The Origins of Cocaine

In the 1960s, the governments of Colombia, Peru, and Bolivia launched

agricultural settlement programs in each country’s vast Amazonian frontier
lowlands. Two decades later, these exact same zones had transformed into
the centers of the illicit cocaine boom of the Americas. Drawing on concepts
from both history and anthropology, The Origins of Cocaine explores how
three countries with divergent mid-century political trajectories ended up
with parallel outcomes in illicit frontier economies and cocalero cultures.
Bringing together transnational, national, and local analyses, the volume
provides an in-depth examination of the deep origins of drug economics in
the Americas. As the first substantial study on the shift from agrarian col-
onization to narcotization, The Origins of Cocaine will appeal to scholars
and postgraduate students of Latin American history, anthropology, global-
ization, development, and environmental studies.

Paul Gootenberg is SUNY Distinguished Professor of History and Sociol-

ogy at Stony Brook University, USA, where he is also currently Chair of
the History Department. He is a former Chair of the Drugs, Security, and
Democracy Program (DSD) of the Social Science Research Council and
Open Society Foundations.

Liliana M. Dávalos is Associate Professor of Conservation Biology at Stony

Brook University, USA. She has advised the United Nations Office of Drug
and Crime on deforestation since 2007 and is a co-author of the 2016 World
Drug Report.
The Origins of Cocaine
Colonization and Failed Development
in the Amazon Andes

Edited by Paul Gootenberg and

Liliana M. Dávalos
First published 2018
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa
business
© 2018 selection and editorial matter, Paul Gootenberg and Liliana
M. Dávalos; individual chapters, the contributors
The right of Paul Gootenberg and Liliana M. Dávalos to be identified
as the authors of the editorial material, and of the authors for their
individual chapters, has been asserted in accordance with sections 77
and 78 of the Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or
reproduced or utilised in any form or by any electronic, mechanical,
or other means, now known or hereafter invented, including
photocopying and recording, or in any information storage or
retrieval system, without permission in writing from the publishers.
Trademark notice: Product or corporate names may be trademarks
or registered trademarks, and are used only for identification and
explanation without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record has been requested for this book

ISBN: 978-1-138-59222-3 (hbk)

ISBN: 978-0-429-48938-9 (ebk)

Typeset in Sabon
by Sunrise Setting Ltd, Brixham, UK
Contents

List of illustrations  vi
List of contributors viii
Acknowledgements x

1 Introduction: Orphans of development: the unanticipated

rise of illicit coca in the Amazon Andes, 1950–19901
PAU L G O OTE N B E RG

2 The ghosts of development past: deforestation and coca

in western Amazonia19
L I L I A N A M . DÁVAL O S

3 Ideas of modernization and territorial transformation: the

case of the Upper Huallaga Valley of Peru53
M A R I TZ A PA R E DE S AN D H E RN Á N MAN RIQUE

4 Creating coca frontiers and cocaleros in Chapare:

Bolivia, 1940 to 199084
A N D R E W C . MIL L IN GTO N

5 Economic development policies in Colombia

(1960s–1990s) and the turn to coca in the Andes Amazon 114
J E N N I F E R S . HO L ME S, VIVE CA PAVÓ N , A N D SHEILA AMIN
G U TI É R R E Z DE P IÑ E RE S

6 The making of a coca frontier: the case of Ariari, Colombia 133

M A R Í A - C L A R A TO RRE S

7 Epilogue: will governments confront coca cultivation,

or its causes?160
A DA M I SAC S O N

Index 173
Illustrations

Maps
5.1 Colombian population growth from 1951 to 2005 (percentage) 120
6.1 Colombian cordilleras and political division 137
6.2 Meta Department and main western rivers 138

Figures
2.1 Stylized trajectory of forest fragmentation at agricultural
frontiers, from old-growth forests to rural fields 23
2.2 Annual deforestation rates from legal crops or from
coca cultivation 24
2.3 Summary of deforestation rates in areas or at times without
illicit crops, or when coca cultivation was present 29
2.4 Left: planned Carretera Marginal de La Selva connecting the
Colombia–Venezuela border to Santa Cruz. Major centers of
illegal coca production. Right: illegal coca cultivation in the
Andean countries for 2014 41
3.1 Inter-Andean valleys and colonization zones 59
4.1 Areas of settlement and forest clearance in the Chapare
lowlands, September 1966 87
4.2 Sample Corona photographs of Bolivian Amazon, 1966 88
5.1 GDP growth rate in constant dollars, including or excluding
the illicit economy of Colombia 123
5.2 GDP estimates of the Colombian national illicit economy in
constant COP 124
6.1 Mid-1960s liberal press presented the AFP agrarian reform
as a showcase of success. The photo caption reads: This
campesino was a few steps away from finding himself in a
conflict zone. Now he is working the land. 2nd March 1964
edition of El Tiempo (Courtesy of El Tiempo)143
 Illustrations  vii
Tables
2.1 Deforestation rates in countries with coca cultivation
recorded since 2000 and neighbors without records
over same period 22
2.2 Local and regional rates of forest loss with and without coca
in Andean countries, in chronological order 31
2.3 Government-sponsored colonization projects in western
Amazonia, 1960s–1970s 42
3.1 Major colonization projects in the Peruvian jungle from 1961
to 1972 65
3.2 List of World Bank loans to the Peruvian government through
the Banco de Fomento Agropecuario (BFA) to expand
agricultural credits (in USD) 70
3.3 Percentage of BFA credits granted in the jungle,
1941–198970
3.4 Debts of cooperatives affiliated with colonization project
Tingo Maria-Tocache-Campanilla for clearing forest 71
3.5 Total population of Valle de Alto Huallaga (VAH),
1940s–1980s73
3.6 Rural population growth rates in Alto Huallaga 74
4.1 Cochabamba Department: provinces with municipalities in
the Chapare lowlands in the 2012 census 90
4.2 Socio-political organization and settlement geography for
Villa Tunari Municipality in the 2001 census 91
4.3 Settlements in the Chapare colonization zone, 1980 96
4.4 Chapare: remaining forest cover by sector between
1966 and 1988 98
4.5 Coca leaf production, 1956–1963 in metric tons 100
4.6 Reports of cocaine factories in Cochabamba, Santa Cruz, and
Tarija departments, 1950–1967 101
4.7 Coca production 1940–1990: Bolivia and Cochabamba and
La Paz departments in metric tons 104
4.8 Coca leaf production, 1956–1958 in metric tons 106
5.1 Population trends and fraction of previous estimates
(percentage)119
5.2 Rural preschool enrollment 120
5.3 Rural preschool teachers 121
5.4 Rural primary school enrollment 121
5.5 Rural primary school teachers 121
5.6 GDP per capita in constant 1975 pesos (1980–1989) 122
5.7 Agriculture, ranching, and fishing as sectoral GDP 126
6.1 Origin of Meta migrants in 1972 145
6.2 Forms of land acquisition, Ariari, 1972 145
6.3 Land distribution in Incora-sponsored projects, 1972 146
Contributors

