hacienda in Urrao, destroying the house and killing the Conser-
vative mayordomo. Next, on February , the guerrillas killed seven sol-
diers in Pabón, burned down eight houses near the town of Betulia, and
torched six more houses and assassinated two conservatives near More-
lia in Concordia. The guerrillas culminated their spree in early March
by leading an ambush on the government’s forces in Urrao, killing three
soldiers, three policemen, and wounding five others.99 No government
action against the guerrillas took place until mid-March, nearly a month
after the guerrillas first initiated their attacks. The government’s failure
to respond to guerrilla assaults on the region prompted Urrao’s Conser-
vative party committee to complain that the death toll of Conservatives
killed by the guerrillas in Urrao had risen to seventy-six and to denounce
Governor Braulio Henao Mejía’s administration for being incapable of
defeating the guerrillas.100
As was true of Conservatives in western Antioquia, southwestern Con-
servatives had been pressing since the time of Gaitán’s assassination for
the right to organize their own armed groups as a way of responding
more rapidly to guerrilla attacks. These groups were already active by 
in towns such as Betulia, but despite frequent threats to ‘‘take matters
into their own hands,’’ few permanent contrachusma groups had actu-
ally emerged in the southwest between  and . Municipal requests
for arms or for additional logistical support to expand already estab-
lished contrachusma forces were given added impetus in , however,
when local Conservatives—like their counterparts in eastern and west-
ern Antioquia—began to lose faith in the state’s ability to defend them
or their properties. Conservative patience with the regional governmen-
tal response to guerrilla attacks wore out around the time that Governor
 Urrao and the Southwest 

Braulio Henao Mejía also began to despair of ever possessing sufficient

troops to adequately patrol the region.
Conservatives in the southwest were also increasingly likely to view
the national police with whom they shared a common political affiliation
as cowardly, poorly armed, and too intent on lining their own pockets
to be reliable allies in the struggle against Liberal guerrillas. Conserva-
tive charges that the police were ineffectual and corrupt were confirmed
just a short time after the guerrilla attacks of early . The townspeople
of Salgar accused the police of ignoring the guerrillas but assassinating
the mayor and engaging in extensive robberies against haciendas in both
Salgar and Bolívar. Locals then interpreted the inadequacy of state forces
through the lens of regional and ethnic xenophobia. The problem, local
officials insisted, was that the police were not Antioqueños: ‘‘When the
police were made up of men from Antioquia who were drawn from the
same areas in which they were posted to serve, these sorts of assassina-
tions and robberies never occurred nor was anyone’s life ever vulnerable
as it is today when one sees every policeman as a potential murderer or
thief. This state of affairs can only be explained by the fact that the police
are not from Antioquia and are brought from other parts of the coun-
try where the life of the citizenry has no value.’’ 101 A month later, when
twenty soldiers were found dead in Concordia, local Conservatives in-
sisted that they not be sent any more ‘‘outsider’’ policemen but rather be
given twenty rifles with which to arm themselves.102
Betulia’s Conservative citizens, moreover, devised what they consid-
ered to be a perfect plan to defeat the guerrillas that avoided dealing with
the police or the military, which they did not trust. They proposed giving
‘‘honorable people’’ a shotgun so that they might defend themselves, and
they pushed for passage of a law that would exempt self-defense killings
from being tried as homicides. They also demanded that rural Liberals
be denied any commodities not produced on the land itself (such as salt,
clothing, and cigarettes). This would make them suffer just as Conser-
vatives displaced from their homes did. They insisted as well that any
Liberals found to act as spies for the guerrillas be treated as criminals just
as the guerrillas were.103
The growing desire for revenge made the struggle between govern-
ment forces and the guerrillas more brutal. In an encounter between
policemen and guerrillas in the corregimiento of El Yerbal in Betulia in
March , men led by Franco’s right-hand man, the guerrilla leader
 Blood and Fire

‘‘El Mister,’’ shot three police and tore out the eyes and tongue of one of
them.104 Such acts made it politically more difficult for Governor Henao
Mejía to resist local demands that he distribute arms ‘‘so that locals might
form chusmas and gangs [pandillas] that would perpetrate the same ac-
tions as the Liberals do and, in this way, apply the law of an ‘‘eye for
an eye’’ [la ley del talión],’’ as one disapproving Conservative public em-
ployee in Titiribí put it.105 Whether he was aware of the danger of endors-
ing what amounted to an official license to mete out justice or not, the
governor eventually capitulated to local demands for arms. Both Con-
cordia and Betulia were among those receiving government-issued guns
later that month.106
The strengthening of paramilitary responses to guerrilla attacks ini-
tially had the same effect in the area surrounding Urrao as it did in other
parts of Antioquia: it increased rather than lessened the intensity of local
conflict. With only a brief respite in July, Urrao’s Liberal forces led as-
saults against La Camara in Salgar, several different hamlets in Betulia
and Urrao, Morelia in Concordia, and the area of Anocozca between
Urrao, Abriaquí, and Caicedo. They cleared these areas of Conservative
settlers between June and August, right before the beginning of the main
coffee harvest, in a deliberate effort to create a broad swath of exclusive
Liberal-controlled territory that emergent contrachusma forces would
find impossible to penetrate. The government, as usual, did not immedi-
ately respond to these attacks. Instead, when the guerrillas returned once
more to their bases to concentrate on the coffee harvest in October, the
military organized a pincer movement to block the guerrillas’ escape
route from Guasabra in Caicedo to La Encarnación in Urrao. This netted
the government  head of stolen cattle, enabled them to destroy all the
guerrilla camps and safe houses along the way, and to kill forty-five men
up in arms.107
At this point, the struggle in Urrao developed in directions that pro-
foundly affected the outcome and long-term consequences of violence in
the region. The elimination of Conservative men who might have swelled
the forces of contrachusma groups based in Betulia, Caicedo, and Con-
cordia ensured that Conservative paramilitary organizations were never
able to achieve the influence, size, or impact that such forces grew to have
in western or eastern Antioquia. By killing men, the guerrillas capital-
ized on the weaknesses of the patriarchal family-farm model that domi-
nated in the southwest. A highly individualistic, extensive nuclear family
 Urrao and the Southwest 

made up of a husband, wife, and many children was the dominant form
of household organization in the areas of traditional settlement within
Antioquia. Once the patriarch of such a family and his eldest sons were
killed, no one remained to exact revenge but a widow who was encum-
bered (in most cases) by several dependent minors. Unable to work or
defend the land, women and their children typically abandoned it and
migrated. This strategy of clearing potential sites of opposition proved
so successful that eventually policemen under the direction of officers
such as Arturo Velásquez adopted it as well. Conservatives in Altamira in
 complained that  widows and their dependent children had been
forced into exile in the corregimiento after the police killed all the Liberal
men in nearby San Mateo.108
The guerrilla policy of killing and displacing Conservative families
and forcing survivors to migrate and abandon their lands might have
been expected to create or renew an upswell of local support for official
attempts to combat the guerrillas. But excesses committed by national
police troops deployed in Betulia, Salgar, and Caicedo, and the police’s
tendency to respond vindictively rather than strategically to guerrilla
assaults, introduced an element of uncertainty and random terror that
ultimately undermined local support for the government even among
Conservative property holders. The police cast their retaliatory net so
broadly that they indiscriminately attacked Liberals and Conservatives
alike; merely inhabiting an area in which the guerrillas operated was
enough to be deemed an accomplice of them. The indiscriminate attacks
led by the government’s police forces ultimately prompted private pro-
ducer associations and members of the regional economic elite with ex-
tensive properties in the southwest to pressure the regional government
to avoid using the police to supervise or protect the local coffee harvest. It
was only in the southwest where powerful institutions such as 
held sway that sufficient political muscle could be exercised to limit the
police presence which had come to be acknowledged throughout Antio-
quia as one of the primary instigators of unrest. This probably saved more
civilian lives than any other single action undertaken during the period
of la Violencia. In contrast, the military’s sporadic but intense campaigns
to eradicate guerrilla headquarters, escape routes, and supply lines—a
policy that had called down upon the governor’s head the vociferous de-
rision of extremist Conservatives—finally began to have some success in
the war against armed insurrection.
 Blood and Fire

The implications of these combined elements in the struggle between

government forces and local insurgents were far-reaching. First, the con-
flict in and around Urrao never evolved into an escalation of paramilitary
forces and crude, indiscriminate killing as it did in western and parts of
eastern Antioquia. The ability to avoid such an outcome was aided by
the relatively tight control wielded by the military over the police troops
assigned to support its tactical maneuvers in the region and by the mili-
tary’s general reluctance to endorse or conduct joint operations with local
contrachusma forces. The influence of powerful landowners in control
of the production of Colombia’s main export commodity was also cru-
cial in this regard. Second, the struggle between insurgents and the army
assumed the characteristics of a war waged between armed equals that,
aside from the conscious elimination of Conservative men by guerrillas
in selected hamlets, tended not to be indiscriminately directed at broad
sections of the civilian population.
Casualty statistics tend to support the conclusion that ‘‘regular,’’ as
opposed to ‘‘irregular,’’ warfare spared civilian lives and made a recon-
ciliation between opposing armed factions possible in the aftermath of
la Violencia. Reconciliation was difficult to achieve in areas where parti-
san conflict evolved in complex directions that included struggles over
cultural, racial, and ethnic identity and where the responsibility for com-
bating guerrillas and maintaining public order was primarily entrusted
to paramilitary forces. Army and church records from  to  indi-
cate a rise in the number of civilian deaths beginning in  that reflects
a general intensification of violence in and around Urrao.109 The army re-
corded only one violent civilian death in , but the number of recorded
violent deaths rose to  in  and  in . Clashes between guerril-
las and the government also appear to have increased in . While the
army recorded no guerrillas among the dead in either  or , for
instance,  and  guerrillas were listed as casualties in Urrao in 
and , respectively. More than half of the total civilian casualties re-
corded by the armed forces over a three-year period occurred in .
Of these, fifty occurred before August when contrachusma forces briefly
expanded and accompanied the police on missions to wreak revenge on
hamlets where no guerrillas were based (see appendix A.).110
While it is certainly possible that the military consciously underesti-
mated the number of civilian deaths and inflated the number of guerrillas
killed to give its struggle greater legitimacy, local church records echo
 Urrao and the Southwest 

the details of official deaths registered by the army. Urrao’s parish priest
noted that  of the  deaths recorded in the parish registry of 
were the result of violence.111 By , the number had grown to , while
in —the year in which the greatest number of violent deaths were
officially recorded in Antioquia—twice that many people () were reg-
istered as having died by violent means. Many of the dead were Liberal
guerrillas. In , in contrast, when most violent activity in the area was
reduced to one-on-one combat between the armed forces and the guerril-
las and direct attacks of guerrilla encampments, official civilian casualties
declined to five.
The latter half of  proved seminal for other reasons as well. In
the year and half before the army initiated aerial strafing, Urrao’s guer-
rillas had been able to capitalize upon divisions within the forces sent
to control public order in the region. However, after mid-, greater
consensus and coordination among the government’s forces limited the
guerrillas’ ability to move easily within the region. Guerrilla access to
support and supplies became increasingly difficult. Cattle-rustling con-
tinued unabated, but the government’s successful attempt to limit the
trade in other basic necessities such as cigarettes, matches, fuel, and salt
forced the guerrillas to overextend themselves. They were obliged to seek
out towns further afield where government checkpoints were less com-
mon or less effective. The search for salt was perhaps the most fraught
with danger and the most urgent. Guerrillas desperately raided salt bins
left out in pastures for cattle only to find that they had been poisoned by
government agents hoping to force the guerrillas to surrender.112 An im-
portant surreptitious exchange between guerrillas and merchants from
towns such as Salgar, Bolívar, Ituango, and Puerto Arquía in the Chocó
had to be negotiated to replace the loss of supplies from within Urrao
itself.
The civilian population was also affected by government control of
basic commodities and restrictions on the transportation and sale of
cattle and agricultural goods that were imposed with growing intensity
between March and October of . The management of day-to-day
survival for Urrao’s civilian inhabitants became increasingly arduous. A
strict system of police roadblocks emerged to limit the mobility of peas-
ants from the surrounding countryside who traveled into Urrao on mar-
ket day to buy and sell their produce, baptize their children, use the coffee
credit agency, and catch up on local news. Those entering and exiting
 Blood and Fire

the town were individually searched to ensure that they did not carry
goods that might be used to supply the guerrillas. Since the government
suspected that ‘‘all of the local merchant community sends the guerril-
las presents and sells them cloth and food,’’ it implemented a system of
rationing in which families were allowed to purchase and carry only the
basic items needed to survive over a period of a few days.113 Urraeños got
around this by selling to third parties, smuggling out items at night, or
transporting goods along the mule paths, contraband routes, and streams
that crisscrossed the town and only locals knew about. But, by the end
of , general hardship and hunger wore down guerrilla and civilian
resistance in Urrao.
The military’s destruction of guerrilla bases in Caríazul on the Cerro
de Pabón and government-imposed rationing gradually forced the guer-
rillas to scale back their activities. They either retreated into the Chocó
or blended back into the agricultural working population in areas such
as La Camara in Salgar. The guerrillas withdrew from open conflict in
the middle of August  and maintained a low profile through Janu-
ary of the following year. The guerrillas’ decline in activity, however, was
countered by a concurrent intensification of police and contrachusma ac-
tivity in the area. The rise of paramilitary forces temporarily shifted the
focus of violence onto the unarmed civilian population. Refugees from
Urrao’s countryside flooded the town, while food restrictions, constant
surveillance, and the gradual intrusion of violence within the town limits
made life increasingly unbearable. Individuals who had been children in
Urrao during la Violencia remembered that mutilated bodies and fetid
corpses became a familiar presence in the early morning along Urrao’s
main street. Children stumbled over dead bodies on the way to school or
witnessed public works trucks as they dumped their macabre load before
their terrified eyes.114
In January , the attacks that had become characteristic of the sea-
sonal cycle of violence in the southwest were resumed, but the guerrillas
were noticeably weakened. They limited their activities to quick cattle
theft and petty robbery of basic staples from haciendas in Caicedo, Urrao,
and Salgar.115 They resumed assassinations of Conservatives and govern-
ment officers in March but conducted them in an almost haphazard fash-
ion. Most casualties occurred when guerrillas, in the course of stealing
cattle or moving their camps inadvertently stumbled upon or were sur-
prised by government troops rather than from a conscious effort to elimi-
 Urrao and the Southwest 

nate members of the opposition.116 In June, there were local reports that
Urrao’s guerrillas were active in the southwest and still engaged in rob-
bing, arson, and assassination, but no official reports of guerrilla violence
were filed with the governor’s office. Guerrilla strength had essentially
dissolved by late . The concerted efforts of the regional government
to block supplies and the movement of cattle—the sale of which was abso-
lutely forbidden between the municipalities affected by violence around
Urrao, much to the despair of local ranchers—ultimately eroded the eco-
nomic basis of survival for armed Liberal revolt.
The end of la Violencia in Urrao was almost anticlimactic. Two months
after the military led a coup against Laureano Gómez’s government in
June , Captain Franco and several guerrilla officers were ambushed
and captured in a local cantina. The guerrilla leader and trusted mem-
bers of the plana mayor had, with considerable trepidation, agreed to
meet a delegation of government representatives to discuss the condi-
tions for a peaceful surrender. Suspecting a possible trap, several guerril-
las had hung back and stationed themselves at various points in the town,
but armed government agents surprised them and took Franco prisoner
on August , .117 Franco and his surviving officers were sent to ‘‘La
Ladera’’ prison in Medellín and, not long after, his followers surrendered
their arms and accepted the political amnesty offered by General Rojas
Pinilla’s government. Most of Urrao’s former guerrillas served sentences
of several months, while Franco—the most sought-after guerrilla leader
in Antioquia—was imprisoned for nearly three years. Upon his release,
his unabated popularity prompted him to launch a career as a Liberal
politician, but the fear that he would usurp and upstage them moved
local Liberal leaders in Urrao to pressure the regional party to channel
Franco’s ambitions elsewhere. For a few months before his death in a mys-
terious drowning accident in the San Jorge River in the Chocó, Juan de
Jesus Franco traveled up and down the Colombian Caribbean as an offi-
cial spokesman (and hero) for the Liberal party.
Inspired by their parish priest, the citizens of Urrao marked the con-
clusion of armed conflict in the zone in a unique fashion. For several
years, Father Ramirez had carefully kept track of the place and cause of
death of his parishioners even when he had not buried them. Those who
died as a result of violence had been denied Catholic burial since , be-
cause they were presumed to have violated divine law in rising up against
a legitimately elected government. The local priest had disagreed with
 Blood and Fire

this policy, but was too prudent and too closely observed by extremist
Conservatives to violate it during la Violencia. Instead, he enlisted the
relatives of the dead to provide information regarding the circumstances
of death and the probable location of the bodies. After the military coup,
the priest drew his parishioners together and instructed them to dig up
the remains of all those who had died without the benefit of Catholic
ritual. The priest and Urrao’s citizens (of both parties) organized a pro-
cession and conducted a collective burial to mark the end of la Violencia.
Two secular rituals were also used to mark the end of la Violencia in
Urrao. First, local Liberals raised money to hire the sculptor who had
originally made the bust of Rafael Uribe Uribe that had graced the town’s
central square before the violent intrusion of the Virgin of Fatima in
October . They commissioned him to make a statue that was iden-
tical to the first. When it was ready, the municipal authorities and party
leaders installed it in the central plaza amid considerable pomp. Second,
the town purchased a bronze plaque and dedicated the main road lead-
ing into Urrao to Captain Juan de Jesus Franco. The entry to Urrao is
still known as the ‘‘Avenida Capitán Franco,’’ a tribute to a rebel leader
without precedent in any other part of Antioquia.
No contrachusma groups survived in Urrao, although remnants
briefly continued to disturb the countryside around Salgar, Bolívar, and
Betulia for several months after the military came to power.118 Still, they
were too weak to constitute the threat to property and power typical of
paramilitary forces active in western and parts of eastern Antioquia after
June . Former guerrillas who remained in the area, in contrast, en-
joyed such widespread respect and legitimacy that several of them later
went on to become town councilmen and officeholders in Betulia, the
center of local Conservative opposition during la Violencia. And, in what
can only be understood as a peculiar twist of history, when violence
once more emerged in Urrao in the late s it was led not by vindic-
tive Conservatives seeking revenge against their Liberal adversaries, but
by members of Alfonso López Michelson’s Movimiento Revolucionario
Liberal () who opposed the agreement to return to civilian govern-
ment known as the National Front.119 Having experienced the hardships
of an outlaw life—‘‘la mala vida,’’ as the guerrillas called it—Franco’s ex-
combatants were unwilling to take up arms again, and their refusal made
them the target of their fellow party members. Ironically, Liberal violence
achieved what local Conservatives during la Violencia had not: former
 Urrao and the Southwest 

guerrillas were forced by the followers of the  to abandon their lands
and move elsewhere, many of them joining the growing waves of rural
migrants leaving for towns such as Itaguí, Envigado, and Medellín.

The Defeat of the Guerrillas

What motivated the regional government to intensify its efforts against

Urrao’s Liberal guerrillas so dramatically in late ? And what moti-
vated them to charge the army, not paramilitary groups or the police,
with the primary responsibility for combating the guerrillas? Several
possible explanations exist, the most significant of which may be the
ignominious assassination of ninety-six soldiers in the Llanos in July
. This event prompted the military to rethink its counterinsurgency
strategy in Colombia and the central government to take more seriously
regional complaints that ‘‘our resources of arms and men continue to be
notoriously insufficient.’’ 120 In the specific case of Urrao, moreover, the
escalation in government attacks against guerrilla forces was also surely
influenced by two other issues: a growing perception among regional
authorities that the guerrillas posed a threat to property arrangements
and production in what was considered the very heartland of Antioquia’s
economy and, also, the guerrillas’ enduring popularity among the local
population.
In June , Governor Henao Mejía informed the minister of gov-
ernment in a coded message that there had been ‘‘a rebirth of politi-
cal violence during the previous two weeks’’ in which Urrao’s guerrillas
had killed first eighteen and then another thirty male peasants.121 The
victims, the governor insisted, were consciously, not randomly chosen.
Aside from the ‘‘war of nerves’’ that these assassinations had produced
in the area, other events and information regarding the guerrillas’ activi-
ties in the region combined to worry the governor that something more
sinister than exclusively partisan violence was afoot. Urrao’s Conserva-
tive committee had warned the governor a year earlier that the guerrilla
leader, Captain Franco, was ‘‘beginning to redistribute lands in the ex-
tensive area between Pabón and its immediate vicinity.’’ 122 The commit-
tee accused one guerrilla of being a professional real estate agent who
was buying up and speculating in devalued lands owned by Conserva-
tives who were forced to abandon the area. At the time, the governor
had paid little attention to these reports.123 In August, however, Urrao’s
 Blood and Fire

Conservative committee members again insisted that men, not women

and children, were the primary targets of guerrilla violence and, further,
that the men were killed only when they ignored the warnings to leave
that were given before hamlets were attacked.124 A month later, all of the
Conservative-owned haciendas in Pabón had been abandoned and taken
over by guerrilla forces. The Caja de Credito Agrario calculated that to
recapture the occupied haciendas and ‘‘pacify’’ the area would require
stationing five hundred soldiers in Pabón.125
Complaints of land usurpation also began to emerge from towns such
as Salgar, where most land was held in large estates. The owner of a large
coffee estate in La Camara complained that guerrillas had taken over pri-
vate properties in the hamlet and appropriated parts of the coffee har-
vest since  and that they were planning to do the same in . In
contrast to what was happening in Urrao, the estates in La Camara were
not being destroyed or taken over directly by the guerrillas. The majority
of La Camara’s inhabitants were Liberal, as were the majority of mayor-
domos, but the estate owners were sometimes Conservative. To impede
outright land confiscation, mayordomos and guerrillas from Urrao had
worked out a deal: in return for not destroying haciendas, mayordomos
received half the coffee production while the owner’s share was given to
the guerrillas.126 Witnesses also began to report that guerrillas were paid
a wage and that they alternated between making war and sowing fields
on empty public lands and usurped properties.127 Alarm began to spread
among Antioquia’s authorities that partisan violence masked more pro-
foundly unsettling and radical objectives. The language employed by the
guerrillas in official communiqués in  contributed to this impression.
In response to a peace initiative mediated by Urrao’s Father Ramirez in
March , Captain Franco signed his decision to decline taking part in
the initiative by referring to himself as the representative of ‘‘guerrillas
and workers.’’ 128
By early , the occupation of lands by guerrillas and the selec-
tive eradication of Conservative inhabitants prompted calls by govern-
ment supporters for a razed-earth policy.129 The replacement of Governor
Braulio Henao Mejía in July  by hard-liner Dionisio Arango Ferrer,
moreover, removed any remaining obstacles to such a plan. Policemen
and contrachusmas led the vanguard in areas perceived to be guerrilla-
‘‘infested.’’ They targeted any area perceived to be under the influence of a
guerrilla band—regardless of the political affiliation of the area’s inhabi-
 Urrao and the Southwest 

tants—since anyone who succeeded in continuing to farm, or who could

still travel between fields and home was automatically assumed to be a
guerrilla sympathizer.130 The extremes to which this policy could be taken
became evident in early . Conservatives who opposed the razed-earth
policy in Betulia complained that contrachusma groups based in Alta-
mira, Betulia and, El Socorro, Concordia, ‘‘do not operate where there
are guerrilla bases, but in peaceful areas where everyone is dedicated to
working their agricultural properties. [This] has discredited the govern-
ment because it is the local authorities in such areas who support those
abominable crimes.’’ 131
The struggle between partisan forces had become a struggle over who
should have the right to inhabit, control, or farm particular areas, and
the determination of that right was increasingly shaped by historical per-
ceptions of conformity and rebellion. Or, as the infamous police chief
of Betulia, Major Arturo Velásquez, succinctly put it when justifying his
decision to raze two settlements and kill everyone in them: ‘‘Seeing as
how the inhabitants of la Mina y la Guamala have provided great support
for bandits since the War of the Thousand Days [–], the under-
signed [Arturo Velásquez Acosta, Jefe Fracción], without consulting my
superiors but acting with the best of intentions, ordered the disoccupa-
tion of those hamlets. . . . I felt quite sorry to do so, but I believe that
where the families of honorable Conservatives can’t live, there is also no
reason to protect the enemy.’’ 132 It is not my intention to underestimate
the significance or depth of partisan feeling that influenced the vindic-
tive desire to sweep clean areas of inhabitants belonging to the opposi-
tion during la Violencia. But several factors make it implausible to argue
that politics alone shaped the emergent conflict over land even in the
supposedly ‘‘nonradical’’ violence of the southwest. First, much of the
property around which struggles occurred was concentrated in large es-
tates. Second, the coffee frontier had effectively closed several decades
earlier and with it the opportunity to own land had become more re-
mote. Third, struggles over access to agricultural employment were in-
tense, which suggests that jobs were increasingly insufficient and hard to
come by in the region. As a whole, the department of Antioquia’s census
of employment by sector recorded a decline of more than two hundred
thousand agricultural jobs between  and .133 While census figures
may reflect errors in calculation, even before la Violencia the number of
worried editorials in Medellín’s newspapers regarding high unemploy-
 Blood and Fire

ment figures, scarce food, and the need to implement price controls on
basic commodities would seem to support the idea that agricultural areas
were experiencing a crisis. Most agricultural employment had histori-
cally been concentrated in the coffee-producing southwest and the east-
ern municipalities nearest Medellín. The presence of adverse economic
and social conditions must necessarily be taken into account in any con-
sideration of the ultimate impact of violence on the region. Additional
data, moreover, reinforce the impression that economic considerations—
however subconscious or ideologically undeveloped they may have been
among the guerrillas—were intimately interwoven into the partisan con-
cerns that underwrote armed conflict in the region.
After a military coup toppled the Conservative government in June
, Urrao’s guerrillas would surrender only after a number of conditions
were met. Among them were several that exclusively addressed the issue
of credits, government aid to promote colonization, tools, seeds, and the
distribution of lands to amnestied guerrillas and to ‘‘poor peasants and
their families.’’ 134 If no social agenda underpinned armed insurrection,
why would demands of this nature have occupied such a significant por-
tion of the terms of political surrender? Guerrillas also demanded that no
individuals who might have exercised authority in the locality as national
policemen, contrachusmas, or extremist public employees be allowed to
resume positions of power in the area or be entitled to reside there. In
addition to the intense animosity that would likely have characterized re-
lations between these forces and the guerrillas, competition over scarce
resources and an attempt to undo the speculative gains which some of the
government-sponsored groups had achieved during la Violencia clearly
preoccupied the guerrillas.
The guerrillas, moreover, did not voluntarily give up their arms as lore
would have it. Indeed, in July of  Captain Franco issued a preemptory
‘‘decree’’ from the mountain strongholds of western Antioquia inform-
ing Colombia’s new military government of his terms for the negotia-
tion of peace and belligerently signed it, ‘‘Mayor Franco, Commander,
The Revolutionary Forces of the West and Southwest, General Guerrilla
Headquarters.’’ The tone of ‘‘Decree no. ’’ is that of one military com-
mander to another, not a subordinate to his superior as the subheading of
the document attests: ‘‘Document by which certain dispositions regard-
ing public order in the southwest, west, and north of Antioquia are de-
creed. These shall also apply to the department of El Chocó. The Guerrilla
 Urrao and the Southwest 

Command exercising the rights conferred upon it by the High Command

of the Main Revolutionaries.’’ 135 Was this bluster? Or had Urrao’s guer-
rilla forces moved in the direction of the Communist-led, Llanos forces
whom they openly admired? 136
But perhaps the most startling piece of evidence pointing to the im-
portance of struggles over land in the evolution of southwestern violence
was the solution for bringing la Violencia to an end that was suggested by
ecclesiastic officials and moderate Conservatives in . In March, the
Conservative daily El Colombiano explicitly stated that violence was the
result of an absence of ‘‘social justice.’’ 137 Certain sectors of Antioquia’s
elite seemed to recognize publicly that mixed in with partisan hatred was
an inchoate but real anger born of economic despair and the declining
possibilities of ownership, opportunity, and mobility in rural Antioquia.
The auxiliary bishop of Medellín, Buenaventura Jaureguí, declared pub-
licly in April that a policy of ‘‘parceling out lands’’ would ensure ‘‘the
salvation of the rural folk [campesino].’’ 138 This pronouncement was fol-
lowed in May by an article in El Colombiano urging a four-point program
for ending violence in the countryside. The newspaper suggested that
the government ‘‘should begin to parcel out large estates [and] popular
and agricultural credits should be increased’’ to build roads and homes
with the aid of loans provided by the Instituto de Credito Territorial
(Land Credit Institute). El Colombiano also suggested that small indus-
tries should be encouraged and subsidized with government assistance
in municipalities suffering from the effects of violence.139
To my knowledge, no other Colombian region raised the issue of land
redistribution as a central determinant of any peace negotiation to end la
Violencia and certainly not at such an early date. Nor is there evidence in
studies of other areas affected by violence of official sources that publicly
attributed la Violencia to social and economic inequity, rather than par-
tisan differences. Yet in Antioquia a sector of the regional elite hailed the
implementation of a land redistribution program as ‘‘a mission of Chris-
tian charity [to guarantee the] patriotic defense of Colombian society and
the preservation of true peace.’’ 140 The newspaper followed up these sug-
gestions with an editorial recommending that ‘‘wherever a situation of
disorder exists and it is possible to parcel out haciendas, this should be
done because nothing makes a man more orderly and responsible than
attachment to the land and knowing himself to be a property owner.’’ 141
This was the regional political ideology of convivialismo writ large.
 Blood and Fire

The pragmatism characteristic of Antioqueño bipartisan politics be-

fore the advent of single-party rule in  once more reasserted itself
in El Colombiano’s analysis of the causes of violence and its needed
solutions. Antioquia’s elite had always perceived a close correlation be-
tween citizenship and property ownership and were convinced from the
days of Acción Catolíca and the encyclicals of Pope Leo XIII that ma-
terial well-being was crucial to the maintenance of social harmony and
political order. At the time of la Violencia, a core of regional political
leaders understood the need to rethink property and economic arrange-
ments and the degree to which so-called partisan conflicts were linked to
broader structural conflicts in Colombian society. In hindsight it seems
nothing short of tragic that they were ignored. How much later blood-
shed might have been averted if a timely reform program had been im-
plemented like that suggested by El Colombiano’s owners in ?
When Governor Braulio Henao Mejía was forced to resign in July be-
cause he had been unable to put an end to violence, El Colombiano reiter-
ated its belief that the catalysts of violence had been neither moral nor
partisan. They insisted on attributing widespread regional unrest instead
to ‘‘the lack of food, housing, [and] education’’ and considered these as
‘‘the true obstacles to achieving peace.’’ 142 Interestingly, El Colombiano
specifically noted that the policy of distributing land to assuage violence
should not be limited only to Urrao. This would seem to suggest that
Urrao had emerged as the prototype for official land distribution schemes
as a solution to violence. Some Conservative extremists suggested that
no amount of land redistribution could alter the fundamentally rebel-
lious nature of the town’s inhabitants and that the only solution to the
area’s problems would be to ‘‘recolonize Pabón with Conservatives from
Marinilla’’ and other eastern Antioqueño towns.143 But the regional gov-
ernment ignored such suggestions and, when in the early s the
National Land Redistribution Institute () was founded as part of
an agrarian reform program, Urrao was among the first municipalities in
Antioquia to break up and redistribute haciendas.
Sometime in late  or early , moreover, the regional govern-
ment commissioned a secret report regarding the nature of unrest in
Urrao and western Antioquia from what appears to have been a United
States government agent.144 After providing an overview of the effect of
la Violencia on Urrao and the surrounding rural hamlets, the agent con-
cluded:
 Urrao and the Southwest 

Undoubtedly in the beginning [of guerrilla activity] there was a touch

of politics in the acts of the bandits, they proclaimed themselves Lib-
eral patriots, and it is almost certain that at that time they were aided
by some Liberals, politics being what they are in Colombia. . . . Later,
once they understood the true nature of the Liberal bandits, the ma-
jority denied them any further support and then they were treated the
same way the Conservatives were. . . . It is very doubtful that there is
an upstanding Liberal [de bien] who is now supporting the bandits.145

The agent went on to describe a ruined local economy in which estate

owners could no longer visit their properties for fear of being killed, the
wealthy had purchased plane tickets and fled, the cost of living had sky-
rocketed, and the only remaining inhabitants were those too poor to
leave town.146 All of this led the author of the report on Urrao to con-
clude that ‘‘the poor are on the verge of dying of hunger. For them it is no
longer a matter of politics but a question of sheer individual survival.’’ 147
But, even worse, the agent insisted that the public was convinced that the
Colombian government had no interest in ending the conflict in Urrao:
‘‘Those who were once strong Conservatives have changed their minds.
They haven’t exactly become Liberals but there is no doubt that many be-
lieve that Urrao has been made a pawn in a political chess game in which
the town has no bet. In other words, that the government isn’t really inter-
ested in eliminating the bandits because as long as these remain active,
the government will have an excuse to maintain the State of Siege.’’ 148
And the agent added, ‘‘Men who have empty stomachs and who see
their relatives dying of hunger are desperate men. . . . Urrao is rapidly get-
ting to the point of being fertile ground to receive Communism; not be-
cause people like it or because they understand what Communism means
but simply because the Communists offer them an exit from their present
difficulties. A similar situation exists in many parts of the world today.’’ 149
Like the moderate Conservatives of El Colombiano, moreover, the author
of this report also believed that addressing people’s material needs had to
be the first step in bringing violence to an end (and inhibiting the spread
of Communism). ‘‘The only solution’’ to the region’s problems, the agent
argued, ‘‘is to obtain work for all those who are unemployed. . . . It is of
the utmost importance to improve the morale of the people. Otherwise,
anything could happen, none of it good.’’ 150 The solution lay in expand-
ing public works construction and military fortifications in the region so
 Blood and Fire

as to hire as many people as possible. The military fortifications were to

be supervised and led by ‘‘the Army Engineering Corps with infantrymen
of Indian descent (Quichua [sic]) from Nariño . . . because (a) they are
accustomed to high altitudes and cold weather; (b) they are good people
in the jungle; (c) by nature are inclined to be strictly disciplined so long as
they have good officers; and (d) they’d probably like the work.’’ 151 While
these fortifications were under construction, all of the peasants in the
area were to be removed. After ‘‘the cleansing [limpieza], these peasants
[campesinos] may return to their lots and keep working without fear.’’ 152
Beyond extensive logistical recommendations intended to guide the
development of a counterinsurgency strategy in the region, the report
concluded its recommendations with a disquisition on the nature of local
guerrilla strength, its connections to Communism, and the dangers that
Communism posed to Colombia and Urrao. Interestingly, the agent be-
lieved that the local guerrillas had never numbered more than ‘‘two hun-
dred altogether and no more than fifty men in any single band,’’ figures
that, while not too distant from the estimate Franco’s own lieutenants
gave of their group, were far below those bandied about by the govern-
ment or even later researchers.153 The agent also noted—as had former
guerrillas and local Conservatives—that it was persistently ‘‘rumored
that the active bandits receive a monthly salary of  pesos. If there is
any truth to this [rumor], it is surely not Colombians who are supplying
the money, but they may nonetheless be acting as the agents of outsiders
[ forasteros].’’ 154
The report identified the centers of Communist infiltration in Colom-
bia as the eastern Llanos and Western Antioquia (from Urabá to Urrao).
But, ‘‘in terms of continental strategy, the Llanos in and of themselves
cannot be considered important, although they may be used to con-
stantly threaten Bogotá and so divert the attention of the government
away from the area where the Communists are actually planning to do
considerable damage.’’ 155 The topography of the Atrato River Valley, the
agent speculated, made it ideal for ‘‘supplying submarines,’’ and ‘‘planes
could also easily be hidden in that area [the valley below the Atrato River].
Based in the Atrato, submarines could harass traffic traveling along the
Panama Canal and, using guided missiles, could damage the Gatún locks.
Amphibian planes pose the same danger.’’ 156 Urrao’s guerrillas would
be the smokescreen behind which Communist agents could hide, but
because of the enormous sums of money involved in mounting such a
 Urrao and the Southwest 

project, a mechanism other than legitimate bank transfers would have to

be found to launder sufficient cash to underwrite the Communist take-
over of Colombia. The agent rightly noted that smuggling would prove to
be the government’s Achilles’ heel, for the absolute lack of customs agents
or other government officials deployed around Urabá’s numerous points
of entry or ports made it unlikely that Communists would get caught.
The locals, for instance, would have little incentive to report the pres-
ence of Communist smugglers: ‘‘they won’t inform for two reasons, first,
for fear that the Communists will take revenge and, second, because they
won’t want to forfeit what will seem to them like very good wages.’’ 157 The
leaders of this smuggling plot to introduce Communism, the author of
the report insisted, ‘‘come from Guatemala, since it is that country where
the Communists’ general headquarters for Latin America is located.’’ 158
This was of course the moment at which U.S. propaganda and military
efforts were under way to topple President Arbenz in Guatemala by ac-
cusing him of being a Communist.
The informant ended his report on the Communist danger lurking in
Urrao with a dire warning to the Colombian government:

It is clear that if the state of misery continues in Urrao and other mu-
nicipalities west of the Cauca River, the Communists will have a huge
reserve of people from which they will be able to recruit whatever
number they need when the time comes to assume control. Once the
war begins, the Communists can deliver arms and officers to train
natives. These will be used only in the jungle to defend submarine
bases, . . . precisely the same model used by Communists in Malaysia,
Indo-China, and Burma, and that they are using in Brazil and Bolivia,
to say nothing of Colombia.159

One can dismiss conclusions such as those included in the agent’s syn-
opsis of the state of public order in Urrao and western Antioquia as little
more than extreme propagandistic reactions in an emergent Cold War era
of which Colombia was not immune. Colombia’s Communist party was
tiny, its followers were mainly concentrated among skilled artisans (type-
setters, shoemakers, tailors) in cities, enclave workers along the Magda-
lena River, and certain urban intellectuals. There is no evidence that there
were any Communist sympathizers in Urrao or its surrounding areas.
But however risible predictions of a Communist takeover may have been,
they nonetheless point to real anxieties about the presence of economic
 Blood and Fire

and social conditions perceived to be conducive to revolutionary activity

in particular areas of Colombia.
Indeed, economic subsidies, technical assistance, the distribution of
public lands, cheap credit, and the creation of additional public works
jobs did in fact come to be adopted by Antioquia’s authorities as part
of a plan developed in  to restore public order and rehabilitate areas
affected by violence. Former guerrillas indicated that the extension of
cheap credits, cattle stock through the Fondo Ganadero (Cattle Fund),
and the distribution of public lands in and around Urrao had made it
possible for them to eventually reinsert themselves into civilian society
and survive economically after la Violencia.
Yet, to those intimately involved with Pabón’s guerrilla forces, the
notion expressed by the government agent that guerrilla objectives had
deviated from the limited agenda endorsed by the official Liberal direc-
torate and evolved into the precursor for Communist takeover in Colom-
bia would have been deeply shocking. The testimony of Graciela Urrego,
Captain Franco’s widowed common-law wife, vividly underscores just
what made the experience of la Violencia in Urrao so different from the
experience of violence in other parts of Antioquia and so difficult to en-
compass in loaded, catchall phrases such as ‘‘radical’’ or ‘‘traditional.’’
In the mid-s, Graciela Urrego lived with an unmarried sister in
a modest house located on the main road leading into Urrao, the same
road named after her famous guerrilla husband, Captain Juan de Jesus
Franco.160 The municipal authorities had given her the house as a token
of gratitude after Franco’s death. But for the house and a precarious in-
come secured by the sale of homemade sausages (chorizos) and popsicles
( paletas), Graciela Urrego would have been penniless. The house stood
not far from the gas station where all the trucks and buses traveling the
intermunicipal roads were serviced, a place where Medellín’s Liberal di-
rectorate and the local guerrillas had exchanged reports and messages via
sympathetic drivers and attendants during la Violencia.
The world described by Graciela Urrego was divided into ‘‘good’’ and
‘‘bad’’ rather than Liberal and Conservative. She noted how there had
been ‘‘good’’ Conservatives, for instance, who helped the guerrillas by
selling needed supplies or providing crucial information and protec-
tion against despised, imported police troops and unscrupulous arri-
vistes who exploited the opportunity for personal gain and offered them-
 Urrao and the Southwest 

selves as informants. ‘‘Bad’’ Conservatives included government troops

who harassed and raped the women who traveled back and forth be-
tween the town and their rural residences. But there were ‘‘good’’ Conser-
vatives among the government ranks, too. Guerrilla sympathizers who
were among the lower ranks of the government troops sent to patrol the
region warned the guerrillas and townspeople in advance about which
officers to avoid or when sweeps were to be conducted against them. In-
deed, Graciela was convinced that well before Franco’s arrival in Urrao, a
pact had been made between Eduardo Villamil, the army commander of
Medellín’s Fourth Brigade and the national Liberal leadership. The army
commander and the Liberal leadership, she was convinced, had agreed to
foment the creation of Liberal guerrilla forces throughout Antioquia and
to ensure that the army would not oppose or engage in combat with them.
Graciela Urrego held as an incontrovertible truth that Liberal leader
Carlos Lleras Restrepo had sanctioned the guerrillas’ existence; his motto
‘‘Fé y Dignidad’’ (Faith and Dignity) was their rallying cry. As ‘‘supreme
commander,’’ Lleras supposedly sent instructions by secret code to mem-
bers of Medellín’s Liberal party directorate who in turn relayed these to
the guerrilla leadership in Urrao. Whether or not this was true is less
important than the fact that such a belief enabled Urrao’s guerrillas to
view themselves as individuals engaged in the legitimate defense of their
party’s interests rather than as mere outlaws or, more dangerously, as
radicals intent upon restructuring the local terms of property tenure
and labor.
Graciela Urrego remembered the period of armed resistance as both
a terrible and wonderful experience. She retained—as did many other
Urraeños—a sense of satisfaction or pride tinged with a sense of pro-
found sacrifice; the taking up of arms was seen as an unfortunate but
ultimately noble enterprise. And, the men and women who took part in
defending their lives and the Liberal party were loyal citizens forced by
circumstances beyond their control to engage in a deeply sacred struggle.
In interviews, former guerrillas and their companions repeatedly con-
trasted the ideals that motivated their armed struggle with contemporary
guerrilla struggles in Colombia. Only those who had never had a choice
but to retreat into the hills, they insisted, would willingly choose such a
life—hence the reason for Urrego’s and other former guerrillas’ deep dis-
approval and contempt for contemporary Marxist-based guerrilla orga-
 Blood and Fire

nizations. And, had she been aware of the ‘‘American’’ informant’s report
that posited Urrao as the next springboard of world Communism, Gra-
ciela Urrego would have been apoplectic.
Graciela Urrego’s faith in her party was unshaken even when a high-
ranking regional Liberal political leader came to see her in  and de-
manded that she hand over any documents that might have been left by
Franco or his command from the ‘‘old days’’ of la Violencia. The Liberal
leader—originally from Peque—insisted to Urrego the need to destroy
any ‘‘compromising’’ materials. She remained loyal even as young Lib-
eral politicians who had been too young to take up arms or had fled to
the safety of Medellín during la Violencia invoked her husband’s name
and lied about their participation in Urrao’s guerrillas to enhance their
chances of being elected. She remained unwavering in her loyalty when
Liberals who opposed the National Front agreement turned upon the
old Liberal guerrillas of la Violencia and forced them to migrate away
from Urrao.
She remained silent and loyal throughout, buoyed by the memory of
collective hardship and loss, of friendships that for a brief period had
erased the barriers of class in Urrao and bound people together in a single
endeavor. She was oblivious to the seminal role drawn for Urrao in the
struggle for world domination. Nor did it occur to her that the guerril-
las’ mild redistribution efforts might have alarmed the government of
the time, just as the takeovers of property led by the Marxist guerrillas
she disapproved of continued to worry the same Colombian state in the
s. To Urrego, la Violencia had only been about the fight to guarantee
the Liberal party’s right to take part in shaping the course of Colombian
history, not about the advancement of self-interest or the first stage in
the creation of a new national order.
Who was right regarding the ultimate objectives and organization of
Urrao’s guerrilla forces or about the meaning of motivations underlying
violence in the town? Graciela Urrego or the U.S. adviser to the Colom-
bian government? If one accepts the inherent duality of historical process,
no contradiction need lie at the heart of the apparent differences in in-
terpretation between Urrego’s account and that of the American agent.
The point of la Violencia, even in supposed areas of ‘‘traditional settle-
ment’’ where partisan objectives were the guiding force behind armed
insurrection, is that it was multifaceted and ambiguous, that politics and
economic considerations can never be considered as discrete forces, that
 Urrao and the Southwest 

class struggle could be waged within the confines of Conservative and

Liberal ideologies without the participants in these conflicts seeing any
contradiction in the demands for land and the restoration of their party
to power.

Conclusions

Urrao’s experience during la Violencia was paradoxically Antioquia’s

most unusual. The town took up arms against the Conservative gov-
ernment reluctantly and as a last resort when all possible democratic
means of protest were exhausted or cut short by the state itself. There
was no major support for Laureano Gómez locally, nor were the town’s
Liberals and Conservatives involved in violent contestation with each
other between  and . No rebellion took place in the town after
the assassination of Jorge Eliécer Gaitán in April . Instead, vio-
lence emerged in response to the concerted attempts of the regional gov-
ernment to undermine the authority and legitimacy of the town’s own
democratically-elected representatives. These attempts were perceived
locally as a transgression of unspoken but powerful expectations gov-
erning the relationship and negotiation of power between municipal au-
thorities, the regional government, and the central state.
Although conflict in Urrao emerged along partisan lines, it was also
more complex than this. Neither the Liberal nor the Conservative parties
were monolithic. The behavior of two of Urrao’s town councilmen and
many of their local cronies are a telling example of the internal divisions
within the parties. Memories of an older, regional tradition of bipartisan
compromise and pragmatism, moreover, shaped local expectations that
partisan differences could be resolved via negotiation rather than arms.
This bipartisan ‘‘memory’’ or tradition, however, came into conflict with
the new politics of extremism and partisan hegemony. Urraeños took up
arms only after a long period of official harassment foreclosed the demo-
cratic possibilities of negotiating an end to conflict. In this, Urrao’s ex-
perience differed from what occurred in the eastern part of Colombia
where, aside from the mining towns of the northeast where unionized
miners appealed to their rights as citizens to contest state harassment, no
municipal tradition of political negotiation between towns and the re-
gional state seemed to operate and where violence was introduced from
outside the department rather than from within.
 Blood and Fire

Although Urrao’s Liberal guerrillas might not seem to have differed

much from the multiple armed groups that sprang up all over Antioquia
during la Violencia, at a deeper level, they were different. Their success
was predicated in part on their ability to intersect with and build upon
a preexisting history of local identity and resistance. Unlike guerrillas in
other parts of Antioquia (except parts of Urabá and western Antioquia
in which local clans also played a strong role), moreover, Urrao’s guerril-
las grew out of and were intimately linked to Urrao’s civilian population
rather than arbitrarily or artificially imposed from outside. This charac-
teristic not only endowed the guerrillas with enormous popular legiti-
macy and support, it ensured that in the aftermath of violence Urrao’s
townspeople were able to feel justified and proud of their activities during
la Violencia. This in turn enabled Urrao to enter into a process of recon-
ciliation, to grieve collectively as no other Antioqueño town did, and, in
the course of this, to transcend the horrors of war and mutual brutality.
The regional and central government’s attitude toward violence in the
southwest also influenced the outcome of partisan conflict in Urrao and
contrasts strongly with the way the regional authorities approached the
issue of partisan unrest elsewhere. First, although various towns within
Urrao’s range demanded and received arms to form contrachusma forces
with which to combat the guerrillas as western and eastern Conservatives
also did, the regional government never delegated public order responsi-
bilities in Urrao to the contrachusma and the police as they did in other
parts of Antioquia. Very early on, and despite the scarcity of trained per-
sonnel, Antioquia’s regional government opted to rely on the army in
their struggle to stem guerrilla activity. The regularization of conflict and
the avoidance of random, vindictive strikes against Liberals in retaliation
for guerrilla attacks became a seminal feature of public order strategy
in Urrao. This distinguished Urrao’s experience from the proliferation of
paramilitary forces which became common in other Antioqueño areas
severely affected by violence and saved the town from the persistent ven-
dettas that became a permanent feature of other formerly violent areas
after .
Violence in Urrao also never gave way to the sorts of landlord-led
paramilitary forces bent on eliminating squatters and consolidating con-
trol over land and resources common to western Antioquia. Nor did
largely autonomous armies of Conservative peasants intent on stripping
jobs and lands away from nonextremists in the east emerge in Urrao. Yet,
 Urrao and the Southwest 

la Violencia in Urrao did grow to involve conflicts that transcended purely

‘‘partisan’’ objectives. Indeed, although on the surface violence in Urrao
seemed to conform to the paradigm of ‘‘safe’’ partisan conflict, in reality
regional authorities recognized the development of armed conflict in the
southwest as at least partially the result of emerging structural inequali-
ties in Antioquia’s social and economic organization and production.
Urabá. Rural folk
flee guerrilla attacks
in the Chocó and are
marshalled by the
Colombian army into
a displaced persons’
camp in Antioquia.
Epilogue

When I’ve been digging and I’m tired and don’t want to do any more, I think
how it could be me in the grave I’m working on. I wouldn’t want someone to
stop digging for me. . . .—Manuel, member of a forensic team in Guatemala in
Michael Ondaatje’s Anil’s Ghost

When Violence Ceases to Be Academic

I finished the research for the project that eventually evolved into this
book between  and . During two long stretches I lived in my par-
ents’ apartment in downtown Medellín, just two blocks from the Parque
de Bolívar and the Metropolitan Cathedral in the heart of Medellín’s com-
mercial and financial district. From my window I heard the daily bustle
of street vendors hawking aquacates and miracle liver-spot creams such
as concha nacar and the constant backfire of noisy buses on the Bello cir-
cuit. The past and present seemed intimately intertwined as peddlers—
the heirs to a regional tradition of snake-oil salesmen—vied comfort-
ably with late-model automobiles and computer-controlled commercial
operations. At the glass-enclosed paean to architectural modernism—
the Edificio Argos—on the corner of Bolivia and the Avenida Oriental,
working people waited patiently to board buses back to distant suburbs.
At night I watched from my balcony as the crowds gathered at the steps
of the cathedral to witness and applaud the boisterous show put on by
local transvestites. On the surface there seemed nothing to suggest that
fifteen minutes away young men in the city’s poorer neighborhoods were
being gunned down on street corners, that prostitutes, the homeless, and
street children were nightly victims of ‘‘social cleansing,’’ or that less than
a few miles outside the city’s limits rural folk were caught in the crossfire
of guerrillas, right-wing death squads, and the armed forces. Businesses
bustled, the streets were clean, the phones worked, glaring examples of
misery were absent. I was struck by the possibility of multiple, disso-
nant, lived realities and often thought that a similar bifurcation must have
characterized the average Antioqueño’s experience during la Violencia. I
 Blood and Fire

began to understand how the memories and trauma of la Violencia were

mediated by such subjective factors as social, physical, and cultural place.
On the eve of the military coup in , violence was one thing to the
urban dweller who was constantly reminded by entrepreneurs and the
authorities of the region’s unstoppable prosperity and of the harmony
that reigned within Medellín’s factories. It was quite another to the rural
folk in peripheral towns where the stench of burned bodies permeated
hamlets and miserable women and children fled a razed-earth policy per-
petuated by paramilitary groups with the support of government agents.
It is this contingent, divergent, and unstable aspect of violence that I have
tried to analyze in this book.
As I concluded my research, narcotics traffickers known as the
‘‘Extraditables’’ (led by the now-defunct Medellín drug lord, Pablo Es-
cobar) stepped up their campaign of assassination targeting policemen
and public figures to protest the threat of extradition to the United States.
The governor of Antioquia, the regional head of the national police, the
Liberal presidential candidate, Luis Carlos Galán, numerous local judges,
and several university professors were brutally killed in the space of a few
months. Every weekend, moreover, the corpses of thirty to forty young
men—inhabitants of the lower-class neighborhoods that ring Medellín
and are popularly known as the comunas—swelled the death count in
addition to the politicians and public employees that constituted the
alarming balance sheet of local terror. The line separating the focus of my
research and daily lived existence increasingly blurred. In the mornings
I worked on analyzing the data I had extracted from various local and
regional archives and tried to make sense of the violence of the s and
s. In the afternoons I took a cab to the Universidad de Antioquia
and conducted closed workshops with researchers examining the emer-
gence of death squads and youth assassin gangs in Medellín and periph-
eral areas such as Urabá, the Magdalena Medio, and the Bajo Cauca. I
tried to see how or if the violence I studied and contemporary violence fit
together. With the help of researchers from the Instituto de Estudios Re-
gionales, we drew up maps that linked the young men from the city’s most
violent neighborhoods with emergent paramilitary violence in periph-
eral towns. At night I returned home to join in the collective evening
experience of the broadcast news, to gaze numbly at bloody images of
lifeless bodies, and to anesthetize the pain with the inanity of the eve-
ning soaps. I would finally fall asleep late in the evening only to leap up
 Epilogue 

in terror hours later at the sound of a bomb that seemed to explode only
a block or two away.
Once a week several professors from the departments of economics,
psychology, history, political science, sociology, and education came
together in a remote seminar room at the Universidad de Antioquia to
try and make sense of the chaos that seemed to engulf us. We wrote posi-
tion papers and acted as informal consultants to the mayor on the subject
of violence in Medellín, the city with the dubious distinction of being the
international ‘‘cocaine capital’’ and having the world’s highest homicide
rate. Several colleagues were in the midst of concluding an oral history
project with inhabitants of the region then most heavily affected by the
narcotics trade and armed violence (the Magdalena Medio). As I read
the transcripts of these interviews I was repeatedly struck at how despite
being asked about violence in the s, those interviewed came back
again and again to their memories of la Violencia. Reports in the daily
papers and the reaction of Antioquia’s political leaders and economic
elite to the contemporary escalation of violence in the region, moreover,
served to reinforce my sense of déjà vu.
During la Violencia and now today, complex, overlapping-yet-distinct
phenomena are indiscriminately attributed to a single catalyst. In the
s, members of the Conservative government dismissed la Violencia as
a partisan-based problem that could be solved through the deployment
of greater force, ignoring the danger posed by locally sponsored armed
groups that arose to fulfill the function of public authority abandoned
by the state. In the last decade of the twentieth century, various Colom-
bian administrations dismissed the problem of contemporary violence
as one that was essentially reducible to a single issue (the emergence of
the narcotics trade or leftist insurgency) and concluded as their predeces-
sors did that selectively directed repression could bring about peace. Like
their predecessors, recent administrations have ignored the danger posed
by the proliferation of publicly endorsed but privately armed groups en-
trusted with the duty of maintaining public order. If during la Violencia
the central culprits of violence were seen by officials as small bands of
Liberal/Communist guerrillas and criminal delinquents, they were often
grouped together under the generic rubric of ‘‘bandits’’ or ‘‘revolution-
aries.’’ Moreover, in the contemporary period, naming has been simi-
larly imprecise. Narcotics trafficking, political dissidence, leftist insur-
gency, criminality, and civil disobedience are indiscriminately lumped
 Blood and Fire

together under the convenient denomination of first the unruly, urban

slum dweller (the sicario) and, more recently, the ‘‘narco-guerrilla.’’ In
both instances civilians are likeliest to bear the brunt of stepped-up cam-
paigns of violence (exercised by the state or its ‘‘enemies’’) rather than
the armed groups supposedly responsible for promoting violence in the
first place.
A comparison of past and present violence—an exercise suggested by
the very people most affected by violence and whose lives have been fun-
damentally altered by the experience of violence in the last fifty years—
highlights the significance of the Antioqueño experience during la Vio-
lencia and the continuities and ruptures between what might on the
surface appear to be an unchanging and endemic situation of constant
strife. For instance, what would it mean if violence tended to recur in
the same places over time? And, if violence today still tends to be largely
concentrated in the areas that were most violent at mid-century—when
in principle the underlying causes of violence were quite different from
what they are now—then what relationship might there be between mid-
century violence and contemporary violence? Is it possible that to under-
stand Colombian violence—not just la Violencia—we must ask different
questions and draw different connections than the ones that have histori-
cally oriented studies of violence?
A great deal of emphasis, for instance, has been placed on the impor-
tance of partisan labels in the formation of Colombian identity and the
incidence of Colombian violence. Indeed, most analysts of the violence
have tended to focus almost exclusively on the nation’s political history,
analyzed from a top-down perspective, and traced through the history of
the Liberal and Conservative parties. This book confirms that the parties
were undeniably important. Partisan conflicts and differences were sig-
nificant in shaping people’s beliefs and comportment in Colombia as well
as in producing the environment in which la Violencia developed. How-
ever, the Antioqueño case also makes evident that partisan identity and
individual notions of citizenship and identification with the state varied
quite considerably within Colombia. Ultimately, the notion of mono-
lithic Colombian parties does not explain how people understood their
participation in the parties, how politics functioned at the local and re-
gional level or why strong identification with a particular party should
have inevitably led to violence.
The variety of experiences characterizing la Violencia in Antioquia in-
 Epilogue 

stead suggests the importance of considering how class, ethnicity, kinship

relations, local power distribution, geography, and concepts of citizen-
ship intersect to shape people’s sense of identity and their willingness
or refusal to accept violence as a legitimate political tool. In Urrao, for
instance, a long history of cooperative relations between a minority of
moderate Conservatives and a majority of moderate Liberals combined
to ensure that local politicians exhausted nonviolent, democratic avenues
of protest against the intrusion of extremist regional appointees intent
on marginalizing the town’s historically recognized leaders. The refusal
of some regional authorities to respect the mandate of locally elected offi-
cials or to negotiate with them to resolve partisan differences eventually
prompted the emergence of armed opposition groups who challenged
the state, but violence was not the town’s first or preferred approach. In
parts of Urabá, in contrast, where no prior history of cooperation with the
regional government existed and where few democratic mechanisms for
the exercise of political opposition functioned, violence often appeared
as the only viable response available to local inhabitants against regional
authorities and policies perceived to be repressive and exclusionary.
Despite my suggestion that the history of a particular population or
area’s relationship with the regional or central state was a determinant
factor in the way la Violencia evolved in Antioquia, the issues at stake dur-
ing the conflict bear a greater relationship to contemporary Colombian
violence than they do to the partisan-based civil wars of the nineteenth
century. The ‘‘empleo-mania’’ (patronage-based hiring frenzy) that his-
torian Charles Bergquist found partially fed the partisan disputes that re-
sulted in Colombia’s greatest nineteenth-century civil war, the War of the
Thousand Days, for instance, appears to have been far less pronounced
in Antioquia than elsewhere in Colombia. Competition for the govern-
ment sinecures and access to patronage positions and influence that were
controlled by the central state through partisan monopoly simply do not
appear to have motivated widespread support for the War of the Thou-
sand Days in Antioquia. But perhaps the main reason why la Violencia
does not appear to have been a direct legacy of nineteenth-century civil
conflict, however, is that it was fueled by a peculiar confluence of fac-
tors that emerged in the twentieth, not the nineteenth, century. Among
these factors was the growing centralization of power in Bogotá (at the
expense of local and regional power); the closing of the coffee frontier
and its promise of social and economic mobility to the poor through land
 Blood and Fire

ownership; the expansion of industrialization without a commensurate

increase in employment by mid-century; and the growing privatization
of land and resources on the periphery (the traditional outlet for disap-
pointed settlers and workers).
The closure or restriction of traditional avenues of mobility, more-
over, coincided with the rise of a distinctive middle sector in both parties
that was anxious to construct an electoral machine explicitly based
on lower-class support. Therefore, ‘‘sectarian populism’’—an important
component, although not necessarily that which led to the greatest in-
tensity of violence in Antioquia—was a phenomenon relatively new to
regional, mid-twentieth-century politics. In other words, though pur-
ported ideological differences between the parties existed in both the
nineteenth and twentieth centuries, the issues that fed off such partisan
differences and enabled la Violencia to develop into intense and even en-
demic strife were recent in Antioquia and geographically circumscribed.
It is not coincidental that the areas most severely affected during la Vio-
lencia were also those experiencing the greatest economic and social dis-
location and transformation during the decades immediately preced-
ing the violence, or that electoral conflict did not necessarily lead to
widespread violence in core areas within Antioquia. Severe violence in
twentieth-century Antioquia was the result of an intersection between
geography and social unrest rooted in structural inequalities in which
partisanship played a role, but not necessarily a determinant one. In con-
trast to the nineteenth century, moreover, the leaders of armed move-
ments during la Violencia were local individuals of middle- and lower-
class origin, not members of the elite (as had generally been true of
nineteenth-century civil wars).
In many instances, mid-twentieth-century violence was not the spon-
taneous result of inherent local partisan conflict but was rather con-
sciously spearheaded by selective sectors of the regional state or tacitly
encouraged by local bosses to advance interests that had little or nothing
to do with ideological differences. Armed groups were privately financed
but operated in the name of the state. These findings are not unique
to Antioquia’s experience of la Violencia. Gonzalo Sánchez’s work on
Tolima, Keith Christie’s work on Caldas, and other regional studies have
suggested the existence of state-directed violence and the connection be-
tween paramilitary organizations (referred to as ‘‘pajaros’’ or ‘‘contra-
chusmas’’) and certain governors, regional bureaucrats, and officers of
 Epilogue 

local and regional Conservative party directorates. But, while regional

studies have suggested the existence of state terror as a generalized policy
and assumed that such a policy was mandated from Bogotá (that is, the
central state) where it enjoyed the unequivocal support of a monolithic
Conservative government, the evidence from Antioquia suggests a far
more complex picture of the evolution of state-sponsored violence. Re-
gional and local authorities rather than a central state played determinant
roles in promoting private terror in Antioquia and shaped the attitude
of both inhabitants and party officials toward the use of violence as a
political tool.
There was no consensus in Antioquia regarding the use of either offi-
cial violence or the tacit endorsement or toleration of privately sponsored
violence for partisan ends, however. Numerous towns and even members
of the ruling Conservative party rejected the use of both official violence
and private forms of terror as legitimate mechanisms for resolving civil
conflict. Violence—terror, repression, homicide, forcible displacement,
and rape—moreover, was selective, and the determination of its use by
official forces was largely shaped by the nature of the relationship be-
tween the regional authorities or state and local populations. This rela-
tionship, moreover, was in turn shaped by ethnic, cultural, and socioeco-
nomic factors, rather than partisan considerations alone. It was far easier,
for instance, to justify and impose a regime of terror against individuals
or areas perceived to deviate from a sociocultural norm of Antioqueñi-
dad than it was to do so in centrally situated towns perceived to embody
idealized regional values and practices. Thus, the direction, form, inci-
dence, and intensity of violence was determined not only by the majority
political affiliation of a particular population, but by a combination of
factors that included the nature of the local economy, the distribution
and use of land, the racial and ethnic identity of the inhabitants, the de-
gree of conformity to a regional ideal of political and social organization,
and the relative strength of either local forms of collective organization
or the state’s own ability to assert it’s authority. For example, miners in
Segovia were able to successfully impede the conservatization of their
municipality and the violent imposition of regional appointees through
active resistance and reliance on a history of radical collective association.
The regional authorities’ prior experience with Segovia’s mining com-
munity deterred even adamant Conservative extremists from deploying
paramilitary forces to repress local Liberals despite the local inhabitants’
 Blood and Fire

political dissidence, militant union activity, and evident failure to con-

form to the values of Antioqueñidad. A history of collective association
and dissident political mobilization was cemented by shared cultural and
ethnic practices in Segovia, while a history of alienation from the regional
authorities reinforced a powerful sense of local identity and a tradition
of resistance. Faced with concerted opposition in an area of strategic re-
sources and production, the regional authorities backed down in their
confrontations with the inhabitants of Segovia.
If other than strictly partisan factors shaped the regional state’s de-
cision to engage in repression rather than negotiation with local popu-
lations during la Violencia and if these factors shaped local perceptions
of the legitimacy of the state and the legitimacy of using violence to re-
sist the intrusion of state-sponsored violence, then we must rethink our
understanding of the causes of violence and of why violence was con-
centrated in some areas while other areas with a similar political outlook
went relatively untouched. Rethinking the causes of violence and why it
took particular forms in particular areas, moreover, also requires a re-
thinking of the seemingly generic and generalized character that has been
historically attributed to la Violencia.
It is difficult to know how representative the pattern of violence ex-
perienced in Antioquia may have been of violence in other parts of
Colombia at mid-century. In part this is true because no other study of la
Violencia has focused on the significance of factors other than partisan-
ship and class as explanations of violence. The conflation of geography,
strategic resources, an absent state, contrasting cultural and ethnic iden-
tities, and political dissidence in the Antioqueño areas where violence
proved most severe, would nonetheless seem to suggest parallels or com-
parisons to other episodes of selectively intense violence in twentieth-
century Latin America. For instance, in its recent inquiry into the thirty-
six-year-old Guatemalan civil war in which some two hundred thousand
Guatemalans were disappeared or killed, the Guatemalan Truth Commis-
sion acknowledged the role that a complicated history of racism, political
intolerance, selective indigenous support for leftist insurgency, and the
protection of specific economic interests played in making some groups
and areas of Guatemala more likely to bear the brunt of counterinsur-
gency terror than others.1 Similarly, the Peruvian state also identified spe-
cific regional populations as hotbeds of radical support for the Maoist
insurgents in its recent struggle against Sendero Luminoso, tailoring its
 Epilogue 

actions in such areas both in relation to the ethnic and cultural identity of
the local population and the level of perceived local support for the state.2
In both the Guatemalan and Peruvian cases, an a priori identification of
particular groups as culturally distinct, coupled with a history of hostility
between state and citizen, contributed to both the intensity of seemingly
generalized violence and the approach applied by the state to stamp out
local insurgent support. The point of the comparison to the Antioqueño
case is not to suggest a simple causal relationship between racism and vio-
lence, but rather to suggest how histories of perceived difference can fun-
damentally shape the relationship between the state and specific groups
and increase the potential for conflicts to be resolved through repression
rather than negotiation. The current conflict in Chiapas, Mexico, also
illustrates the intersection of ethnicity, geography, and violence. There,
an area historically rich in resources, in which ownership of the land is
largely concentrated in a few hands, houses a population at the margins of
economic survival that is also perceived as ethnically and culturally dis-
tinct from the inhabitants of central Mexico. Much as was true in Colom-
bia during la Violencia, in contemporary Chiapas, regional authorities
direct, or tacitly endorse, privately sponsored armed groups to defend
capitalist production and a traditional party system against indigenous
insurgents and the threat of Communism.3
In Colombia, moreover, the coincidence of ethnicity and culture,
geography, and class has persisted beyond the period of la Violencia to
define the parameters of contemporary violence and its primary victims.
The Solidarity Network, a Colombian government agency that deals with
the problem of internal displacement, estimates that fighting between
paramilitary groups, guerrillas, and the army has forced . million of
Colombia’s  million people to relocate since . Half of these dis-
placements have taken place between  and .4 Human rights ana-
lysts estimate that the number of displaced people in Colombia is the
greatest in the Western Hemisphere, but in contrast to the internal refu-
gees in similarly war-torn societies such as Chechnya or Kosovo, Colom-
bian refugees ‘‘are not readily distinguishable by their ethnicity, language,
or religion.’’ Paul Oquist estimated in his  study of la Violencia in
Colombia that some two million people were displaced by violence be-
tween  and , a time span and number of refugees comparable to
that characterizing contemporary Colombian violence. The individuals
forced to relocate during la Violencia, like contemporary refugees, were
 Blood and Fire

also not considered to have been distinguishable by ‘‘their ethnicity, lan-

guage, or religion,’’ although the Violencia refugees were in their majority
distinguished by their partisan identity. In reality, however, the notion
that refugees both during la Violencia and in contemporary Colombia
are indistinguishable from the nondisplaced population is only partially
true. Several organizations that track contemporary human displace-
ment have noted the disproportionate presence of individuals of Afro-
Caribbean and indigenous descent (from the Chocó, Magdalena River
Valley, and Urabá, for instance)—many of whom are women and chil-
dren—among the refugees of violence in Antioquia and Colombia as a
whole.5 Increasingly, moreover, those forced to relocate to cities such as
Medellín eschew internment in refugee camps or the official label of ‘‘dis-
placed person’’ for fear of the term’s negative associations. Current refu-
gees from peripheral areas fear that their towns of origin are assumed by
urban inhabitants to be ‘‘nests’’ of leftist sympathizers and that their pres-
ence in particular neighborhoods will provoke retaliation by right-wing
paramilitary groups.6
During la Violencia competition for scarce employment and resources
in cities such as Medellín influenced tensions between established in-
habitants and newly arrived rural migrants fleeing violence. But urban
tensions were also shaped by perceived cultural differences between an
established population drawn primarily from nearby, centrally situated
municipalities and refugees from peripheral towns in western, north-
ern, northeastern, and far eastern towns (Ituango, Frontino, Dabeiba,
Caicedo, Urrao, Remedios, Caucasia, Puerto Berrío, and the port towns
along the Magdalena River). These towns were thought to be ‘‘unruly’’
and ‘‘different’’ even before the advent of violence. In Alonso Sala-
zar’s interview-based account of the emergent youth assassin groups in
Medellín, his respondents repeatedly allude to both a history of physi-
cal displacement from peripheral towns in the aftermath of la Violen-
cia and the reproduction of violence within their families as factors that
influenced their decision to become hired guns for the narcotics trade.7
What is significant is not whether a direct correlation in fact existed, but
that the perception of a relationship between geographic origin, cultural
identity, and violence should have persisted and been reproduced among
the very people likeliest to take part in and be affected by violence. For
Salazar, the coincidence of these factors confirms the existence of a ‘‘cul-
ture of violence’’ in Antioquia. But I would suggest something different.
 Epilogue 

I see a continuation of the perception of dissidence and resistance attrib-

uted by the regional authorities to inhabitants from the periphery dur-
ing la Violencia that was used to justify repression against them. Rather
than a ‘‘culture’’ of violence, Antioqueño violence must be read against
the backdrop of a history of internal colonialism and exclusion in which
multiple points of difference between regional authorities and local in-
habitants existed over such issues as land and resources, access to political
power, and the right to self-determination. These differences have been
conveniently reduced to a question of inherent rebelliousness perceived
by regional authorities to threaten some idealized Antioqueño stability,
regardless of the time period in question.
In current analyses of the problem of violence and internal displace-
ment in Colombia, the victims of violence are also often characterized as
‘‘ordinary civilians’’ who find themselves caught in the crossfire of what
are considered by international observers to be ‘‘armed groups that have
been warring with increasing ferocity since the ’s.’’ 8 But this per-
ception—that the problem of civilian displacement in which ‘‘ordinary’’
people are forced to relocate dates mainly from the emergence of leftist
insurgency—is largely erroneous. La Violencia, not the s, is the point
of departure for understanding current violence. Indeed, another simi-
larity between the contemporary period and la Violencia is that then, as
now, the majority of those displaced are less ‘‘victims caught in the cross-
fire’’ between armed groups, than the primary targets of those armed
groups.9 In other words, only very recently in Colombia have the para-
military groups, guerrillas, and the Colombian armed forces begun to
consciously combat each other, rather than directing their actions, as has
historically been the case, against an unarmed civilian population sus-
pected of aiding or tolerating the presence of the ‘‘enemy.’’
As was also true during la Violencia (at least in Antioquia), violence
in contemporary Colombia is the responsibility of both the left and the
right, but the majority of those currently displaced have been forced to
move by the presence of right-wing paramilitary groups, while leftist
groups (the equivalent of la Violencia’s guerrilla groups) are responsible
for approximately a third of all displacements, and the Colombian Army
for less than  percent.10 In both periods, moreover, paramilitary groups
have grown up with the tacit support of selected members of the regional
government and even the armed forces.11 Indeed, the precedent for the
current right-wing paramilitary phenomenon in Antioquia—which has
 Blood and Fire

now spread to other parts of Colombia, but which was first explicitly en-
dorsed and supported by an Antioqueño governor and other regional
officials and members of the departmental bourgeoisie as a quasi-official
policy—may be traced back to the endorsement of the contrachusma and
the legal arguments made by Antioqueño governors to justify the arming
of independently organized civilian groups during la Violencia. The cata-
lysts for the emergence of privately organized terror groups both during
la Violencia and now, moreover, are nearly identical: the defense of stra-
tegically located valuable resources in areas where the state exerts little
authority and does not command the loyalty or identification of the ma-
jority of the local population.
While the growth of paramilitary organizations is hardly unique
to Antioquia, there are aspects of regional political practice and self-
perception that have contributed to the enthusiasm with which privately
organized forms of public order maintenance have been embraced in the
department from the period of la Violencia to the present. Long before the
advent of partisan strife in the s, Antioquia’s affluence, demographic
and physical size, strategic resources, and pronounced sense of regional
identity enabled it to act with considerable autonomy vis-à-vis the cen-
tral state and even the national directorates of the two main parties. Re-
gional autonomy coupled with a deeply ingrained sense of local pride
persuaded Antioqueños and their leaders that they were better able to
rule their department’s fortunes and determine its policies than authori-
ties based in Bogotá. A refusal to follow Bogotá’s lead may paradoxically
have enabled much of Antioquia to elude the partisan conflict that en-
gulfed and paralyzed other Colombian regions between  and .
But regional defiance of central government dictates and disagreements
within the leadership of the Conservative party eventually meant that
when violence did become severe in the department, governors such as
Braulio Henao Mejía found themselves isolated and unable to obtain
needed arms and support from the central government with which to
squash armed guerrilla groups operating in peripheral areas. Faced with
an absolute incapacity to maintain public order in far-flung areas where
violence threatened regional control and resources strategic to Antio-
quia’s economy, regional officers—some reluctantly, others enthusiasti-
cally—endorsed the creation of local paramilitary forces to combat guer-
rilla insurgency. Even after it became apparent that such groups preferred
to attack civilians rather than pursue guerrillas and that they were no
 Epilogue 

less prone to collusion and theft than their supposed enemies, Antio-
quia’s authorities continued to justify the brutal actions of such groups
by appealing to notions of regional identity and the defense of regional
interests. The contrachusma, it was alleged, were Antioqueños through
and through—unlike the national police or members of the armed forces
whose ‘‘alien’’ customs and origins (that is from the Afro-Caribbean coast
or the indigenous areas of Boyacá) endangered the very stability of Antio-
quia’s cultural identity. Racism, regional isolationism, and desperation
combined to justify terror in the name of defending departmental sov-
ereignty and honor during la Violencia. This set a dangerous precedent
whose reverberations may still be felt in the regional endorsement of rural
and urban autodefensas and convivirs, the privately financed and pub-
licly endorsed security or self-defense organizations that have sprung up
throughout Antioquia in the wake of the armed forces’ perceived inability
to protect the lives and interests of rural and urban capitalists.
In the s and s, Antioquia once more felt abandoned by the
central government and was forced to face alone an escalation of violence
due to narcotics trafficking. As it became apparent to regional officials
that the army and police—poorly trained and paid and too scattered to
seriously engage in combat with leftist guerrillas or insufficiently moti-
vated to go after popular narcotics dealers—would not be able to defend
against a rising tide of kidnappings, homicides, and theft, Antioquia’s
authorities and selected members of the regional bourgeoisie once more
decided that only regionally paid and organized forces made up of ‘‘dyed-
in-the-wool’’ Antioqueños could hope to control violence. Monthly con-
tributions elicited from harried cattle ranchers in guerrilla-controlled
territories have gradually given rise to paramilitary forces whose bru-
tal efficiency far surpasses that of the Colombian army or police. Like
their predecessor organization—the contrachusma—paramilitary forces
in contemporary Antioquia and Colombia have gone far beyond the ini-
tially limited mandate of self-defense to ‘‘cleanse’’ entire populations per-
ceived for a variety of reasons to ‘‘deviate’’ from the accepted norms of
political belief, economic organization, or personal association.12
The coincidence between the areas most affected by violence in the
s and the areas currently most affected by violence in Antioquia
today, and the state’s endorsement of a paramilitary solution to violence
in those areas both then and now, is striking.13 Although the issues that
have spurred conflict in those zones have shifted over time, many of the
 Blood and Fire