Liliana M. Dávalos is Associate Professor of Conservation Biology at Stony

Brook University (New York). Dávalos is a 2012 National Academies of
Sciences Education Fellow in the Life Sciences, a 2013 Kavli Frontiers of
Science Fellow for outstanding early career, and has advised the United
Nations Office of Drug and Crime on deforestation since 2007. She is a
coauthor of the 2016 World Drug Report.
Paul Gootenberg is SUNY Distinguished Professor of History and Sociology
at Stony Brook University (New York), and current Chair of the History
Department. He is author of Andean Cocaine: The Making of a Global
Drug (UNC Press, 2008) and editor of Cocaine: Global Histories (Rout-
ledge UK, 1999). From 2011 to 2014, he chaired the Drugs, Security, and
Democracy Program (DSD) of the Social Science Research Council and
Open Society Foundations.
Adam Isacson is the Senior Associate for Defense Oversight at the Washing-
ton Office on Latin America (Washington, DC). At WOLA, his Defense
Oversight Program monitors security trends and U.S. military coopera-
tion with the Western Hemisphere, with a particular focus on Colombia.
He has published and co-written dozens of reports and articles, testified
before Congress several times, and led several congressional delegations.
Sheila Amin Gutiérrez de Piñeres is the Dean of The Burnett Honors Col-
lege and Professor of Public Administration at The University of Central
Florida. Gutiérrez de Piñeres is a 2013 Founding Fellow of the University
of Texas System Academy of Distinguished Teachers, and was a Fulbright
Research Scholar in Colombia. She is author, with Jennifer S. Holmes and
Kevin Curtin, of Guns, Thugs, and Development in Colombia (University
of Texas Press, 2009) and, with Michael Ferrantino, of Export Dynam-
ics and Economic Growth in Latin America: A Comparative Perspective
(Ashgate Publishers, 2000).
Jennifer S. Holmes is Professor and Head of Political Science and Political
Economy at the University of Texas at Dallas. Holmes received a 2011
Regents’ Outstanding Teaching Award, University of Texas System. She is
 Contributors  ix
author, with Sheila Amin Gutiérrez de Piñeres and Kevin Curtin, of Guns,
Thugs, and Development in Colombia (University of Texas Press, 2009)
and author of Terrorism and Democratic Stability (Manchester Univer-
sity Press, 2008).
Hernán Manrique is a Teaching Assistant of Sociology at the Pontificia Uni-
versidad Católica del Perú and Research Assistant at the Centro de Inves-
tigación de la Universidad del Pacífico (Lima). He has published several
articles on drug trafficking and drug policy in Peru. His thesis studied
state building and alternative development in the Upper Huallaga Valley.
Andrew C. Millington is Professor of Land Change Science and Distinguished
Scholar at Flinders University, Adelaide, Australia. He has worked on a
land use, conservation, and natural resources projects in Bolivia, funded
by various international agencies, since 1995. He is the co-author of the
both Land Change Science in the Tropics: Changing Agricultural Land-
scapes (Springer, 2008) and The Sage Handbook of Biogeography (Sage
2011).
Maritza Paredes is Associate Professor of Sociology at La Pontificia Univer-
sidad Católica del Perú (Lima).  She is the co-author of Resource Booms
and Institutional Pathways. The Case of Peru (Palgrave, 2017). From
2015 to 2016, she was a Research Fellow of the Drugs, Security, and
Democracy Program (DSD) of the Social Science Research Council and
Open Society Foundations.
Viveca Pavón is a candidate for the PhD in Public Policy and Political Econ-
omy at the University of Texas at Dallas, and holds a BS in Economics
and an MBA from Texas A&M University. Her major area of interest is
violence and organized crime with an emphasis on Latin America, espe-
cially El Salvador, Guatemala, and Honduras.
María-Clara Torres is a PhD candidate in the History Department at Stony
Brook University and a former researcher at CINEP, the human rights
NGO in Colombia. Her work has been funded by Colciencias, CLACSO,
the Tinker Foundation, the Social Science Research Council, and a Grass-
roots Development Fellowship of the Inter-American Foundation. She is
the author of Estado y coca en la frontera colombiana: el caso de Putu-
mayo (CINEP, 2011).
Acknowledgements

An edited book is obviously built on intensive collaboration, and in the best

cases grows rather than ruins friendships. Paul Gootenberg wishes above
all to thank his co-editor, Liliana M. Dávalos. Liliana and I come from
astonishingly different cultural origins, and, on the surface, radically dis-
tinct fields (myself, an interdisciplinary historian; Liliana, an interdisciplin-
ary biologist), notwithstanding our overlapping passion for understanding
coca/cocaine. We are also, almost by accident, colleagues at the same New
York public research university, Stony Brook. It has been an absolute delight
to work with, and learn from, someone as brilliant and serious as Liliana!
Throughout this process I think both of us have been surprised by our “syn-
chronicity” on both the book’s theme and in our editorial eye.
This book was also informed by Gootenberg’s extramural work at the
Social Science Research Council/Open Society Foundation’s “Drugs, Secu-
rity, and Democracy” (DSD) interdisciplinary research program on Latin
America, which he was fortunate enough to chair from 2011 to 2014. Two
of the authors in this volume, Peruvian sociologist Maritza Paredes and
Colombian historian (and current Stony Brook doctoral candidate) María
Clara Torres, were DSD fellows. The remarkable DSD network also engaged
my friend and pioneer Andean rights ethnographer Kimberly Theidon (now
at the Fletcher School at Tufts University). Kim was a close original col-
laborator on this book who helped convene our symposium at LASA16 in
New York, which first brought together our team of researchers. Kim: My
warmest gratitude!
Liliana M. Dávalos thanks her co-editor, Paul Gootenberg. Paul and I
have been discussing the history and geography of coca from distinct van-
tage points: his, grounded in historical analysis; mine, focused on the forest
and its disappearance. At the time I began to understand Paul’s research
better, my work on the forests of the Andes Amazon yielded a puzzling con-
clusion. While places with coca lost more forest as more people moved to
them, the amount of coca did not relate to the number of migrants. By
then, H. Leonardo Correa of the United Nations Office on Drugs and Crime
(UNODC) in Bogotá asked me to take on similar questions for the entire
world. This was the opportunity needed to collect data for the whole of the
 Acknowledgements  xi
Amazon Andes, and to take on a longer-term view. Borrowing a page from
Paul, I looked for papers dating back to before the coca boom. The results
came to Paul’s attention thanks to the serendipity of the Long Island Rail
Road, and he was generous enough to accept my bid to co-edit this volume.
But to focus only on the professional would leave out an important source
of inspiration. In his twilight, and only once or twice, my grandfather Crispín
Álvarez pined for “his land in the Caquetá.” He and Teófila Sánchez, both
campesinos, were ejected from the inter-Andean valley in Tolima in the giant
mid-twentieth-century spasm known as La Violencia. Fleeing violence and
hunger they landed on a small, fast-growing city. But it was mere accident
that their daughter, Faridha Álvarez, and my father moved out to the Meta
frontier outside Villavicencio as part of the modernizing wave that recruited
professionals into research centers working for the Green Revolution. For
Teófila never let Crispín set foot in the Amazon Andes. Instead she steered
him to go where there were schools for their growing family, but no land to
settle. As heir to that choice, I thank them.
1 Introduction
Orphans of development:
the unanticipated rise of illicit coca
in the Amazon Andes, 1950–1990
Paul Gootenberg