elements that currently contribute to making some areas more violent

than others—and the state’s role in promoting that violence—are de-
cades old. Then as now, violent areas constitute important vectors of con-
tact and exchange with both the outside world and other Colombian de-
partments and are poorly policed and regulated. Mutual mistrust and a
history of indifference or incapacity to make the rule of law felt in such
areas condemn relations between the state and local inhabitants to failure.
These features ensure that the inhabitants and resources located in these
areas continue to be the foci of brutal competition for control between
the state and illicit forces—whether made up of political insurgents, para-
military forces, or narcotics and contraband traffickers—just as they were
during la Violencia. The main avenues of access for illicit arms, goods, and
narcotics, for instance, continue to be centered in northwestern Colom-
bia in the regions of Urabá and parts of the Chocó. In these areas, forces
on either side of the ideological divide confront each other to dominate
power in specific territories, often at the expense, but also with the tacit
support of the state and its official bureaucracy. In fact, just as was true
during la Violencia, it is sometimes difficult to discern exactly who is col-
laborating with whom and for what purpose in these contemporary epi-
sodes of violence in peripheral areas.
The situation of the Magdalena Medio area of Antioquia underscores
both the continuities in violence over time and the complexity of the
issues contributing to the endemic character of violence in peripheral
zones. When narcotics traffickers emerged as a powerful force in Colom-
bian society in the s and increasingly laundered their illicit profits
through the purchase of land in departments like Antioquia, they sought
out areas such as the Magdalena Medio. Here, the Pablo Escobars and the
Rodriquez Gachas of the underworld came into conflict with leftist guer-
rillas embroiled in a much older battle with large property owners of es-
tates and extractive industries like mining. As newly enriched, large land-
owners with unparalleled access to state-of-the-art arms and abundant
personnel, narcotics traffickers were able to ally with local landowners
(and members of a beleaguered military) to form ruthlessly effective and
extensive paramilitary forces with which to eradicate potential peasant
protest and combat leftist insurgents. The result has been extraordinarily
bloody. But even in its modern manifestations, contemporary violence
in the area, which may seem far removed from la Violencia, continues
to bear vestiges of that earlier conflict. The local inhabitants in mining
 Epilogue 

and ranching communities such as Segovia, Remedios, and Puerto Berrío

still identify with dissident political parties and militant labor unions just
as they did at mid-century.14 They still consider themselves to possess
a cultural identity different from that of Antioqueño officials based in
Medellín, and they still appear to those officials as defiant, godless, and
rebellious peoples who invite the massacres and repeated political assas-
sinations that have been their recurrent fate.
The case of la Violencia in Antioquia thus further suggests the need
to rethink the definition and use of certain analytical categories such
as ‘‘region,’’ ‘‘state,’’ and ‘‘citizen’’ in studies of violence. To begin, at
least two Antioquias existed. One—an idealized Antioquia—was a place
where people had a modicum of access to social and economic mo-
bility; land ownership was a real possibility; communities were linked
by strong common beliefs, Catholicism, and specific social practices and
values; and people preferred to do business rather than take up arms.
The idealized Antioquia seemed an unlikely candidate for an outbreak
of widespread conflict at mid-century and, in fact, never experienced
significant levels of violence. The ‘‘other’’ Antioquia—where land was
concentrated in a few hands, where agribusiness estates and cattle ranches
dominated production, where landless migrants and seasonal workers
defined the labor market, and where extractive industries not subsistence
farming was the norm—is an Antioquia whose existence is barely ac-
knowledged. The residents of this other Antioquia have historically been
viewed with distrust and have been excluded from real participation in
regional power. For these Antioqueños, both la Violencia and contempo-
rary violence represent but a more extreme manifestation of the hostility
and alienation that has historically characterized their relationship to the
regional state.
As a category of analysis, ‘‘region’’ is usually defined in relation to
seemingly unproblematic administrative and physical boundaries with-
out regard for the complicated cultural, economic, ethnic, and social
implications that may shape perceptions of regional identity and define
exclusion or inclusion in regional and central structures of power.15 Yet
regions and departments are not homogeneous and neither are the rela-
tions and structures of authority that operate in them. The Antioqueño
case points to the need to restructure the typologies and analytical cate-
gories by which the study of violence has traditionally been approached.
This would call for a remapping of violence in Colombia that would likely
 Blood and Fire

reveal nodes of conflict with shared characteristics and commonalities

that transcend ‘‘departmental’’ boundaries in the official definition of the
term. If the criteria by which violence in regions were redefined, it would
be possible to discern continuities over time even when some of the fac-
tors contributing to the emergence of violence shift or expand. Surely
such knowledge would give us a much better basis from which to address
the issues that contribute to producing recurrent episodes of violence in
particular areas and give greater promise of arriving at nonviolent solu-
tions in the future.
In this study I have suggested that not everyone included within the
physical parameters of the state of Antioquia necessarily or automatically
could lay claim to Antioqueño identity. I have also suggested that the con-
struction of regional identity—and hence the right to claim citizenship
and take part in decision-making or make demands of the state—was
predicated on a constantly negotiated, unstable, and highly exclusionary
notion of sociocultural propriety. The import of this redefinition of the
concept of region and citizenship is multiple. Many analyses of la Violen-
cia (and certainly two of the most influential ones—Daniel Pécaut’s and
Paul Oquist’s) have drawn attention to the importance of the power (or
absence) of the state in Colombian society and in the definition of Colom-
bian identity. Leaving aside for the moment the question of whether a
central state existed at all in Colombia or whether it collapsed in the face
of violence, what if no consensus existed among Colombians as to what
constituted the state or what their relationship to it might be? What if
more than one notion of the state existed and this were heavily influ-
enced by local and regional experiences? This study of Antioquia is not
the first to suggest that the concept of the central state in the context of the
Colombian Violencia is a potentially problematic or even irrelevant cate-
gory of analysis for understanding individual perceptions of authority or
power on the ground. For most Colombians the central state was an ab-
stract concept and power was largely exercised and determined locally or
regionally, not in Bogotá. In the Antioqueño case, moreover, it is evident
that the central state possessed very few means of making its authority felt
at the regional or municipal level, and that it relied in large measure on
the complicity or cooperation of local and regional authorities for even
its tenuous presence on a day-to-day basis.
But the Antioqueño case also suggests that local inhabitants developed
and deployed notions of the central state that were inextricably shaped
 Epilogue 

by their understandings of the workings of power in their localities and

within the regional state. These understandings in turn influenced their
ability to claim citizenship and access power structures. Perhaps one
of the most remarkable aspects of studying the course of la Violencia
in Antioquia is the discovery of the degree to which people—ordinary
people who sometimes couldn’t write their names and had never been to
Medellín or had much contact with government—believed in the politi-
cal system and made use of it. Regional inhabitants insisted on writing
petitions, vocalizing their dissatisfactions, and articulating their notions
of the proper function of the state and its obligation to its citizens even
when the individuals who held regional positions of power belonged to
the opposition. These same people—and they were surprisingly numer-
ous—regardless of their party affiliation, often and fearlessly made clear
to local and regional officers their disagreement with the use of violence as
a political tool. They often went so far as to include their national identity
numbers after their names on countless letters and petitions so that they
could be easily identified and located. They engaged in acts of civil dis-
obedience by refusing to pay the policemen who were imposed by the re-
gional government and were empowered to divest locally elected leaders
from their rightful municipal offices. Ordinary people also risked their
lives to save the lives of neighbors and family members targeted by vio-
lence, just as many Colombian journalists, judges, academics, and ordi-
nary folk do today. And they saw through the claims of political legiti-
macy made by armed groups on both sides of the ideological fence as
motivated by little more than ill-disguised ambition rather than honor.
The experience of Antioquia during la Violencia suggests that Colom-
bians were much more sophisticated about politics and their expectations
of the state than they are usually given credit for. It is significant—and
I have felt compelled to highlight the phenomenon in this study—how
many people resisted partisan violence and refused to take part in para-
military violence (including landowners who might have been expected
to benefit from it). They did so even when the state offered them the
possibility of doing so with impunity. It is equally significant that many
Antioqueños attempted to use the law and nonviolent mechanisms—
such as voting, petitions, editorials, civil disobedience—to influence the
state and its policies, and it is important to acknowledge that mechanisms
did exist to mediate negotiated solutions to conflict. But it is equally im-
portant to note that such mechanisms of nonviolent resolution to conflict
 Blood and Fire

were not present or accessible everywhere and that a direct correlation

appears to exist between the possibility of a negotiated solution to vio-
lence (or even its preclusion) and the impossibility of any response but
repression on the part of the state or armed resistance on the part of a
local population.
It is tempting to yearn for a period in Colombian history when the dif-
ferences that generated violence seemed black and white—easy to under-
stand and pinpoint. It is equally tempting to want to draw a clear dis-
tinction between the origins and purpose of violence in the s and
that which affects Colombia today, to want to view the two as sepa-
rate and distinct and violence as a whole as isolated, chaotic, and ex-
traordinary. Finally, it is tempting to believe the powerful and by now
largely internalized idea that there is something unique or peculiar to
the Colombian case, some intractable cultural propensity that makes vio-
lence there inevitable and endemic. But there is no such animal. If one
listens carefully, if one ignores the cacophonous dissonance of simplistic
assessments, echoes of the stories of violence recounted here can be heard
amid the survivors of the conflicts in Rwanda, Northern Ireland, Bosnia,
Chechnya, Guatemala, or Sri Lanka. I have tried to hear the stories be-
hind the stories, to piece together fragments of lost stories and stories too
horrifying to want to remember or retell. At times I have wanted to shut
out those stories, to cover my ears, to run away from violence, includ-
ing the potential for it that lies unrecognized in all of us. But the dead
surrounded and tugged at me, while the living would not let me rest, re-
minding me that even when I no longer wanted to that I must tell their
stories. Finally, I hope that I have betrayed neither those who died nor
those who survived and entrusted their tales to me over these many years.
Appendix A: Tables

Table A.1. Deaths due to violence in  most violent municipalities, –

Percentage of Cumulative Population, Deaths per 

of
Municipality Deaths total deaths percentage     population

Dabeiba  
.
 
.
 ,   .

Puerto Berrío  .  . ,


. 
Urrao  
.
.

, 
.
Cañasgordas
 .
. ,  
.
Remedios   .
 . , .

Frontino 
.


.

, .

Yolombó 
.

.
, .
Betulia 

.

.
, .
Antioquia 
.
 
. , 
.
Caucasia 
.
 .
,
.

Peque 
.
  . , . 
Anzá  .  . , 
.
Cisneros  .

.
 , .

Salgar  . .
 , .

Sabanalarga  .
 .
,  .
Buriticá  . . , .
Ituango  . .  ,
.

Cáceres  . .
 ,  .

Maceo  . . ,  . 
Turbo  .
 . ,
 .

San Luis  .

. ,
 .

Caicedo  .
 . ,
.

Cocorná  .
.  ,
 .
Bolívar  . .

, .
Abriaquí  .

.

,
 .

All others 
.  

,,

.

Department
total , 

,,
.

(Source: , , vol. , ‘‘Informe sobre la acción del bandolerismo, –,’’ Medellín,  May ;
, Panorama Estadístico de Antioquia, siglos xix y xx, table ..., pp. –)
 Appendix A

Table A.2. Deaths due to violence, –, by subregion

Percentage Cumula- Popula- Deaths per
of total tive per- tion, 

of  
Region Deaths deaths centage   population

Magdalena 
 .
.
,
 .
Occidente ,
 
. .  ,
 
.
Bajo Cauca 
 . 
.  ,

.
Urabá  .
. ,

.
Nordeste  .
 . 
, .
Suroeste 
  . .
  ,

.

Norte 
.
. ,

.

Oriente 
.
 .  ,
.
Sur

.
 . ,
.

Central 

.
 


,
.

Department total , 

,,
.

(Source: , , vol. , ‘‘Informe sobre la acción del bandolerismo, –,’’ Medel-
lín,  May ; , Panorama Estadístico de Antioquia, siglos xix y xx, table ..., pp.
–)
 Appendix A 

Table A.3. Percentage population change by subregion, –

Region  –   –
 
–   – 

Urabá .

.
.


.

Norte 
.

.

.

.

Bajo Cauca .

.

.
.

Nordeste .
 .
.
.

Magdalena .
 .

.
.

Occidente  .
.
.
.

Suroeste 
.

.
.
.

Sur .
.

.
.

Oriente 
.

.
.
.

Central
.
.
.
.

Departmental .
.

.
.

(Source: , Panorama Estadístico de Antioquia, siglos xix y xx, table ..., pp. –)
 Appendix A

Table A.4. Civilian and ‘‘bandit’’ deaths, –, by subregion

Region Civilians ‘‘Bandits’’ Totals

Urabá 
 
Norte   

Bajo Cauca   


Nordeste    
Magdalena

 

Occidente  ,
 ,

Suroeste
  

Sur



Oriente 
 
Central   

Totals ,  ,

,

(Source: , , vol. , ‘‘Informe sobre la acción del bandolerismo, –,’’ Medellín,
 May )
 Appendix A 

Table A.5. Deaths by year and subregion, –

Number of deaths Percentage of total
Region    
     
Total    
     
  –

Bajo Cauca 

  
 

  
Central      
 

Magdalena



 


   
Nordeste 


   

   
Norte 
 


 

Occidente


 
  ,



  

Oriente
  
 

 

Sur   


 

Suroeste   


 

  
 


Urabá 

  

 
Department  

, 
, 

(Source: , , vol. , ‘‘Informe sobre la acción del bandolerismo, –,’’ Medellín,  May )
  Appendix A

Table A.6. Political prisoners, La Ladera Prison, Medellín, –

Number of prisoners Percentage of total prisoners from
during year place of origin by year
Town of origin  
     
 
– 
 
     
 
– 

Amalfí



.

.

.

.

Antioquia Vieja

  
.

.
.
.

El Bagre


 
.

.

.

Betania



.

.

.

.

Betulia 


 .

.

.

Buriticá


 
.

.

.

Cañasgordas



.

.

.

.

.

Caucasia      
.

.
.

.

.

Chigorodó





.

.

.

.

Chocó

 
.

.

.

.
.

Dabeiba    

.

.
.

.
.

Frontino



.

.
.

.

.

Ituango   
 
.

.

.

.

.

Liborina





.

.

.

Medellín

  
.

.

.

.

Mutatá     
.
.

.

.
.

Necoclí 

 .

.

.

.

.

Neguá (Chocó)





.

.

.

.

Peque





.

.

.

.

Puerto Berrío 
 
 

.

.
.
.
.

Puerto Perales

  
.

.

.
.
.

Quibdó


.

.

.
.
.

Quinchía (Caldas)





.

.
.

Remedios



.

.

.

.

Riosucio (Chocó)


 
.

.

.

San Luis


 
.

.

.

Segovia





.

.

.

.

San Juan de Urabá  
  .

.

.

.

.

Tagachí (Chocó)


 
.

.

.

 Appendix A 

Table A.6. Continued

Number of prisoners Percentage of total prisoners from
during year place of origin by year
Town of origin  
     
 
– 
 
     
 
– 

Titiribí



.

.

.

.

Turbo 
   
  .
.
.
.

.

Uramita
 

.

.

.

.

.

Urrao    
 .

.
.

.
.

Yalí



.

.

.

.

Total   
  

(Source: , , vol. , ‘‘Presos políticos liberales en la cárcel ‘La Ladera’ por orden de la Cuarta Bri-
gada,’’  May )
 Appendix A

Table A.7. Number of property holdings by municipality, , ,


Municipios      

Abejorral ,
,  ,


Abriaquí
 
 
Alejandría 
  
Amagá
,


,

,
Amalfí , 
, ,

Andes ,  ,  ,

Angelópolis  ,
 ,

Angostura ,
 ,
 ,

Anorí  , 
,
Antioquia 
, ,
Anzá  ,
 ,

Arboletes ,

,
Armenia ,

Barbosa ,

,

,
Bello , ,
 
,
Belmira   ,

 ,


Betania ,

,


Betulia , ,  ,

Bolívar ,

,
 ,

Buriticá   ,
,

Cáceres
 ,

,

Caicedo 
 ,
,

Caldas ,

,

,
Campamento 
,
 ,
Cañasgordas , , ,

Caramanta 
 ,
,

Carmen de Viboral ,
 , ,


Carolina ,

, ,
Caucasia , ,
Chigorodó   ,
 Appendix A 

Table A.7. Continued

Municipios      

Cisneros ,  , 

Cocorná
,

, ,

Concepción  , ,
Concordia ,
, ,



Copacabana ,
,
,

Dabeiba ,
 , , 
Don Matías ,
 ,
 ,

Ebéjico ,
 ,
 ,

Entrerríos  ,

,

Envigado ,  ,
 ,
Fredonia ,

 ,

 
Frontino , , 
,
Giraldo
 ,
 ,
Girardota , , ,

Gómez Plata ,

, 

Granada ,

,
,

Guarne ,

,
 ,


Guatapé  ,

,

Heliconia ,
 , ,

Itagui ,
 ,

,
Ituango
,
 ,  ,
Jardín ,  ,

,

Jericó ,

, ,
La Ceja , ,

 ,

La Estrella , ,
 , 

La Unión  ,

,

Liborina ,
,
 ,


Margento , ,

Marinilla
, , ,
Medellín  ,
,

 ,
Montebello , ,
, 
 Appendix A

Table A.7. Continued

Municipios      

Murindó 
,

Mutatá 

Nariño ,
, 

Nechí 
,

Olaya 
 

Pavarandocito  ,

Peñol ,
,
 
Peque   , 

Pueblorrico , 
,
,

Puerto Berrío ,
,
 ,

Remedios  , 
,
Retiro ,
 ,
, 
Rionegro ,
 , ,
Sabanalarga
 ,
 ,
Salgar
,
 , 
San Andrés  ,
 , 
San Carlos ,


, ,

San Jerónimo ,
 ,

,

San Luis , , ,
San Pedro ,

, 
,

San Rafael , ,


, 

San Roque
, ,

,

San Vicente , ,
 ,


Santa Bárbara ,

,
,

Santa Rosa de Osos ,

 , ,

Santo Domingo ,

,

,

Santuario ,
, ,


Segovia ,


, ,

Sonsón , ,


,

Sopetrán ,
 ,

,

Támesis
,
,  
 Appendix A 

Table A.7. Continued

Municipios      

Tarso   ,


Titiribí , ,  ,

Toledo
 ,  ,
Turbo
 ,  ,

Urrao , ,
 ,

Valdivia 
,  ,
Valparaiso
 , ,

Venecia , ,  ,

Yarumal ,
, ,

Yolombó
,  , ,
Zaragoza  , ,

Department 
,
,

,

(Source: , Panorama Estadístico de Antioquia, siglos xix y xx, table ..., pp. –
)
  Appendix A

Table A.8. Change in average property values in  most violent municipalities, –
Average
Average Percentage change in annual percentage
Deaths property value average value change in value
Municipality   – 
       –   –   –   – 

Frontino 
 , ,



,
  
Caicedo 

,


,    ,



Sabanalarga  
, ,
 
  

Cocorná  , ,
,  
  

Dabeiba   ,
 , 
  

Cisneros  
 , ,

 

 
Antioquia  ,

,  ,


   
Caucasia  n.d. ,  ,
 n.d.  n.d. 
Buriticá   , 
, ,

  
Urrao  
,

 ,
    
Peque   , 
,    
Cañasgordas
 ,
 ,
,
   
Betulia 
 , 
,


,   

Abriaquí  , ,  ,



 
San Luis   ,
 , 


Yolombó  ,
,
,
   
 
Salgar 
, ,  ,
 
 (
)

()
Puerto Berrío  , 
, , 
 ()  ()
Bolívar  , 
, 
, ,




Cáceres  ,
, , 
 ( ) 
(
)
Ituango   
,
 
,  ,

 

Remedios   
 
, ,
 ,
 
 
Anzá   ,

 
,  , 
 
Turbo   ,
 ,
 ,
 (

) 
 ()
Maceo  n.d. n.d. n.d. n.d. n.d. n.d. n.d.
Department , ,

,
,
  




(Source: , Panorama Estadístico de Antioquia, siglos xix y xx, table ..., pp. –)
Appendix B: Maps

Map B.. Antioquia: Rivers and highways

(Source: Instituto Geográfico Augustín Codazzi)
 Appendix B

Map B.. Contrachusma arms purchases,  and 

(Source: Archivo de la Secretaría de Gobierno de Antioquia,  and )
 Appendix B 

Map B.. Land invasions by guerrillas in Western Antioquia

(Source: Instituto Geográfico Augustín Codazzi; Archivo de la Secretaría de Go-
bierno de Antioquia, , vol. , Caja de Crédito Agrario, April , )
 Appendix B

Map B.. Contrachusma, 

(Source: Archivo de la Secretaría de Gobierno de Antioquia, )
 Notes

Introduction
 Colombia is divided into administrative units known as ‘‘departments’’ that are
equivalent to ‘‘states.’’ In this book I will use the terms ‘‘province,’’ ‘‘department’’
and ‘‘region’’ interchangeably to refer to Antioquia and other Colombian states.
 La Violencia refers to the episodic phenomenon of Colombian civil unrest that
occurred between  and . No exact consensus exists regarding the be-
ginning or endpoint of the Violence. Some scholars argue that violence began
as early as the s, others date its inception to the election of Conservative
Mariano Ospina Pérez as president in , and many mark the assassination
of Liberal leader Jorge Eliécer Gaitán as the ‘‘real’’ beginning of la Violencia.
Regardless of the exact date, most scholars agree that the ‘‘classic’’ period of
violence—waged between Conservatives and Liberals primarily in defense of
traditional nineteenth-century partisan banners—ended with the military coup
of General Rojas Pinilla on June , . Subsequent years of violence are gen-
erally divided into stages characterized by military government (–), the
substitution of partisan concerns with social banditry (the mid-s through
the early s), and, ultimately, the emergence of contemporary armed leftist
guerrilla movements (after ). A fuller elaboration of the complexity of la
Violencia is provided below and in note  below.
 Scholars dedicated to the study of violence in Colombia are informally known
as ‘‘violentologists.’’
 Departamento Administrativo Nacional de Estadística (), Colombia esta-
dística (Bogotá, ), table , ; , Panorama estadístico de Antioquia
(hereafter, ) (Bogotá, ), table ..., ‘‘Población por sexo, según muni-
cipios: censos de –,’’ pp. –.
 Scholars debate the total number of deaths caused by la Violencia. The lowest
estimate of deaths is , while other estimates are as high as ,. I
have relied on the statistics calculated by Paul Oquist in Violence, Conflict, and
Politics in Colombia (New York: Academic Press, ), table ., ‘‘Distribution
of Violencia Fatalities for the Ten Most Affected Departments, –,’’ p. .
Nationally, the provinces of Viejo Caldas (listed by Oquist as ‘‘Caldas Antiguo’’
and currently divided into the provinces of Caldas, Quindío, and Risaralda),
Tolima, Norte de Santander, and Santander were ranked respectively as first,
second, fourth, and fifth in terms of total casualties.
 Oquist, Violence, Conflict, and Politics, table ., ‘‘Migrations, Fatalities, and
Land Parcels Lost due to La Violencia,’’ p. .
 Notes to Introduction

 Charles Bergquist is perhaps the scholar who most forcefully suggested the
existence of a link between coffee and violence. See Bergquist, Labor in Latin
America: Comparative Essays on Chile, Argentina, Venezuela, and Colombia
(Stanford: Stanford University Press, ), p. . Bergquist’s assumption was
not unreasonable; several of the regions suffering the greatest number of casu-
alties during la Violencia (Tolima, Viejo Caldas, and Antioquia), for instance,
were also important centers of coffee production. Coffee production, however,
may have been a less important variable in determining the intensity of vio-
lence in particular localities than other factors such as geographic location, the
nature of the relationship between localities and the regional and central states,
and the local structure of land tenure and relations of production. I discuss the
intersection of production, geography, and violence later in this chapter.
 PEA, table .., ‘‘Defunciones, según causas –,’’ p. ; table .., ‘‘De-
funciones, según causas –,’’ p. ; and table .., ‘‘Defunciones, según
causas –,’’ pp. –.
 National census records for the years between  and  are inconsistent and,
when compared to statistics kept during the same period by the governor’s office
and the armed forces, appear to be seriously flawed. The censuses of  and
, for instance, do not include the category ‘‘homicide,’’ although the latter is
included in census statistics of deaths after . There is, moreover, a big jump
(more than ,) in the number of deaths ruled ‘‘mal definidas’’ (ill-defined) by
the census between the years  and . See , table .., ‘‘Difunciones,
según causas –,’’ p. . This category continued to be included in the
census after , but the statistical jump recorded between  and  de-
clined by the same amount (,) between  and . Between  and
 the number of deaths deemed ‘‘ill-defined’’ grew by , or more per year
and reached a peak of , in  (or two and a half times the number regis-
tered in ). In the meantime, the category ‘‘homicide’’ registered an increase
from  in  to  in . The peak of registered homicides () occurred
in , the year in which violence-related statistics kept by the governor’s office
also peaked. An additional category of deaths called ‘‘Otras muertes violentas
o accidentales’’ remained relatively constant between  and , declined
between  and  (except in ), and then rose back to  levels (ap-
proximately  or so per year) after . See , table .., ‘‘Difunciones,
según causas –,’’ p. . Selected parish death registry records, such as
that of the municipality of Urrao (one of the five towns in which the number of
officially registered deaths due to violence was highest) confirm the impression
that violence-related deaths began to occur during the last months of . This
parish’s records are particularly useful because the priest noted whether a death
was the result of violence and where and how it had taken place. See Archivo
Parroquia, Registro de funciones, ‘‘Partidas de funciones,’’ –, vol. ,
Urrao, Antioquia.
 Notes to Introduction 

 Archivo Privado del Señor Gobernador de Antioquia (hereafter, AGA), ,

vol. , ‘‘Informe sobre la acción del bandolerismo,’’ –, Medellín,  May
. The report divided the dead into ‘‘civiles’’ (civilians) and ‘‘bandoleros’’
(bandits).
 I am not suggesting that towns where no casualties occurred did not experience
partisan tensions; however, these tensions generally did not result in a signifi-
cant number of casualties.
 As scholars have argued, this might seem to suggest that the violence in the
southwest was some of the worst in Antioquia. But Urrao, as I explain below,
was in many ways atypical for the southwest: it produced little coffee and was
connected more directly to the Chocó and western Antioquia than to the cof-
fee zone to the south. Urrao, moreover, was but one town in the southwest; the
majority of other southwestern towns were not severely affected by the violence
between  and .
 The government usually referred to Liberals who took up arms as bandole-
ros (bandits) or chusma (rabble), not guerrillas. Achieving state legal recogni-
tion that these armed citizens were politically motivated individuals rather than
criminals became a serious point of dispute between Liberal and Conservative
officers and the Colombian state. In this text I refer to armed guerrilla groups
with an explicit political objective as ‘‘guerrillas.’’
 The term ‘‘paisa,’’ a variation on ‘‘paisano,’’ or fellow countryman, is colloquially
used to refer to the inhabitants of Antioquia.
 David Bushnell, The Making of Modern Colombia: A Nation in Spite of Itself
(Berkeley: University of California Press, ), p. vii.
 The relationship between rural violence, identity, modernization, and the con-
struction of the nation-state in the late nineteenth and early twentieth cen-
turies is explored in a number of works, among them: Ana Maria Alonso,
Thread of Blood: Colonialism, Revolution, and Gender on Mexico’s Northern Fron-
tier (Tucson: University of Arizona Press, ); Euclides da Cunha, Rebel-
lion in the Backlands, trans. Samuel Putnam (Chicago: University of Chicago
Press, ); John Lynch, Argentine Dictator: Juan Manuel Rosas, – (New
York: Oxford University Press, ); Florencia E. Mallon, Peasant and Nation:
The Making of Postcolonial Mexico and Peru (Berkeley: University of Califor-
nia Press, ); Domingo Sarmiento, Facundo, or, Civilization and Barbarism
(New York: Penguin, ); Richard W. Slatta, Gauchos and the Vanishing Fron-
tier (Lincoln: University of Nebraska Press, ); Paul J. Vanderwood, Disorder
and Progress: Bandits, Police, and Mexican Development (Wilmington, Del.: SR
Books, ).
 Among the many works examining the effects of anti-Communist ‘‘dirty wars’’
and counterinsurgency in Latin America, see Kees Koonings and Dirk Kruijt,
eds., Societies of Fear: The Legacy of Civil War, Violence, and Terror in Latin
America (New York: St. Martin’s Press, ); Jennifer Schirmer, The Guate-
 Notes to Introduction

malan Military Project: A Violence Called Democracy (Philadelphia: Univer-

sity of Pennsylvania Press, ); Robert M. Carmack, ed., Harvest of Violence:
The Maya Indian and the Guatemalan Crisis (Norman: University of Okla-
homa Press, ); Brian Loveman and Thomas M. Davies Jr., ed., The Politics
of AntiPolitics: The Military in Latin America, rev. ed. (Wilmington, Del.: SR
Books, ); Timothy P. Wickham-Crowley, Guerrillas and Revolution in Latin
America: A Comparative Study of Insurgents and Regimes since  (Princeton:
Princeton University Press, ).
 Examples of Latin American conflict in which race, ethnicity, geography, and
politics intersected to produce violence of genocidal proportions before the
second half of the twentieth century may be found in Robin L. H. Derby and
Richard Turits, ‘‘Histories of Terror and the Terrors of History: The  Hai-
tian Massacre in the Dominican Republic’’ (paper presented to the Committee
on Latin American History at the annual meeting of the American Historical
Association, New York,  December ); and Nelson Reed, The Caste War of
the Yucatan (Stanford: Stanford University Press, ).
 The most striking parallel of a coterminous and comparably drawn-out interne-
cine struggle to the Colombian case is the war between Catholics and Protestants
in Northern Ireland. See Allen Feldman, Formations of Violence: The Narrative
of the Body and Political Terror in Northern Ireland (Chicago: University of Chi-
cago Press, ), and ‘‘Violence and Vision: The Prosthetics and Aesthetics of
Terror,’’ Public Culture , no.  (): –. In terms of participants, magni-
tude, geographical scope, and complexity, other recent examples of brutal in-
ternal conflict such as that between Hutu and Tutsi or Serb and Croat might
also be compared to the Colombian case. See Liisa Malkki, Purity and Exile:
Violence, Memory, and National Cosmology among Hutu Refugees in Tanzania
(Chicago: University of Chicago Press, ); and Rob Nixon, ‘‘Of Balkans and
Bantustans: ‘Ethnic Cleansing’ and the Crisis in National Legitimation,’’ Tran-
sition  (): –. The latter piece reproduces a photograph of a Muslim
soldier holding the severed head of a Serb—a pose identical to that captured
many times during la Violencia. See also Arjun Appadurai, ‘‘Dead Certainty:
Ethnic Violence in the Era of Globalization,’’ Public Culture , no.  (): –
. Appadurai reminds us of the need to include an approach to ethnic violence
that relies on ideas about the stranger—a figure that is repeatedly alluded to in
tales of la Violencia—not just the notion of collective violence that is ‘‘a product
of propaganda, rumor, prejudice, and memory.’’
 Bushnell, The Making of Modern Colombia, p. .
 Helen Delpar, Red against Blue: The Liberal Party in Colombian Politics, –
 (Tuscaloosa: University of Alabama Press, ); Charles W. Bergquist,
Coffee and Conflict in Colombia, – (Durham, N.C.: Duke University
Press, ); and Daniel Pécaut, Orden y violencia: Colombia, –,  vols.
(Bogotá: Siglo Veintiuno Editores, ).
 Notes to Introduction 

 Bushnell, The Making of Modern Colombia, p. ; and David Bushnell and Neill
Macauley, The Emergence of Latin America in the Nineteenth Century (New York:
Oxford University Press, ).
 Bushnell, The Making of Modern Colombia, pp. –.
 For a lucid discussion of Colombia’s two-party system and the civil wars of the
nineteenth century, see Bergquist, Labor in Latin America, pp. –.
 Bushnell has pointed out that no one knows exactly how a death count of
, was originally arrived at and suggests that it may well be ‘‘too high,’’
see The Making of Modern Colombia, p. .
 The following account of the Reyes’s administration relies heavily on chapter 
of Bergquist’s seminal study, Coffee and Conflict in Colombia. For a compre-
hensive overview of the development of the Colombian economy, particularly
the evolution of the export economy, see José Antonio Ocampo, Colombia y la
economía mundial, – (Mexico, D.F.: Siglo Veintiuno Editores, ), and
‘‘Los orígenes de la industria cafetera, –,’’ in Nueva historia de Colombia,
ed. Jaime Jaramillo Uribe, vol.  (Bogotá: Planeta, ), pp. –.
 Bergquist, Coffee and Conflict, pp. , .
 Bergquist, Coffee and Conflict, p. . For a different perspective on the politi-
cal and social impact of coffee production on the consciousness and political
mobilization of coffee workers employed by large estates, see Michael Jiménez,
‘‘The Limits of Export Capitalism: Economic Structure, Class, and Politics in
a Colombian Coffee Municipality, –’’ (Ph.D. diss., Harvard Univer-
sity, ).
 Luis Ospina Vásquez, Industria y protección en Colombia, –, d ed.
(Medellín: FAES, Editorial Lealon, ), pp. –. Ospina notes that between
 and  ‘‘the situation was such that any political party would have had to
think twice before declaring opposition to this era of ‘progress.’ ’’ And, he adds,
‘‘the most influential men within Liberalism, some of whom did so from their
positions as directors which they had accepted when the practice of mixed [bi-
partisan] ministries were in effect, supported protectionism.’’ On the effects of
bipartisan policy-making in Antioquia, see my own ‘‘Genesis and Evolution of
La Violencia in Antioquia, Colombia, –’’ (Ph.D. diss., Harvard Univer-
sity, ), chapters  and .
 See Bergquist, Labor in Latin America, pp. –, and Miguel Urrutia, The De-
velopment of the Colombian Labor Movement (New Haven: Yale University Press,
), p. , for discussions of how the Banana Strike of  discredited the
Conservative government; and Gonzalo Restrepo Jaramillo, El pensamiento con-
servador (Medellín: Tipografía Bedout, ), pp. –, on the reaction of some
Antioqueño Conservatives to the issue of the clergy’s ‘‘intrusion’’ in the election
of .
 Mariano Arango, El café en Colombia, –: Producción, circulación y polí-
tica (Bogotá: Carlos Valencia Editores, ), p. , and Café e industria, –
 Notes to Introduction

, d ed. (Bogotá: Carlos Valencia Editores, ); Jesús Antonio Bejarano,
Ensayos de historia agraria colombiana (Bogotá: Fondo Editorial CEREC, );
and José Antonio Ocampo and Santiago Montenegro, ed., Crisis mundial, pro-
tección e industrialización (Bogotá: Fondo Editorial CEREC, ).
 Richard E. Sharpless, Gaitán of Colombia: A Political Biography (Pittsburgh:
Pittsburgh University Press, ), pp. –.
 Pécaut, Orden y violencia, :. For an in-depth analysis of Alfonso López
Pumarejo’s first administration, see Alvaro Tirado Mejía, Aspectos políticos del
primer gobierno de Alfonso López Pumarejo, –, Instituto Colombiano de
la Cultura (Bogotá: Grafica Cabrera e Hijos, ).
 Pécaut, Orden y violencia, :.
 Gonzalo Sánchez, Las ligas campesinas en Colombia (Bogotá: Editorial Tiempo
Presente, ), p. , and Hermés Tovar Pinzón, El movimiento campesino en
Colombia durante los siglos xix y xx (Bogotá, ), pp. –.
 The most thorough examination of the evolution of Colombian land laws, de-
velopment of the frontier, and emergence of conflict over public lands may be
found in Catherine LeGrand, Frontier Expansion and Peasant Protest in Colom-
bia, – (Albuquerque: University of New Mexico Press, ), p. . For
an examination of labor unrest on large coffee estates in Tolima and Cundina-
marca, see Bergquist, Labor in Latin America, pp. –.
 A thoughtful treatment of the López administration may be found in Richard
Stoller, ‘‘Alfonso López Pumarejo and Liberal Radicalism in s Colombia,’’
Journal of Latin American Studies , no.  (): –.
 Bushnell, The Making of Modern Colombia, p. ; LeGrand, Frontier Expansion,
p. ; Sánchez, Las ligas, p. .
 Daniel Pécaut, Política y sindicalismo en Colombia, d ed. (Bogotá: Ediciones
Culturales, ), p. . Ann Farnsworth-Alvear explores the growth of textile
production during these years in Medellín and the response of workers and fac-
tory owners to the emergence of Communist labor mobilization in Dulcinea in
the Factory: Myths, Morals, Men, and Women in Colombia’s Industrial Experi-
ment, – (Durham, N.C.: Duke University Press, ), especially chap-
ter .
 Bergquist, Labor in Latin America, p. .
 Arango, El café en Colombia, p. ; El Colombiano,  Feb. ,  Feb. ,
 Feb. , and  Feb. . Coltejer’s profits increased by a factor of twenty
between  and  from , pesos to ,,. See ‘‘Coltejer Profits
Reports,’’ in Vidas y empresas de Antioquia: Diccionario biográfico, bibliográfico
y económico (hereafter, VEA), ed. Alfonso Mejía Robledo (Medellín: Imprenta
Departamental de Antioquia, ), p. .
 Miguel Urrutia, Historia del sindicalismo en Colombia, d ed. (Medellín: Edito-
rial Lealon, ), p. –.
 John Martz, Colombia: A Contemporary Political Survey (Chapel Hill: University
of North Carolina Press, ), p. .
 Notes to Introduction 

 Vernon Lee Fluharty, Dance of the Millions: Military Rule and the Social Revo-
lution in Colombia, – (Pittsburgh: University of Pittsburgh Press, ),
p. .
 Pécaut, Política y sindicalismo, p. , and Urrutia, Development of the Colombian
Labor Movement, p. .
 Pécaut, Política y sindicalismo, p. .
 Herbert Braun has provided the most persuasive and thorough exploration of
the rise of emergent, non-elite politicians in the period preceding la Violencia,
and especially the impact of Jorge Eliécer Gaitán on Colombia’s elite-dominated
political arena in The Assassination of Gaitán: Public Life and Urban Violence
in Colombia (Madison: University of Wisconsin Press, ). See also Carlos
Miguel Ortíz Sarmiento, Estado y subversión en Colombia: La Violencia en el
Quindió, años  (Bogotá: Fondo Editorial CEREC, ), pp. –, for details
of the rise of local political hopefuls around the figure of Conservative Gilberto
Alzate Avendaño. In Antioquia, a number of these non-elite political figures
emerged: Eduardo Berrío González, governor from  to , who began
his own dissident movement to champion municipal interests in Oriente (the
east); and Aníbal Vallejo Arbeláez, who was raised a Liberal but switched over
to Laureano Gómez’s faction of the Conservative party. See Christopher Abel,
Política, iglesia y partidos en Colombia (Bogotá: FAES–Universidad Nacional de
Colombia, ), p. .
 Martz, Colombia, p. .
 Braun, Assassination of Gaitán, pp. –, estimates the material cost of dam-
age in  to be nearly  million pesos, while estimates of the dead range be-
tween  and ,. For a visual catalogue of the impact of the Bogotazo, see
Carlos Delgado, ed., El  de abril en fotos (Bogotá: El Ancora Editores, ).
 Arturo Alape, El Bogotazo: Memorias del olvido (Bogotá: Editorial Pluma, ),
pp. – and pp. –.
 VEA, ‘‘Instituto Colombiano de Seguros,’’ p. ; Abel, Política, p. ; Martz,
Colombia, p. .
 Colombia, Decreto del Estado de Sitio, Nov. .
 Laureano Gómez was an extremely controversial figure in Colombian politics.
He has been depicted as both a brilliant defender of Conservative values and the
intellectual author of violence against the Liberal opposition during la Violen-
cia. The earliest attempt to understand Gómez took a psychoanalytic turn, see
José Francisco Socarras, Laureano Gómez: Psicoanalisis de un resentido (Bogotá:
Ediciones Librería Siglo XX, ). No thorough analysis of Gómez’s presidency
exists, but James Henderson’s Las ideas de Laureano Gómez (Bogotá: Tercer
Mundo, ) provides a good intellectual biography of Gómez. For a more re-
cent treatment of Gómez as a historical figure, see Arturo Abella Rodriguéz,
Laureano Gómez (Bogotá: Espara, ).
 José Gutiérrez Gómez, president of , cited in El Colombiano, Medellín,
; cited again in El Colombiano, Medellín,  April .
 Notes to Introduction

 General Gustavo Rojas Pinilla’s regime has prompted surprisingly few scholarly
analyses, but see Fluherty, Dance of the Millions, and Silvia Galvis and Alberto
Donadío, El Jefe Supremo: Rojas Pinilla en la violencia y en el poder (Bogotá:
Planeta, ).
 Richard Weinert, ‘‘Violence in Pre-Modern Societies: Rural Colombia,’’ Ameri-
can Political Science Review  (June ): –; Fluharty, Dance of the Mil-
lions; Robert Williamson, ‘‘Toward a Theory of Political Violence: The Case of
Rural Colombia,’’ Western Political Quarterly (March ); and James L. Payne,
Patterns of Conflict in Colombia (New Haven: Yale University Press, ). The
persistence of banditry and feuds in Latin American countries such as Brazil
and Mexico at the end of the nineteenth and in the early twentieth centuries
was often cited, for instance, to illustrate the struggle between a more tradi-
tional social, political, and economic order to one guided by secular values
and identified with modernization; see da Cunha, Rebellion in the Backlands;
Todd A. Diacon, Millenarian Vision, Capitalist Reality: Brazil’s Contestado Re-
bellion (Durham, N.C.: Duke University Press, ); Eric Hobsbawm, Primi-
tive Rebels: Studies in Archaic Forms of Social Movement in the Nineteenth and
Twentieth Centuries (New York: W. W. Norton, ); Gilbert M. Joseph, ‘‘On
the Trail of Latin American Bandits: A Reexamination of Peasant Resistance,’’
Latin American Research Review , no.  (): –; Linda Lewin, ‘‘The Oli-
garchical Limitations of Social Banditry in Brazil: The Case of the ‘Good’ Thief
Antonio Silvino,’’ Past and Present : –; Richard W. Slatta, Bandidos: The
Varieties of Latin American Banditry (New York: Greenwood, ); and Vander-
wood, Disorder and Progress. On aggression fueled by status deprivation, see
Williamson, ‘‘Toward a Theory of Political Violence,’’ p. . Florencia Mallon
suggests that perhaps the strong relationship forged between peasants and spe-
cific parties during the nineteenth century was equally present in the Mexico
and Peru, see Peasant and Nation.
 The literature on la Violencia is too vast to mention in its entirety here, but
the most significant studies written since the s include: Charles Berg-
quist, Ricardo Peñaranda, and Gonzalo Sánchez, eds., Violence in Colombia: The
Contemporary Crisis in Historical Perspective (Wilmington, Del.: Scholarly Re-
sources, ); Braun, The Assassination of Gaitán; Comisión de estudios sobre
la violencia, Colombia: Violencia y democracia, Informe presentado al Ministro de
Gobierno (Bogotá: Universidad Nacional, ); Medófilo Medina, La protesta
urbana en Colombia en el siglo veinte (Bogotá: Ediciones Aurora, ); Once
ensayos sobre La Violencia (Bogotá: Fondo Editorial CEREC y Centro Gaitán,
); Oquist, Violence, Conflict, and Politics in Colombia; Pécaut, Orden y Vio-
lencia en Colombia; Gonzalo Sánchez and Donny Meertens, Bandoleros, gamo-
nales y campesinos: El caso de la Violencia en Colombia (Bogotá: El Ancora, );
Gonzalo Sánchez, Los días de la revolución: Gaitanismo y  de abril en provin-
cia (Bogotá: Centro Gaitán, ); Gonzalo Sánchez and Ricardo Peñaranda,
 Notes to Introduction 

ed., Pasado y presente de la Violencia en Colombia (Bogotá: Fondo Editorial

CEREC, ).
Studies of la Violencia in specific regional or municipal contexts include:
Jaime Arocha, La Violencia en el Quindío: Determinantes ecológicos y económicos
del homicidio en un municipio caficultor (Bogotá: ); Darío Betancourt and
Martha L. García, Matones y cuadrilleros: Origen y evolución de la Violencia en
el occidente colombiano, – (Bogotá: Tercer Mundo Editores, ); Julio
Casas Aguilar, La Violencia en los Llanos Orientales (Bogotá: ECOE Ediciones,
); Ulises Casas, De la guerrilla liberal a la guerrilla comunista (Bogotá, );
Apolinar Díaz Callejas, El  de abril  en Barrancabermeja: Diez días de poder
popular (Bogotá: El Labrador, ); Darío Fajardo, Violencia y desarrollo: Trans-
formaciones sociales en tres regiones cafetaleras del Tolima, – (Bogotá:
Fondo Editorial Suramérica, ); James D. Henderson, When Colombia Bled:
A History of the ‘‘Violencia’’ in Tolima (Tuscaloosa: University of Alabama Press,
); Carlos Miguel Ortíz Sarmiento, Estado y Subversion en Colombia; and
Maria Victoria Uribe, Matar, rematar y contramatar: Las masacres de la Violencia
en el Tolima, – (Bogotá: CINEP, ).
For overviews of the literature on la Violencia see Catherine LeGrand, ‘‘La
política y la Violencia en Colombia, –: Interpretaciones en la década de
los ochenta,’’ Memoria y Sociedad , no.  (): –; Carlos Miguel Ortíz
Sarmiento, ‘‘Violencia política de los ochenta: Elementos para una reflexión his-
torica,’’ Anuario Colombiano de Historia Social y de la Cultura – (–):
–; and Gonzalo Sánchez, ‘‘La Violencia in Colombia: New Research, New
Questions,’’ Hispanic American Historical Review , no.  (): –.
 The historiography of the Mexican Revolution is extraordinarily rich and too
complex to summarize here. Among the more significant attempts to inter-
pret the causes and significance of the revolution, see Alan Knight, ‘‘The Mexi-
can Revolution: Bourgeois? Nationalist? or Just a ‘Great Rebellion’?’’ Bulletin
of Latin American Research , no.  (): –; Jaime O. Rodríguez, ed., The
Revolutionary Process in Mexico: Essays on Political and Social Change, –
 (Los Angeles: Latin American Center, UCLA, ); and Gilbert M. Joseph
and Daniel Nugent, eds., Everyday Forms of State Formation: Revolution and
the Negotiation of Rule in Modern Mexico (Durham, N.C.: Duke University
Press, ).
 Martz, Colombia; Robert H. Dix, Colombia: The Political Dimensions of Change
(New Haven: Yale University Press, ) and Alvaro Echeverri Uruburu, Elites
y proceso político en Colombia, – (Bogotá: FUAC, ).
 Ortíz, Estado y Subversión, pp. –, suggests the importance of regional resi-
dence and class in the formation of partisan identity in the Quindío, but no
study of la Violencia has pursued the issue of the relationship between ethnic
and racial origin and politics.
 This was the central premise of scholars such as Weinert and Williamson, and
 Notes to Introduction

also Steffen W. Schmidt in ‘‘La Violencia Revisited: The Clientelist Bases of

Political Violence in Colombia,’’ Journal of Latin American Studies – (May
–): –, and ‘‘The Transformation of Clientelism in Rural Colom-
bia,’’ in Friends, Followers, and Factions, ed. Steffen W. Schmidt et al. (Berkeley:
University of California Press, ); and Pécaut, Orden y violencia.
 The main proponent of this thesis was Paul Oquist, but Richard Maullin in Sol-
diers, Guerrillas, and Politics in Colombia (Lexington, Mass.: D. C. Heath, )
also insisted that the central government exercised very little power at the local
level and over extensive parts of Colombia.
 Pécaut ascribed to this theory as did Ortíz Sarmiento.
 Braun, The Assassination of Gaitán; Ortíz Sarmiento, Estado y subversión.
 Both Sánchez and Schmidt argued that Gaitán had introduced the question of
class into politics.
 Sánchez, Los días de la revolución; Díaz Callejas, El  de abril  en Barranca-
bermeja.
 Arocha, La Violencia en el Quindío, and Tovar Pinzón, El movimiento campesino
en Colombia argued that struggles over land were important features of la Vio-
lencia in both coffee-producing and frontier cattle areas. Gonzalo Sánchez, ‘‘La
Violencia y sus efectos en el sistema político colombiano,’’ Cuadernos Colom-
bianos  (January–April ): –, argued that organized labor was crushed
as a result of violence after , and Pécaut, Política y sindicalismo, agreed.
 Oquist and other analysts of Colombian politics use the term ‘‘hegemony’’
to describe the monopolistic control of the central state by a single political
party. Such a strategy precludes the possibility of collaboration by the political
opposition and, in fact, entails their exclusion from all government and public,
patronage-determined offices or jobs.
Some Colombian scholars have suggested that too little is known about the
degree of integration between local, regional, and national spheres of power in
Colombia to support Oquist’s notion of the significance of ‘‘breakdown’’ of the
national state on the emergence of violence. See Jesús Antonio Bejarano and
Marco Palacios in Once ensayos sobre la violencia, pp. –.
 Pécaut, ‘‘Acerca de la Violencia de los años cincuenta,’’ p. .
 Braun, The Assassination of Gaitán, pp. –.
 Ibid., p. .
 Ibid., pp. –. Braun notes, ‘‘Everything that was physical about Gaitán be-
came deeply symbolic. . . . While the political leaders were repelled by Gaitán’s
behavior, his followers rejoiced at the earthy exploits that made him look like
one of them,’’ p. .
 Ibid., p. .
 Ibid., p. .
 Arocha, La Violencia en el Quindío; Betancourt and García, Matones y cua-
drilleros; Fajardo, Violencia y desarrollo; and Henderson, Cuando Colombia se
desangró.
 Notes to Introduction 

 Sánchez, ‘‘La Violencia y sus efectos,’’ p. .

 Ibid., p. .
 Ibid., p. –.
 Ibid., p. ; Fajardo, Violencia y desarrollo, p. .
 Ortíz Sarmiento, Estado y subversión en Colombia.
 Ibid., pp.  and .
 Ibid., pp.  and .
 A vereda is the equivalent of a village. A municipio (municipality) is a rural
county governed by a mayor and municipal council.
 Ortíz Sarmiento, Estado y subversión en Colombia, pp.  and .
 Ibid., p. .
 Ibid., pp. –.
 Philip Corrigan, ‘‘State Formation,’’ in Joseph and Nugent, eds., Everyday Forms
of State Formation, p. xvii, emphasis in the original; Derek Sayer, ‘‘Everyday
Forms of State Formation: Some Dissident Remarks on ‘Hegemony,’ ’’ in Every-
day Forms of State Formation, p. , emphasis in the original; Philip Abrams,
‘‘Notes on the Difficulty of Studying the State,’’ Journal of Historical Sociology ,
no.  (): –.
 Ibid., p. .
 Ibid., p. , .
 Many authors have noticed differences between the behavior of Liberals and
Conservatives in Antioquia and those in other parts of Colombia and attrib-
uted these differences in large measure to the region’s economy, the role of the
Church, the region’s mountainous location, and the cohesive nature of a pri-
marily merchant elite. See, for instance, Bergquist, Coffee and Conflict in Co-
lombia; Delpar, Red against Blue; Luis H. Fajardo, The Protestant Ethic of the
Antioqueños?: Social Structure and Personality (Cali: Ediciones Departamento
de Sociología, Universidad del Valle, n.d.); and Jorge Orlando Melo, ‘‘La polí-
tica de  a ,’’ in Historia de Antioquia, ed. Jorge Orlando Melo (Medellín:
Editorial Presencia and Suramericana de Seguros), pp. –.
 For the importance of these ‘‘consociational’’ forms of elite organization and
their influence on state policy, see Jonathan Hartlyn, ‘‘Consociational Politics
in Colombia: Confrontation and Accommodation in Comparative Perspective’’
(Ph.D. diss., Yale University, ); Hartlyn, ‘‘Producer Associations, the Politi-
cal Regime, and Policy Processes in Contemporary Colombia,’’ Latin Ameri-
can Research Review , no.  (): –; Echeverri Uruburu, Elites y pro-
ceso político en Colombia; and Eduardo Sáenz Rovner, La ofensiva empresarial:
Industriales, políticos y violencia en los años  en Colombia (Bogotá: Tercer
Mundo, ).
 Roldán, ‘‘Genesis and Evolution of La Violencia,’’ chapter , pp. –, and
chapter , pp. –.
 SGA , vol. , letter from Horacio Correa M. to Governor Dionisio Arango
Ferrer (hereafter, ), Medellín,  May .
 Notes to Introduction

 Ibid.
 This is evident in hacienda and mining hiring patterns in Antioquia where
workers belonged to whatever party happened to predominate in the geo-
graphic area regardless of the partisan affiliation of the owners.
 La Defensa,  Feb. .
 This phenomenon was not limited to Antioquia. Betancourt and García,
Matones y Cuadrilleros, p. , describe a similar situation in their study of Valle,
although their emphasis and conclusions differ from mine.
 LeGrand, Frontier Expansion and Peasant Protest in Colombia, p.  and ap-
pendix D: ‘‘Public Land Grants and Conflicts by Municipality, –,’’ pp.
–; and Fajardo, Violencia y desarrollo. In a twist on this theme, Betancourt
and García argue that the municipios in northern Valle where agrarian conflicts
emerged in the s were those where there had been Liberal-directed violence
in the s. See Betancourt and García, Matones y Cuadrilleros, p. .
 See LeGrand, Frontier Expansion, appendix D for a detailed list of Antioqueño
municipios reporting public land conflicts in the first three decades of the twen-
tieth century.
 Betancourt and García, Matones y Cuadrilleros, p. , for instance, note that vio-
lence in Valle was most intense along the coffee-producing mountain zones colo-
nized by Antioqueños. They suggest that this violence had more of the character
of ‘‘partisan confrontation’’ and seemed to be organized along patron-client net-
works that observed kinship and cultural ties between dependents and coffee
estate owners where both sectors belonged to the same party. Although the au-
thors do not make this argument, the problem of violence may be both partisan
and cultural, that is, waged by Antioqueños belonging to different classes but
sharing similar values and beliefs and a similar embeddedness in proprietary
relations against those who were non-Antioqueño in origin.
 Manuel Uribe Angel, Geografía general del estado de Antioquia (Paris, ).
 Progreso,  Dec. , . The magazine Progreso was the organ of Medellín’s
Sociedad de Mejoras Públicas (Society of Public Improvements). Its editorial
board and contributors included Antioquia’s most prominent statesmen, mer-
chants, and policy makers. In its pages, capitalism, urbanization, rural society,
moral values, class conflicts, and concepts of regional identity, autonomy, and
power were explicitly and vocally debated. As such it provides an unparalleled
window into the thought and discourse of Antioquia’s bourgeois leadership,
who shaped both public policy and social mores in the region.
 Hermés Tovar was the first to note the impact of the  coffee crisis on migra-
tion and ensuing land struggles in new frontier areas as a result of these newly
displaced colonists; see Tovar, El movimiento campesino, p. . Interviews con-
ducted by researchers in  with individuals who were first-generation resi-
dents in eastern peripheral towns such as Cocorná, Yolombó, and San Roque
confirm this phenomenon. Transcripts and cassette recordings of these inter-
 Notes to Introduction 

views may be accessed at the Instituto de Estudios Regionales (), Univer-

sidad de Antioquia, Medellín.
 Tierras baldías or terrenos baldíos are ‘‘public domain land, the ownership of
which is vested in the nation.’’ This is Catherine LeGrand’s definition in Frontier
Expansion, p. . See also ibid., appendix D.
 The literature on colonialism is too great to summarize here. Among the
works that have directly influenced my understanding of the phenomenon are:
Homi K. Bhabha, ‘‘The Other Question: Difference, Discrimination, and the
Discourse of Colonialism,’’ in Out There: Marginalization and Contemporary
Cultures, ed. Russell Ferguson et al. (Cambridge: MIT Press, ); Jean Co-
maroff and John Comaroff, Ethnography and the Historical Imagination (Boul-
der: Westview Press, ), pp. – and pp. –; Frederick Cooper and
Ann Laura Stoler, Introduction to American Ethnologist  (), special issue
‘‘Tensions of Empire: Colonial Control and Visions of Rule’’; Ann Laura Stoler,
‘‘Making Empire Respectable: The Politics of Race and Sexual Morality in
Twentieth-Century Colonial Cultures,’’ American Ethnologist  (): –,
and ‘‘Sexual Affronts and Racial Frontiers: European Identities and the Cultural
Politics of Exclusion in Colonial Southeast Asia,’’ Comparative Studies in Society
and History , no.  (): –; and Nicholas Thomas, Colonialism’s Culture,
Anthropology, Travel, and Government (Princeton: Princeton University Press,
).
 For a discussion of the construction of preconceived ‘‘signs’’ or ‘‘historically
determined sets of root-metaphors,’’ see Sander L. Gilman, Difference and Pa-
thology: Stereotypes of Sexuality, Race, and Madness (Ithaca: Cornell University
Press, ), p. . Gilman posits that three basic categories of difference ‘‘reflect
our preoccupation with the self and the control that the self must have over the
world’’: sexuality, illness, and race. Traits specifically associated with sexuality
(concubinage, promiscuity, prostitution, free union, and so on), race (fetish-
ism, superstition, sloth, maliciousness, darkness, filth) and illness (insalubrious
climate, fevers, malaria, ‘‘weakness’’) are all present in the Antioqueño construc-
tion of frontier difference. For an example of the ways in which such categories
are applied and implicated in situations of violence, see Malkki, Purity and Exile.
 For an analysis of the issue of race in shaping regional identity and cross-
departmental relations in Colombia, particularly between the Chocó and Antio-
quia, see Peter Wade, ‘‘Race and Class: The Case of South American Blacks,’’
Ethnic and Racial Studies , no.  (April ): –, and Blackness and Race
Mixture: The Dynamics of Racial Identity in Colombia (Baltimore: Johns Hop-
kins University Press, ). More recently Nancy Appelbaum has explored the
coincidence of racial imaginings and geography in the construction of identity
for the department of Caldas. See ‘‘Remembering Riosucio: Region, Ethnicity,
and Community in Western Antioquia, –’’ (Ph.D. diss., University of
Wisconsin, Madison, ).
 Notes to Introduction

 For the relationship between middle-class values and citizenship in Europe,
see George Mosse, Nationalism and Sexuality: Middle-Class Morality and Sexual
Norms in Modern Europe (Madison: University of Wisconsin Press, ). Sander
Gilman has shown how stereotypes of difference (the creation of a category of
the ‘‘other’’) are intimately linked to the question of control: ‘‘No matter how
this sense of control is articulated, whether as political power, social status, reli-
gious mission, or geographic or economic domination, it provides an appropri-
ate vocabulary for the sense of difference.’’ See Gilman, Difference and Pathol-
ogy, pp. –. The relationship between specific traits and places or groups of
individuals associated with a specific geographic or cultural space and its im-
pact in shaping notions of identity and policy are explored as well by Andrew
Parker et al. in Nationalisms and Sexualities (London: Routledge, ), and in
Doris Sommer, Foundational Fictions: The National Romances of Latin America
(Berkeley: University of California Press, ). Perhaps the classic Latin Ameri-
can text to make explicit both the nature of a bourgeois code of values and the
relationship between race, culture, and place is Domingo F. Sarmiento, Facundo,
or, Civilization and Barbarism (New York: Penguin, ). The last quote in this
sentence is from Stoler, ‘‘Sexual Affronts and Racial Frontiers,’’ p. .
 Progreso,  Oct. , . The term ‘‘sana’’ is used here to denote both physical
and moral health. For more information regarding the composition and activi-
ties of Antioquia’s bourgeoisie and the overlapping of public and private power
in the province, see Roldán, ‘‘Genesis and Evolution of La Violencia,’’ especially
chapters  and .
 Uribe Angel, Geografía de Antioquia, pp.  and . See also Uribe Angel’s de-
scription of Remedios, p. , Zaragoza, p. , Frontino, p. , and Cocorná,
p. . Uribe Angel wrote before the region of Urabá came under Antioquia’s
jurisdictional control, but, for the purposes of his analysis, he defines the re-
gion of Frontino as incorporating Urabá. Uribe Angel’s formulations were re-
produced and widely disseminated through such classic regional tracts as Mono-
grafías de Antioquia (Medellín, ) and in foreign descriptions and analyses
of Antioquia such as P. L. Bell, Colombia: A Commercial and Industrial Hand-
book, U.S. Department of Commerce, Bureau of Foreign and Domestic Com-
merce, Special Agents Series no.  (Washington, D.C.: Government Printing
Office, ). In her study of family and culture in Colombia, Virginia Gutiérrez
de Pineda assesses and deconstructs the relationship between ethnicity, cultural
differences, and geography embedded in regional identity and mentality. See
Virginia Gutiérrez de Pineda, Familia y Cultura en Colombia (Bogotá: Instituto
Colombiano de Cultura, DANE, ), pp. –, –. See also her tables
and charts relating illegitimacy, free union, and Catholic marriage by ethnicity
and geographic location. Last quote in the sentence by Restrepo Jaramillo, El
pensamiento conservador, pp. –.
 I understand hegemony in a Gramscian sense as ‘‘[depending] for its hold not
only on its expression of the interests of a ruling class but also on its accep-
 Notes to Chapter One 

tance as ‘normal reality’ or ‘commonsense’ by those in practice subordinated to

it,’’ Raymond Williams, Keywords: A Vocabulary of Culture and Society (Oxford:
Oxford University Press, ), p. . Gramsci argues that for hegemony to
work ‘‘the leading group should make sacrifices of an economic-corporate kind.
But there is also no doubt that such sacrifices and such a compromise cannot
touch the essential; for though hegemony is ethical-political, it must also be
economic, must necessarily be based on the decisive function exercised by the
leading group in the decisive nucleus of economic activity,’’ Antonio Gramsci,
Selections from the Prison Notebooks of Antonio Gramsci, ed. Quintin Hoare and
Geoffrey Nowell Smith (New York: International Publishers, ), p. .
 James Parsons, La colonización antioqueña en el occidente de Colombia, d ed.
(Bogotá: Carlos Valencia Editores, ), p. .
 Ibid., and see table , ‘‘Composición Racial en Antioquia,’’ p. .
 Anderson, Imagined Communities, p. . Anderson adds that ‘‘regardless of actual
inequality and exploitation that may prevail . . . the nation is always conceived
as a deep, horizontal comradeship.’’
 Stoler and Jean and John Comaroff have shown in their work on Indonesia and
South Africa, respectively, that the process of colonial identity formation is a
two-way street: colonizers construct the other (the colonized) in relation to their
desired sense of self, but the latter is also transformed and shaped by the colonial
experience.
 Peter Sahlins has noted the important role which the intermediate area between
France and Spain known as Cerdanza played in the gradual emergence and con-
solidation of national identity on either side of and within this contested site.
See Peter Sahlins, Boundaries: The Making of France and Spain in the Pyrenees
(Berkeley: University of California Press, ), pp. –.
 I sometimes use the term ‘‘middle sector’’ to describe the status and rank of
emergent professional politicians in mid-century Antioquia. Some, but not
all, of these men belonged to the middle class. More than a class designation,
‘‘middle sector’’ indicates the middle rank these rising politicians occupied in
the party hierarchy.

. Medellín and Core Municipalities

 For a representative look at how Antioquia’s elite and statesmen liked to repre-
sent themselves and be represented, see Mejía Robledo, VEA; Academia An-
tioqueña de Historia, Un Siglo de Gobierno en Antioquia, – (Medellín,
); Tulio Ospina, Protocolo Hispanoamericano de la urbandidad y el buen
tono, d ed. (Medellín: Felix de Bedout e Hijos, ), especially chapter , ‘‘El
Civismo’’; and the magazine Progreso, –, various issues.
 The following analysis of the political, social, economic, and ideological differ-
ences that divided emergent professional politicians from the elite in Antioquia
relies extensively on my ‘‘Genesis and Evolution of La Violencia,’’ chapter .
 See Gómez’s scathing indictment of Antioqueño congressmen in his parliamen-
 Notes to Chapter One

tary address of , ‘‘Ser Congresista Liberal y Serlo Conservador,’’ pp. –
and ‘‘El peor enemigo: El moderado,’’ p.  in Colombia, Camara de Repre-
sentantes, Laureano Gómez: Obras Selectas, part , vol.  (Medellín: Editorial
Bedout, ).
 Farnsworth-Alvear analyzes the anti-Communist fears of the Liberal press and
regional politicians during López’s first administration in Antioquia in Dulcinea
in the Factory, pp. –.
 Bergquist, Coffee and Conflict in Colombia, p. ; and Melo, ‘‘La política de –
,’’ p. .
 Liberal Jesús Echeverri Duque, for instance, was reluctant to countenance
worker militancy and was almost lynched by the unemployed in Medellín dur-
ing his term as governor. He was eulogized in the terms often used to refer to
Conservative politicians affiliated with or influenced by the Jesuits. See Un siglo
de gobierno, p. –. Even public works were infused with religious meaning in
the political discourse used by the pre- generation of leaders such as Liberal
governor Camilo C. Restrepo. Celebrating the construction of the Carretera al
Mar; Restrepo declared to the regional Asamblea in , ‘‘With Gonzalo Mejía
[a Liberal] and Tobón Quintero, Father Arteaga and Father Máximo, [who have
been] apostles of this work, we shout ‘to the sea, to the sea,’ without pause, with
faith in our destiny, and to the sea shall we go, ‘viva God!’ we shall go,’’ quoted
in Un siglo de gobierno, p. .
 PEA, table ..., ‘‘Censos de –,’’ pp. –; ‘‘Censos de –,’’ pp.
–; ‘‘Censos de –,’’ pp. –.
 PEA, table .., ‘‘Población por zonas, según municipios, censos –,’’
pp. –.
 Alexander Payne, ‘‘Crecimiento y cambio social en Medellín, –,’’ Estu-
dios Sociales  (September ): . Payne notes that a significant number of
those migrating to Medellín were university students, some from regions other
than Antioquia as well.
 Helg, La educación en Colombia, pp. –.
 Many elite Liberals felt the same way. Elite members of his own party joined
moderate Conservatives to organize protests against López’s education project
and to protect the elite monopoly over education. Tomás Rueda Vargas and
Agustín Nieto (elite Liberals) both opposed López’s official education plan of
. See Helg, La educación en Colombia, pp. – and p. .
 Helg, La educación en Colombia; and Miguel Angel Builes, Cartas Pastorales del
Excelentisimo Señor Miguel Angel Builes Obispo de Santa Rosa de Osos,  vols.
(Bogotá: Empresa Nacional de Publicaciones, ).
 Helg, La educación en Colombia, p. . Such an interpretation jibes with Aline
Helg’s own conclusions about the destruction of Catholic schools and institu-
tions during the riots after Gaitán’s assassination in Bogotá. She argues that
such violence was intended to express a popular repudiation of the elitist values
 Notes to Chapter One 

of such institutions whose continued domination of education was predicated

upon and ensured the exclusion of non-elites from power. Helg’s argument
is particularly persuasive since she demonstrates how elite Liberals had been
among the most adamant opponents of López’s educational reforms precisely
because they perceived these as an attack against elite power.
 Various interviews conducted by the author with Emma Londoño de Mejía,
the widow of José Mejía y Mejía, Froilan Montoya Mazo, Aníbal Vallejo Alva-
rez, and Manuel Mejía Vallejo, Medellín, –. See also Roldán, ‘‘La polí-
tica en Antioquia, –,’’ in Historia de Antioquia, ed. Jorge Orlando Melo
(Bogotá: Editorial Presencia, ).
 This is an impression drawn from numerous interviews with individuals active
in politics during this period such as Aníbal Vallejo Alvarez, Froilan Montoya
Mazo, and Bernardo Ospina Román, and by the perusal of the personal corre-
spondence of these and other individuals such as José Mejía y Mejía, his widow,
Emma Londoño de Mejía, Fernando Gómez Martínez, and Belisario Betancur.
 Ibid.
 Pécaut, Orden y violencia, vol. , p. ; and Mauricio Archila, ‘‘Paternalism and
Social Relations in the Early Stages of Colombian Industrialization,’’ paper pre-
sented at the Latin American Labor History Conference, Yale University,  April
, p. .
 Ibid., p. . See also statistics on food prices in Medellín, –, compiled
from reports on the ‘‘Movimiento de la Feria de Ganados Verificada el día  de
octubre ,’’ put together by the Secretaría de Agricultura y Fomento de Antio-
quia, Sección de Estadística y Publicaciones in AGA, , vol. ; and El Colom-
biano,  Oct. ;  Sept. ;  Oct. ; and  Nov. . For reports of
government-instituted food price controls, see El Colombiano,  July .
 Archila, ‘‘Paternalism and Social Relations,’’ p. .
 Colombia, Departamento de Antioquia, Anuario estadístico, años , ,
, appendix /, ‘‘Estadística electoral.’’
 PEA, table ..., ‘‘Población ocupada, por ramas de actividad económica,
Censos –,’’ p. .
 La Defensa is a Catholic newspaper founded in Medellín in  whose con-
tributors identified themselves first as Catholics and second as Conservatives. In
the s La Defensa often acted as the official mouthpiece of Acción Católica.
It was an outspoken critic of bipartisan elite cooperation and Liberal control of
public sector jobs and patronage.
 Governors were appointed by the president and mayors were appointed by gov-
ernors. Municipal councils were popularly elected, and they in turn appointed
and paid the municipal police, town clerk ( personero) and other municipal
offices such as town scribe (escríbano). Police inspectors and work inspectors
were appointed by the governor.
 On the coffee zone, see AGA, , ‘‘Telegramas,’’ no. ,  Oct. , reporting
 Notes to Chapter One

irregularities in an all-Conservative hamlet in Andes; AGA, , ‘‘Telegramas,’’

no. ,  July  regarding violence during the coffee harvest in Jerico; AGA,
, ‘‘Telegramas,’’ no. ,  Sept. , reporting violence waged by a drunken,
Conservative mayor against Liberal citizens while the police stood by in Betu-
lia; and AGA,  ‘‘Telegramas,’’ no. , regarding violence in Bolívar,  July
. On the industrial belt, see AGA, , ‘‘Telegramas,’’ no. ,  Sept.
, naming military mayors to towns such as Envigado and Bello which re-
ported electoral disturbances. The northern towns included Valdivia, Ituango,
Sopetran, and Ebejico. Aníbal Vallejo Alvarez, interview with the author, and
AGA, , ‘‘Telegramas,’’ no. , San Jeronimo,  Oct. .
 Colombia, Memoria de hacienda, Francisco de Paula Pérez, , , 
(Bogotá: Banco de la República, ), ‘‘Auxilios,’’ , pp. –.
 Colombia, DANE, Estadística Electoral,  March  and  Oct. , pp. –
;  June , pp. –; and  Nov. , pp. –.
 AGA, , ‘‘Telegramas,’’ no. , from a member of the Jurado Electoral (elec-
toral jury) appointed to oversee voter registration, Caramanta,  Sept. ;
AGA, , ‘‘Telegramas,’’ no. , from the mayor of Andes who complained
that the Jurado suspended the registration of Conservative voters in the corregi-
miento of San José de Buenos Aires, Andes,  Oct. ; AGA, , ‘‘Telegramas,’’
number illegible, Yolombó,  Oct. ; and AGA, , ‘‘Telegramas,’’ no. ,
Itagüí,  Oct. .
 Centro Cultural Jorge Eliécer Gaitán, Bogotá, Correspondencia de Jorge Eliécer
Gaitán, letter from Aureliano Guerra, Daniel Guerra, and Francisco Espinosa,
Antioquia,  Jan. . Gaitán won a seat on the regional asamblea, as did five
other men (Conrado Sosa, Jesús García, Luis Atehortúa, Amador de J. Guerra,
and Luis Octavio Velásquez) and their suplentes (replacements or stand-ins).
Colombia, Departamento de Antioquia, Anales de la Asamblea de Antioquia,
Medellín,  April , Año :.
In the asamblea, senate and representative races of March ,  percent
of the Liberal voters in Andes voted for Gaitán; this dropped to  percent by
the concejo (municipal council) elections in October . In March,  per-
cent of Bolívar’s Liberal voters voted for Gaitán, but none voted for gaitanista
candidates in October. In Ituango in March  percent of the Liberal vote went
to Gaitán, but no votes were cast for gaitanistas in local concejo elections in
October. The same pattern occurred in Olaya and Caramanta where the March
election garnered, respectively,  percent and  percent of the Liberal vote
for Gaitán and then none in October’s contest. Ebéjico had a more consistent
loyalty to gaitanistas, voting for this movement’s candidates in local, regional,
and national elections from  through  ( percent of the Liberal vote
in March ,  percent in October , and  percent of the Liberal vote
in March ). DANE, Estadística Electoral,  March ,  Oct. , and
 June .
Segovia did not vote for Gaitán in the March  elections but delivered
 Notes to Chapter One 

 percent of its Liberal votes to gaitanistas in local elections in October .