Coca contexts
For decades, conventional wisdom as to why so many poor farmers in the
“global south” get lured into illegal drug crops—poppy, coca, cannabis—
revolves around a broad mix of ideas: their marginality from “weak” or
absent states, grinding poverty or immiseration, or proximity to porous,
ungovernable borders and endemic armed conflicts that foster smuggling
activities. Each of these explanations has a grain of truth for the myriad of
zones in which drug crops thrive, and each has been used to design policies
to discourage illicit crops. But this volume zeroes in on a more specific ori-
gin for the massive and unprecedented turn to illicit coca crops across the
western Amazon during the middle of the last century, which jump-started
the still thriving global cocaine boom of the mid-1970s and beyond. Thou-
sands of migrating peasants, in Peru, Bolivia, and Colombia alike, adopted
coca and became the modern cocalero (coca grower) classes of the Andes in
the aftermath of failed or retreating state-led, mid-century modernizing col-
onization projects to develop the Amazon. The take-off of the Andean drug
trades was thus a vast classic “unintended consequence of social action”
(Merton 1936). Its actors on the ground were, in effect, the orphans of
development.
This striking pattern of flight from failed development in the Andes is
detailed here around three finely grained case studies of frontier sites that
became hotbeds of cocaine trafficking by the 1980s: The Upper Huallaga
Valley of east-central Peru, Bolivia’s lowland Chapare region, and south-
east Colombia’s tropical Meta Department. These geographically disparate
resettlement and agrarian modernization projects were related: connected
by post-war dreams of building a vast Carretera Marginal de la Selva across
the western Amazon from Venezuela to Santa Cruz in eastern Bolivia, pro-
moted by national governments and United States foreign aid and such mul-
tilateral agencies as the Inter-American Development Bank (IDB) and the
World Bank. For example, Peru’s reformist President Fernando Belaúnde
Terry of the mid-1960s celebrated road-building and agricultural stations
across Peru’s western Amazon as the “Conquest of Peru by Peruvians,” a
2  Paul Gootenberg
militant 1959 colonizing metaphor in which fertile valleys such as the Hual-
laga would become the long-dreamed-of breadbasket for teeming coastal
cities and exhausted sierran agriculture. This was before bitter agricultural
experience and new-found environmental awareness belied longstanding
myths of boundless Amazonian productivity.
Another political element at play across all three nations was the Cold
War. All the Andean nations hosted an array of Cold War regimes across
these four decades (1950–1990), most closely aligned with the United States,
and all three political classes also feared restless campesino masses press-
ing against the old hacienda systems of their upland interior (Pike 1977).
Population movement into a developing Amazon was viewed as a friction-
less way of appeasing poor and rebellious rural folk as well as keeping
them out of revolutionary shanty towns in cities (as they were also wrongly
perceived at the time). The Cuban Revolution of 1960 added ideological
urgency to this project. Troubling guerrilla focos (bands) quickly broke out
in remote parts of Colombia (sparking the half-century struggle with the
FARC) and Peru (more quickly squashed from above); in Bolivia’s Santa
Cruz jungles, the iconic revolutionary Ernesto “Che” Guevara landed with
the mistaken notion that Bolivian colonos (lowland colonists), some already
with coca plots, would respond to his revolutionary zeal. Agrarian reform—
of a non-Communist i.e., reformist, gradualist, non-redistributive kind—
became the officially sanctioned policy of the brief but pivotal Kennedy-era
U.S. “Alliance for Progress” (1961–1964). In Bolivia, the breakup of tra-
ditional estates often occurred erratically as peasants fled or empowered
themselves to land after the 1950s urban revolution. Colombia enacted the
narrowest of agrarian reforms, dubbed by Liberal politicians as “Making
Tortillas without Breaking Eggs,” an obviously impossible political recipe.
Confident, stage-based “Modernization Theory” was at its heyday in U.S.
academia and governmental circles (Engerman et al. 2003; Latham 2000),
and such abstract Cold War ideas were read and used by national regimes
for their own purposes, often concretized (in contrast to industrializing
plans) on such tropical development zones. In each country, by the end of
the 1960s, tens of thousands of poor farmer colonos streamed into these
jungles, either to hastily organized settlement poles with formal credits,
land titles, warehouses, and other services, or “informally” spontaneously
on their own, wherever the often unfinished roads led. To take the Peruvian
example again, Huánuco Department, where the Río Huallaga starts its
course towards the Amazon, became by far the country’s fastest-growing
department from the 1950s through the 1970s, outside the coastal capital
of Lima (Werlich 1968).
Then, in different ways and timings, these projects failed or were dis-
banded, with little to show in marketable agricultural goods or stable
farming livelihoods. By the late 1970s, this left hundreds of thousands of
impoverished non-indigenous peasants essentially stranded in the ­Amazon
without the government credits, schools, social and judicial services,
 Introduction  3
c­ommunications, and business opportunities or jobs originally promised
by colonization schemes (Aramburú et al. 1982). In Peru, one of the coun-
try’s first modern feature-length films, Armando Robles Godoy’s La Muralla
Verde (The Green Wall, 1970) captured the insecure mood of Amazonian
colonists disillusioned with the distant state. In Bolivia’s tropical Chapare,
below the highland Cochabamba Valley, many communities became self-­
regulated instead by “sindicatos,” “federaciones,” and “centrales.” Reveal-
ing of migrant roots in unionized altiplano mining camps (Sanabria 1993),
this autonomous organization still marks Bolivia’s politically active lowland
agriculturalists today. In Colombia, the promised electricity and sanitation
never arrived and peasants drifted into itinerant subsistence plots on fragile
soils, joined by migrants escaping the violent conflicts of the highlands on
their own. As suggested in this volume, from the peasant vantage, a colonist
culture of “dependency on the state” meant both autonomy from traditional
power holders and an eventual deep sense of abandonment by authorities.
By the mid-1970s, in any case, a common and powerful ideological
drift was perceptibly underway as well, away from democratic panacea of
“development” to other Cold War projects (also emanating from the United
States) of internal national security. Some militaries adopted a top-down
developmental ideal, notably Peru’s left-wing military regime of 1968–1973.
General Velasco’s far more radical agrarian reform—swift expropriation of
the landed “Oligarchy”—aimed for “modernizing” highland cooperative
estates, but ended up instead accelerating an outflow of landless farm labor-
ers, many into Amazonia (Meyer 2009). But tropical frontiers were also
fading from the states’ narrowing political vision and capacity. The dramatic
debt and bankruptcy crises of the 1970s and 1980s felt across the Andes
made national governments increasingly hard-pressed for funding any wider
territorial or deeper social expansion. Instead “neo-liberalism,” the new gos-
pel mandating the reduction of state roles, and the magic of the market
instead, became dominant throughout the region by the 1980s, followed
after 1989 (and the receding Cold War) by the mantra of “Globalization”
over state-led bounded. National development. In short, the state became in
a historical process increasingly “absent” in these now-populated lowland
pockets.
These were the larger political and global conditions that by the mid-
1970s led to coca’s rise as one of the most attractive goods to alienated
refugees of development along the “Andean Ridge.” The majority of these
campesinos had no earlier horticultural or cultural exposure to the coca
shrub, Erythroxylum coca. Each of these tropical peasant zones, in another
part of the story, had to link up to nascent drug trafficking and processing
rings coalescing since the early 1950s, first in Bolivia, then Peru, and finally
linked to coca in Colombia in the 1980s, a country bereft of previous con-
tact with the shrub, given its small indigenous population using coca-leaf.
Each group of autonomous farmers had to learn to plant, tend, harvest,
hide and chemically process coca-leaf into PBC—pasta básica de cocaína
4  Paul Gootenberg
or coca paste—an easily transportable and sellable form of cocaine sulfates
conveniently made from common “developmental” wholesale goods like
kerosene, plastic sheeting, and cement lime (León and Castro de la Vega
1989). By the mid-1960s it was the most easily marketable of goods in
Chapare, linked to drug syndicates in Cochabamba and La Paz and fun-
neled to newly rising recreational cocaine markets abroad through north-
ern Chile. Paradoxically or not, coca became the perfect market—daresay
“neo-liberal”-era—good, even if most peasants (as opposed to moneymen
higher up the commodity chain) were barely eking a living from it in this
highly unequal illicit economy. By the mid-1970s PBC acquired a similar
role in the explosively expanding “green wall” of coca in the Huallaga, most
of it flowing north through frontier Leticia to swiftly consolidating urban
Colombian processing and distributive groups in Medellín, Bogotá, and Cali
(Gootenberg 2008). Then between the mid-1980s and 1990s, Colombian
colonos themselves, in a set of complex transformations outlined in Chapter 5,
began to swiftly adopt the crop, completing the vertical integration of
Colombia’s regional “cartel” cocaine industry.
In all three Andean nations, the epicenters for illicit cocaine—their geog-
raphies set by the 1980s—were the three central lowland ex-colonization
zones: the Chapare, the Upper Huallaga, and the arc defined by Meta in
the north and Putumayo in the south. By the 1990s, the Chapare alone had
absorbed some 350,000 migrants, with its chief economic activity, coca cul-
tivation, covering 55,000 hectares (the traditional Yungas coca-growing area
near La Paz never re-oriented to cocaine). Further east, the lowland frontiers
of Santa Cruz and Beni (adjacent to Brazilian and Paraguayan border smug-
gling) also experienced a lesser cocaine boom by the late 1970s, significantly,
in the wake of oil and cotton development. Many traffickers of Bolivia’s
notorious 1980s “narco-state” originated and operated from Santa Cruz,
Bolivia’s wealthiest region. The Peruvian Huallaga, with its center around
the former development pole of Tocache, by the early 1990s climaxed at
120,000 hectares in coca, before a sharp fall partly due to the swift Colom-
bian peasant adoption of the crop. (Today’s Peruvian cocaine resurgence in
the southern Valle de los Ríos Apurímac, Ene y Mantaro or VRAEM system
is geographically apart from cocaine’s birthplace to its north). In Colombia,
a more varied geography of coca ensued, starting from the initial former
colonization zones of Guaviare and Sur de Meta (studied here in the Ariari)
and in the 1990s into the southwestern rainforests of Putumayo (along fad-
ing frontier roads of former petroleum and rubber extraction) and Caquetá,
and beyond, to areas like Vaupés and the middle Magdalena Valley. By the
year 2000, Colombia’s 60–80,000 cocalero families supplied two-thirds of
the world’s coca and cocaine, enough for about 700 exportable tons of the
drug. Colombia’s pattern of more scattered coca likely stemmed from rival
trafficker pole promotion of coca planting and the “balloon” dispersal effect
of being the most consistent target of U.S.-funded coca eradication cam-
paigns (at best intermittently enforced in Peru and Bolivia) by the 1990s.
 Introduction  5
To reiterate the core thesis here: the initial historical path from Amazonian
development to cocainization appears systematic and even in some sense of
the word “structural.” It occurred in an archipelago of sites across the Ama-
zonian Andes with shared social traits but in highly distinctive nation-state
configurations. The three countries in play were marked by three contrast-
ing ethnic regimes and longer cultural relationships to coca leaf—Bolivia,
majority Aymara-Quechua indigenous and with widespread popular use
of coca; Peru, by the 1950s becoming a majority cholo (mestizo) culture,
with legal coca commercialized from sites in southern Cusco; and Colombia
with few and scattered indigenous groups such as the Nasi of the Cauca
region. It happened despite three different historical relationships with the
industrial medicinal drug cocaine: Peru, with a long legal (1880s–1940s)
cocaine-­processing sector in the central high Amazon; Bolivia, with strong
coca-leaf cultivation in the Yungas of La Paz for peasant and miner con-
sumption but no cocaine whatsoever until illicit drugs erupted in the 1950s;
and ­Colombia, with neither relationship prior to the 1970s. It unfolded
under highly varying degrees and politics of U.S.-influenced drug enforce-
ment. The three nations also underwent a gamut of agrarian reform trans-
formations after mid-century: Colombia’s constricted under elite control,
Bolivia’s a spontaneous, barely channeled takeover of estates by peasant
groups, and Peru’s, after peasant turmoil in the early 1960s, undertaken
by reformist and then radical government fiat. And cocainization occurred
across a bewildering array of political regimes: from Peruvian populist and
revolutionary reformism, to Colombia’s elite Liberal-Conservative National
Front; and Bolivia’s path from Movimiento Nacionalista Revolucionario or
MNR revolutionaries to the military and rightist regimes that followed. The
exception likely even proves the rule: Ecuador, the fourth Andean nation
that has yet to suffer a major coca boom in its eastern frontiers, had few
active colonization programs in the 1950s–1960s (and these towards the
Pacific coast), with virtually no Amazonian road-building, at least until
petroleum pipelines began in the 1970s. In short, the colonization path into
the upper Amazon was the defining factor in the rise of cocalero complexes.

Methodologies and chapters

This volume builds from newly researched historical case studies of Bolivia,
Peru, and Colombia, but its thesis—that Amazonian colonization led to
cocainization—originates from a surreptitious convergence of methodol-
ogies and colleagues in the field. The book thus adopts a robust “mixed
methods” approach, ranging from bottom-up ethnography and oral history,
political economy and political science, the transnational history of ideas
and institutions to computational spatial and mapping analysis, as well as
addressing policy questions crossing environmental concerns with drug con-
trol issues. On one hand, one editor, Paul Gootenberg is an archival-based
economic historian of Peru who in 2008 published Andean Cocaine: The
6  Paul Gootenberg
Making of a Global Drug, a book that used local sources to examine the
global processes at work in the long-term rise of the commodity cocaine.
One surprise in this research was the key role of the post-war, U.S.-assisted
tropical agricultural research and extension station and colonizing site in
the Upper Huallaga Valley near the then tiny hamlet of Tingo María. This
site inadvertently launched the sudden widespread peasant adoption of coca
in the valley during the 1970s. The findings then raised suspicions of a sim-
ilar story in the Chapare of Bolivia (Gootenberg 2008, 292–300). Gooten-
berg convened a group of specialists to put together case studies comparing
three key Andean cocaine zones. The other editor, Liliana M. Dávalos, is a
conservation biologist at the same university with an interest in the impact
of coca on tropical forests (along with evolutionary research on tropical
fauna). Around the same time Gootenberg’s research group first met at the
Latin American Studies Association (at LASA16, New York),1 Dávalos was
working on a commissioned scientific advisory report for the 2016 UNODC
World Drug Report on the effects of drug crops, trafficking, and eradica-
tion on deforestation rates in the Andes, Myanmar, Laos, and Afghanistan,
responding to growing public interest in drugs and environmental harms.
Dávalos adopted two methodologies: first, she excavated and closely com-
pared detailed maps of Amazonian development projects prepared by
­German-language academics in the mid-1970s and found a striking simi-
larity to 2014 clusters of cocaine production in all three countries, with an
almost perfect resemblance by the 1990s. Second, she designed a spatially
explicit (quantitative multivariate) landscape model using data from 984
Amazonian municipalities in all Amazonian nations which showed the same
lasting historical imprint of past development projects. The results, which
also question the assumption that coca cultivation instead of tropical devel-
opment per se drives Amazonian deforestation, were published in BioScience
(Dávalos et al. 2016). Some of her suggestions about development and drugs
also made it into the UN World Drug Report (UNODC 2016, chapter 2),
the reference work of record on global drug issues.2 By this point, Gooten-
berg and Dávalos realized they were independently converging on the same
problem—peasant colonies and cocaine—and combined their distinctive
research methods into the present volume.
The title of this book refers to western Amazonia (the elevated Amazonian
escarpment of the Andean chain) as the “Andes Amazon.” This term is gain-
ing currency in the scientific community to encompass this vast and excep-
tionally bio-diverse swath of the American tropics stretching from northern
Colombia to southern Bolivia. Each Andean country has its own national
lexicon for these zones where highland and lowland ecosystems converge
and Erythroxylum coca thrives. In Colombia, it is the oriente or selva (jun-
gle); Peruvians use the evocative term Ceja de la Montaña (“Eyebrow of the
Jungle”) or in shorthand Montaña (montane region), though it usually sig-
nals one of twenty or so major basins flowing eastward down the Andes to
distant tributaries of the Amazon River system. In Bolivia, the term yungas
 Introduction  7
prevails for steeply cut ravine zones or subtropical valleys associated with
particular upland centers (for example, the traditional Yungas de La Paz, in
shorthand the Yungas; or the Yungas de Cochabamba, adjacent to lowland
Chapare, in past usage the Chapare). The U.S. drug authorities monitoring
cocaine like to lump it all as the “Andean Ridge.” The Amazon Andes is a
useful term for social scientists and historians to adopt because it under-
scores the intensive social and historical linkages and human exchanges that
have shaped these regions since pre-Columbian times. The Amazon Andes is
not, to paraphrase anthropologist Eric Wolf (1982), some imagined pristine
region “without History.” Indeed, in the mid-twentieth century, these social
and political and ecological connections conditioned, as readers will see, the
rise of illicit cocaine.
The following chapter “The ghosts of development past: deforestation
and coca in western Amazonia,” by Colombian-born biologist Liliana M.
Dávalos. complements this Introduction, and sets the problem in environ-
mental terms. Amazonian forests are still the world’s single largest tropical
frontier, though quickly vanishing. The Andean countries share a third of
this Amazon forest and have attempted to tap their resources for centuries,
but it was only in the 1960s that states—Colombia, Peru, and Bolivia—
coordinated efforts by connecting roads and inaugurating large-scale set-
tlement projects. These investments involved international development
agencies and donors, and ultimately opened the Amazon frontier to millions
of Andean campesinos. The long-term results, now visible from space, are
massive wedges into the forest. By the 1970s, with dwindling prospects for
the originally envisioned agricultural bounty, these centers of colonization
became the global source of cocaine for the international market. By analyz-
ing the spatial distribution of these projects and coca cultivation, and sifting
through a bevy of scientific studies and spatial models, Dávalos shows how
these development efforts relate to deforestation and then to coca produc-
tion. The Amazon Andes frontier where settled lands abut standing forests is
today a moving front of smallholder agriculture that often includes coca (at
first on hillsides, but increasingly further downstream) and largely replaces
old-growth rainforests with low-productivity pastures. Although anti-drug
programs increasingly assume (and legitimate themselves by claims) that
suppressing coca protects biodiversity, both the region’s environmental his-
tory and the lingering spatial signature of past development projects imply
that deforestation dynamics exist even without coca. Coca, in Dávalos
words, is a “symptom” of this development, not its cause. Whatever the case,
experts argue that a strong shift to alternative development is long overdue,
in light of decades of failure to reduce drug supplies through coca eradi-
cation (Mejía 2016). In Colombia’s case, it is now specifically mandated
on an unprecedented scale by the 2016 FARC peace treaty, which directly
engages the country’s cocaleros. Given the magnitude of counter-narcotics
campaigns and drug crops, it is essential to learn from past development
mistakes. We know now that deforestation fronts and coca share a common
8  Paul Gootenberg
origin in colonization and infrastructure projects whose economic payoffs
proved ephemeral. To overcome the twin legacies of deforestation and coca,
new alternative projects should set explicit conservation goals embedded in
decades-long financial commitments.
Chapter 3, by two Peruvian sociologists, Maritza Paredes and Hernán
Manrique, is “Ideas of modernization and territorial transformation in the
rise of coca: the case of the Upper Huallaga Valley, Peru.” The chapter illus-
trates how modernization theories and national political pressures filtered
into agrarian reform and then into colonization programs. Unexpectedly,
they contend, the failure of these ideas and projects contributed to the