Remedios, like Segovia, did not vote for Gaitán in March , but gave local
gaitanista candidates  percent of the Liberal vote in the concejo election of
October.
 DANE, Estadística Electoral,  March ,  Oct. , and  June . See
also this volume appendix C for statistics on dissident voting in Antioquia in
the  election for asamblea, concejo, congress, and the senate.
 These towns had the misfortune to be located near Conservative redoubts. Olaya
and Pueblorrico were the two most severely affected by the conservatization
process. Olaya went from having an electorate that was  percent Liberal in
 to one that was only  percent Liberal in . Pueblorrico dropped from
a Liberal total of  percent in  to  percent in .
 AGA, , ‘‘Telegramas,’’ no. , from José María Bernal (hereafter JMB) to
Carlos Echeverri, Jesús Saldarriaga et al., Caramanta,  June .
 PEA, p. . The coffee farms of Titiribí, in fact, were the most concentrated in
Antioquia, averaging , coffee trees per farm. In contrast, the average cof-
fee plantation in Fredonia, the region’s largest volume producer of the bean, had
, trees. See PEA, table ...., ‘‘Número de plantaciones, cafetos y maqui-
naria según municipios,’’ p. .
 AGA, , ‘‘Telegramas,’’ no. , from Rafael Azula Barrera, Secretario Gen-
eral de la Presidencia to JMB,  June .
 Cáceres had existed since colonial times when it attracted settlers searching for
gold mines, but it experienced only sluggish growth until the early to mid-
twentieth century when the expansion of important cattle haciendas owned by
members of Medellín’s industrial bourgeoisie and members of the Ospina family
awakened an interest in the town, Monografías de Antioquia, p. . The town
had one of the region’s lowest turnouts by eligible voters, see Colombia, De-
partamento de Antioquia, Anuario estadístico, años , , , appendix
, , ‘‘Estadística electoral: Resultado de las elecciones para Presidente de la Re-
pública en el Departmento de Antioquia, el día  de Mayo de .’’
 AGA, , ‘‘Telegramas,’’ no. , from Eduardo Berrío González (hereafter
EBG) to Rafael Azula Barrera,  Aug. .
 Colombia, Departmento de Antioquia, Anuario estadístico, años –, (d)
Café—volumen aproximado de la producción cafetera en Antioquia—
(sacos de pergamino de  / kilos), p. .
 Melo, ‘‘La Política en Antioquia,’’ p. .
 AGA, , ‘‘Telegramas,’’ no. , from JMB to the mayor of Fredonia,  July
; AGA , ‘‘Telegramas,’’ no. , from JMB to Mariano Ospina Pérez
(hereafter, MOP),  July .
 AGA, , ‘‘Orden Público,’’ (hereafter, O.P.), telegram to the Governor,  June
.
 AGA, , ‘‘O.P.,’’ telegram to the Governor,  June .
 El Colombiano,  July ; the Junta de Rentas oversaw the operation and col-
 Notes to Chapter One

lection of perhaps the single greatest source of regional revenue—the tax on

the state liquor and tobacco monopoly—and as such wielded considerable clout
in the region’s administrative affairs. About  percent of Antioquia’s revenues
were generated by the liquor and tobacco tax.
 AGA, , ‘‘O.P.,’’ Governor to the Ministro de Gobierno (hereafter, MinGob),
 Sept. .
 Urrutia, Development of the Colombian Labor Movement, p.  and pp. –;
and Pécaut, Orden y violencia, pp. –, pp. –; p. , n. , and p. ,
n. .
 Ivan Darío Osorio, Historia del sindicalismo antioqueño, – (Medellín:
Tipografía y Litografía Sigifredo, n.d.), p. .
 Letter from Froilan Montoya Mazo in Medellín to Jorge Eliécer Gaitán, Centro
Cultural, Jorge Eliécer Gaitán, Correspondencia de Jorge Eliécer Gaitán, Centro
Gaitán, Bogotá,  Sept. .
 All but two of these towns—Betulia and Ebéjico—were located on the periph-
ery.
 Pécaut, Política y sindicalismo, p. .
 Ibid., p. , n. .
 AGA, , ‘‘Telegramas,’’ no. , June . While the government’s response
to the workers’ demands was clearly motivated by the desire to simultaneously
weaken the union and replace Liberal workers with Conservatives, it bears re-
membering that a precedent for such actions was established during the 
railroad strike in Antioquia, when a Liberal governor, Captain Julián Uribe
Gaviria, declared that strike illegal, dismissed workers by accusing them of being
Communist sympathizers, and then replaced them with hand-picked Liberal
workers loyal to the mainstream and elite-led wing of the party.
 AGA, , ‘‘Telegramas,’’ no. , Governor to Shell Oil Company, Casabe,
 Aug. .
 Like many other contemporaneous Latin American authoritarian governments
—Perón and Vargas in Argentina and Brazil spring to mind—the laureanistas
used a nationalist rhetoric and nationalist principles to justify both the suppres-
sion of ‘‘internationalist’’ groups such as the Communist Party and the promo-
tion of native industrialist and economic interests against international compe-
tition.
 PEA, p. . AGA, , vol. , ‘‘Memorándum’’ describing the sources of reve-
nue to pay for regional public works projects, n.d. .
 AGA, , vol. , n.d.  and .
 Quoted in Osorio, Historia del sindicalismo antioqueño, p. .
 Ibid.
 Other towns where Gaitán’s assassination sparked unrest were Rionegro, Reme-
dios, Turbo, Segovia, Zaragoza, San Jerónimo, Caucasia, Amagá, Angelopolis,
and Cañasgordas.
 Varones Ilustres, p. ; ‘‘Mensaje del Gobernador,’’ in Francisco Duque Betan-
 Notes to Chapter One 

cur, Historia del Departamento de Antioquia, d ed. (Medellín: Editorial Albon-

Interprint S.A., ), p. ; and Sánchez, Los dias de la revolución, pp. –.
 AGA, , ‘‘O.P.,’’ Bolívar, telegram to the Governor,  April .
 Ibid.
 AGA, , ‘‘O.P.,’’ El Carmen del Atrato, telegram to the Governor,  April
.
 AGA, , ‘‘O.P.,’’ Andes, telegram to the Governor,  April .
 AGA, , ‘‘O.P.,’’ Dionisio Arango Ferrer to Darío Echandía, Bogotá, regarding
the Shell Oil camp in Casabe,  April .
 AGA, , ‘‘O.P.,’’ President, Bogotá, telegram to the Governor,  June .
 Ibid.,  June .
 El Colombiano,  July .
 Interview with Carlos Uribe Echeverri, quoted in VEA, pp. –.
 VEA, p. ; and Academia Antioqueña de Historia, Un siglo de gobierno, pp.
–.
 Academia Antioqueña de Historia, Un siglo de gobierno, pp. –.
 It is best not to take accusations of ‘‘Communism’’ or the constant use of the
term ‘‘Communist’’ too seriously, particularly in the Antioqueño context where
the party was small. Liberals were persistently conflated with Communists in
Conservative discourse during this period as were any groups making popular
demands or critical of Conservative rule. Regardless of the actual ideological
orientation, accusations of being a Communist were bandied about freely when-
ever the status quo appeared threatened or popular sectors rebelled. Academia
Antioqueña de Historia, Un siglo de gobierno, p. .
 Ibid.
 AGA, , vol. , telegram from Braulio Henao Mejía (hereafter, BHM) to the
Ministro de Gobierno, Bogotá,  Aug. .
 SGA, , vol. , communication from Dionisio Arango Ferrer (hereafter,
DAF) to Visitadores Administrativos,  July .
 These are too numerous to list here, but representative complaints may be found
in SGA, , vol. , telegram from Bello,  Aug. ; SGA, , vol. , tele-
gram from Betulia,  Dec. ; SGA, , vol. , telegram from Yolombó,
 Dec. ; AGA, , ‘‘Papeles del Señor Gobernador (–),’’ telegram
from Betulia,  Jan. ; AGA, , vol. , telegram from Medellín,  May
; AGA, , vol. , telegram from Caicedo,  April ; AGA, , vol. ,
telegram from Giraldo,  July ; AGA, , vol. , El Colombiano complains
of police brutality, Medellín, (no specific date) ; AGA, , vol. , telegram
from Puerto Berrío,  Oct. ; AGA, , vol. , telegram from the local
inspector, Puerto Nare, calling customs officers ‘‘a public menace, a corps of
assassins,’’  Oct. .
 The visitador administrativo is a holdover from the Spanish colonial period when
agents of the Spanish king were appointed to conduct a visita, or assessment,
after a viceroy or governor had finished his term in office. Colombian visitadores
 Notes to Chapter One

acted as government agents sent to collect information and submit reports on

topics as varied as public order, malfeasance, or party discipline.
 SGA, , vol. , communiqué from the departmental detectives to Jorge Sala-
zar, Jefe del Detectivismo, regarding Jardín,  April .
 Ibid.
 El Colombiano,  Jan. ; El Colombiano,  Feb. ; and El Colombiano,
 Feb. .
 La Defensa,  Feb. . ‘‘Nueve abrileños’’ was the colloquial term by which
right-wing Conservatives referred to Gaitán’s followers.
 Ibid.
 SGA, , vol. , Federación Nacional de Comerciantes to the Governor,  Feb.
.
 Ibid.
 La Defensa,  March ,  March , and  March .
 Marceliano Vélez, quoted in Varones Ilustres de Antioquia: Biografías de los aca-
démicos de números fallecidos (Medellín: Editorial Universo, ), p. , and
Clodomiro Ramírez quoted in Varones Ilustres de Antioquia, p. .
 Alejandro López quoted in Alberto Mayor Mora, Etica, trabajo y productividad,
d ed. (Medellín: Ediciones Tercer Mundo, ), p. ; and see Tulio Ospina,
Protocolo Hispanoamericano de la urbinidad y el buen tono, d ed. (Medellín:
Felix de Bedout e Hijos).
 Mayor Mora, Etica, trabajo y productividad, p. .
 El Colombiano,  Feb. .
 El Colombiano,  March .
 El Colombiano,  March  and  March .
 El Colombiano,  March .
 El Colombiano,  March .
 El Colombiano,  April .
 Alfonso López Pumarejo’s son had been involved in a scandal concerning a
boxer named Mamatoco a few years before this remark, and I am assuming that
referring to the governor by this name in this context was intended to suggest
that he kowtowed to the Liberals. AGA, , ‘‘Archivo Privado del Señor Gober-
nador de Antioquia’’ (hereafter, APSG), letter from a visitador administrativo
to Fernando Gómez Martínez (hereafter, FGM), Medellín,  March .
 SGA, , vol. , telegram from EBG to FGM regarding Salgar’s mayor,  Feb.
.
 AGA, , ‘‘APSG,’’ telegram from Puerto Berrío,  May ; telegram from
Remedios,  May ; AGA, , ‘‘MinGob,’’ telegram regarding Sopetran,
 May .
 AGA, , ‘‘APSG,’’ telegram, regarding Cisneros,  April , and AGA, ,
‘‘MinGob,’’ telegram regarding Itagüí,  April .
 AGA, , ‘‘MinGob,’’ telegrams regarding Itagüí,  April  and  May .
 Notes to Chapter One 

 AGA, , ‘‘MinGob,’’ telegram from Bogotá,  May .

 SGA, , vol. , Sindicato Ferrovario de Antioquia to the Governor,  Jan.
.
 SGA, , vol. , Sindicato de Trabajadores Frontino Gold Mines, Segovia to
the Governor,  March , emphasis in the original.
 AGA, , vol. , Public road workers in Yolombó to the Governor,  April
, and AGA, , vol. , Public road workers in Yolombó and Segovia to the
Governor,  April .
 AGA, , vol. , Bogotá, Ministro Gobierno to the Governor regarding public
road workers in Yolombó and Segovia,  April .
 AGA, , ‘‘MinGob,’’ Bogotá to Medellín,  May .
 SGA, , vol. , FEDETA to the Governor,  May .
 SGA, , vol. , Conservative Committee, Yarumal to the Secretaría de Go-
bierno, Antioquia,  May .
 AGA, , ‘‘MinGob,’’ telegram from Bogotá,  April .
 AGA, , ‘‘MinGob,’’ telegram from Andes,  May .
 AGA, , vol. , telegram from Bolívar,  May .
 AGA, , telegram from Yarumal to FGM,  May .
 AGA, , telegram from San Roque,  May ; AGA, , ‘‘MinGob,’’
telegram from Caracolí,  May ; AGA, , telegram from San Roque,
 May .
 SGA, , no volume number or date, Jairo de Bedout to the Secretario de Go-
bierno de Antioquia (hereafter, SecGobAnt) regarding Olaya,  March .
 SGA, , vol. , Detectives to Jorge Salazar Restrepo, Jefe Departamento de
Investigación Criminal, regarding Jardín,  April .
 El Colombiano,  May .
 AGA, , ‘‘APSG,’’ telegram from Ricardo Moreno to FGM regarding Reme-
dios,  May .
 El Colombiano,  May .
 AGA, , ‘‘APSG,’’ telegram from Puerto Berrío,  May .
 AGA, , telegram from Caracolí,  June ; AGA, , telegram from San
Rafael, June .
 AGA, , telegram from Titiribí,  June .
 El Colombiano,  June .
 El Colombiano,  April ,  May , and  May .
 El Colombiano,  Aug. .
 Liberal majorities in some towns also swelled in response to Liberals fleeing from
nearby areas where they did not constitute a majority or where Conservative
intimidation was more severe or effective. Such ‘‘target towns,’’ places where the
increase in the number of Liberal voters was disproportionate to the natural in-
crease of the local population between  and , reflected the possibility
that there was an influx of voters from elsewhere or fraud. The increased pres-
 Notes to Chapter One

ence of Liberal voters in certain towns was also a result of cross-departmental

migration, particularly in Puerto Berrío where ports such as Puerto Nare were
attracting fleeing Liberals from Santander and Boyacá.
 Aníbal Vallejo Alvarez (a regional laureanista leader), interview with the author,
Medellín, .
 For instance, Bolívar and Betulia exported violence to Bolombolo, Venecia;
Santa Barbara exported violence to Valparaiso; Yolombó exported violence to
Amalfi; Bello exported violence to Rionegro; and so on.
 These towns were: Abriaquí, Alejandría, Angostura, Belmira, El Carmen del
Viboral, Entrerríos, Granada, Ituango, Jardín, Jericó, Liborina, Nariño, San
Andrés, San Luis, San Pedro, Santa Rosa, San Vicente, and Yarumal.
 Consequently, the majority of conservatized towns were located in the south-
ern and southwestern coffee-producing areas (Andes, Armenia, Bolívar, Cara-
manta, Fredonia, Montebello, Tarso, Valparaiso) while the rest were scattered
in the northeast (Anorí, San Roque, and Yolombó), west (Olaya, Sabanalarga,
and Sopetrán), and center (Barbosa and Bello).
 El Colombiano,  Aug. .
 AGA, , ‘‘MinGob,’’ Directorio Liberal to Colonel Régulo Gaitán,  June
.
 AGA, , vol. , mayor of Bolívar to FGM,  July .
 AGA, , ‘‘MinGob,’’ Directorio Liberal to Colonel Régulo Gaitán,  June
.
 AGA, , ‘‘MinGob,’’ MinGob Bogotá to EBG regarding Bolívar,  Aug. .
 AGA, , ‘‘Telegramas,’’ Santuario’s mayor to FGM,  July .
 El Colombiano,  Aug.  and  Aug. . Jorge Aristizábal, John Gómez,
Darío Ramírez Gaviria, Jaime Posada Angel, and Gilberto Zapata Lotero repre-
sented the Liberals, while Manuel Yepes, Darío Londoño Villa, Francisco Ospina
Pérez, Jorge Escobar, Mario Saldarriaga, and Eduardo Isaza Martínez repre-
sented the Conservatives.
 AGA, , ‘‘APSG,’’ Juez Tercero, Medellín to EBG,  Sept. .
 AGA, , ‘‘MinGob,’’ telegram from Puerto Berrío to MinGob, Col. Régulo
Gaitán, Sept. .
 SGA, , vol. , Sindicato Ferrovario de Antioquia to the Governor,  Oct.
.
 Ibid.
 SGA, , vol. , Venecia, Fuerzas Armadas to Comandante, th Brigade,
Medellín,  Dec. .
 AGA, , vol. , telegram,  Oct. , and El Colombiano,  Oct. .
 AGA, , vol. , telegram from the president of the Federation of Coffee
Growers to EBG,  Oct. .
 Sergeant Bedoya would later be sent to the Magdalena Medio region to organize
contrachusma forces on the haciendas there.
 AGA, , ‘‘APSG,’’ telegram from MinGob Andrade to EBG, Nov. .
 Notes to Chapter Two 

 AGA, , ‘‘APSG,’’ telegram from MinGob Andrade to EBG,  Nov. .
 AGA, , ‘‘APSG,’’ telegram from MinGob Andrade to EBG,  Nov. .
 AGA, , ‘‘Telegramas,’’ Conservative mayor to EBG,  Nov. ; AGA, ,
vol. , Departamento de Seguridad to EBG,  Nov. .
 El Colombiano,  Nov.  and  Nov. .
 El Colombiano,  Oct. .
 El Colombiano,  Oct. .
 El Colombiano,  Nov. .
 El Colombiano,  Nov. ,  Nov. , and  Nov. .
 Decree no. ,  July .
 AGA, , vol. , telegram,  Nov.  and  Nov. ; AGA, , ‘‘Tele-
gramas,’’ telegram,  Nov. ; AGA, , ‘‘APSG,’’ Rentas to EBG,  Dec.
; and AGA, , vol. , telegram regarding troops for Medellín,  Dec. .
 AGA, , vol. , telegram,  Nov.  and AGA, , ‘‘MinGob,’’ telegram
from MinGob to EBG,  Dec. .
 AGA, , ‘‘Telegramas,’’ telegram,  Dec. .
 AGA, , ‘‘APSG,’’ telegram reporting on events in Concordia and Betulia,
 Dec. .
 AGA, , ‘‘Telegramas,’’ the priest of San Juan de Urabá to EBG,  Dec. .
 SGA, , vol. , Directorio Conservador de Antioquia (hereafter, DiConsAnt)
to the SecGobAnt,  Dec. .
 AGA, , ‘‘APSG,’’ Sindicato del Ferropacifico to the Governor,  Nov. .
 SGA, , vol. , circular no. /A, Colonel Carlos Bejarano, Director General
de la Policia Nacional, Bogotá, to the Governor,  Jan. .
 AGA, , vol. , MinGob to the Governor regarding Venecia,  June .
 AGA, , vol. , Colonel Villamil to the Governor regarding Betulia,  July
.
 AGA, , ‘‘APSG,’’ Governor Valencia, Chocó to Eduardo Berrío,  Nov. .
 AGA, , ‘‘APSG,’’ Colonel Villamil to MinGob,  Jan. .
 AGA, , vol. , Medellín, Ricardo Posada to Maríano Ospina Pérez,  May
.
 Ibid.
 El Colombiano,  Feb. .
 Aníbal Vallejo Alvarez, interview with the author, Medellín, .
 Oquist, Violence, Conflict, and Politics, p. .
 Rosca means ‘‘inner circle’’ or ‘‘clique,’’ with the added sense of dirty politics or
favoritism behind it.

. Bajo Cauca, Magdalena Medio, and the Northeast

 Ernesto Guhl, Colombia: Bosquejo de su geografía tropical, vol.  (Bogotá: Biblio-
teca Básica Colombiana, ); James Parsons, Antioqueño Colonization in West-
ern Colombia, d ed. (Berkeley: University of California Press, ).
 Gloria Isabel Ocampo, ‘‘Hacienda, parentesco y mentalidad: La colonización
 Notes to Chapter Two

antioqueña en el Sinu,’’ Revista Colombiana de Antropología,  (–): –

. Ocampo raises questions similar to my own regarding the usefulness of ob-
serving official administrative distinctions when analyzing the organization of
production, labor, and political comportment in regions such as the Bajo Cauca.
 SGA, , vol. , DiConsAnt to Secretario de Gobierno de Antioquia (hereafter,
‘‘SecGob’’), Aníbal Vallejo A.,  Nov. .
 SGA, , vol. , Visitador Administrativo to Governor,  July .
 Ibid. and Monografías de Antioquia (Medellín; n.p., ), p. .
 La Defensa,  Feb. .
 La Defensa,  Feb. .
 SGA, , vol. , Visitador Administrativo to Governor,  March .
 SGA, , vol. , Visitador Administrativo, Caucasia to Governor,  March
.
 Ibid.
 SGA, , vol. , Conservatives, Zaragoza to Governor,  April ; SGA ,
vol. , ‘‘O.P.,’’ Colonel Eduardo Villamil to Governor regarding Puerto Berrío,
 April ; SGA , vol. , Jairo de Bedout regarding Remedios to Governor,
 April .
 SGA, , vol. , DiConsAnt to Governor,  May .
 AGA, , ‘‘Telegramas,’’ Presidente de la República to Governor,  May .
 SGA, , vol. , Directorio Conservador, Caucasia to Governor,  April .
 SGA, , vol. , Rafael O’Brien Cadavid to comandante de la Polinal,  Aug.
.
 SGA, , vol. , Antioquia, Secretaría de Obras Publicas, ‘‘Informe para el Sr.
Gobernador de Antioquia,’’ Medellín,  Nov. . The composition of workers
in the gold mines and Shell Oil’s Casabe camp was described as ‘‘ percent
costeños and santandereanos.’’ AGA, , vol. , Shell Oil Camp, Yondó,
Remedios,  Nov. .
 SGA, , vol. , Asamblea General Del Sindicato de Trabajadores de Pato Con-
solidated Gold Dredging Ltd., ‘‘Proposicion #,’’ January .
 Ibid.
 Similar claims around citizenship were made by Mexican workers in the Parral
district in the early twentieth century. See William E. French, ‘‘Progreso For-
zado: Workers and the Inculcation of the Capitalist Work Ethic in the Parral
Mining District,’’ in Rituals of Rule, Rituals of Resistance: Public Celebrations and
Popular Culture in Mexico, ed. William Beezeley, Cheryl English Martin, and
William E. French (Wilmington, Del.: Scholarly Resources, ), p. .
 AGA, , ‘‘O.P.,’’ telegram to Governor,  Sept. .
 AGA, , ‘‘O.P.,’’ telegram,  Nov. .
 AGA, , ‘‘APSG,’’ Comité Conservador to Governor,  June .
 AGA, , ‘‘APSG,’’ telegram to Governor,  July .
 AGA, , ‘‘APSG,’’ Manuel Marín to Governor,  Sept. .
 Notes to Chapter Two 

 AGA, , vol. , ‘‘Informe sobre la acción del bandolerismo,’’ Medellín,  May
.
 AGA, , ‘‘MinGob,’’ MinGob to Governor and to SecGob, Sept. .
 SGA, , vol. , Polinal to Governor,  Sept. .
 SGA, , vol. , Sindicato Ferrovario to Governor,  Oct. .
 AGA, , ‘‘Telegramas,’’ Presidencia de la República,  Nov. .
 SGA, , vol. , ‘‘O.P.,’’  Nov. .
 AGA, , ‘‘Teleg-Presidencia de la República,’’ part of a report dated  Nov.
, but filed by the Governor on  Dec. .
 SGA, , vol. , Fuerzas Armadas de Colombia to Governor,  Dec. .
 SGA, , vol. , ‘‘O.P.,’’  Dec. .
 AGA, , vol. , Sindicato de Trabajadores de la Shell to Governor,  Jan. .
 AGA, , vol. , Shell Oil legal representative, Remedios to Governor,  June
.
 AGA, , ‘‘APSG,’’ DiConsAnt to Governor,  June .
 AGA, , ‘‘Telegramas,’’ Colonel Eduardo Villamil to Governor,  Aug. .
 AGA, , ‘‘Telegramas,’’ telegrapher to Governor,  Aug. .
 AGA, , ‘‘Telegramas,’’ Colonel Villamil to Ministro de Guerra (hereafter,
MinGuerra),  Aug. .
 AGA, , ‘‘Telegramas,’’ Colonel Villamil to Governor,  Aug. .
 AGA, , ‘‘APSG,’’ Inspector de Policía, Montelibano to Governor,  July .
 AGA, , ‘‘APSG,’’ Bolívar to Antioquia,  July .
 SGA, , vol. , Governor to Polinal, Bogotá,  July .
 AGA, , vol. , DiConsAnt to Governor,  Sept. .
 AGA, , ‘‘APSG,’’  Nov. .
 AGA, , vol. , Turbo,  Dec. ; SGA, , vol. , Urrao,  Dec. ;
AGA, , vol. , Puerto Perales,  Dec. .
 AGA, , vol. ,  Feb. .
 Aplanchar is the verb used to describe beating someone with the flat part of a
machete; aplanchamiento is the noun describing the action itself.
 AGA, , vol. , registrador municipal to the Governor, Jan. .
 AGA, , vol. , Departamento de Seguridad to the Governor,  Jan. .
 AGA, , vol. , Arturo to José, Colorado,  March .
 AGA, , vol. , Juan de Dios Arango to Governor,  Dec. .
 AGA, , vol. , Conservative Committee to Governor,  Dec. .
 SGA, , vol. , Administrador de Rentas, Zaragoza to Governor,  Nov. .
 AGA, , vol. , Caucasia, José María Brand, registrador municipal to the
Governor, Jan. .
 AGA, , vol. , DiConsAnt to Governor,  March .
 AGA, , vol. , BHM to Gonzálo Restrepo Jaramillo, Ministro de Relaciones
Exteriores, Bogotá,  March . In this and the following paragraph, all quo-
tations refer to this letter.
 Notes to Chapter Two

 The narrative that follows is based on the testimony provided by Angela Rosa
Arango to the authorities in Medellín. The full transcript of her testimony may
be found in SGA, , vol. , ‘‘Declaración de Angela Rosa Arango,’’ Departa-
mento de Investigación Criminal, Medellín,  May . Much of Angela Rosa’s
testimony is independently corroborated by archival sources and interviews
with former guerrillas conducted by the author. Whenever a specific piece of
information in her account can be verified by other sources, I include these in
the notes to the text.
 The hacienda ‘‘Nuevo Mundo’’ was located in Puerto Berrío and was the re-
peated target of guerrilla cattle rustling, see AGA , vol. , Puerto Berrío,
 June . The mayor of Puerto Berrío reported that four policemen had been
sent to the paraje ‘‘Nuevo Mundo’’ to capture the delinquent Manuel Mona
who along with other men had attacked the police and killed one policeman.
Mona was captured. See AGA, , vol. , mayor of Puerto Berrío to Governor,
 March .
 AGA, , vol. , Yolombó,  June . This is a report that the guerrillas
were camped on a farm owned by Luis Segundo and Gabriel Alvarez and that
they were linked to a Liberal leader from Yolombó and to a dentist and Liberal
jefe from Yalí. The latter owned a finca called ‘‘La Sellada’’ thought to be part
of a major supply route for guerrillas. On August , , moreover, Pielroja’s
appearance in Maceo was reported by Liberal administrators of Conservative-
owned haciendas in the area who reported that workers were called in by Piel-
roja and asked why the local Liberal leadership hadn’t offered them material
support. See AGA, , vol. ,  Aug. .
 Angela Rosa never mentioned how they supported themselves as they moved
between one safe house and another through an area covering three munici-
palities. But we can infer from the mention in other parts of her testimony that
the women were generally hired to provide domestic help and the men to clear
and sow fields by the mayordomos or administrators left in charge of the large
haciendas held by absentee landlords in the region.
 AGA, , vol. ,  July ; Pielroja’s presence is first reported along the ha-
ciendas of the railroad line.
 LeGrand, Frontier Expansion, p. , defines a corregimiento as ‘‘the smallest
administrative unit in the Colombian countryside, a subdivision of a munici-
pality.’’
 AGA, , vol. ,  Nov. , corroborates Angela Rosa’s version of this event.
 AGA, , vol. ,  Nov. ; the army reported killing  guerrillas found
with explosives near Virginias (on the railroad line) and on  Nov.  re-
ported killing  ‘‘revolutionaries’’ and capturing the guerrilla leader, Rangel.
The last report proved to be false.
 Captain Corneta admitted that his band stole indiscriminately from Liberal and
Conservative haciendas alike. Captain Corneta (Francisco Montoya), interview
with the author, Medellín, spring .
 Notes to Chapter Two 

 Ibid.
 Ibid. Indeed, Corneta reported that the majority of the men under his com-
mand were not Antioqueños but Santandereanos, so there was no reason for the
women to have known them.
 Ibid. This conforms to information given by Corneta who said that all his men,
except for butchers and cooks, were armed.
 Ibid. Corneta corroborated stealing uniforms from the police.
 AGA, , vol. , original message in code, but deciphered on the reverse,
 Feb. .
 AGA, , vol. , Yolombó,  Feb.  and  Feb. .
 LeGrand, Frontier Expansion, p. , defines caserio as ‘‘a small rural village that
has no administrative status.’’ Regarding the attack, see AGA, , vol. , Nechí,
Caucasia,  Feb. .
 AGA, , vol. , Governor to notary of Caucasia,  Feb. .
 AGA, , vol. , police inspector, Puerto Nare, to Governor,  April .
 AGA, , vol. , Sta Rita, Municipio de San Luis,  April .
 AGA, , vol. , San Luis,  April .
 AGA, , vol. , Puerto Nare,  April .
 AGA, , vol. ,  March ; AGA, , vol. , Oficio no. ,  March
.
 AGA, , vol. , Guarumo, Cáceres,  March .
 AGA, , vol. , Segovia,  May ; AGA, , vol. , Conservative Com-
mittee, Segovia to Governor,  May .
 SGA, , vol. , Virginias,  May .
 AGA, , vol. ,  May ; SGA, , vol. , Caracolí,  June .
 SGA, , vol. , Puerto Berrío,  June .
 SGA, , vol. , Cristalina,  June ; AGA, , vol. , Remedios/Puerto
Berrío,  June .
 SGA, , vol. , Yalí/Yolombó,  June .
 AGA, , vol. , El Jardín and Remedios,  June ; AGA, , vol. , ‘‘La
Alicia,’’ Puerto Berrío,  July ; AGA, , vol. , ‘‘El Delirio,’’ Maceo,  July
; and AGA, , vol. , Virginias, Puerto Berrío, and Maceo,  July .
AGA, , vol. , Casabe,  July ; AGA, , vol. , El Bagre, Zaragoza,
 July .
 AGA, , vol. , Luis Andrade, MinGob to Governor,  April .
 AGA, , vol. ,  April .
 AGA, , vol. , Conservatives to Governor,  April .
 AGA, , vol. , Governor to municipios,  April .
 San Roque, interview with Instituto de Estudios Regionales (hereafter, INER),
Universidad de Antioquia,  Sept. .
 San Roque, interview with INER,  Aug. .
 San Carlos, interview with INER,  June .
 Ibid.
 Notes to Chapter Two

 AGA, , vol. , Directorio Liberal de Antioquia (hereafter, DiLib) to Gover-

nor,  Sept. .
 AGA, , vol. ,  April .
 AGA, , vol. , various petitioners to Governor,  June .
 AGA, , vol. , de Bedout to Colonel Luis Abadía, Commander of the Fourth
Brigade, Medellín,  June .
 Ibid.
 AGA, , vol. , various signatories headed by Félix de Bedout to Governor,
 June .
 SGA, , vol. , Amalfi/Yolombó, ‘‘El Tigre,’’  Dec. .
 SGA, , vol. , telegram no. , EBG to President, Bogotá, July .
 SGA, , vol. , Amalfi, personero to Governor,  Feb. .
 AGA, , ‘‘O.P.,’’  June .
 AGA, , ‘‘O.P.,’’ telegram from Yolombó to Governor,  Aug. .
 AGA, , vol. , Conservative Committee, El Tigre, Amalfi to Governor,
 Nov. .
 AGA, , vol. , El Tigre to Governor regarding Yolombó contrachusma at-
tacks,  Dec. .
 AGA, , vol. , priest to Governor,  Nov. .
 SGA, , vol. , Caja de Credito Agrario to Governor,  Feb. .
 ANCB, v.  f. ; v.  f. ; v.  f. ; v.  f. ; v.  f. ; v.  f. ; v. .
f.  (Puerto Berrío); v.  f. ; v.  f.  (Remedios); v.  f.  (Yolombó);
v.  f.  (Puerto Nare/San Luis); v.  f.  (San Roque); v.  f.  (Caracolí).
See also LeGrand, Frontier Expansion. I am most grateful to Catherine LeGrand
who generously shared with me her archival notes regarding colono disputes in
Antioqueño towns.
 Captain Corneta, interview with the author, Medellín, .
 SGA, , vol. , San Roque,  Jan. .
 SGA, , vol. , Puerto Berrío,  Jan. .
 SGA, , vol. , Bedout to Governor,  Feb. .
 SGA, , vol. , Inspector de Trabajo, Puerto Triunfo,  April .
 Ibid.
 SGA, , vol. ,  June .
 Ibid.
 AGA, , vol. , priest, Yolombó to Governor,  Oct. .
 Granada, interview with INER,  June ; Alejandría, interview with INER,
 Sept. ; and San Rafael, interview with INER,  June .
 For an explicit and detailed examination of the supposed racial and sexual threat
posed by the overwhelmingly black and non-Antioqueño population that had
settled the northeast and Magdalena Medio, see SGA, , vol. , Antioquia,
Secretaría de Obras Publicas, ‘‘Informe para el Sr. Gobernador de Antioquia,’’
Medellín,  Nov. . Female schoolteachers at the camp school were per-
ceived to be endangered by the proximity of ‘‘the sort of class of racial types
 Notes to Chapter Two 

such as Casabe’s: the upper echelons of the hierarchy is in the hands of English-
men, and  percent of Labor is made up of Caribbean coastal people [costeños]
and people from Santander [santandereanos]. We would need to have lay female
teachers who are awfully saintly to be able to resist the insinuations of this en-
vironment.’’
 SGA, , vol. , Fuerzas Armadas, Puerto Berrío to Governor,  June .
 Ibid.
 SGA, , vol. , Maceo,  July .
 SGA, , vol. , Maceo,  Aug. .
 AGA, , vol. , Maceo,  Aug. .
 AGA, , vol. , Colonel Gustavo Berrío, commander Fourth Brigade to DAF
in a missive entitled, ‘‘Relaciones entre Comando Brigada y la Gobernación,’’
 Jan. ; and AGA , vol. , DAF to Colonel Gustavo Berrío,  January
.
 AGA, , vol. , Governor to municipal mayors,  Aug. .
 AGA, , vol. , Fuerzas Armadas, circular marked ‘‘Reservado,’’  Sept. .
 AGA, , vol. , Remedios to Governor,  Sept. .
 AGA, , vol. , Caracolí to Governor,  Oct. .
 AGA, , vol. , mayor, Yolombó to Governor,  Oct. .
 AGA, , vol. , Remedios, Conservative committee to Governor,  Oct. .
 AGA, , vol. , police inspector to Governor,  Oct. .
 SGA, , vol. , Cisneros,  June .
 AGA, , vol. , Colorado to Governor,  Nov. .
 Ibid.
 Former president and Antioqueño native son, Maríano Ospina Pérez, and his
moderate supporters, among them prominent Antioqueño Conservatives, were
crucial to mobilizing the coup against Laureano Gómez. See Bushnell, The
Making of Modern Colombia, p. .
 AGA, , vol. , Puerto Berrío, priest to Governor,  Dec. .
 AGA, , vol. , ‘‘Informe sobre la acción del bandolerismo,’’  May .
 SGA, , vol. ,  Jan. .
 SGA, , vol. ,  Feb. ; SGA, , vol. ,  March .
 SGA, , vol. , La Dorada,  May .
 SGA, , vol. , priest, Amalfi to Governor,  June ; SGA, , vol. ,
Conservative miner, Nús to Governor,  June .
 AGA, , vol. , Yalí,  April .
 AGA, , vol. , Remedios,  July .
 SGA, , vol. , DiLibAnt to Governor Píoquinto Rengifo,  Aug. .
 AGA, , vol. , Huila to Governor,  May .
 Sánchez, ‘‘La Violencia y sus efectos,’’ p. .
 San Roque, Alejandría, San Carlos, San Rafael, and Granada, various interviews
with INER, .
 AGA, , vol. , telegramas, San Roque,  Aug. ; Puerto Nare,  June
 Notes to Chapter Three

; Puerto Triunfo,  June ; Amalfi,  Aug.  and  Sept. ; Puerto
Perales, Sept. ; Yolombó,  Oct. ; Remedios,  Oct. ; and Cisneros,
 Dec. .
 Captain Corneta (Francisco Montoya), interview with the author, Medellín,
.
 SGA, , vol. , former guerrillas in ‘‘El Tigre,’’ Amalfi to DiLibAnt,  Sept.
.

. Urabá and Western Antioquia

 PEA, table .., ‘‘Población por sexo, según municipios,’’ pp. –.
 Ibid.
 The following section draws heavily on material in my article, ‘‘Violencia, colo-
nización y la geografía de la diferencia en Colombia,’’ Analísis Político  (Sept.–
Dec. ): –. For an examination of the catalysts to violence in the area (cur-
rently part of the department of Córdoba) that bordered on Urabá and from
which many of the armed groups operating during la Violencia in Antioquia
came, see Mauricio Romero, ‘‘Tierra y violencia en Córdoba,’’ Revista Análisis 
(): –, and Rural Transformation and Political Violence in Colombia, –
. A Comparative Study of Intra-Regional Change in a Non-Exporting Zone,
Center for Studies in Social Change (CSSC), New School for Social Research,
Working Paper Series, Working Paper no.  (March ): –.
 Maria Teresa Uribe de H., Urabá, Territorio o región? (Medellín, ), pp. –.
LeGrand, Frontier Expansion and Peasant Protest, p. . Hermes Tovar Pinzón,
El movimiento campesino en Colombia, p. ; and Uribe de H., Urabá, Territo-
rio o región?, p.  and pp. –; Gloria Isabel Ocampo, ‘‘Entre la localidad y
la Nación: Aspectos políticos de la construcción de identidades en un contexto
regional,’’ Revista Colombiana de Antropología , pp. –, details diverse
conceptions of identity and citizenship among inhabitants of different regions
in Colombia.
 AGA, , ‘‘Telegramas,’’ O.P.,  April . AGA, , ‘‘Telegramas,’’ O.P.,
 April ; and telegram,  April .
 AGA, , ‘‘Telegramas,’’ O.P.,  April .
 AGA, , Jorge Bejarano, Ministro de Higiene to Governor,  Aug. .
 AGA, , ‘‘Telegramas,’’ O.P.,  Nov. .
 AGA, , vol. , Departamento de Investigación Criminal to the Governor,
 Nov. , telegram regarding the behavior of the Colombian vice consul in
Panama, Otila Salazar de Puerta, who allegedly used her diplomatic post to
smuggle illicit arms to Liberal guerrillas in Urabá.
 AGA, , ‘‘ASPG,’’ Turbo,  May .
 AGA, , ‘‘MinGob,’’ presidential delegates, Turbo to Governor, June ;
AGA, , ‘‘MinGob,’’ Sopetran,  May .
 Aníbal Vallejo Alvarez, interview with the author, Medellín, spring .
 Notes to Chapter Three 

 AGA, , vol. , Colonel Eduardo Villamil, commander Cuarta Brigada,

Medellín to EBG,  Nov. .
 AGA, , ‘‘Telegramas,’’ Turbo, Nov. .
 SGA, , vol. , Dabeiba, Conservative committee to the Governor,  Feb.
.
 The regional government had already bought and shipped a thousand revolvers
to its supporters in January .
 SGA, , vol. , Smith and Wesson representative, Cali to the Governor,  Feb.
.
 SGA, , vol. , Caicedo,  April .
 AGA, , vol. , Jorge Salazar, Jefe Departamento de Investigación Criminal
to Governor,  May .
 SGA, , vol. , ‘‘Mapa de campamentos de bandoleros, Antadó.’’
 SGA, , vol. , ‘‘Asociación e Instigación para delinquir y demas delitos
conexos. Autos de Detención,’’ Ministerio de Justicia, Juzgado  de Instruc-
ción Criminal, Oficio no. , ‘‘A las autoridades de Policía de la República,’’
 Feb. .
 SGA, , vol. , oficio no.  regarding Frontino to Governor, May , .
 AGA, , vol. , Camparusia/Dabeiba,  May .
 SGA, , vol. , n.d.; SGA , vol. , ‘‘Asociación e Instigación para delin-
quir y demas delitos conexos. Autos de Detención,’’ Ministro de Justicia, Juzgado
 de Instucción Criminal, oficio no. , ‘‘A las autoridades de Policía de la
República,’’  Feb. .
 SGA, , vol. , José Parra Salazar, Visitador Administrativo Departamental
to SecGob, Medellín, Oficio no. ,  Dec. .
 Ibid.
 Bernardo Ospina Román, interview with the author, Medellín, spring .
Ospina Román was a prominent regional gaitanista figure in Antioquia who
helped the Liberal guerrillas and mediated between them and the party in
Medellín.
 AGA, , vol. , Departamento de Investigación Criminal (hereafter, DIC) to
Governor,  Nov. ; AGA, , vol. , Tnte. Col. Hernandez Pardo, Secre-
tario MinGuerra to Governor,  May ; AGA, , vol. , police inspector
to Governor,  June ; and AGA, , vol. , police inspector to Governor,
 June .
 AGA, , vol. , police inspector, Frontino to Governor,  June .
 AGA, , vol. , Luis Vásquez to Governor,  June .
 AGA, , vol. , Gustavo Echavarría to Governor, ‘‘Memorandum sobre la
situación de Orden Publico que se presenta en la región de Dabeiba a Chigo-
rodó,’’  Feb. .
 AGA, , ‘‘APSG,’’ Quibdó (Chocó) to Governor,  July ; AGA, ,
‘‘APSG,’’ Governor Valencia, Chocó to Governor,  July .
 Notes to Chapter Three

 SGA, , vol. , Uramita,  April .

 AGA, , vol. , Tucura/Dabeiba, BriCom-Cartagena to Governor,  July
.
 AGA, , vol. , Urrao, Father Zapata to Governor,  July ; AGA, ,
‘‘APSG,’’ La Camara (Salgar), MinGob to Governor,  July .
 AGA, , vol. , Colonel Villamil, Fourth Brigade, Medellín to Jorge Quintero,
Subteniente, Dabeiba, Mutatá, Chigorodó,  July .
 AGA, , vol. , Dabeiba,  July .
 AGA, , ‘‘APSG,’’ military commander, Caicedo to Governor,  Aug. .
 AGA, , vol. , Colonel Villamil to the Governor,  Aug. ; AGA, ,
‘‘APSG,’’ Jorge Salazar, Jefe DIC to Governor,  Aug. .
 AGA, , ‘‘APSG,’’ Caucasia, telegrapher to Governor,  Aug. ; SGA, ,
vol. , Ituango,  Aug. .
 AGA, , ‘‘APSG,’’ Secretario General, DiConsAnt to Governor,  Aug. .
 AGA, , ‘‘APSG,’’ Domingo Saresty to Governor,  Aug. .
 AGA, , vol. , ‘‘Informe sobre la acción del bandolerismo entre  y mayo
 de ,’’  May .
 AGA, , ‘‘APSG,’’ Urabá,  Aug. ; AGA, , ‘‘APSG,’’  Aug. ; and
AGA, , ‘‘APSG,’’ Rafael Posada, mayor to Governor,  Aug. .
 AGA, –, oficio no. , José A. Echavarría, police inspector to Governor,
 Aug. .
 AGA, , vol. , Mayor Infantería Marina Luis Millán Vargas, Jefe Polinal Di-
visión Antioquia, oficio no.  to Governor,  April .
 SGA, , vol. , Conservative committee, Turbo to Governor,  Aug. .
 AGA, , ‘‘APSG,’’ Colonel Villamil to Governor,  Aug. .
 AGA, , ‘‘APSG,’’ EBG to Laureano Gómez,  Aug. .
 AGA, , ‘‘APSG,’’ BHM to Laureano Gómez,  Aug. .
 AGA, , vol. , ‘‘Informe de Comisión Región Urabá,’’  Oct. .
 To assuage them, the colonel ordered five soldiers, an officer, and a light ma-
chine gun to complement the local police whose sole weapons were two used
machetes.
 LeGrand, Frontier Expansion, appendix D; and PEA, pp. –.
 AGA, , vol. , ‘‘Informe de Comisión Región Urabá,’’  Oct. .
 Ibid.
 AGA, , ‘‘Papeles,’’ telegram from mayor, San Juan de Urabá to Governor,
 Aug. ; and ‘‘Papeles,’’  Aug. .
 AGA, , vol. , ‘‘Informe de Comisión Región Urabá,’’  Oct. .
 At this time, the large concessions of land to transnational banana companies
and to Colombian subcontractors had not yet begun. These would eventually
accelerate and transform Urabá into one of Latin America’s leading exporters
of bananas and a site of considerable organized labor and leftist struggle after
, see Fernando Botero Herrera, Urabá: Colonización, violencia, y crisis del
 Notes to Chapter Three 

estado (Medellín: Editorial Universidad de Antioquia, ); AGA, , vol. ,

‘‘Informe de Comisión Región Urabá,’’  Oct. .
 AGA, , vol. , ‘‘Informe de Comisión Región Urabá,’’  Oct. .
 Ibid.
 AGA, , vol. , Turbo,  Nov. ; AGA, , vol. , Puerto Abaldía,  Nov.
.
 AGA, , vol. , Governor to Secretaría de Hacienda, Chocó,  Nov. ;
AGA, , vol. , Cartagena to Governor,  Dec. .
 SGA, , vol. , Urabá,  Dec. ; AGA, , vol. , Nicolás Gaviria to
MinGob, Bogotá, forwarded to Governor,  Dec. .
 AGA, , vol. , Antonio Moreno Mosquera, governor of Chocó to Governor,
 Nov. .
 AGA, , vol. , Braulio Henao to Roberto Cavalier, governor of Bolívar,
 Dec. ; AGA, , vol. , Capt. Salcedo, Fourth Brigade, Medellín to the
Governor,  Jan. .
 SGA, , vol. , Nicolás Gaviria, Cañasgordas to Directorio Nacional Conser-
vador and MinGob,  Dec. .
 SGA, , vol. , Caja de Credito Agrario to Governor,  Dec. .
 Ibid.
 Ibid.
 AGA, , vol. , Carlos Arrubla, subjefe seguridad to Governor,  Dec.
.
 AGA, , vol. , visitador ejecutivo to Governor,  Jan. .
 SGA, , vol. , Caicedo to Governor, Jan. .
 Ibid.
 AGA, , vol. , Laureano Gómez to BHM (in code and deciphered on the
back of the document),  Jan. .
 AGA, , vol. , BHM to Gonzalo Restrepo Jaramillo, marked ‘‘confidential,’’
 Feb. .
 AGA, , vol. , detectives to Governor,  Jan. .
 AGA, , vol. , Mutatá/Carepa, ‘‘Informe de Orden Publico,’’  Feb. .
 AGA, , vol. , Inspector Nacional de Bosques to Governor, Jan. .
 AGA, , vol. , ‘‘O.P.,’’  Dec. .
 PEA, p. .
 See Roldán, ‘‘Violencia, colonización y geografía de la diferencia en Colombia,’’
pp. –.
 SGA, , vol. , Jefe de Seguridad to Governor,  March ; SGA, , vol. ,
subjefe Seguridad to Governor,  April .
 AGA, , vol. , mayor, Turbo to Governor,  Jan. .
 AGA, , vol. , Colonel Luis Abadía to Governor,  Feb. .
 AGA, , vol. ,  Feb. .
 AGA, , vol. , Major González to Governor,  Jan. .
 Notes to Chapter Three

 AGA, , vol. , ‘‘Memorándum sobre la situación de Orden Público,’’  Feb.

.
 AGA, , vol. , BHM to Laureano Gómez, rough draft of letter, n.d. .
 AGA, , vol. , Conservative citizens, Abriaquí to Governor,  June .
 AGA, , vol. , Anzá,  Jan. ; AGA, , vol. , Colonel Abadía to Gov-
ernor,  Jan. .
 AGA, , vol. , ‘‘Memorándum sobre la situación de Orden Público,’’ n.d.
Jan. .
 AGA, , vol. , Laureano Gómez to Governor (original in code, but deci-
phered on the reverse),  Jan. ; AGA, , vol. , n.d. .
 AGA, , vol. , BHM to Gonzalo Restrepo Jaramillo,  Jan. .
 SGA, , vol. , ‘‘Plan de Orden Público #,’’  Feb. .
 AGA, , vol. , BHM to Gonzalo Restrepo Jaramillo,  Feb. .
 Ibid.
 AGA, , vol. , visitador administrativo to Governor,  Feb. .
 AGA, , vol. , Governor to Fuerzas Armadas,  March .
 AGA, , vol. , El Carmelo,  March .
 AGA, , vol. , Turbo,  March .
 AGA, , vol. , recolector de Hacienda, Chocó to Governor,  March .
 AGA, , vol. , ‘‘Informe Orden Publico,’’ Urabá,  March .
 AGA, , vol. ,  March .
 AGA, , vol. , Dabeiba,  March .
 AGA, , vol. , mayor, Frontino to Bogotá,  May ; AGA, , vol. ,
Frontino,  June .
 AGA, , vol. , Frontino,  July .
 AGA, , vol. , Dabeiba,  Aug. .
 AGA, , vol. , Tabacál, Buriticá,  Aug. .
 AGA, , vol. , Caja de Credito,  Sept. .
 Ibid.
 AGA, , vol. , mayor, Ituango to Governor,  Jan. .
 SGA, , vol. , DiConsAnt to Governor,  Sept. .
 AGA, , vol. , Luis Andrade, Bishop of Santa Fé de Antioquia to Governor,
 Feb. .
 AGA, , vol. , Bishop of Santa Fé de Antioquia to Governor,  July .
 AGA, , vol. , Bishop of Santa Fé de Antioquia to Governor,  Feb. .
 SGA, , vol. , ‘‘Informe Confidencial,’’ Medellín to Governor,  July .
 SGA, , vol. ,  March .
 AGA, , vol. , Peque,  May .
 AGA, , vol. , Anzá,  June ; AGA, , vol. , Sabanalarga,  May .
 AGA, , vol. , priest, Sabanalarga to Governor,  June .
 AGA, , vol. , priest, Sabanalarga to Governor,  June .
 AGA, , vol. , Colonel Abadía to Governor,  June .
 Notes to Chapter Three 

 SGA, , vol. , Fuerzas Armadas, marked ‘‘Reservado’’ to Governor,  July
.
 AGA, , vol. , Peque,  Aug. .
 SGA, , vol. , ‘‘Encargado Telefonos,’’ Jefe del Departamento de Seguridad
to Governor,  Aug. .
 AGA, , vol. , Peque,  Aug. ; AGA, , vol. , Peque,  Aug. .
 AGA, , vol. , Father Yepes, Sabanalarga to Governor,  Aug. .
 AGA, , vol. , Carretera al Mar/Urabá, Fuerzas Armadas, oficio no.  to
Governor,  Aug. .
 AGA, , vol. , , Antioquia to Governor,  Sept. .
 AGA, , vol. , Frontino/Salgar/Urrao, DiLibAnt,  Sept. ; AGA, ,
vol. , Buriticá,  Sept. ; AGA, , vol. , Dabeiba,  Sept. ; and
AGA, , vol. , Colonel Abadía regarding Juntas de Uramita to the Governor,
 Oct. .
 AGA, , vol. , lawyer for the Jefe de Orden Público, report on Frontino,
 Oct. .
 SGA, , vol. ,  to the Governor regarding Cañasgordas,  Nov.
.
 AGA, , vol. , Father Hugo Helmar, Caicedo to Governor,  Feb. .
 AGA, , vol. , Caicedo,  April .
 AGA, , vol. ,  April .
 SGA, , vol. , priest, Ituango, letter forwarded to Director General de la
Policía by Abogado Jefe de Orden Público,  May .
 SGA, , vol. , Uramita,  May .
 SGA, , vol. , Frontino,  July .
 AGA, , vol. , Fuerzas Armadas, Fourth Brigade, Medellín to Governor,
 June .
 AGA, , vol. , Conservative committee, Toledo to DiConsAnt,  Sept. .
 AGA, , vol. , Rioverde/Murrí,  Nov. .
 AGA, , vol. , Polinal oficio no. , Girardota,  Feb. .
 AGA, , vol. , Anzá,  March ; AGA, , vol. , Buriticá,  March ;
and AGA, , vol. , Dabeiba,  March .
 AGA, , vol. , Dabeiba,  March .
 SGA, , vol. , Dabeiba,  June .
 SGA, , vol. , Caja de Credito Agrario to Governor,  April .
 SGA, , vol. , Camparusia (Dabeiba),  July .
 AGA, , vol. , ‘‘Informe sobre la acción del bandolerismo,’’ May .
 AGA, , vol. , Abriaquí,  April ; AGA, , vol. , Anocozca, Cai-
cedo to Abriaquí,  April .
 Ibid.
 SGA, , vol. , mayor, Abriaquí to Nicolás Carmona Bernal, Aug. .
 AGA, , vol. , ‘‘Informe sobre la acción del bandolerismo,’’  May .
 Notes to Chapter Three

 AGA, , vol. , Governor to municipal mayors,  Sept. .

 AGA, , vol. , mayor, Toledo to Governor,  Oct. .
 AGA, , vol. , mayor, Toledo to Governor,  Oct. .
 AGA, , vol. , Conservatives, Uramita/Cañasgordas to Governor,  Oct.
.
 AGA, , vol. , mayor, Cañasgordas to Governor,  Oct. .
 AGA, , vol. , mayor, Dabeiba to Governor,  Dec. .
 SGA, , vol. , mayor, Frontino to Governor, Jan. .
 SGA, , vol. , Caja de Crédito Agrario regarding Frontino to Governor,
 Feb. .
 SGA, , vol. , Mario Javier Mesa, Visitador, ‘‘Informe correspondiente al
conservatismo de Dabeiba,’’ ‘‘Anexo,’’  Dec. .
 SGA, , vol. , Jefe del Departamento Juridico to SecGobAnt,  March ;
SGA, , vol. , Caja de Crédito Agrario, El Carmen (Chocó),  March .
 AGA, , vol. , Anzá,  March ; AGA, , vol. , mayor, Riogrande/
Turbo to Governor,  April .
 AGA, , vol. , mayor, Caicedo to Governor,  April .
 SGA, , vol. , Chigorodó,  April .
 AGA, , vol. , military mayor, Chigorodó to Governor,  March .
 AGA, , vol. , Gilberto Saldarriaga, Urabá to Governor,  April .
 SGA, , vol. , Dimas Navarro Marín, Anzá to Governor,  May .
 SGA, , vol. , Ramón Castaño, ex–municipal Conservative committee presi-
dent, Dabeiba to Governor,  June .
 AGA, , vol. , ‘‘Informe sobre el bandolerismo,’’  May . AGA, ,
Polinal, ‘‘Antioquia: Número de muertos desde agosto de  hasta enero ()
de ,’’ n.d. May .
 AGA, , vol. , ‘‘Informe sobre la acción del bandolerismo,’’ n.d. May .
 SGA, , vol. , ‘‘Presos Políticos Liberales en la Carcel de ‘‘La Ladera’’ por
orden de la Cuarta Brigada,’’  May .
 SGA, , vol. , Father Blandón Berrío, Peque/Uramita to Governor,  June
.
 AGA, , vol. , mayor, Liborina to Governor,  May .
 SGA, , vol. , FGM (owner of El Colombiano) to DAF, sent via private letter
because the original article was censored from publication,  June .
 AGA, , vol. , Governor orders all mayors to collect arms distributed under
previous regime,  Aug. .
 SGA, , vol. , Father Hugo Helmer, Caicedo to Governor,  July .
 SGA, , vol. , mayor, San Jerónimo complains of abuses against workers
on finca ‘‘Las Estancias,’’  July .
 SGA, , vol. , DiLibAnt to Governor,  Aug. .
 AGA, , vol. , Urama/Dabeiba to Governor,  Oct. .
 AGA, , vol. , Caja de Credito Agrario regarding Uramita/Cañasgordas and
Frontino to Governor,  Nov. .
 Notes to Chapter Four 

 SGA, , vol. , DiLibAnt to Governor,  Nov. .

 SGA, , vol. , Polinal, Buriticá to Governor,  Dec. .
 Ibid.
 SGA, , vol. , Auditoría General del Departmento de Antioquia to Governor,
 Oct. .
 SGA, , vol. , DiLibAnt to Governor,  Nov. .
 This and all subsequent land-value figures were calculated from PEA, table
..., ‘‘Número y avalúo catastral de la propiedad gravada y no gravada, –
,’’ pp. –; and table ..., ‘‘Número y avalúo catastral de la propiedad
gravada y no gravada ,’’ pp. –.

. Urrao and the Southwest

 Graciela Urrego, interview with the author, Urrao,  Aug. .
 Other southwestern towns such as Venecia and Andes also experienced vio-
lent incidents from  to , but these were episodic and primarily involved
unionized workers employed by the railroad and public road construction and
did not involve armed Liberal resistance against the state.
 For an examination of the association between frontier and ‘‘barbarism’’ in other
parts of Latin America, see Sarmiento, Life in the Argentine Republic; Silvio R.
Duncan Baretta and John Markoff, ‘‘Civilization and Barbarism: Cattle Frontiers
in Latin America,’’ Comparative Studies in Society and History , no.  ():
–; and Slatta, Gauchos and the Vanishing Frontier.
 Virginia Gutiérrez de Pineda, Familia y Cultura en Colombia, ‘‘Matrimonio
Catolico y Union Libre ,’’ p. , and ‘‘Ilegitimidad ,’’ p. .
 I am using the term here as Natalie Zemon Davis does in her essay ‘‘The Reasons
of Misrule,’’ in Society and Culture in Early Modern France (Stanford: Stanford
University Press, ), pp. –.
 Joaquin Montoya Escobar, Liberal committee president and president of the
municipal council during la Violencia, interview with the author, Urrao,  Aug.
.
 AGA, , ‘‘Telegramas,’’ O.P.,  April .
 AGA, , ‘‘APSG,’’ telegram,  Feb. .
 Joaquin Montoya Escobar, interview with the author, Urrao,  Aug. .
 AGA, , ‘‘Telegramas,’’ Valparaiso,  March .
 AGA, , ‘‘MinGob,’’ Jericó,  April ; AGA, , ‘‘MinGob,’’ Bolívar,
 April  and  April .
 Various individuals, informal interview with the author, Urrao, August .
 Joaquin Montoya Escobar, interview with the author, Urrao,  Aug. ; Jose-
fina Montoya Montoya, interview with the author, Urrao,  Aug. .
 Alcaldía Municipal de Urrao, ‘‘Sumarios,’’  Dec. .
 For the use of drunkenness as a guise to resistance, see William Taylor, Drink-
ing, Homicide, and Resistance in Colonial Mexican Villages (Stanford: Stanford
University Press, ).
 Notes to Chapter Four

 AGA, , vol. , letter from Joaquin Montoya, Eugenio Arroyave, Pepe Arenas,
Francisco Luis Guzman, and others, Urrao to Governor,  Oct. .
 AGA, , vol. , Urrao,  Oct. .
 AGA, , vol. ,  Nov. ; Aníbal Vallejo Alvarez, interview with the
author, Medellín, ; Joaquin Montoya Escobar, interview with the author,
Urrao,  Aug. .
 It is not coincidental that in current dialogues with Marxist guerrilla organiza-
tions all warring factions have agreed that church officials should act as inter-
mediaries in the resolution of conflict in Colombia.
 Luis Andrade Valderrama, Obispo Santa Fé de Antioquia, Cartas Pastorales,
(Medellín, n.d.).
 AGA, , no vol. no., anon. telegram to Governor, n.d.
 AGA, , vol. , letter from Colonel Eduardo Villamil to Governor,  Jan.
. Colonel Villamil made explicit his discomfort with the escalation of vio-
lence and confided that ‘‘the struggle there was tormenting me mentally.’’
 Medianalga (Avelino Urrego), interview with the author, Urrao,  Aug. .
Army neutrality in Urrao is confirmed in confidential documents in the gov-
ernor’s archive in which members of the Conservative administration com-
plained that the army aided the guerrillas and turned a blind eye to their ac-
tivities in Urrao. See AGA , vol. , ‘‘Declaración de José Nicanor Arboleda
Rodriguez,’’ Departamento de Seguridad, Identificación y Extranjeriá, Medellín,
 April .
 Medianalga (Avelino Urrego), interview with the author, Urrao,  Aug. .
 Ibid.
 Franco’s letter to Pioquinto Rengifo was reproduced in the magazine Cromos,
 May , pp. –. Froilan Montoya Mazo was the real author of Franco’s let-
ter to the governor. Froilan Montoya Mazo, interview with the author, Medellín,
fall .
 Medianalga (Avelino Urrego), interview with the author, Urrao,  Aug. .
 AGA, , vol. , Secretario General, Bogotá to Governor Braulio Henao Mejía,
 Jan. ; AGA, , vol. , Colonel Abadía to Governor Braulio Henao Mejía,
 Jan. .
 AGA, , vol. , telegram from José Gallego, mayor, Urrao to Governor,
 Sept. .
 Medianalga (Avelino Urrego), interview with the author, Urrao,  Aug. ;
Tarzán (Herlindo Montoya), interview with the author, Betulia,  Aug. ;
and Graciela Urrego, interview with the author, Urrao,  Aug. .
 Tarzán (Herlindo Montoya), interview with the author, Betulia,  Aug. .
 By , Franco was signing his missives to the government and other docu-
ments as ‘‘Major’’ Franco. The plana mayor was created some time after the ini-
tial organization of the guerrillas when some members of the group felt that
Franco had become too dictatorial and that power should be shared. Tarzán
(Herlindo Montoya), Betulia,  Aug. .
 Notes to Chapter Four 

 AGA, , vol. , Departamento de Seguridad, Identificación y Extrangería,

Medellín, ‘‘Declaración de José Nicanor Arboleda Rodriguez,’’  April ;
AGA, , vol. , Alcaldía Municipal de Urrao, ‘‘Copia de la Declaración Ren-
dida Por el Señor Argemiro Herrera Correa,’’  May ; Medianalga (Avelino
Urrego), interview with the author, Urrao,  Aug. ; Tarzán (Herlindo Mon-
toya), interview with the author, Betulia,  Aug. ; and Graciela Urrego,
interview with the author, Urrao,  Aug. .
 Graciela Urrego, interview with the author, Urrao,  Aug. .
 The trisagio is a prayer to the Holy Trinity usually recited during the adoration
of the Eucharist during Mass.
 Medianalga (Avelino Urrego), interview with the author, Urrao,  Aug. .
On the day I interviewed Avelino Urrego, his wife and sister-in-law (Debora
and Judith) were also present, and they contributed their memories of Franco’s
guerrillas, too.
 Debora and Judith, interview with the author, Urrao,  Aug. .
 Medianalga (Avelino Urrego), interview with the author, Urrao,  Aug. .
 AGA, vol. , , Alcaldía Municipal de Urrao, ‘‘Copia de la Declaración Ren-
dida Por el Señor Argemiro Herrera Correa,’’  May .
 Graciela Urrego, interview with the author, Urrao,  Aug. ; AGA, ,
vol. , Report of Colonel Luis Abadía, Comandante de la a Brigada to the Gov-
ernor,  Aug. ; AGA, , vol. , Alcaldía Municipal de Urrao, ‘‘Declaración
del Señor Cruz Antonio Higuita Cossio,’’  May .
 Indeed, perhaps the closest parallel to Urrao’s guerrilla forces would be the army
of peasants organized by Zapata in Morelos during the Mexican Revolution.
 AGA, vol.  (), judicial testimony,  May .
 Sara Ruddick has argued persuasively in ‘‘Toward a Feminist Peace Politics,’’
Gendering War Talk, ed. Miriam Cooke and Angela Woollacott (Princeton:
Princeton University Press, ), pp. –, that ‘‘in most cultures, war’s mas-
culinity is constructed in tandem with a distinctively military femininity,’’ which
may include ‘‘masking or denying the sexual assaultiveness of ‘our troops.’ ’’
Certainly Ruddick’s assertion that the ‘‘loyal military female . . . eroticizes ‘our’
heroes’’ accurately describes the prevailing attitude of the female companions
of the guerrillas in Urrao, but I would also suggest that because such women, in
contrast to those of Ruddick’s analysis, were at the battlefront with their men,
they may have made the eroticized aggression that is generally so important an
aspect of ‘‘assaultive masculinity’’ in times of war, less of a motivating factor in
guerrilla warfare in Urrao. I am not arguing that women are inherently more
moral, but rather that the circumstances of Urrao’s guerrilla organization where
war and domesticity were not mutually exclusive may have shaped a policy that
repudiated sexual assault as a valid weapon of resistance among the guerrilla
command.
 There were no reports of sexual abduction, gang rape, or battles over control
of concubines among Urrao’s guerrillas. See also Germán Guzmán Campos,
 Notes to Chapter Four

Orlando Fals Borda, and Eduardo Umaña, La Violencia en Colombia, vol. , th
ed. (Bogotá: Carlos Valencia Editores).
 Medianalga (Avelino Urrego), interview with the author, Urrao,  Aug. ;
Josefina Montoya Montoya, interview with the author, Urrao,  Aug. .
 It is interesting to note that until recently in Colombia even those who dis-
agreed with the ideology and tactics of leftist guerrillas were wont to hope
that if one of their loved ones were kidnapped it would be the guerrillas and
not the army, right-wing paramilitary groups, or common delinquents who
were responsible. It was widely believed that the guerrillas were likelier to treat
their prisoners humanely than either the government or groups associated with
the right would. See Herbert Braun, Our Guerrillas, Our Sidewalks: A Jour-
ney into the Violence of Colombia (Niwot, Colo.: University Press of Colorado,
); and Gabriel García Márquez, News of a Kidnapping (New York: Penguin
Press, ).
 Josefina Montoya, interview with the author, Urrao,  Aug. .
 Absalón Machado, interview with the author, Bogotá, fall .
 AGA, , vol. , Colonel Luis Abadía A., commander, a Brigada to Governor,
 Aug. .
 AGA, , vol. , mayor, Urrao to Governor,  Oct. .
 AGA, , ‘‘APSG,’’ Ministro de Justicia to Governor,  Dec. .
 AGA, , vol. , telegram from soldiers, Betulia to their superior, Medellín,
 Dec. .
 AGA, , ‘‘APSG,’’ telegram, Ricardo Mejía, Administrador de Rentas and
Ruben Vélez, Tesorero, Salgar accusing the army of ‘‘convivencia liberalismo’’
to Governor,  Dec. .
 AGA, , ‘‘APSG,’’ telegram, Colonel Villamil, Fourth Brigade to MinGob,
 Jan. .
 AGA, , ‘‘APSG,’’ report of assassination of police inspector and mayor, Isleta,
 March .
 SGA, , vol. ,  June .
 AGA, , ‘‘Papeles,’’ telegram, Directorio Conservador Medellín to MinGob,
Bogotá,  July .
 AGA, , vol. , administrator, Concordia to Francisco Ospina Pérez,  July
.
 AGA, , vol. , telegram, Colonel Villamil to Governor,  July .
 AGA, , vol. , Father Zapata, Urrao to Governor,  July ; AGA, ,
vol. , mayor, Salgar to Governor,  July ; and AGA, , ‘‘APSG,’’ message
forwarded to Governor by MinGob, Bogotá,  July .
 AGA, , vol. , Conservative committee, Salgar to Governor,  July .
 AGA, , ‘‘APSG,’’ Colonel Villamil to Governor,  Aug. ; AGA ,
‘‘APSG,’’ ‘‘Informe de Orden Público, Urrao,’’  Aug. .
 AGA, , ‘‘Papeles,’’ telegram, Colonel Villamil to Governor,  Aug. .
 Notes to Chapter Four 

 SGA, , vol. , list of towns buying arms from the regional government:
Támesis, Venecia, Titiribí, Anzá, Betania, and Pueblorrico,  July .
 The dates of guerrilla attacks in these different locales never differed by more
than a couple of days although the objects of attack sometimes varied from
area to area—more concentrated on the state, its offices, and representatives in
places like Urabá, and more directed at individual Conservative families and
their properties in places like Urrao, the Bajo Cauca, and western Antioquia.
The coincidence of guerrilla attacks between  and  in different areas of
Antioquia was too exact to have been mere chance; indeed, in several cases the
guerrillas themselves, or those warning the government that an attack was im-
minent, indicated that these were coordinated. Whether the coordination was
engineered in Medellín at Liberal party headquarters or in Bogotá is unclear,
but that there was coordination is simply beyond doubt.
 AGA, , ‘‘APSG,’’ Guasabra, Antioquia,  August .
 AGA, , ‘‘Papeles,’’ José Toro, Directorio Conservador Urrao to Secretario
General del Presidente,  Aug. .
 AGA, , ‘‘APSG,’’ Jesús Arroyave, Urrao to Governor,  Aug. .
 SGA, , vol. , ‘‘Informe,’’ no. , from Rafael Osorio, Director Departa-
mento Jurídico to Governor,  Dec. .
 AGA, , vol. , Urrao and Betulia,  Dec. ; AGA, , vol. , Urrao,
 Dec. .
 AGA, , ‘‘Papeles,’’ Directorio Conservador, Betulia to Governor,  Aug.
.
 SGA, , vol. , ‘‘Informe,’’ no. , from Rafael Osorio, Director Departa-
mento Juridico to Governor,  Dec. .
 This runs contrary to what researchers such as Guzmán, Fals Borda, and Umaña
argue regarding the barbaric quality of violence in Antioquia. I have no explana-
tion for why I found so few explicit references to sexual mutilation and torture
in the period from  to  (five among thousands of references). It is pos-
sible that such violence was more characteristic of the pajaro or contrachusma
violence which emerged in particular areas of Antioquia after  or that it
was simply so commonplace as to not merit explicit mention in the reports or
complaints filed with the authorities. See Guzmán, Fals Borda, and Umaña, La
Violencia en Colombia, vol. , p. .
 The dichotomization of troops into ‘‘us’’ and ‘‘them’’ and the attribution of bar-
barity to the enemy are commonly invoked and necessary tropes in the justi-
fication of war. Sara Ruddick notes, for instance, ‘‘typically, masculinities are
also divided between the enemy and ‘our troops.’ ‘We’ are the just warrior-
protectors. By contrast, a particularly malignant form of swaggering mascu-
linity—a criminalized, sexualized aggression—is attributed to the enemy. When
enemy males are racialized as predators from whom innocent countries or
women and children need protection, they become killable killers ready to be
 Notes to Chapter Four

burned and buried in their trenches.’’ See Ruddick, ‘‘Toward a Feminist Peace
Politics,’’ p. .
 SGA, , vol. , ‘‘Informe,’’ no. , from Rafael Osorio, Director Departa-
mento Jurídico to Governor,  Dec. .
 AGA, , ‘‘Papeles,’’ Directorio Conservador, Salgar to MinGob, Bogotá,
 July ; AGA, , ‘‘Papeles,’’ police inspector, Buchadó, Chocó to Gov-
ernor,  March ; AGA, , ‘‘Papeles,’’ Carmen, Chocó to Governor,  July
; AGA, , vol. , telegram from Concordia to Governor,  July ; AGA,
, vol. , Conservative committee, Salgar,  July  and  Aug. .
 Indeed, this author found amid the records of Urrao’s municipal archive a form
letter sent out by Germán Saldarriaga (whose pseudonym was Testis Fidelis),
the author of De Caín a Pilatos (Medellín, ), a grisly account of Liberal bar-
barity executed between  and . Saldarriaga’s letter explicitly solicited
‘‘gruesome examples of guerrilla/Communist violence, the more hair-raising,
the better’’ for inclusion in a second edition to his book. Collected testimonial
literature such as De Cain a Pilatos resembled in content and propagandistic in-
tent the Brazilian literatura de cordel recounting famous bandit exploits or Fox’s
Book of Martyrs issued by Protestants during the reign of the Catholic English
queen, Mary Tudor, in the sixteenth century.
 AGA, , vol. , Colonel Villamil to Teniente Quintero,  July .
 AGA, , vol. , Governor Braulio Henao Mejía to Laureano Gómez,  Feb.
.
 SGA, , vol. , telegram from Governor to president Conservative committee,
Urrao,  June .
 AGA, , vol. , BHM to Laureano Gómez, undated rough draft of letter.
 AGA, , vol. , letter from Felipe Chica, engineer, Secretaría de Obras Pub-
licas, Antioquia to Director General de Caminos, Concordia,  April .
 Ibid.
 AGA, , vol. , ‘‘Memorandum sobre la situación de Orden Público que se
presenta en la región de Dabeiba y Chigorodó,’’  Feb. . Salgar reported an
attack of two hundred men on  July , while Betulia reported that one
thousand guerrillas were stationed on a nearby hacienda, see AGA , vol. ,
telegram, Salgar,  Aug. ; and AGA, , vol. , mayor, Betulia to the Gov-
ernor,  Dec. and  Dec. .
 AGA, , vol. , telegram, Governor to MinGob Urdaneta Arbelaez,  Aug.
.
 AGA, , vol. , Governor to Laureano Gómez,  April .
 AGA, , vol. , ‘‘Declaración de José Nicanor Arboleda Rodríguez,’’ Departa-
mento de Seguridad, Identificación y Extrangería, Medellín,  April .
 Medianalga (Avelino Urrego) interview with the author, Urrao,  Aug. .
 AGA, , vol. , Conservative committee, Urrao to Governor,  Aug. .
 AGA, , vol. , Caja de Crédito Agrario to Governor,  Sept. .
 Notes to Chapter Four 

 AGA, , vol. , Colonel Abadía, a Brigada to Governor,  May .