formation of areas of drug production (especially the around the former
Proyecto de Colonización Tingo María-Tocache-Campanilla) in the Upper
Huallaga Valley of Peru. Throughout the twentieth century, the Peruvian
state promoted the settlement of the high Amazon rainforest as a solution
for intensifying sierran land conflicts. The government built roads and pro-
vided incentives for migrants to move into the “free land.” The collapse of
these organized initiatives in the 1970s, however, contributed to massive
deforestation and an isolated “enclave” of poor peasants, with little access to
markets and dependent on state intervention to make their local economies
sustainable. The ideas that guided these transformations, however, were not
new or unique to Peru. The colonization of the Amazon Andes was heavily
influenced by the Cold War politics of the United States. During the early
1960s, the international community urged Latin American states to seek
rapid developmental transformations through state initiatives to counter the
Communist threat posed by the Cuban Revolution (Latham 2000). This
chapter follows the pathways of such developmental concepts through the
recommendation of the UN-sponsored Economic Commission on Latin
America (ECLAC) and the Inter-American Development Bank (IDB), which
proved essential to promoting failure-prone colonization in Peru’s central
Amazon.
Chapter 4, by British geographer and veteran Bolivia scholar Andrew C.
Millington, is “Creating coca frontiers and cocaleros in Chapare: Bolivia,
1940 to 1990.” This chapter weaves together threads of published sources,
new methodologies, and archival research to provide an updated rigorous
geographical and spatial interpretation of the establishment of Bolivia’s
most notorious and politicized coca frontier: Chapare. The time frame spans
World War II to 1990, which changed Chapare’s traditional role as impene-
trable borderlands between upland areas of Cochabamba and Bolivia’s low-
land tropical expanses towards the Brazilian border. The chapter closely
traces the precise routing and patterns taken by migrants, their settlements,
and social networks, while they started to cut the forests of the Chapare into
a novel and dynamic coca frontier. A mix of government-sanctioned settle-
ment and protracted social breakdown drew settlers down into the tropics.
The chapter ends up providing a precise periodization of the emergence of
the Chapare—dedicated to illicit coca for new cocaine exports—over the
 Introduction  9
traditional Yungas coca regions near La Paz, during and after the Bolivian
Revolution of 1952. In a certain sense, the Chapare is emblematic of the
Bolivian state’s failure to provide a viable agrarian reform for its peasant
majority as well as basic social services to migrants, who began early on to
organize themselves on their own in Chapare instead. The chapter also aptly
exploits a new kind of Cold War era data set to study colonization fron-
tiers in South America. It maps the pace of forest clearance after 1966 from
declassified intelligence photographs from the U.S. Corona H-4 spy satellite
photography missions of the high Cold War.
Chapter 5, by the political economy oriented team comprising political
scientist Jennifer S. Holmes, graduate student Viveca Pavón, and economist
Sheila Amin Gutiérrez de Piñeres, is “Economic development policies in
Colombia (1960s–1990s) and the turn to coca in the Andes Amazon.” It
provides an overview of the economic policies and economic cycles between
the 1960s and 1990s that helped push peripheral, indigent Colombian farm-
ers into coca production, virtually a new national crop. First, it reveals that
beyond classic issues of Colombia’s political vulnerability to illegal drugs
and violence stressed by previous studies (Thoumi 1992), political economy
was at play. Colombia became ripe for illicit coca because of the specific
dynamics of economic inequalities and insecurities created by the country’s
cyclical political economy of coffee, and by its associated elite practices
of boom investing and business illegality. The second part of the chapter
examines impacts on the three core lowland areas where migrants flocked
after the mid-1970s for coca livelihoods: the new Amazonian departments
of Guaviare, Putumayo, and Caquetá. These case studies reveal poorly sus-
tained governmental support for colonists in such basic local services as edu-
cation and security, coca’s scarce legal alternatives, and poverty and social
inequality conditions in refuge regions too. The next chapter follows up on
this backdrop with a close-knit ethnographic history of peasant migrants to
Meta Department, Colombia’s fourth initial coca zone.
Chapter 6, by the Colombian ethnographer and historian María-Clara
Torres, is “The making of a coca frontier: the case of Ariari, Colombia”
It zeroes in on the Ariari settlement pole, a tropical belt of Meta Depart-
ment in southeastern Colombia. Torres demonstrates continuities between
Colombia’s 1950s La Violencia civil war (which drove politically persecuted
peasants out of mountainous coffee lands and inter-Andean valleys of the
interior), the 1960s Alliance for Progress, and the delayed rise of coca plots
in Ariari by the 1980s. This chapter frames the rise of illicit coca—­hitherto
unknown—as the aftermath of a minuscule agrarian reform that only inten-
sified Colombia’s protracted rural social conflicts. In Cold War Colombia,
divided elites engineered a severely limited form of land reform. The chapter
excavates the early 1960s political debates and challenges faced by both
Colombian elites and American diplomats in designing the U.S.-sponsored
Alliance for Progress’ land reform, which vaunted colonizing lightly settled
public lands in Amazonia. It explores ground-up, from campesino reactions
10  Paul Gootenberg
and memories of the settlement zone, how agricultural development policy
crystallized into the failure-plagued, state-run migration program known
as “Project Meta N°1.” Finally, the chapter shows how the protagonists of
this migration from the highlands, displaced and destitute mestizo peasants,
with a negligible cultural history of coca, began their early involvements
with the shrub. This site too became a hotbed of illicit drug production by
the 1980s, and still is today.
Chapter 7, the Epilogue by Adam Isacson, an experienced ­Washington-based
drug and defense policy analyst, is “Will governments confront coca cultiva-
tion, or its causes?” Isacson dissects the current Andean coca conjuncture in
light of these new studies of cocaine’s historical origins. While to policy spe-
cialists the four-decade drug war against Andean coca and cocaine appears
to be utterly exhausted, Andean nations, according to Isacson, still need to
choose between solutions that take obvious “short-cuts” (such as coercive
aerial fumigation of coca) and durable solutions to lowering drug exports
that integrate civic society, especially socially marginalized cocaleros, into
genuine alternative livelihoods. The present moment is precarious, not only
because of the advent of the Trump regime in the United States—which
sends such uncertain signals about drug policy—but because of the loom-
ing coca/cocaine crisis in Colombia, where for a complex set of reasons
illicit coca has quickly rebounded since 2014 to its peak 2000 levels. Isacson
lays out two scenarios, one optimistic towards drug-reform-style change,
which specifically cuts down Colombian production, the other a return to
past short-sighted drug-war eradication strategies. Consolidation of gains
in rural services is at the core of the optimistic scenario, while short-lived
declarations of victory characterize short-term paths. Calling for a balance
of enhanced state services and developmental commitments to cocaleros in
the Amazon, these ideas need to be revised in light of discoveries about the
origins of cocaine. Neither a stronger footprint of the state, nor develop-
ing the Amazon are in and of themselves an insurance policy against illicit
drugs, particularly if projects are unsustained, environmentally destructive,
or insulated from peasant participation.

Literatures and implications

This is the first book to systematically connect the dots between peasant
colonization, Cold War modernization drives, and the origins of illicit drugs,
but it draws on and intersects with several key literatures. The archival-based
history of drugs in the region is still in the making (Gootenberg and Campos
2015), except for abundant journalistic writings on traffickers and “cartels”
and more serious new social science literatures about states, elites, regional
insurgencies, violence, and narcotics trades. There are some excellent exam-
inations of Amazonian development (García Jordán 2001; Santos Granero
and Barkley 2000) and of colonization (Aramburú et al. 1982; Werlich 1968,
to note a few). More specifically, each of the three major catchment areas in
 Introduction  11
our lens has suggestive prior 1980s-era ethnographies and oral histories on
peasant migration into Amazonia: Harry Sanabria’s study of Bolivian alti-
plano miners’ moves into the Chapare, Edmundo Morales’ ethnography
of Huallaga peasants during the initial cocaine boom, or Alfredo Molano
Bravo’s narrative of Guaviare settlements in Colombia (Molano Bravo
­
1987; Morales 1989; Sanabria 1993).
A long-studied topic is the Cold War in Latin America and U.S. over-
seas development policy, which is now spawning a boom of works re-­
examining the political roles of “modernization theory” (Brands 2012;
Cooper and Packard 1997; Engerman et al. 2003; Latham 2000) and
lately, the Latin American developmental interpretations of it, for exam-
ple at CEPAL’s headquarters in mid-century Chile (Beigel 2010). Histor-
ical and ethnographic critiques of economic development abound—the
“anti-­politics machine”—though to misquote Mrs. Thatcher, no one has
yet offered poorer people a viable “alternative” (Ferguson 1990; Escobar
1995). Other critiques have introduced concepts like “Deviant Globaliza-
tion” (Gilman et al. 2011): that the spread of global market-led devel-
opment has itself spawned an array of interconnected illicit economies,
much as seen in the Amazon. To date, however, only one such work,
­Daniel Weimer’s Seeing Drugs: Modernization, Counterinsurgency, and
U.S. ­Narcotics Control in the Third World, 1969-1976 (Weimer 2011),
deals explicitly with illicit drugs, arguing, for example, in Mexico’s case
that escalating repressive U.S. anti-drug politics abroad in the pivotal
1970s reflected Cold War securitization goals. Indeed, most of the liter-
ature on the origins of cocaine in the Andes is still mired in ever-popular
Cold War “conspiracy theories,” inspired by Alfred W. McCoy’s classic
heroin saga of Southeast Asia (McCoy 1991). That is, cocaine is mostly
read as a by-product of Cold War alliances and secret wars against left-
wing guerrilla movements, including the infamous “Contra” episode of the
1980s that allegedly created North America’s urban “crack” scene (i.e.,
Marcy 2010; Scott and Marshall 1988). The alternative theory, found in
many official polemics and books of the 1990s with titles invoking the
threat of “narco-terrorism” is that cocaine was simply a tool of Commu-
nist guerrillas (like Peru’s 1980s Sendero Luminoso or Colombia’s FARC),
used to build war chests by taxing and coercing hapless campesinos to do
their dirty work (Tarazona-Sevellano with Reuter 1990; cf. Felbab-Brown
2010). One big problem with all of these interpretations (as well as those
that overplay the vertical power of drug “cartels”) (critiqued in Kenney
2007) is that they focus on easy villains. And they ignore the very real
social and political forces, such as social inequalities, that drive masses of
peasants into growing illicit drugs, and others into trafficking, or for that
matter the social contexts that lure millions into avidly consuming drugs
in the metropole. This literature also absolves the badly construed interna-
tional drug policies that accentuate the profitability and violence of illicit
trades. They are short on both context and causality.3
12  Paul Gootenberg
A more serious in-depth social science literature obviously exists on why
peasants and other social actors take up drugs, often closely linked to for-
mulating still problematic strategies of “alternative development” or the
contested concept of developmental “conflict goods” (Collier 2007; Cooper
2001; Snyder 2006). Little of this work, however, goes deeply into history. An
exception of note here is an overlooked 1992 book from the British Panos
Institute, edited by Michael L. Smith, Why People Grow Drugs: Narcotics
and Development in the Third World (1992), which encompasses develop-
mental case studies of Thailand, Pakistan, Colombia, Peru, and Bolivia. The
chapters on Peru and Colombia (written by anthropologists Roger Rumrrill
and journalist/sociologist Alfredo Molano Bravo respectively) did not over-
look the impact of prior developmental projects in these regions, although
the book mostly advances a drugs driven-by-poverty thesis. This is still
debatable because the vast majority of poor folk around the world, even
when given opportunities, avoid risky illicit goods. Finally, worth stressing
is that early analyses of drugs in the Andes, such as Colombian economist
Francisco Thoumi’s classic 1992 essay, “Why the Illegal Psychoactive Drugs
Industry Grew in Colombia,” canonized the widespread notion that illicit
activities are largely a result of an absent, weak, fragmented, or politically
illegitimate state (Thoumi 1992), a thesis embraced by most NGOs, govern-
ments, and the United Nations—one this book hopes to challenge histori-
cally. Recent scholarship from Colombia is starting to recognize that drug
conflicts may have contributed to frontier state building as much as resulted
from its absence (Duncan 2014). Finally, an incipient rich area of Latin
American drug history is their “environmental impact,” but these works are
too early in gestation to survey here.
This Introduction will end here by bundling three potential implications of
studies like these (bearing in mind that Dávalos’s and Isacson’s chapters will
explicitly broach policy problems). First, this historical finding about the gen-
esis of Andean cocaine may have possible generalizable or comparative value.
To be sure, drug trades likely have a myriad of particular origins, from social
inequalities and post-colonial frontier frictions to conflict and smuggling cul-
ture zones, and historically debilitated or compromised states. But the pat-
tern identified here—as an aftermath of failed development—easily finds a
few notable analogues across Latin America and beyond. For example, many
corners of rural Mexico, besides proximity to contraband routes into the
United States, experienced the retreat of its “strong state” developmentalist
project—the 1930s national agrarian reform—by the 1970s and1980s, when
modern peasant drug cropping and rural-born trafficking intensified (Knight
2012). One Mexican anthropologist identifies the retreat of state-fostered
rural development in lowland “Tierra Caliente” Michoacán around Uru-
apan as the starting point to widespread ejido migrant turn to marijuana
growing and trafficking there in the 1970s–1980s (Maldonado 2013). It is
worth noting a big puzzle of comparative politics—that both abjectly “weak
state” Colombia and by Latin American standards exceptionally “strong
 Introduction  13
state” Mexico emerged as the region’s two leading narco-economies (i.e.,
Snyder and Durán-Martínez 2009). An intervening factor may have been
the local impacts of failed development sites and state-expansion projects,
which led peasantries into informal commercial outlets. Another example:
Paraguay’s modern cannabis export sector, now deeply entrenched in the
country’s politics and economy, took off from the 1960s–1970s bankruptcies
of firms (some American-owned) unable to compete with Brazilian coffee
in the northern Amambay and Canendiyu frontier with Brazil. Or Colom-
bia’s earlier bonanza marimbera (the “Colombian gold” marijuana boom of
the 1960s–1970s) in its northern Guajira peninsula was intertwined with
regional agro-industrial cotton projects and cattle elites (Britto 2015). Even
further afield, Lebanon’s Beqaa Valley, long a site of export-quality hash-
ish production, intensified its role during French colonial rule that brought
modern irrigation projects to the expansive valley during the 1920s–1930s.
This trade flourished for years under local political bosses after colonial-
ism, until the destruction of its national state in the 1970s turned Lebanon
into what some consider the world’s first modern “narco-state” (Marshall
2012). Finally, Helmand Province in southern Afghanistan, now the volatile
source of over half the world’s heroin poppy, offers a remarkable mirror
to Andean cocaine, notwithstanding the whirlwind of Soviet, Taliban, and
CIA politics around it. A recent historical study (Bradford 2018) clarifies the
strategic region’s descent from a model 1960s–1970s-style U.S. AID develop-
ment zone—with large-scale irrigation that accelerated labor migration for
cotton farming—to its later environmental salinity crisis and the country’s
1980s national political breakdown. Failed modernization was prelude to
Helmand’s 1990s take-off as a labor-intensive illicit drug export platform.
Second, these findings about the origins of Andean cocaine contest or
rather nuance a number of shibboliths about drug control and alternative
drug policies. First, illicit cocaine did not, as often presumed, erupt from
a pristine state of lawlessness or statelessness. In fact, its centers were
among the most state-affected areas of the Amazon Andes, and often mar-
keted through bustling urban hubs of entrepreneurial “modernity” such as
Medellín in Colombia (Roldán 1999), or Bolivia’s Santa Cruz de la Sierra.4
It was the retraction of state services, promises, and expectations—incident-
ally, a sociological factor offered for many armed insurgencies as well—that
led to rampant illicit activities and alternative forms of local governance
(Arjona 2016).5 Such rural “informality” is far less recognized than the
organization of the urban “informal sector” outside the state that drew the
rapt attention of Latin American sociology by the 1970s. Therefore, “build-
ing” the state and its legitimacy in such sociological “brown areas” (O’Don-
nell 2004) cannot be taken as a perfect, much less novel, recipe for arresting
drug trades, though it will likely help, particularly stable and legitimate
institutions of local justice.
Second, in light of the Andean drug war’s abject failure after forty years
of costly suppressive U.S. and national state policies, experts agree there is
14  Paul Gootenberg
no alternative to “alternative development.” But officials and stakeholders
seemed to have no consensus what this means and how it can be success-
fully implemented and maintained. The same caveat applies: these birthing
areas of coca trades were in fact relatively “modern,” “developed” zones of
the Andean Amazon, which then failed to deliver sustained livelihoods to
peasants. A larger paradox is that while “development” (in pointed quotes)
has repeatedly been declared “dead” for decades among a range of social
sciences from Economics to Anthropology (Escobar 1995; Ferguson 1990;
Lal 1983/2000; Rist 1997/2002), it is still being “discovered” anew and then
sanctified in drug control and global drug reformer circles alike. This para-
dox of timing aside—the 1960–1970s Age of Development is long past—new
commitments to “small-d” development need to be long, genuine, and just,
not simply the handover of peasant-opened lands to large companies and
extractive industries. It is likely affordable, however, because as economists
like Colombian Daniel Mejía have demonstrated repeatedly, the marginal
cost of eradicating coca as the central form of drug control is irrationally
astronomical (Mejía 2016). Finally, Liliana Dávalos and others will con-
tend that the long-term environmental history of these fragile eco-systems
where the Amazon meets the Andes—and where the world’s most valuable
bio-diverse rainforests have been turned into mediocre short-term harvests
and then low-productivity cattle pasture—shows that “development” per se
may also be wrongheaded, and should be contained as much as deepened.
In Peru, quality smallholder cacao is heralded as a solution but in parts of
lowland Colombia coca has been succeeded by capital-intensive and labor-­
exploitative agribusiness, such as corporate palm plantations. Better roads
and infrastructure should be carefully implemented. Policies should also
promote forest and biological conservation, parks, scientifically designed
wildlife corridors, and local stakes in biodiversity such as ecotourism.
Third, phenomena like the origins of Andean cocaine underline a num-
ber of understudied areas in history and the social sciences. For example,
they strongly suggest that licit and illicit economic, commercial, and social
worlds go hand in hand (Andreas 2013; Nordstrom 2007; van Schendel
and Abraham 2005). They cannot be studied well in moral separation, one
“good” the other “bad.” In the Amazon Andes, this is driven home by the
increasing blurred lines between conflict-ridden activities like drugs and
“illegal” mining (mainly gold), logging, gun-running, counterfeit products,
wildlife poaching, and the wider developmental and social impacts of elite
money laundering. These strict distinctions also make for poor policies, for
example, by criminalizing large swaths of local populations, such as the
cocaleros of Colombia and Peru (Bolivia’s now furthest along on a path to
decriminalization and Colombia pledged to legitimate these farmers under
the 2016 FARC-ERP peace accord). And while James C. Scott and others
have underscored the patent failures to peoples and environments of many
of the large-scale “high modernist” developmental projects of the twentieth
century (Scott 2000), under both socialist and capitalist regimes, the peopled
 Introduction  15
aftermaths of such failures get seemingly short-shrifted (James Ferguson’s
work on copper-belt Zambia is a salient exception) (Ferguson 1999), par-
ticularly those fostered by global market forces or the globally triumphant
Cold War. To return to the Amazon, what happened to the many other pri-
vate and public “Fordlandias” that litter the landscape of the Americas?
(Grandin 2010). The refugees of development are likely not just found in
illicit drugs.