 AGA, , vol. , Bocas de Peque,  Sept. .
 AGA, , vol. , Colonel Abadía, a Brigada to Governor,  Oct. .
 AGA, , vol. , J. V. C. to Governor,  Nov. .
 Medianalga (Avelino Urrego), interview with the author, Urrao,  Aug. .
An ex-guerrilla confirmed Urrego’s statement, see AGA , vol. , ‘‘Declara-
ción de José Nicanor Arboleda Rodríguez,’’ Departamento de Seguridad, Iden-
tificacion y Extrangería, Medellín,  April .
 Ibid.; and Tarzán (Herlindo Montoya), interview with the author, Betulia,
 Aug. .
 AGA, , vol. , Urrao,  Nov. .
 AGA, , vol. , Urrao,  Sept. ; AGA, , vol. , Salgar,  Sept. ; AGA,
, vol. , Salgar,  Sept. .
 AGA, , vol. , Betulia,  Jan. ; AGA, , vol. , Urrao,  Feb.
; AGA, , vol. , Concordia,  Feb. ; AGA, , vol. , Conser-
vative committee, Betulia to Governor,  Feb. ; AGA, , vol. , Urrao,
 March .
 AGA, , vol. , Conservative committee, Urrao to Governor,  Feb. .
 AGA, , vol. , Emilio Botero, Cartagena, Bolívar to Governor,  Jan. .
 AGA, , vol. , Conservative citizens, Concordia to Governor,  Feb. .
 AGA, , vol. , Conservative committee, Betulia to Governor,  March .
 AGA, , vol. , oficio no. , Betulia,  March .
 AGA, , vol. , Conservative official, Titiribí to Governor,  April .
 AGA, , vol. , Medellín,  April .
 AGA, , vol. , Las Azules,  Oct. .
 SGA, , vol. , Conservatives, Betulia to Governor,  May .
 While such records are clearly not exhaustive and probably severely underesti-
mate the total number of dead, since many bodies were flung into rivers, dis-
membered, or buried in unmarked graves before body counts could be done,
they do provide a suggestive indication of the relative number of casualties
among each of Antioquia’s towns. AGA, , vol. , ‘‘Informe sobre la acción
del bandolerismo,’’ May .
 Ibid.
 No deaths were attributed to violence in parish ledgers for , but this may
simply be because Father Zapata made no distinction between violent and other
deaths. In fact, his successor’s (Father Ramirez) tendency to separate deaths due
to violence from other deaths seems to have constituted a fairly isolated prac-
tice in Antioquia. My research uncovered no other parish registries with similar
notations.
 Tarzán (Herlindo Montoya), interview with the author, Urrao,  Aug. ;
Graciela Urrego, interview with the author, Urrao,  Aug. .
 AGA, , vol. , Colonel Luis Abadía A., a Brigada to Governor,  Aug.
 Notes to Chapter Four

; AGA, , vol. , telegram from mayor José Gómez, Urrao to Governor,
 Oct. .
 Absalón Machado, interview with the author, Bogotá, fall .
 AGA, , vol. , Caicedo,  Jan. ; AGA, , vol. , Caicedo,  Jan. ;
AGA, , vol. , Caicedo,  Jan. ; and AGA, , vol. , Urrao,  Jan. .
 AGA, , vol. , Salgar,  March ; AGA, , vol. , Montebello, Salgar,
 March ; and AGA, , vol. , La García, Caicedo,  March .
 AGA, , vol. , Urrao,  Aug. .
 Criminal bands dedicated to the theft of coffee proliferated in towns to the south
of Urrao in the later years of the s, but it is not clear—with the exception
of Jardín where a famous contrachusma band operated openly—to what degree
these groups had any overt political affiliation or objectives after .
 Tarzán, interview with the author, Betulia,  Aug. ; Medianalga, interview
with the author, Urrao,  Aug. ; and Graciela Urrego, interview with the
author, Urrao,  Aug. .
 AGA, , vol. , Governor to MinGob Andrade (in code but deciphered on
the reverse), Urrao,  June .
 Ibid.
 AGA, , ‘‘Informe,’’ Directorio Conservador, Medellín to Governor, July .
 AGA, , vol. , Comité Conservador, Urrao to Governor, July .
 AGA, , vol. , Comité Conservador, Urrao to Governor,  Aug. .
 AGA, , vol. , Urrao,  Sept. ; AGA, , vol. , Caja de Crédito Agrario
to Governor,  Sept. .
 AGA, , vol. , Salgar,  Feb. .
 AGA, , vol. , Urrao,  April , . Guerrillas were supposedly paid seventy
to one hundred pesos per month plus bonuses (although Franco was alleged to
have promised recruits a salary of six hundred pesos per month). See also, AGA,
, vol. , Capellán Fuerzas Armadas, Pereira to the Governor,  Jan. ,
for accusations that the guerrillas received a salary.
 AGA, , vol. , Urrao forwarded to Governor,  May .
 AGA, , vol. , Conservatives, Betulia to Governor,  March .
 AGA, , vol. , Major Luis Millán, Director General, Polinal, Bogotá to
Governor,  Feb. . The major justified burning the entire area around the
guerrilla camp of Camparusia in Dabeiba, arguing that despite the presence
of twenty Conservative inhabitants, they were suspected of colluding with the
guerrillas since ‘‘they freely travel without anyone bothering them.’’
 SGA, , vol. , Betulia, May , .
 SGA, , vol. , report of mission to eradicate guerrillas in Betulia and Urrao
undertaken by Arturo Velásquez to the Governor,  May .
 PEA, table ..., ‘‘Población ocupada, por ramas de actividad económica,
Censos –,’’ p. .
 Decree no.  issued by ‘‘The Revolutionary Forces of the West and the South-
 Notes to Chapter Four 

west,’’ General Guerrilla Headquarters, Pabón, Urrao,  July , Froilan Mon-
toya Mazo (Private Archive), Medellín.
 ‘‘Decree no.  re: Public Order,’’ signed Major Franco, Pabón,  July , Froi-
lan Montoya Mazo (Private Archive), Medellín. Froilan Montoya Mazo was a
gaitanista Liberal leader who maintained close contact with Urrao’s guerrillas
throughout la Violencia.
 Graciela Urrego, interview with the author, Urrao,  Aug. ; Medianalga
(Avelino Urrego), interview with the author,  Aug. ; and Tarzán (Herlindo
Montoya), Betulia,  Aug. . In addition to repeatedly indicating that the
Llanos guerrillas constituted Urrao’s role model during la Violencia, Graciela
Urrego insisted that the guerrillas had a communication system set up between
Urrao and the Llanos made up of human relay runners who traveled from post
to post from the west to the east to confer with their Llanos counterparts on
strategy and orders.
 El Colombiano,  July .
 El Colombiano,  April .
 El Colombiano,  May .
 Ibid.
 El Colombiano,  May .
 El Colombiano,  May .
 AGA, , vol. , Capellán Fuerzas Armadas, Pereira to Governor,  Jan. .
 AGA, , vol. , ‘‘O.P.,’’ ‘‘Algunas Notas Relativo [sic] a los Bandidos de
Urrao,’’ n.d. I am deducing that this was written by an English speaker because
of the syntactical errors and literal translations scattered throughout the text
and the obsession with Cold War ideology and the infiltration of Communism
into the region.
 Ibid., . These observations were delivered in the form of ‘‘bullets’’ numbered
from  through . The above quote includes points  to  in the original report.
 Ibid., points  through  under the heading ‘‘Situación Actual.’’
 Ibid., point  under the heading ‘‘Las Resultas’’ [sic].
 Ibid., point .
 Ibid., points  and .
 Ibid., point  under the heading ‘‘La Solución.’’
 Ibid., point .
 Ibid., point .
 Guzmán, Fals Borda, and Umaña, in La Violencia en Colombia, estimated the
number of active guerrillas in Antioquia at five thousand and many other re-
searchers have reproduced this figure unquestioningly.
 AGA, , vol. , ‘‘Algunas notas,’’ point  under the heading ‘‘Conclusiones:
El bandolerismo en Colombia ahora no es lo que fuera a [sic] empezar.’’
 Ibid., point  under the heading ‘‘Conclusiones.’’
 Ibid., point .
 Notes to Chapter Epilogue

 Ibid., point .

 Ibid., point .
 Ibid., point .
 The ensuing account is based on Graciela Urrego, interview with the author,
Urrao,  Aug. .

Epilogue
 Greg Grandin discusses the findings and deliberations of the Guatemalan Truth
Commission, particularly regarding the classification of violence in the Guate-
malan case as genocide and highlights the historical antecedents that contrib-
uted to the state’s targeting of particular groups of Mayan Indians during its
counterinsurgency campaigns in an unpublished paper (), ‘‘Insoluble Acts
and Historical Solutions: Law, History, and Latin American Cold War Terror.’’ I
am grateful to the author for bringing his paper to my attention. For interviews
with military commanders during the Guatemalan civil war and an exploration
of the logic behind their counterinsurgency campaign, see Jennifer Schirmer,
The Guatemalan Military Project.
 Steve J. Stern, ed., Shining and Other Paths: War and Society in Peru, –
(Durham, N.C.: Duke University Press, ), especially Nelson Manrique, ‘‘The
War for the Central Sierra,’’ pp. –. The intersection of violence, race, and
ethnicity in the Peruvian case is more explicitly developed in Manrique, ‘‘Politi-
cal Violence, Ethnicity, and Racism in Peru in the Time of the War,’’ Journal
of Latin American Studies , no. ; and, on this same theme, see also ‘‘Vietnam
in the Andes: ‘Pancho,’ ’’ in The Peru Reader: History, Culture, Politics, ed. Orin
Starn, Carlos Iván DeGregori, and Robin Kirk (Durham, N.C.: Duke University
Press, ), pp. –.
 Neil Harvey, The Chiapas Rebellion: The Struggle for Land and Democracy (Dur-
ham, N.C.: Duke University Press, ); John Womack Jr., ed., Rebellion in
Chiapas: A Historical Reader (New York: New Press, ); Shadows of Tender
Fury: The Letters and Communiqués of Sub-Comandante Marcos and the Zapa-
tista Army of National Liberation, ed. Frank Bardacke, Leslie López, and the
Watsonville, California, Human Rights Committee (New York: Monthly Re-
view, ).
 New York Times, ‘‘Crisis in Colombia as Civil Strife Uproots Peasants,’’  Oct.
.
 The regional paper, El Colombiano, published a series of articles focusing on
the problem of human displacement in Antioquia and Colombia and the large
number of refugees from Urabá and the Magdalena Medio, among them ‘‘Cam-
pesinos rumbo al destierro: Pavarandó, el refugio,’’  Dec. .
 These fears are not baseless, see Carlos Castaño’s warning that there would be
many more massacres in towns perceived to be leftist supporters, in El Tiempo,
‘‘Va a haber muchos más mapiripanes,’’  Sept. . For evidence that he has
 Notes to Epilogue 

made good on his promise, see New York Times, ‘‘Rightist Squads in Colombia
Beating the Rebels,’’  Dec. .
 Alonso Salazar, Born to Die in Medellín (London: Verso, ).
 Ibid.
 Elaine Scarry, The Body in Pain: The Making and Unmaking of the World (New
York: Oxford University Press, ) refers to this as the euphemistic and erro-
neous notion of the so-called indirect consequences of war.
 New York Times,  Oct. .
 See the issues of the Colombian political magazine, Alternativa, devoted to
Antioqueño Governor Alvaro Uribe Vélez’s endorsement of a paramilitary solu-
tion to violence and the effects of this policy in Urabá in ‘‘Mano Dura? O
Tenaza Militar?’’ in Alternativa  ( Nov.–  Dec. ): –; and the special
issue devoted to analyzing the paramilitary phenomenon and the Colombian
Supreme Court’s debate regarding the constitutionality of privately financed
armed groups with a special emphasis on the organization of Convivirs in Antio-
quia in ‘‘Retornan los pajáros,’’ Alternativa  ( Dec. – Feb. ): –.
 ‘‘Se debe acabar con la mentira oficial,’’ Alternativa  ( Jan.– Feb. ): –
(interview with ex–army colonel Carlos Alfonso Velásquez, former commander
of Urabá, regarding the spread of paramilitary organizations and their collu-
sion with the armed forces in Colombia); Cynthia Aronson and Robin Kirk,
State of War: Political Violence and Counterinsurgency in Colombia (New York:
Human Rights Watch/Americas, ); and Human Rights Watch/Americas–
Arms Project, Colombia’s Killer Networks: The Military-Paramilitary Partnership
and the United States (New York: Human Rights Watch/Americas, ).
 Manuel Enrique López B., ‘‘Implicaciones del conflicto armado en el modelo de
economía cafetera. Aproximación al caso del suroeste antioqueño,’’ in Conflic-
tos Regionales: La Crisis del Eje Cafetero, ed. Gonzalo Sánchez G., Jorge Enrique
Robledo Castillo, Absalón Machado C., Manuel E. López, and Christopher Lon-
don (Bogotá: FESCOL and IEPRI, ), anexo  and , pp. –.
 During the s and s political dissent in these towns was expressed
through support of the Unión Patriotica, a Communist affiliate, the majority of
whose members have been assassinated.
 The debate regarding the validity of ‘‘region’’ as an analytical category is one
that has recently received a good deal of attention from Colombian scholars
and is widely debated by scholars of Latin America in general. Among the rele-
vant works for Colombia are Miguel Borja, Estado, Sociedad y Ordenamiento
Territorial en Colombia (Bogotá: CEREC, ); and Fabio Zambrano Pantója,
Colombia, País de Regiones (Bogotá: CINEP/Colciencias, ).
Bibliography

Archives and Manuscript Collections

Note on archival sources: There were  volumes of correspondence in the
Archivo de la Gobernación de Antioquia for the period from  to . None of
the material in the archive is indexed, nor were all of the volumes numbered. The
absence of a volume number in the citations therefore is not an oversight. The
material contained in the Archivo de la Gobernación, moreover, was not always in
chronological order. Thus, it was possible for material from say, March , to be
included in different volumes of the  correspondence or sometimes even in
volumes for the correspondence of another year.

Archivo de la Secretaría de Gobierno de Antioquia, –,  vols. ()

Archivo Nacional de Colombia, Ministerio de Industrias, Correspodencia de
Baldíos ()
Archivo Parroquial, Registro de funciones, –, Urrao, Antioquia
Archivo Privado del Señor Gobernador de Antioquia, –,  vols. ()
Archivo Privado, Froilan Montoya Mazo, –, Medellín
Archivo Privado, José Mejía y Mejía, Medellín
Centro Cultural Jorge Eliécer Gaitán, Correspondencia de Jorge Eliécer Gaitán,
–, Bogotá
Newspaper collection, library of the Universidad de Antioquia, Medellín

Newspapers and Magazines

Acción Católica Colombiana
Alternativa
El Colombiano
El Correo
Cromos
Crónica Municipal, Medellín
La Defensa
El Diario
New York Times
El  de Abril
El Obrero Católico
El Poder
El Pueblo
Revista de la Policía Nacional, Bogotá
Progreso
 Bibliography

Revista del Estado Mayor

El Siglo
El Tiempo

Government Publications
Antioquia. Alcaldía de Medellín. . Convocatoria a la paz. Medellín: Imprenta
Municipal.
Antioquia. Departamento Administrativo de Planeación. . Distribución de
los predios rurales en Antioquia según su tamaño . Medellín: Imprenta
Departamental.
Antioquia. Dirección de Desarrollo. . Región Suroeste: Analísis de la situ-
ación actual. Informe Básico, Inventario y Carectización. Medellín: Imprenta
Departamental.
Colombia. Banco de la Repúbica. . Memoria de Hacienda, Francisco de
Paula Pérez, , , . Bogotá.
Colombia. Camara de Representantes. . Laureano Gómez: Obras Selectas.
Part , vol. . Medellín: Editorial Bedout.
Colombia. Camara de Representantes. n.d. Román Gómez: Obras selectas.
Bogotá: n.p.
Colombia. Departamento Administrativo Nacional de Estadística. . Censo
nacional agropecuario, Antioquia. Bogotá.
Colombia. Departamento Administrativo Nacional de Estadistíca. . Colom-
bia Estadística. Bogotá.
Colombia. . Decreto del Estado de Sitio. Bogotá.
Colombia. Departamento Administrativo Nacional de Estadística. . Pano-
rama estadístico de Antioquia. Bogotá. ()
Colombia. Departamento Administrativo Nacional de Estadística. Estadística
Electoral, –. Bogotá.
Colombia. Departamento de Antioquia. Anales de la Asamblea de Antioquia.
Medellín: Imprenta Departamental.
Colombia. Departamento de Antioquia. Anuario estadístico, años , ,
.
Colombia. Departamento de Antioquia. Anuario estadístico, años , , .
Colombia. Departamento de Antioquia. Anuario estadístico, años –.
Colombia. Ministerio de Gobierno, Memorias , , , , , ,
, , , . Bogotá.

Interviews
   
Manuel Mejía Vallejo, Medellín, Aug. .
Medianalga [Avelino Urrego], Urrao, Antioquia,  Aug. .
Judith and Debora, Urrao,  Aug. .
 Bibliography 

Tarzán [Herlindo Montoya], Betulia, Antioquia,  Aug. .

Joaquín Montoya Escobar, Urrao, Antioquia,  Aug. .
Josefina Montoya, Urrao, Antioquia,  Aug. .
Graciela Urrego, Urrao, Antioquia,  Aug. .
Iván J. Cadavid (priest), Bogotá, fall .
Absalón Machado, Bogotá, fall .
Froilan Montoya Mazo, Medellín, fall .
Emma Londoño de Mejía, Medellín, Oct. .
Bernardo Ospina Román, Medellín, Nov. .
Capitán Corneta [Francisco Montoya], Medellín, spring .
Aníbal Vallejo Alvarez, Medellín, April .

       

 (),   , 
Alejandría,  Sept. .
Granada,  June .
San Carlos,  June .
San Rafael,  June .
San Roque,  June,  Aug., and  Sept. .

Secondary Sources
Abel, Christopher. . Política, iglesia y partidos en Colombia. Bogotá: –
Universidad Nacional de Colombia.
Abella Rodríguez, Arturo. . Laureano Gómez. Bogotá: Espara.
Abrams, Philip. . ‘‘Notes on the Difficulty of Studying the State.’’ Journal of
Historical Sociology  (): –.
Academia Antioqueña de Historia. . Un siglo de gobierno en Antioquia, –
. Medellín: Secretaría de Servicios Administrativos del Departamento de
Antioquia.
Alape, Arturo. . El Bogotazo: Memorias del olvido. Bogotá: Editorial Pluma.
. . Noche de Pájaros. Bogotá: Planeta Colombiana Editorial, S.A.
Alexander, Robert. . Communism in Latin America. New Brunswick, N.J.:
Rutgers University Press.
Alonso, Ana Maria. . Thread of Blood: Colonialism, Revolution, and Gender
on Mexico’s Northern Frontier. Tucson: University of Arizona Press.
Alvarez Gardeazábal, Gustavo. . Condores no entierran todos los días. Bogotá:
Plaza y Janes Editores Colombia.
Anderson, Benedict. . Imagined Communities: Reflections on the Origin and
Spread of Nationalism. London: Verso.
Andrade Valderrama, Luis. n.d. Cartas Pastorales. Medellín: n.p.
Appadurai, Arjun. . ‘‘Dead Certainty: Ethnic Violence in the Era of Global-
ization.’’ Public Culture  (): –.
 Bibliography

Appelbaum, Nancy. . ‘‘Remembering Riosucio: Region, Ethnicity, and

Community in Western Antioquia, –.’’ Ph.D. diss. University of
Wisconsin, Madison.
Arango, Mariano. . Café e industria, –. Bogotá: Carlos Valencia
Editores.
. . El café en Colombia, –: Producción, circulación y política.
Bogotá: Carlos Valencia Editores.
Archila, Mauricio. . ‘‘Paternalism and Social Relations in the Early Stages of
Colombian Industrialization.’’ Paper presented at the Latin American Labor
History Conference, Yale University,  April .
. . Cultura e identidad obrera: Colombia, –. Bogotá: .
Arocha, Jaime. . La Violencia en el Quindío: Determinantes ecológicos y
económicos del homicidio en un municipio caficultor. Bogotá: n.p.
Aronson, Cynthia, and Robin Kirk. . State of War: Political Violence and
Counterinsurgency in Colombia. New York: Human Rights Watch/Americas.
Balibar, Etienne, and Immanuel Wallerstein. . Race, Nation, Class: Ambigu-
ous Identities. New York: Verso.
Bardecke, Frank, Leslie Lopez, and the Watsonville, California, Human Rights
Committee. . Shadows of Tender Fury: The Letters and Communiqués of
Sub-Comandante Marcos and the Zapatista Army of National Liberation. New
York: Monthly Review.
Baretta, Silvio R. Duncan, and John Markoff. . ‘‘Civilization and Barbarism:
Cattle Frontiers in Latin America.’’ Comparative Studies in Society and History
 (): –.
Bayer, Tulio. . Carretera al mar. Medellín: n.p.
Bejarano, Jesús Antonio. . ‘‘La economía.’’ In Manual de Historia de Colom-
bia. d ed. Ed. Jaime Jaramillo Uribe. Bogotá: Procultura S.A., Instituto
Colombiano de Cultura.
. . Economía y Poder. Bogotá: Fondo Editorial .
. . Ensayos de historia agraria colombiana. Bogotá: .
Bell, P. L. . Colombia: A Commercial and Industrial Handbook. U.S. Depart-
ment of Commerce, Bureau of Foreign and Domestic Commerce. Special
Agent Series no. . Washington, D.C.: Government Printing Office.
Bergquist, Charles W. . Coffee and Conflict in Colombia, –. Durham,
N.C.: Duke University Press.
. . Labor in Latin America: Comparative Essays on Chile, Argentina,
Venezuela, and Colombia. Stanford: Stanford University Press.
Bergquist, Charles, Ricardo Peñaranda, and Gonzalo Sánchez, eds. . Violence
in Colombia: The Contemporary Crisis in Historical Perspective. Wilmington,
Del.: Scholarly Resources.
Berry, Albert R., and Ronald G. Hellman, ed. . Politics of Compromise:
Coalition Government in Colombia. New Brunswick, N.J.: Transaction Books.
 Bibliography 

Betancourt, Darío, and Martha L. García. . Matones y cuadrilleros: Origen y

evolución de la Violencia en el occidente colombiano, –. Bogotá: Tercer
Mundo Editores.
Bhabha, Homi K. . ‘‘The Other Question: Difference, Discrimination, and
the Discourse of Colonialism.’’ In Out There: Marginalization and Contempo-
rary Cultures. Eds. Russell Ferguson et al. Cambridge:  Press.
Borja, Miguel. . Estado, Sociedad y Ordenamiento Territorial en Colombia.
Bogotá: .
Botero Herrera, Fernando. . Urabá: Colonización, violencia, y crisis del
estado. Medellín: Editorial Universidad de Antioquia.
Bourdieu, Pierre. . ‘‘What Makes a Social Class? On the Theoretical and
Practical Existence of Groups.’’ Berkeley Journal of Sociology : –.
Braun, Herbert. . The Assassination of Gaitán: Public Life and Urban Violence
in Colombia. Madison: University of Wisconsin Press.
. . Our Guerrillas, Our Sidewalks: A Journey into the Violence of
Colombia. Niwot, Colo.: University Press of Colorado.
Brew, Roger. . ‘‘Aspects of Politics in Antioquia, –.’’ B.A. Phil.
Thesis. Oxford University, St. Anthony’s College.
. . El desarrollo económico de Antioquia desde la independencia hasta
. Bogotá: Banco de la República.
Buci-Glucksmann, Christine. . Gramsci and the State. London: Lawrence and
Wishart.
Builes, Miguel Angel. . Cartas Pastorales del Excelentisimo Señor Miguel
Angel Builes Obispo de Santa Rosa de Osos.  vols. Bogotá: Empresa Nacional
de Publicaciones.
Buitrago Salazar, Evelio. . Zarpazo the Bandit: Memoirs of an Undercover
Agent of the Colombian Army. Mobile: University of Alabama Press.
Bushnell, David. . Eduardo Santos and the Good Neighbor Policy, –.
Gainesville: University of Florida Press.
Bushnell, David, and Neill Macauley. . The Emergence of Latin America in
the Nineteenth Century. New York: Oxford University Press.
. . The Making of Modern Colombia: A Nation in Spite of Itself.
Berkeley: University of California Press.
Campo, Urbano. . Urbanización y violencia en el Valle. Bogotá: Ediciones
Armadillo.
Canal Ramírez, Gonzalo. . Estámpas y testimonios de violencia. Bogotá: n.p.
Cardona Grisales, Guillermo. . Para un estudio sobre la Violencia en Colom-
bia. Bogotá: .
Carmack, Robert M., ed. . Harvest of Violence: The Mayan Indian and the
Guatemalan Crisis. Norman: University of Oklahoma Press.
Casas Aguilar, Julio. . La Violencia en los Llanos Orientales. Bogotá: 
Ediciones.
 Bibliography

Casas, Ulises. . De la guerrilla liberal a la guerrilla comunista. Bogotá: n.p.

Cepeda, Manuel José. . Cómo son las elecciones en Colombia: Guía de votante.
Bogotá: , , .
Christie, Keith. . ‘‘Antioqueño Colonization in Western Colombia: A Re-
appraisal.’’ Hispanic American Historical Review  (May): –.
. . Oligarcas, campesinos, y política en Colombia. Bogotá: Universidad
Nacional de Colombia.
Colmenares, Germán. n.d. Partidos políticos y clases sociales en Colombia. Bogotá:
n.p.
Colombia en cifras. . Bogotá: El Mes Financiero y Económico.
Comaroff, Jean, and John Comaroff. . Ethnography and the Historical Imagi-
nation. Boulder: Westview Press.
Comisión de estudios sobre la violencia. . Colombia: Violencia y democracia,
Informe presentado al Ministerio de Gobierno. Bogotá: Universidad Nacional
de Colombia.
Conniff, Michael. . Latin American Populism in Comparative Perspective.
Albuquerque: University of New Mexico Press.
Cooper, Frederick, and Ann Laura Stoler. . ‘‘Introduction: Tensions of
Empire.’’ American Ethnologist : –.
Corrigan, Philip. . ‘‘State Formation.’’ In Everyday Forms of State Formation.
Eds. Gilbert Joseph and Daniel Nugent. Durham, N.C.: Duke University
Press.
da Cunha, Euclides. . Rebellion in the Backlands. Trans. Samuel Putnam.
Chicago: University of Chicago Press.
Davis, Natalie Zemon. . Society and Culture in Early Modern France. Stan-
ford: Stanford University Press.
Delgado, Carlos, ed. . El  de abril en fotos. Bogotá: El Ancora Editores.
Delpar, Helen. . Red against Blue: The Liberal Party in Colombian Politics,
–. Tuscaloosa: University of Alabama Press.
Derby, Robin L. H., and Richard Turits. . ‘‘Histories of Terror and the Ter-
rors of History: The  Haitian Massacre in the Dominican Republic.’’
Paper presented to the Committee on Latin American History at the annual
meeting of the American Historical Association, New York,  December
.
Diacon, Todd A. . Millenarian Vision, Capitalist Reality: Brazil’s Contestado
Rebellion. Durham, N.C.: Duke University Press.
Díaz Callejas, Apolinar. . El  de abríl  en Barrancabermeja: Diez días de
poder popular. Bogotá: El Labrador.
Dix, Robert H. . Colombia: The Political Dimensions of Change. New Haven:
Yale University Press.
Duque Betancourt, Francisco. . Historia del Departamento de Antioquia. d
ed. Medellín: Editorial Albon Interprint.
 Bibliography 

Echavarría, Guillermo. . Camilo C. Restrepo. Medellín: Semigráficos.

Echeverri Uruburu, Alvaro. . Elites y proceso político en Colombia, –.
Bogotá: .
Fajardo, Darío. . ‘‘La Violencia y las estructuras agrarias en tres muncipios
cafeteros de Tolima: –.’’ In El agro en el desarrollo histórico colom-
biano: Ensayos de economía. Ed. Francisco Leal Buitrago et al. Bogotá: Punta
de Lanza.
. . Violencia y desarrollo: Transformaciones sociales en tres regiones
cafetaleras del Tolima, –. Bogotá: Fondo Editorial Suramérica.
Fajardo, Luis H., n.d. The Protestant Ethic of the Antioqueños?: Social Structure
and Personality. Cali: Ediciones Departamento de Sociología, Universidad
del Valle.
Farnsworth-Alvear, Ann. . Dulcinea in the Factory: Myths, Morals, Men, and
Women in Colombia’s Industrial Experiment, –. Durham, N.C.: Duke
University Press.
Feldman, Allen. . Formations of Violence: The Narrative of the Body and
Political Terror in Northern Ireland. Chicago: University of Chicago Press.
. . ‘‘Violence and Vision: The Prosthetics and Aesthetics of Terror.’’
Public Culture  (): –.
Fidelis, Testis [pseud.]. . De Cain a Pilatos (Prologue by Juan Roca Lemus
[Rubayata] and an endorsement by the Bishop of Santa Rosa de Osos, Miguel
Angel Builes). Medellín: n.p.
Flórez-Malagón, Alberto Guillermo. . ‘‘Peasants, Landlords, and Political
Culture in Colombia: The Ubaté Valley during ‘La Violencia,’ –.’’
Ph.D. diss. State University of New York at Stony Brook.
Fluharty, Vernon Lee. . Dance of the Millions: Military Rule and the Social
Revolution in Colombia, –. Pittsburgh: University of Pittsburgh Press.
French, William E. . ‘‘Progreso Forzado: Workers and the Inculcation of
the Capitalist Work Ethic in the Parral Mining District.’’ In Rituals of Rule,
Rituals of Resistance. Eds. William Beezeley, Cheryl English Martin, and
William E. French. Wilmington, Del.: Scholarly Resources.
Franco Isaza, Eduardo. . Las guerrillas del Llano. d ed. Medellín: Ediciones
Hombre Nuevo.
Gaitán de Valencia, Gloria. . Colombia: la lucha por la tierra en la década del
treinta. Bogotá: n.p.
Galvis, Silvia, and Alberto Donadio. . Colombia Nazi –: Espionaje
alemán, la cacería del FBI, Santos, López y los pactos secretos. Bogotá: Planeta
Colombiana Editorial.
. . El Jefe Supremo: Rojas Pinilla en la violencia y en el poder. Bogotá:
Planeta.
García, Antonio. . Geografía económica de Caldas. d ed. Bogotá: Banco de
la República.
 Bibliography

García Márquez, Gabriel. . La mala hora. Bogotá: Editorial La Oveja Negra.
. . News of a Kidnapping. New York: Penguin Press.
Gilhodes, Pierre. . ‘‘Agrarian Struggles in Colombia.’’ In Agrarian Problems
and Peasant Movements in Latin America. Ed. Rodolfo Stavenhagen. Garden
City, N.Y.: Doubleday.
. . La Question agraire en Colombie, –. Paris: n.p.
Gilman, Sander L. . Difference and Pathology: Stereotypes of Sexuality, Race,
and Madness. Ithaca: Cornell University Press.
Gómez, Laureano. n.d. El basilisco en acción. Bogotá: n.p.
Gómez Martínez, Fernando. n.d. Mordaza, diario secreto de un escritor público.
d ed. Medellín: Edición de El Colombiano.
González Arias, José Jairo, and Elsy Marulanda Alvarez. . Historias de
frontera: Colonización y guerras en el Sumapaz. Bogotá: .
Gónzalez Restrepo, Jorge. . Por los caminos de la muerte: Un recuento de
La Violencia en Salgar. MS, Fundación Antioqueña de Estudios Sociales,
Medellín.
Gracian, Luis. n.d. Laureano Gómez, un hombre, un partido, una nación. Bogotá:
Ediciones Nuevo Mundo.
Gramsci, Antonio. . Selections from the Prison Notebooks of Antonio Gramsci.
Eds. Quintin Hoare and Geoffrey Nowell Smith. New York: International
Publishers.
Granados, Wilson. . ‘‘La Violencia en Urrao, Antioquia, –.’’ Tesis de
grado. Universidad de Antioquia.
Grandin, Greg. . The Blood of Guatemala: A History of Race and Nation.
Durham, N.C.: Duke University Press.
. . ‘‘Insoluble Acts and Historical Solutions: Law, History, and Latin
American Cold War Terror.’’ Unpublished manuscript.
Green, John W. . ‘‘ ‘Vibrations of the Collective’: The Popular Ideology of
Gaitanismo on Colombia’s Atlantic Coast, –.’’ Hispanic American
Historical Review  (): –.
Guhl, Ernesto. . Colombia: Bosquejo de su geografía tropical.  vols. Bogotá:
Biblioteca Básica Colombiana.
Gutiérrez de Pineda, Virginia. . Familia y Cultura en Colombia. Bogotá: In-
stituto Colombiano de Cultura, Departamento Administrativo Nacional de
Estadística.
Guzmán Campos, Germán, Orlando Fals Borda, and Eduardo Umaña. . La
Violencia en Colombia.  vols. th ed. Bogotá: Carlos Valencia Editores.
Hagen, Everett E. . On the Theory of Social Change: How Economic Growth
Begins. Homewood, Ill.: Dorsey Press.
Hartlyn, Jonathan. . ‘‘Consociational Politics in Colombia: Confrontation
and Accomodation in Comparative Perspective.’’ Ph.D. diss. Yale University.
. . ‘‘Producer Associations, the Political Regime, and Policy Processes
in Contemporary Colombia.’’ Latin American Research Review  (): –.
 Bibliography 

. . The Politics of Coalition Rule in Colombia. New York: Cambridge

University Press.
Harvey, Neil. . The Chiapas Rebellion: The Struggle for Land and Democracy.
Durham, N.C.: Duke University Press.
Held, David. ed. . States and Societies. New York: New York University Press.
Helg, Aline. . La educación en Colombia, –: Una historia social,
económica y política. d ed. Bogotá: Fondo Editorial .
Henderson, James. . Las ideas de Laureano Gómez. Bogotá: Tercer Mundo.
. . When Colombia Bled: A History of the ‘‘Violencia’’ in Tolima.
Tuscaloosa: University of Alabama Press.
Hennessy, Alistair. . The Frontier in Latin American History. Albuquerque:
University of New Mexico Press.
Herrera, Ernesto León [Fidel Blandón Berrío]. . Lo que el cielo no perdona.
Bogotá: Editorial .
Herrera Soto, Roberto. . Antología del pensamiento conservador en Colombia.
 vols. Bogotá: n.p.
Hobsbawm, Eric J. . Primitive Rebels: Studies in Archaic Forms of Social
Movement in the Nineteenth and Twentieth Centuries. New York: W. W.
Norton.
. . Bandits. Rev. ed. New York: Pantheon Books.
Human Rights Watch/Americas–Arms Project. . Colombia’s Killer Net-
works: The Military-Paramilitary Partnership and the United States. New York:
Human Rights Watch/Americas.
Jaramillo, Roberto Luis. . ‘‘La otra cara de la colonización antioqueña hacia
el sur.’’ Revista de Extensión Cultural  (December).
Jaramillo Uribe, Jaime. . El pensamiento colombiano en el siglo xix. Bogotá:
Editorial Temis.
. ed. . Manual de historia de Colombia.  vols. d ed. Bogotá: Pro-
cultura S.A., Instituto Colombiano de Cultura.
Jiménez, Michael. . ‘‘The Limits of Export Capitalism: Economic Structure,
Class, and Politics in a Colombian Coffee Municipality, –.’’ Ph.D.
diss., Harvard University.
Joseph, Gilbert M. . ‘‘On the Trail of Latin American Bandits: A Reex-
amination of Peasant Resistance.’’ Latin American Research Review  ():
–.
Joseph, Gilbert M., and Daniel Nugent, eds. . Everyday Forms of State For-
mation: Revolution and the Negotiation of Rule in Modern Mexico. Durham,
N.C.: Duke University Press.
Kalmanowitz, Salomón. . Desarrollo de la agricultura en Colombia. Bogotá:
n.p.
. . Economía y nación: Una breve historia de Colombia. Bogotá: Siglo
Veintiuno Editores, .
 Bibliography

Knight, Alan. . ‘‘The Mexican Revolution: Bourgeois? Nationalist? or Just a

‘Great Rebellion’?’’ Bulletin of Latin American Research  (): –.
Koonings, Kee, and Dirk Kruijt, eds. . Societies of Fear: The Legacy of Civil
War, Violence, and Terror in Latin America. New York: St. Martin’s Press.
Latorre, Emilio. . Transporte y crecimiento regional en Colombia. Bogotá:
Editorial Presencia.
Leal Buitrago, Francisco. . Estado y política en Colombia. Bogotá: Siglo
Veintiuno Editores de Colombia, .
LeGrand, Catherine. . Frontier Expansion and Peasant Protest in Colombia,
–. Albuquerque: University of New Mexico Press.
. . ‘‘La política y la Violencia en Colombia, –: Interpreta-
ciones en la década de los ochenta.’’ Memoria y Sociedad  (): –.
Lewin, Linda. . ‘‘The Oligarchical Limitations of Social Banditry in Brazil:
The Case of the ‘Good’ Thief Antonio Silvino.’’ Past and Present : –.
Lievano Aguirre, Indalecio. . Los grandes conflictos sociales y economicos de
nuestra historia.  vols. th ed. Bogotá: Tercer Mundo.
López, Alejandro. . Idearium liberal. Paris: n.p.
. . Escritos escogidos. Bogotá: Biblioteca Básica Colombiana.
López B., Manuel Enrique. . ‘‘Implicaciones del conflicto armado en el mo-
delo de economiía cafetera. Aproximación al caso del suroeste antioqueño.’’
In Conflictos Regionales: La Crisis del Eje Cafetero, ed. Gonzalo Sánchez G.
et al. Bogota: , .
López C., Hugo. . ‘‘La inflación en Colombia en la década de los veintes.’’
Cuadernos Colombianos .
López de Mesa, Luis. . De cómo se ha formado la nación colombiana. d ed.
Medellín: n.p.
López Giraldo, Fermín. . El apostol desnudo: O dos años al lado de un mito.
Manizales: Editorial Zapata.
López Toro, Alvaro. . Migración y cambio social en Antioquia. Medellín:
Ediciones Hombre Nuevo.
Loveman, Brian, and Thomas M. Davies Jr., eds. . The Politics of Antipoli-
tics: The Military in Latin America. Rev. ed. Wilmington, Del.: Scholarly
Resources Books.
Lynch, John. . Argentine Dictator: Juan Manuel Rosas, –. New York:
Oxford University Press.
Machado, Absalón. . El café: De la aparcería al capitalismo. Bogotá: Tercer
Mundo Editores.
Malkki, Liisa H. . Purity and Exile, Violence, Memory, and National Cos-
mology among Hutu Refugees in Tanzania. Chicago: University of Chicago
Press.
Mallon, Florencia E. . Peasant and Nation: The Making of Postcolonial Mexico
and Peru. Berkeley: University of California Press.
 Bibliography 

Marcucci, Olindo. n.d. La revolución agraria en Colombia: Partido social agrario

nacionalista. Bogotá: Editores Minerva.
Martz, John. . Colombia: A Contemporary Political Survey. Chapel Hill:
University of North Carolina Press.
Maullin, Richard. . Soldiers, Guerillas, and Politics in Colombia. Lexington,
Mass.: D. C. Heath.
Mayor Mora, Alberto. . Etica, trabajo y productividad. d ed. Medellín:
Ediciones Tercer Mundo.
McGreevey, William Paul. . An Economic History of Colombia, –.
Cambridge: Cambridge University Press.
Medina, Medófilo. . Historia del Partido Comunista de Colombia. Vol. .
Bogotá: n.p.
. . La protesta urbana en Colombia en el siglo veinte. Bogotá: Edi-
ciones Aurora.
Mejía Robledo, Alfonso, ed. . Vidas y empresas de Antioquia: Diccionario
biográfico, bibliográfico y económico. Medellín: Imprenta Departamental de
Antioquia.
Mejía V., Gilberto. n.d. El comunismo en Antioquia (María Cano): Memorias.
Medellín: Ediciones Pepe.
Mejía Vallejo, Manuel. . El día señalado. Barcelona: Destino.
Melo, Jorge Orlando. . ‘‘La política de  a .’’ In Historia de Antioquia.
Ed. Jorge Orlando Melo. Bogotá: Editorial Presencia y Suramericana de
Seguros.
Molano, Alfredo. . Los años del tropel: Relatos de la Violencia. Bogotá:
Editorial Presencia.
Molina, Gerardo. . Las ideas liberales en Colombia, –. Bogotá: n.p.
. . Las ideas liberales en Colombia, –. Bogotá: n.p.
Monografías de Antioquia. . Medellín: n.p.
Mosse, George L. . Nationalism and Sexuality: Middle-Class Morality and
Sexual Norms in Modern Europe. Madison: University of Wisconsin Press.
. . The Nationalization of the Masses: Political Symbolism and Mass
Movements in Germany from the Napoleonic Wars through the Third Reich.
Ithaca: Cornell University Press [].
Murray, Pamela S. . ‘‘Forging a Technocratic Elite in Colombia: A History
of the Escuela Nacional de Minas of Medellín, –.’’ Ph.D. diss. Tulane
University.
Nixon, Rob. . ‘‘Of Balkans and Bantustans: ‘Ethnic Cleansing’ and the Crisis
in National Legitimation.’’ Transition : –.
Ocampo, Gloria Isabel. –. ‘‘Hacienda, parentesco y mentalidad: La colo-
nización antioqueña en el Sinu.’’ Revista Colombiana de Antropología :
–.
. . ‘‘Entre la localidad y la Nación: Aspectos políticos de la con-
 Bibliography

strucción de identidades en un contexto regional.’’ Revista Colombiana de

Antropología : –.
Ocampo, José Antonio. . Colombia y la economía mundial, –.
Mexico, D.F.: Siglo Veintiuno Editores.
. . ‘‘Los orígenes de la industria cafetera, –.’’ In Nueva
historia de Colombia. Vol. . Ed. Jaime Jaramillo Uribe. Bogotá: Planeta.
Ocampo, José Antonio, and Santiago Montenegro, eds. . Crisis mundial,
protección e industrialización. Bogotá: Fondo Editorial .
Once Ensayos Sobre la Violencia. . Bogotá: Fondo Editorial  and
Centrogaitán.
Oquist, Paul. . Violence, Conflict, and Politics in Colombia. New York: Aca-
demic Press.
Ortiz Mesa, Luis Javier. . ‘‘Antioquia bajo la regeneración, –.’’ In
Historia de Antioquia. Ed. Jorge Orlando Melo. Bogotá: Editorial Presencia.
Ortiz Sarmiento, Carlos Miguel. . ‘‘Fundadores y negociantes en la coloniza-
ción del Quindío.’’ Lecturas de Economía  (January–April): –.
. . Estado y subversión en Colombia: La Violencia en el Quindío, años
. Bogotá: Fondo Editorial .
. . ‘‘La Violencia y los negocios: Quindió años  y .’’ In Pasado
y presente de la Violencia en Colombia. Ed. Gonzalo Sánchez and Ricardo
Peñaranda. Bogotá: Fondo Editorial .
. –. ‘‘Violencia política de los ochenta: Elementos para una re-
flexión historica.’’ Anuario Colombiano de Historia Social y de la Cultura
–: –.
Osorio, Iván Darío. n.d. Historia del sindicalismo antioqueño, –.
Medellín: Tipografía y Litografía Sigifredo.
Ospina, Tulio. . Protocolo Hispanoamericano de la urbanidad y el buen tono.
d ed. Medellín: Félix de Bedout e Hijos.
Ospina Vásquez, Luis. . Industria y protección en Colombia, –. d ed.
Medellín: , Editorial Lealon.
Palacios, Marco. . El café en Colombia, –: Una historia económica,
social y política. d ed. Bogotá: Siglo Veintiuno Editores.
. . Entre la legitimidad y la violencia: Colombia, –. Bogotá:
Norma.
Pardo Pardo, Alberto. . Geografía económica y humana de Colombia. d ed.
Bogotá: Ediciones Tercer Mundo.
Park, James William. . Rafael Núñez and the Politics of Colombian Regional-
ism, –. Baton Rouge: Louisiana State University Press.
Parker, Andrew, et al. . Nationalisms and Sexualities. London: Routledge.
Parsons, James. . Antioqueño Colonization in Western Colombia. d ed.
Berkeley: University of California Press.
. . La colonización antioqueña en el occidente de Colombia. d ed.
Bogotá: Carlos Valencia Editores.
 Bibliography 

. n.d. Urabá, salida de Antioquia al mar. Medellín: Corpourabá.