Notes
1 Their original LASA16 essay focused on specifics of the Proyecto de Coloni-
zación Tingo María-Tocache-Campanilla and is now a forthcoming piece in the
Journal of Latin American Studies (Paredes and Manrique 2016).
2 For part of Dávalos’s extended UN-sponsored research (re-contextualized by
the UNODC) see: (www.researchgate.net/publication/304347613_The_world_
drug_problem_and_sustainable_development).
3 These varied theories exhibit, in other words, Hofstadter’s classic black and
white “paranoid style” of American political thought.
4 The urban locus of drug trafficking and national drug consumption is in many
cases also a complication of the modernization thesis. Medellín (Antioquia) was
Colombia’s modern industrial capital of the post-war era, with its factories and
workforce (many of them rural migrants) in depression by the 1970s; Cali was
also a rapidly expanding commercial gateway to the Pacific. Similarly, in Mexico,
Guadalajara, Mazatlán, and border cities like Ciudad Juárez and Tijuana were
anything but economic backwaters when they became centers of drug trades
in the 1980s, which fed off of rapidly growing, informalized workforces. Even
some notorious sites of urban drug consumption and violence (such as Mexican
shanty towns, Jamaican slums like Trenchtown, the Brazilian favelas of Rio) are
best read as refuge zones from mid-century urban renewal, political, or modern-
ization projects.
5 To distract from the argument here, if state projects are expanded to encompass
mobilizations and armed social movements from below, the pattern is even wider.
For example, the demobilized or failed 1980s revolutions of Central America in
Guatemala and El Salvador may link to the explosion of later regional criminal-
ity and illicit activities, including drugs (Nicaragua a telling exception). Similarly,
the rise of the VRAEM drug region in Peru (mostly in lowland Ayacucho) after
2000 may be a refugee aftermath of the 1990s failure of the Sendero Luminoso
revolt that originated in highland Ayacucho.