Partido Comunista de Colombia. n.d. Treinta años de lucha del Partido Comu-
nista de Colombia. Bogotá: n.p.
Payne, Alexander. . ‘‘Crecimiento y cambio social en Medellín, –.’’
Estudios Sociales  ().
Payne, James L. . Patterns of Conflict in Colombia. New Haven: Yale Univer-
sity Press.
Pécaut, Daniel. . Política y sindicalismo en Colombia. d ed. Bogotá: Edi-
ciones Culturales.
. . ‘‘Acerca de la Violencia de los años cincuenta.’’ Boletín Socioeconó-
mico .
. . Orden y violencia: Colombia, –.  vols. Bogotá: Siglo
Veintiuno Editores.
Pineda Giraldo, Roberto. . El impacto de la violencia en el tolima, el caso del
Líbano. Bogotá: Universidad Nacional de Colombia.
Pinzón, Martin Alonso. . Historia del conservatismo. d ed. Bogotá: Tercer
Mundo.
Posada, Francisco. . Colombia: Violencia y subdesarrollo. Bogotá: n.p.
Ramsey, Russell W. . ‘‘Critical Bibliography on La Violencia in Colombia.’’
Latin American Research Review  (): –.
. . Guerrilleros y soldados. Bogotá: Editorial Tercer Mundo.
Reed, Nelson. . The Caste War of the Yucatan. Stanford: Stanford University
Press.
Restrepo Jaramillo, Gonzalo. . El pensamiento conservador. Medellín: Tipo-
grafía Bedout.
Rippy, J. Fred. . The Capitalists and Colombia. New York: n.p.
Rock, David, ed. . Latin America in the s: War and Post-War Transitions.
Berkeley: University of California Press.
Rodríguez, Jaime O., ed. . The Revolutionary Process in Mexico: Essays on
Political and Social Change, –. Los Angeles: Latin American Center,
.
Roldán, Mary. . ‘‘La política en Antioquia, –.’’ In Historia de Antio-
quia. Ed. Jorge Orlando Melo. Bogotá: Editorial Presencia.
. . ‘‘Guerrillas, contrachusmas y caudillos: La Violencia en Antioquia,
–.’’ Estudios Sociales  (March): –.
. . ‘‘Genesis and Evolution of La Violencia in Antioquia, Colombia,
–.’’ Ph.D. diss. Harvard University.
. . ‘‘Violencia, colonización y la geografía de la diferencia en Colom-
bia.’’ Análisis Político  (Sept.–Dec.): –.
Romero Aguirre, Alfonso. . El partido conservador ante la historia. Bogotá:
Librería America.
Romero, Mauricio. . ‘‘Tierra y violencia en Córdoba.’’ Revista Análisis :
–.
 Bibliography

. . Rural Transformation and Political Violence in Colombia, –:

A Comparative Study of Intra-Regional Change in a Non-Exporting Zone. Cen-
ter for Studies in Social Change, New School for Social Research, Working
Paper no. .
Ruddick, Sara. . ‘‘Toward a Feminist Peace Politics.’’ In Gendering War Talk.
Ed. Miriam Cooke and Angela Woollacott. Princeton: Princeton University
Press.
Saénz Rovner, Eduardo. . La ofensiva empresarial: Industriales, políticos y
violencia en los años cuarenta en Colombia. Bogotá: Tercer Mundo.
Safford, Frank. . ‘‘Significación de los antiqueños en el desarrollo econó-
mico colombiano: Un examen crítico de la tesis de Everett Hagen.’’ Anuario
Colombiano de Historia Social y de la Cultura, pp. –.
. . The Ideal of the Practical: Colombia’s Struggle to Form a Technical
Elite. Austin: University of Texas Press.
. . ‘‘Race, Integration and Progress: Elite Attitudes and the Indian in
Colombia, –.’’ Hispanic American Historical Review  (): –.
Sahlins, Peter. . Boundaries: The Making of France and Spain in the Pyrenees.
Berkeley: University of California Press.
Salazar, Alonso. . Born to Die in Medellín. London: Verso.
Sánchez, Gonzalo. . Los ‘‘Bolsheviques del Líbano’’ (Tolima). Bogotá: n.p.
. . ‘‘La Violencia y sus efectos en el sistema político colombiano.’’
Cuadernos Colombianos  (Jan.–April): –.
. . Las ligas campesinas en Colombia. Bogotá: Editorial Tiempo
Presente.
. . Los días de la revolucíon: Gaitanismo y  de abril en provincia.
Bogotá: Centro Gaitán.
. . ‘‘La Violencia in Colombia: New Research, New Questions.’’
Hispanic American Historical Review  (): –.
. . ‘‘Los estudios sobre la Violencia: Balance y perspectivas.’’ In
Pasado y presente de la Violencia en Colombia. Ed. Gonzalo Sánchez and
Ricardo Peñaranda. Bogotá: Fondo Editorial .
Sánchez, Gonzalo, and Donny Meertens. . Bandoleros, gamonales y campesi-
nos: El caso de la Violencia en Colombia. Bogotá: El Ancora.
Sanín Echeverri, Jaime. . Ospina supo esperar. Medellín: Editorial Andes.
Sarmiento, Domingo. . Facundo, or, Civilization and Barbarism. New York:
Penguin.
Scarry, Elaine. . The Body in Pain: The Making and Unmaking of the World.
New York: Oxford University Press.
Schirmer, Jennifer. . The Guatemalan Military Project: A Violence Called
Democracy. Philadelphia: University of Pennsylvania Press.
Schmidt, Steffen W. –. ‘‘La Violencia Revisited: The Clientelist Bases
of Political Violence in Colombia.’’ Journal of Latin American Studies –
(May): –.
 Bibliography 

. . ‘‘The Transformation of Clientelism in Rural Colombia.’’ In

Friends, Followers, and Factions. Ed. Steffen W. Schmidt et al. Berkeley:
University of California Press.
Scott, James C. . Domination and the Art of Resistance: Hidden Transcripts.
New Haven: Yale University Press.
Sharpless, Richard E. . Gaitán of Colombia: A Political Biography. Pittsburgh:
University of Pittsburgh Press.
Silva, Renán. ed. . Territorios, Regiones, Sociedades. Bogotá: .
Slatta, Richard W. . Gauchos and the Vanishing Frontier. Lincoln: University
of Nebraska Press.
. . Bandidos: The Varieties of Latin American Banditry. New York:
Greenwood Press.
Socarras, José Francisco. . Laureano Gómez: Psicoanalisis de un resentido.
Bogotá: Ediciones Librería Siglo XX.
Sommer, Doris. . Foundational Fictions: The National Romances of Latin
America. Berkeley: University of California Press.
Stallybrass, Peter, and Allon White. . The Politics and Poetics of Transgres-
sion. Ithaca: Cornell University Press.
Starn, Orin, Carlos Iván DeGregori, and Robin Kirk. . The Peru Reader:
History, Culture, Politics. Durham, N.C.: Duke University Press.
Stern, Steve J., ed. . Shining and Other Paths: War and Society in Peru,
–. Durham, N.C.: Duke University Press.
Stoler, Ann Laura. . ‘‘Making Empire Respectable: The Politics of Race
and Sexual Morality in Twentieth-Century Colonial Cultures.’’ American
Ethnologist : –.
. . ‘‘Sexual Affronts and Racial Frontiers: European Identities and
the Cultural Politics of Exclusion in Colonial Southeast Asia.’’ Comparative
Studies in Society and History  (): –.
Stoller, Richard. . ‘‘Alfonso López Pumarejo and Liberal Radicalism in s
Colombia.’’ Journal of Latin American Studies  (): –.
Taussig, Michael. . ‘‘Terror as Usual: Walter Benjamin’s Theory of History as
a State of Siege.’’ Social Text  (fall/winter): –.
Taylor, William. . Drinking, Homicide, and Resistance in Colonial Mexican
Villages. Stanford: Stanford University Press.
Thomas, Nicholas. . Colonialism’s Culture, Anthropology, Travel, and Govern-
ment. Princeton: Princeton University Press.
Tirado Mejía, Alvaro. . Aspectos sociales de las guerras civiles en Colombia.
Instituto Colombiano de la Cultura. Bogotá: Editorial Andes.
. . Aspectos políticos del primer gobierno de Alfonso López Pumarejo,
–. Instituto Colombiano de la Cultura. Bogotá: Grafíca Cabrera
e Hijos.
. . Descentralización y centralismo en Colombia. Bogotá: Editorial
 Bibliography

Oveja Negra Serie Monográfica de Ciencias Sociales y Política de la Funda-

ción Friedrich Naumann.
Torres Giraldo, Ignacio. . La huelga general en Medellín. Medellín: Editores
Bedout.
. n.d. Los inconformes. Bogotá: Cooperativa Editores.
Tovar Pinzón, Hermes. . El movimiento campesino en Colombia durante los
siglos xix y xx. Bogotá: n.p.
Uribe Angel, Manuel. . Geografía general del estado de Antioquia. Paris: n.p.
Uribe de H., Maria Teresa. . Urabá, Región o Territorio? Un análisis en el
contexto de la política, la historia y la etnicidad. Medellín: n.p.
Uribe de Hincapié, María Teresa, and Jesús María Alvarez. . Poderes y re-
giones: Problemas en la constitución de la nación colombiana, –.
Medellín: Centro de Investigaciones de las Ciencias Sociales, Universidad de
Antioquia.
Uribe, Maria Victoria. . Matar, rematar y contramatar: Las masacres de la
Violencia en el Tolima, –. Bogotá: .
Urrutia, Miguel. . The Development of the Colombian Labor Movement. New
Haven: Yale University Press.
. . Historia del sindicalismo en Colombia. d ed. Medellín: Editorial
Lealon.
Vanderwood, Paul J. . Disorder and Progress: Bandits, Police, and Mexican
Development. Wilmington, Del.: Scholarly Resources Books.
Vargas Velásquez, Alejo. . Colonización y Conflicto Armado. Bogotá: .
Varones ilustres de Antioquia: Biografías de los academicos de números fallecidos.
. Medellín: Editorial Universo.
Vega Cantor, Renan. . Crisis y caída de la República Liberal, –.
Bogotá: Editorial Mohan.
Wade, Peter. . ‘‘Race and Class: The Case of South American Blacks.’’ Ethnic
and Racial Studies  (): –.
. . Blackness and Race Mixture: The Dynamics of Racial Identity in
Colombia. Baltimore: Johns Hopkins University Press.
Waller, Altina L. . Feud: Hatfields, McCoys, and Social Change in Appalachia,
–. Chapel Hill: University of North Carolina Press.
Walton, John. . Reluctant Rebels: Comparative Studies of Revolution and
Underdevelopment. New York: Columbia University Press.
Weinert, Richard. . ‘‘Violence in Pre-Modern Societies: Rural Colombia.’’
American Political Science Review  (June): –.
West, Robert C. . The Pacific Lowlands of Colombia: A Negroid Area in the
American Tropics. Baton Rouge: Louisiana State University Press.
Wickham-Crowley, Timothy P. . Guerrillas and Revolution in Latin America:
A Comparative Study of Insurgents and Regimes since . Princeton: Prince-
ton University Press.
 Bibliography 

Williams, Raymond. . Keywords: A Vocabulary of Culture and Society.

Oxford: Oxford University Press.
Williamson, Robert. . ‘‘Toward a Theory of Political Violence: The Case of
Rural Colombia.’’ Western Political Quarterly  (March).
Womack, John Jr. . Zapata and the Mexican Revolution. New York: Vintage.
, ed. . Rebellion in Chiapas: A Historical Reader. New York: New
Press.
Zambrano Pantója, Fabio. . Colombia, País de Regiones. Bogotá: ,
Colciencias.
Zamosc, Leon. . The Agrarian Question and the Peasant Movement in Colom-
bia, –. New York: Cambridge University Press.
Zapata Restrepo, Miguel. . La mitra azul, Miguel Angel Builes: el hombre, el
obispo, el caudillo. Medellín: Editorial Beta.
. . El obispo tropezó  veces. Medellín: Editorial Bedout.
Index

Abadía, Luis (colonel), – , , ; and patriarchal family
Abriaquí, –, ; and Conservative model, ; and regional discourse,
refugees,  –, ; religious differences in,
Acción Católica, , ,  ; and situation of workers, , ;
Alejandría,  southwest of, –, , , ;
Altamira, , ,  and state employment, ; violence
Amalfi, –, , , –, –, in, , , 
,  Apartadó, , 
Anderson, Benedict,  Aplanchamientos: of Liberals, , 
Andes, , , , , , – Arango Ferrer, Dionisio, , , –
Andrade, Luis, – , , , , –, , –,
Andrade Valderrama, Luis (Bishop of –, –, 
Santa Fé de Antioquia), –,  Arboletes, 
Antioqueñidad, , –, ,  Armed groups, , , –, –,
Antioqueño(s): and Catholic Church, ; , , –. See also Bandits;
description of, –, , , , , Contrachusmas; Guerrillas
, , , –, –, –, Army/armed forces. See Military
; and hacendados, ; and López Arocha, Jaime, 
Pumarejo, ; and partisan politics, Arrivistes: definition of, ; and local
, , , , ; and period between affairs, ; and personal gain, 
–, ; and relationship to Atrato River, 
state and parties, , , , , 
Antioquia: case of, –, –, , Bajo Cauca, , , , , –, –
, ; coffee sector in, , , , – , , –, , , –, ,
, ; and control of Urabá, , , , 
; deaths in, , , , , , , Bandits, , , , –, , ,
; description of population in, , , –, –, –, –,
; and ‘‘disorder,’’ , ; in the s –, , , , , , –,
and s, ; industrial growth in, ; 
and inhabitants in the periphery, – Bandoleros. See Bandits
, , ; and its elites, –, –, Barbacoas, , 
–, , , , , , , –; Barrancabermeja, Santander, 
as a locus of violence, ; migrations Bergquist, Charles, , ,  n.
as a result of violence in, , –; and Bernal, José María, –, –, , ,
partisan violence, , , , , , , , , 
 Index

Berrío González, Eduardo, , , , , El Carmelo, , , 
–, , , –, , , –, Carretera al Mar, –, –,
, –, – –, , 
Betania, ,  Cartel: of Medellín, 
Betulia, , , , –, –, Casualties: and bandoleros, , ; and
–, – civilians, –, ; of civil wars, ;
Bipartisanship, , , . See also impact of, in Antioquia, , ; in 
Convivialismo and , , ; percentage of, ,
Blacks,  , , ; statistics of, , , ,
Bogotá, , –, , , , , , ; violence related, –, , ,
, , –, , –; and , , , –, 
exclusive elite, ; and minister of Catholic Church, , , ; Catholicism,
government, , , ,  ; and education, ; and marriage,
Bogotazo,  ; and militancy, ; and trade
Bolívar (department), , –, –, unions, 
–, –, –, , ; and Cattle, , –, , –, , ,
black Liberal soldiers, – ; expropriation of, ; and hacien-
Bolívar (town), , , , , , , das, , , , ; rustling of, ,
, –, , , ,  , , , , –, , –,
Bolombolo, , , , –,  , –, –, –, –,
Boyacá (department), , , , ; –
and indigenous population,  Caucasia, , , , , , –, –
Braun, Herbert, –,  n. , , –, –, 
Buriticá, , –, – Centrally based municipios/municipali-
Bushnell, David,  ties. See Core municipalities
Central state, , , , , , , –
Cáceres, –, , –, , –, , , , –, , , , ,
 , , , , –, , ,
Caicedo, , , , –, –, , –, , , , –,
, , –, –,  , , –, –, –;
Caja de Crédito Agrario y Minero and Antioquía, , , , , ;
(Agrarian and Mining Credit in Antioquia at mid-century, ,
Agency), , –, , , , , , , ; and authority out-
, –,  side of Bogotá, , ; as a ‘‘claim,’’
Caldas (department), ,  , , ; conflict between regional
Camparusia (guerrilla camp), , , state and, , , , , , , ;
, – and employment, , , , ; and
Cañasgordas, , , , –, , expansion of authority, , , ;
–, –; deaths in, , ,  formation of, ; and hegemony, ;
Cantinas, –, ,  and Liberal party’s rise to power, ;
Caracolí, –, –,  and López Pumarejo, , , ; mean-
Caramanta, ,  ing of, ; and ‘‘monopoly of force,’’
 Index 

, , , , ; in s, ; during la Violencia, ; pre-Violencia,
power of, , ; and regions (not –; southwestern frontier, ; and
Antioquia), ; theoretical framework urbanization, ; violence and, , ,
of, ; and violence, , –; weak 
presence of, , ,  Colombian Petroleum Workers Union,
Chavarriaga, Manuel,  
Chigorodó, , –, –, , , Colonos, , , , , , , 
– Colorado, 
Chocó (department), , –, – Communist party, , , , ; and
, –, , , –, –, leaders, , , , ; as a left wing
–, , , , ; and Afro- of Liberal party, , , ; and the
Colombians, ; Carmen del Atrato, Llanos, ; members of, , –;
; Quibdó,  and spread of Communism, –,
Chusmeros/chusma (rabble), –, , 
, , , , , . See also Concordia, , –, 
Bandits; Contrachusmas Confederation of Colombian Workers
Cisneros, ,  (), , , –; Conservative
Citizenship, , , , , –, attacks on, –; crushing of, ;
– harassment of, –; and Liberal
Civil wars: la Violencia as a civil war, , brokers, ; and militant unions, ;
, ; nineteenth-century, , ; violence against, 
War of the Thousand Days, , , Conservative party, , , , , –,
. See also La Violencia , , , –, , –, , ,
Clientelism, – , , , , –, –, ,
Cocaine, ,  , , , 
Cocorná,  Conservatives: and Antioquia, , ;
Coffee: for export, , , , ; and and Communism, –, , –,
frontier in Antioquía, , , , ; , , , ,  n.; and Conser-
and haciendas, –, , ; and vatization of Liberal municipalities,
harvest of, , –, ; heartland , , , –, ; and factions,
of, , ; and land tenure patterns, ; as a forastero (outsider), , ;
; producers and merchants, , , as godos, , ; for moderation,
, ; producing municipalities, , –, , –, –, , –
, , , , ; and violence, , , , , , , ; after ,
–, –, ,  , , , ; as observers, ;
Colombia, xi, , , , , , , , and policemen, , , ; and
, , , –, –; areas Rafael Reyes, ; and relationship
untouched by la Violencia, ; and with engineers, , , . See also
change in suffrage law, , –; Conservative party
compared to Latin America, –, ; Contrachusmas, –, –, –, –
conflict in, , ; and Constitution , –, , , , , –,
of , ; contemporary crisis, , ; –, –, , , –, –
 Index

Contrachusmas (continued ) Elías Calle, Ramón (guerrilla leader),

, –, –, , , –; , 
and attitude toward, – El Mister (guerrilla leader), , 
Convivialismo, , , , ,  Escobar, Aparicio (guerrilla leader), 
Core municipalities, , –, , , ,
, ; and Conservative party, , Fajardo, Darío, , 
; and definition of, ; and people Federation of Antioqueño workers
of peripheral areas, ; and violence, (), ; and state repression, 
,  Federation of Coffee Growers
Corneta (Capítan or Captain), –, (), , , ; and dan-
 ger to coffee harvest, –; and
Correa, José,  influence of, , ; and partisan
Corrigan, Philip, – conflict, ; and policy to impede
Corruption,  violence, –
Cosmopolita. See Non-Antioqueños Franco, Juan Jesús (Liberal guerrilla),
Costeño (Caribbean Coastal), –, , –, , , , , ; as Cap-
,  tain Franco, , , , , ;
Cundinamarca (department), , ; and distribution of land, ; and his
and indigenous settlement,  armed group, , , , , ,
Customs Revenue Board, ; and offi- , , 
cers/agents, –, , , , , Fredonia, , , , , , 
–, –, , , , , ; Frente Nacional, , , ; con-
and smuggling, ,  flict between, –, ; and Pax
Conservadora (–), 
Dabeiba, , , , , , –, , Frontino, , –, –, , –
–, , –, –, ; as , –, –, , , ; as a
a buffer zone, ; deaths in, –, , buffer zone, 
, ,  Frontino Gold Mines Union, , ; and
Damaquiel, ,  company, , , 
Dance of the Millions, 
David, Sergio (guerilla leader),  Gaitán, Jorge Eliécer: assassination
Deaths. See Casualties of, , , , , , –, , ,
Decree ,  –, –, , , –, –
Displacements (human, forcible), , , , , ; as a dissident, ; and
, , , – gaitainismo, ; and la Violencia, ,
Drugs, ; and Pablo Escobar, ; and ; as a Liberal populist, , ; in the
trafficking,  s, , –, , , , ; as a
petit bourgeois, ; and revolution, 
Ebéjico, ,  Gaitanistas, , –, , –;
Echandía Darío, ,  gaitanista leaders, xi, ; and munici-
Electoral violence, –, –, –, pal councils, , , , , , ; as
, –, , , ,  workers, , 
 Index 

Gamonales. See Political brokers , , , , , , , ,
García, Trino (guerrilla leader),  –, –, , ; and request
Geographically peripheral areas. See for troops, , 
Periphery History: of Antioquía, ; of Colombia,
Gold, , , , , ,  ; and Gaitán, 
Gómez, Laureano (Conservative party
leader/ president), , –, –, Identity: alternative conception of, ;
, , , –, –, , , and national sovereignty, , ; of
, , , , , , , , the region, , , –, –; and
, ,  n.; and Spain’s Franco, role of parties, , , , 
 Ituango, , , –, , , –
Gómez Martinez, Fernando, , –, , , , 
, , , 
Guarumo,  Jardín, , , , , 
Guatemala, , ; and civil war,  Jesuits: education and elites, ; and Leo
Guerrillas: and amnesty of, , , , XIII’s Rerum Novarum, 
; and attacks, –, –, ,
, , , –, , –, , Labor unions, , , –, , , –
–, ; as bodyguards, ; and , –, , , –, –, ,
camps, , , , –, , ; , , 
and Catholic ritual, ; and family La Cucaracha (leader of guerrilla group),
clans, , , ; and insurgency, ; –; and mutilation of police, 
interviews with, xi; of the Left, , , Land: and agrarian unrest, ; and colo-
, , –; of the Liberal party, nists, , ; conflict over, in the
, , , , , , –, – s, , ; invasion of, ; Law 
, , –, , , , –, of , ; as public property, ,
–, , –, , , , , , , ; since s, , ;
; as political prisoners, ; and usurpation of, –, , ; and
rape, , ; recruitment of, ; violence, , , , –, 
relationship with soldiers, , , Latin America, ; anticommunist state
; and salt, ,  terrorism, ; leftist insurgency in, ;
violence in, , 
Hacendados, , , –, , Laureanistas (followers of Laureano
– Gómez), , , –, –, –
Haciendas, , , –, , , , , , , –; Jose Mejía y
, , , –, ; and agri- Mejía and Belisario Betancur, 
business, ; break up of, ; and La Violencia: as an aberration, ; and
parceling out, ; and settlers, ; as Colombian partisan differences, ,
site of struggle, , ; and workers, , , , , , , , ; and
, , , , ,  cultural differences, –, , ,
Henao, Hernán, , ; death of,  ; development of, , , ; and
Henao Mejía, Braulio, –, –, economic expansion, ; evolution of
 Index

La Violencia (continued ) Magdalena (department), 

in s, ; first period of, , , , Magdalena Medio (middle Magdalena
, ; as a hegemonic project, , ; Valley), –, –, –, , ,
idealized version of, –; impact –, –, , , –, 
of, on land values, ; interpretation Magdalena River, –, –, –,
of, –; and killing of peasants, , , , , , –, –, 
, , ; and lower class mo- Magdalena Transport Workers
bilization, ; and Marxist inspired (), , 
struggles, ; and Mexican Revolu- Massacres, , 
tion, ; rape of women and, , , Mayordomo (estate manager), ,
, , , , ; scholarship on, –, , , 
; second phase of, , ; studies Mayors, –, –, –, –,
of, –, , ; summary of,  , –, , –, –, –
n. , , , ; appointment of, ,
Law  of , , – –, , , , –, , , ;
LeGrand, Catherine, xii, ,  n. and homicide and assault, ; reports
Liberals: in alliance with army, ; as of violence, , , , 
followers of Gaitán, , , , . Medellín (city/capital of Antioquía), ,
See also Liberal party; Gaitán, Jorge , , , –, –, –, –,
Eliécer –, , , –, , , ,
Liberal party, , –, , , , – , , , , –, , –,
, , , , , , , –; –, –, , –, , –
change in relationship with workers, , –, –, , –; and
, ; conflict with Conservatives, Bello municipality, ; and Fourth
, , , , , , ; and Con- Brigade, , , , , , ;
servatives, , , , , , –, , growth of, ; and its elite, , , ,
, , , , , , , , ; and its newspapers, –; and
, , –, ; disagreement refugees, , ; and Sonsón, ;
within, , ; harassment of, , , and textile mills, ; and violence, ,
, , ; and Resistance, , , , 
, ; rise of non-elite politicians, Mejía, Vicente (alias el Dormido), ,
 
Lleras Camargo, Alberto, ,  Merchants, , , , –, 
Lleras Restrepo, Carlos, , ,  Mexico, ; Chiapas, ; Díaz, Porfirio,
López Pumarejo, Alfonso: compared ; and formation of state, ; and the
to Cardenas administration, ; and Revolution, 
educational reform, –; and Lib- Middle sector of professional politicians:
eral abstention, ; Revolution on the emergence of, , , ; impact of,
March (–), –, –, ; , –, , , –; of non-elite
second term of,  origin, , , –, , 
Migrants, , , , , , , , ,
Maceo, , , –, –; and , , ; as cosmopolitans, ;
authorities,  definition of, ; as the ‘‘other,’’ 
 Index 

Military, , –, –, –, , Nechí, , 

, , , –, , , –, Necoclí, , , 
–, –, –, , , Ninth of April , , , ; and nueve
–, , , –; absence of abrileños, 
powerful, ; and amnesty, , ; Non-Antioqueños, –, , , –,
and bombing, ; Cartagena’s Sec- , . See also Migrants
ond Brigade, ; defeat of, , ; Nuñez, Rafael, 
and deployment of heavy armament,
; and landowners, –, , Oil, , , , , ; and workers for
; militarization of regions, , Shell, 
; and Ministries of War, Govern- Olaya Herrera, Enrique, , 
ment, and Justice, ; officers, , ; Oquist, Paul, –, , , 
request for, , ,  Organized labor. See Labor unions
Military coup, , , –, –, Organized workers, , , , 
; failure of, ; support for, ; in Ortíz Sarmiento, Carlos, , , , 
twentieth century, ,  Ospina, Pedro Nel, 
Mining: and mines, , ; robbing of, Ospina Pérez, Mariano, , , –,
; and security force, –, ; as –, , –, –, , –,
a strategic sector, ; towns, ; and –, , ; and bi-partisan com-
victims, ; and workers, , , , promise, , ; and state of siege in
, –, , , , , ,  , , , 
Ministry of Labor, 
Movimiento Revolucionario Liberal Pabón, –, –, , –
(), ,  Paisas. See Antioqueños
Municipalities: in Antioquia’s periph- Panama: and contraband arms, , ,
ery, , , ; and colonists, ; and ; country of, ; and danger to
deaths, ; and state power, ; and canal, ; loss of, , ; strategic
workers’ mobilization,  location of Turbo, 
Murindó, ,  Paramilitary forces, , , , –, ,
Murrí, , , , ,  –, , , , –, ,
Mutatá, , –, , –, , , –, , , –; and
,  Conservative leaders, , , –;
as a response, , ; as right-wing
Narcotics. See Drugs death squads, 
National Association of Industrialists Paramilitary groups. See Paramilitary
(), ; influence of, ; and state forces
of economy in ,  Parish priest, , –, , , ,
National Federation of Merchants , , –, , –, –,
(), , , ; influence of, –
 Parsons, James: and la raza Antioqueña,
National government. See Central State 
National Land Redistribution Institute Parties (Liberal and Conservative), ,
(),  –, , , , –, ; agree-
 Index

Parties (continued ) ; and separate jurisdiction, ; as

ment between, . See also Frente a target, ; See also Liberals
Nacional Political brokers, , , ; and la Vio-
Partisan affiliation: and Antioquia before lencia, ; and national politicians, ;
the s, , ; meaning of,  and public patronage jobs, , –,
Partisan politics: and violence, –, , 
, , , , ,  Presidential elections of , ; and
Pato Consolidated Mining Company, campaign, , 
–,  Presidential election of , –, ,
Pavarandocito, ,  , , 
Peasants, ; assassination of, –, Pro Paz initiative, 
; and complaints, ; and expul- Public lands. See Land
sion of, , –; and parties, , Public sector workers, , –, ,
; and resettlements, ; restriction –, –, –, 
to mobility of, ; as victims, , , Public works, , –, , , , ;
, ,  and workers, , –, , , ,
Pécaut, Daniel, ,  –
Peque, , –, –, , , , Pueblorrico, , 
, , , , ,  Puerto Antioquia, 
Periphery: and antagonism, ; in Antio- Puerto Berrío, , , , –, , ,
quia, , , –, , , ; and –, –, –, –, –
mines, –; and regional value sys- , –, , , ; and deaths,
tem, ; and violence, –, , , , , –, ; and port, ; and
, –, –, , ,  refugees, ; and zone, 
Peru: and Sendero Luminoso,  Puerto Boyacá, 
Pielroja (guerrilla leader), – Puerto Nare, , –, –, ,
Police, –, , –, –, , –
–, –, –, , –, Puerto Perales, , , , 
–, –, –, , , Puerto Triunfo, , , , , –
, ; and abuse of peasants, , 
, –, ; appointment of, ;
and armed encounters with guerrillas, Ramírez, Father, , –
–, , , , , , , Ranching. See Cattle
–; and Bejarano, head of, ; Razed-earth policy, –, 
as a civilian force, , , , , , Regional authorities. See Regional
, –; ethnic composition of, in government
Urabá and Western Antioquía, , Regional government: of Antioquia,
–, ; and ineffectuality of, , , , –, –, –, –,
, ; and Liberal towns, , , , –, , –, –, ,
, , ; and number of agents, –, , –, –, –
; purging of Liberals in, ; and , –, –, –, –;
railroad workers, , ; removal of, and Catholicism, ; as a ‘‘claim,’’ ;
 Index 

and conformity to values, ; and Saresty, Domingo (minister of govern-

Conservative party, , , ; and ment), –
hegemonic bargain, , , ; in the Sayer, Derek, –, 
periphery, ,  School of Mines, , 
Regional State. See Regional government Seasonal laborer, 
Remedios, , –, , –, , – Segovia, , –, , , , ,
, , –, , ; and zone, –, 
,  Shell Casabe Camp, , –, 
Restrepo, Carlos E.,  Socialist party, 
Restrepo Jaramillo, Gonzalo (minister of Sociedad de Agricultores Colombianos
foreign relations), – (), 
Revolutionary juntas, , ,  Soldiers. See Military
Reyes, Rafael, – Spanish Civil War, 
Rionegro,  Strikes: and Lleras Camargo, ; and
Rojas Pinilla, Gustavo (General), , , public sector, , –, , ; and
, ,  United Fruit Company, 

Sabanalarga, , , –, , , Tarazá, 

– Tarso, , , , , 
Salgar, , , , –, –, Terror, , , , , –, , ,
, , – , , , 
Salvoconductos (Permits), –, – Tierras Baldias. See Land
Samaná,  El Tigre (corregimiento), 
San Carlos, ,  El Tigre (hacienda): case of, , , ;
Sánchez, Gonzalo, –,  and workers, 
San Jerónimo,  Titiribí, –, , , , 
San Jorge Road, – Tolima (department): and violence, ,
San José de Providencia,  
San Juan de Urabá, , –, ; Transportation, ; and workers, ,
port of,  –, , 
San Luis, –, –,  Turbay Ayala, Gabriel, , , 
San Pedro de Urabá,  Turbo, , , , –, , –,
San Roque, , , –, ,  –, –, –, 
Santa Barbara, , 
Santa Fé de Antioquía, , , , , Unión de Trabajadores Antioqueños
 (), , 
Santander (department), , –, Union of Colombian Workers (), 
– United States, , 
Santander (guerrilla leader), – United States government agent: and his
Santa Rita,  report, –
Santos, Eduardo: and labor unrest,  Universidad de Antioquia (University of
Santuario,  Antioquia), xi, , , , , –
 Index

Universidad Javeriana,  –, , –, –, , ,

Universidad Pontificia Bolivariana,  , , ; intersection of, with
Urabá: description of, ; and smug- economic interests, –, , ,
gling of arms, –, ,  ; ‘‘official story’’ of, , ; and
Uribe Gaviria, Julian, , , ,  ospinistas, ; and public employees,
Uribe Uribe, Rafael (bust), ,  –, , , , , , –,
Urraeños (inhabitants of Urrao), , ; in Quindío, ; related statistics,
, ,  ; relationship of, to partisan affilia-
Urrao, xi, , , , , , , , – tion, , , , –, , , ,
, , , , –, –, ; , , , , , ; role of
as a benchmark, , ; and contra- class in, , , –; role of cultural
band, , ; description of, –; differences in, , –, , , –
and population of African descent, ; , ; role of ethnicity and race in,
as a prototype, ; and womenfolk, , –, , , –; in rural areas,
,  , ; stories of, ; study of, ,
Urrego, Graciela (guerrilla companion), , ,  n.; in the s, ;
, , – trajectory of, –, , , , , ,
Usaga, Patricio (guerrilla leader), – , , , , ; turning point of,
,  in , ; unstable aspect of, ;
urban manifestation of, ; use of, ,
Valle (department of Valle del Cauca), , , –, –, , –, ,
, ; Cali, ; and violence,  , , , , , , , ;
Velásquez Acosta, Arturo, , , , victims of, –, 
 Violentologos, ,  n.
Venecia, , , ,  Virginias, , 
Viejo Caldas (the contemporary de- Virgin of Fatima, , , , 
partments of Quindío, Caldas, and
Risaralda),  Western Antioquia: description of, ;
Villamil, Eduardo, , , , , , and growth in real estate values, 
, , , 
Violation of human rights,  Yalí, , , –, 
Violence: in Antioquía, , , , , Yarumal, , –
–, , , , , , – Yolombó, , , , ; and its ha-
, –; on the cattle frontier, cienda owners, –, –, ;
; causes of, , , , ; in zone, , 
coffee-producing towns, , ; in
Colombia, xii, , , , ; in Colom- Zapata, Father Manuel, , , , ,
bian regions, , , ; as an elite , 
response, , ; geography of, , , Zarazoga:, , , –, , , 
, –; and guerrillas, –,
Mary Roldán is Associate Professor of Latin
American History at Cornell University.

Library of Congress Cataloging-in-Publication Data

Roldán, Mary
Blood and fire : La Violencia in Antioquia,
Colombia, – / Mary Roldán.
p. cm. Includes bibliographical references and index.
 --- (cloth : alk. paper)
 --- (pbk. : alk. paper)
. Violence—Colombia—Antioquia (Dept.)
. Colombia—Politics and government—–.
. Political violence—Colombia—Antioquia
(Dept.)—History. I. Title.
.   .'—dc 