References
Andreas, Peter. 2013. Smuggler Nation: How Illicit Trade Made America. New York:
Oxford University Press.
Aramburú, Carlos E., Enrique Bedoya, and Jorge Recharte, eds. 1982. Colonización
en la Amazonía. Lima: Editorial CIPA.
Arjona, Ana. 2016. Rebelocracy: Social Order in the Colombian Civil War. Cam-
bridge: Cambridge University Press.
Beigel, Fernanda. 2010. “Dependency Theory: The Creation of a New Social Theory
in Latin America.” In The International Handbook on Diverse Sociological Tra-
ditions, edited by Sujata Patel, 188-200. London: Sage Press.
16  Paul Gootenberg
Bradford, James D. 2018 (forthcoming). Poppies, Politics, and Power: A Global
History of Drugs, Diplomacy, and Drug Policy in Twentieth-Century Century
Afghanistan. Ithaca, NY: Cornell University Press.
Brands, Hal. 2012. Latin America’s Cold War. Cambridge, MA: Harvard University
Press.
Britto, Lina. 2015. “Hurricane Winds: Vallenato Music and Marijuana Traffic in
Colombia’s First Illegal Drugs Boom.” Hispanic American Historical Review
95(1): 71–102.
Collier, Paul. 2007. “Economic Causes of Civil Conflict and their Implications for
Policy.” In Unleashing the Dogs of War: Conflict Management in a Divided World,
edited by Chester A. Crocker, Fen Osler Hampson and Pamela Aall, 197–218.
Washington, DC: ISIP Press.
Cooper, Frederick, and Randall Packard, eds. 1997. International Development and
the Social Sciences: Essays on the History and Politics of Knowledge. Berkeley,
CA: University of California Press.
Cooper, Neil. 2001. “Conflict Goods: The Challenges for Peacekeeping and Conflict
Prevention.” International Peacekeeping 8(3): 21–38.
Dávalos, Liliana M., Karina M. Sanchez, and Dolors Armenteras. 2016. “Deforesta-
tion and Coca Rooted in Twentieth-Century Development Projects.” BioScience
66(11): 974–82.
Duncan, Gustavo. 2014. Más que plata o plomo: El poder político del Narcotráfico
en Colombia y México. Bogotá: Random House-Debate.
Engerman, David, Nils Gilman, Mark H. Haefele, and Michael E. Latham, eds.
2003. Staging Growth: Modernization, Development and the Global Cold War.
Amherst, MA: University of Massachusetts Press.
Escobar, Arturo. 1995. Encountering Development: The Making and Unmaking of
the Third World. Princeton, NJ: Princeton University Press.
Felbab-Brown, Vanda. 2010. Shooting Up: Counterinsurgency and the War on
Drugs. Washington, DC: Brookings Institution Press.
Ferguson, James. 1990. The Anti-Politics Machine: “Development,” Depoliticization,
and Bureaucratic Power in Lethoso. New York: Cambridge University Press.
———. 1999. Expectations of Modernity: Myths and Meanings of Urban Life on the
Zambian Copper-Belt. Berkeley, CA: University of California Press.
García Jordán, Pilar. 2001. Cruz y arado, fúsiles y discusos: La construcción
de los orientes en el Perú y Bolivia, 1820–1920. Lima: Instituto de Estudios
Peruanos.
Gilman, Nils, Jesse Goldhammer, and Steven Webber, eds. 2011. Deviant Global-
ization: Black Market Economy in the 21st Century. New York: Continuum
Books.
Gootenberg, Paul. 2008. Andean Cocaine: The Making of a Global Drug. Chapel
Hill, NC: The University of North Carolina Press.
Gootenberg, Paul, and Isaac Campos. 2015. “Towards a New Drug History of Latin
America: A Research Frontier at the Center of Debates.” Hispanic American His-
torical Review 95(1): 1–36.
Grandin, Greg. 2010. Fordlandia: The Rise and Fall of Henry Ford’s Forgotten Jun-
gle City. New York: Picador Press.
Kenney, Michael. 2007. From Pablo to Osama: Trafficking and Terrorist Networks,
Government Bureaucracies, and Competitive Adaptation. University Park, PA:
Penn State Press.
 Introduction  17
Knight, Alan. 2012. “Narco-Politics and the State in Modern Mexico.” In Violence,
Coercion, and State-Making in Twentieth-Century Mexico, edited by Wil G.
Pransters, 115–34. Stanford, CA: Stanford University Press.
Lal, Deepak. 2000 (orig: 1983). The Poverty of Developmental Economics.
­Cambridge, MA: MIT Press.
Latham, David. 2000. Modernization as Ideology: American Social Science and
“Nation-Building” in the Kennedy Era. Chapel Hill, NC: University of North
Carolina Press.
León, Féderico, and Ramiro Castro de la Vega. 1989. Pasta básica de cocaína: Un
Estudio Interdisciplinario. Lima: CEDRO.
McCoy, Alfred W. 1991 (orig: 1972). The Politics of Heroin in Southeast Asia.
­Chicago, IL: Lawrence Books.
Maldonado Aranda, Salvador. 2013. “Stories of Drug Trafficking in Rural Mexico:
Territories, Drugs, and Cartels in Michoacán.” European Review of Latin Amer-
ican and Caribbean Studies 94: 43–66.
Marcy, William L. 2010. The Politics of Cocaine: How U.S. Foreign Policy has
­Created a Thriving Drug Industry in Central and South America. Chicago, IL:
Lawrence Hill Books.
Marshall, Jonathan V. 2012. The Lebanese Connection: Corruption, Civil War, and
the International Drug Traffic. Stanford, CA: Stanford University Press.
Mejía, Daniel. 2016 (orig: 2011). “Anti-Drug Policies Under Plan Colombia: Costs,
Effectiveness, and Efficiency.” In Anti-Drug Policies in Colombia: Successes, Fail-
ures, and Wrong Turns, edited by Alejandro Gaviria and Daniel Mejía, 51–65.
Nashville, TN: Vanderbilt University Press.
Merton, Robert K. 1936. “The Unanticipated Consequences of Purposeful Social
Action.” American Sociological Review 1(6): 894–904.
Meyer, Enrique. 2009. Ugly Stories of the Peruvian Agrarian Reform. Durham, NC:
Duke University Press.
Molano Bravo, Alfredo. 1987. Selva adentro: Una historia oral de la colonización del
Guaviare. Bogotá: Ancora Eds.
Morales, Edmundo. 1989. Cocaine: White Gold Rush in the Andes. Tucson, AZ:
University of Arizona Press.
Nordstrom, Carolyn. 2007. Global Outlaws: Crime, Money, and Power in the Con-
temporary World. Berkeley, CA: University of California Press.
O’Donnell, Guillermo. 2004. “The Quality of Democracy: Why the Rule of Law
Matters.” Journal of Democracy 15(4): 32–46.
Paredes, Maritza, and Hernán Manrique. 2016. “Desarrollismo estatal y auge del
cultivo de la coca en la Amazonía peruana.” Paper presented to LASA16 (New
York, May 2016). English version forthcoming in Journal of Latin American
Studies (2018).
Pike, Frederick B. 1977. The United States and the Andean Republics. Cambridge,
MA: Harvard University Press.
Rist, Gilbert. 2002 (orig: 1997). The History of Development: From Western Origins
to Global Faith. London: Zed Books.
Roldán, Mary. 1999. “Colombia: Cocaine and the Miracle of ‘Modernity’ in
Medellín.” In Cocaine: Global Histories, edited by Paul Gootenberg, 165–82.
London: Routledge UK.
Sanabria, Harry. 1993. The Coca Boom and Rural Social Change in Bolivia. Ann
Arbor, MI: University of Michigan Press.
18  Paul Gootenberg
Santos Granero, Fernando, and Féderica Barkley. 2000. Tamed Frontiers: Economy,
Society, and Civil Rights in Upper Amazonia. Boulder, CO: Westview Press.
van Schendel, Willem, and Itty Abraham, eds. 2005. Illicit Flows and Criminal
Things: States, Borders, and the Other Side of Globalization. Bloomington IN:
Indiana University Press.
Scott, James C. 2000. Seeing Like a State: How Certain Schemes to Improve the
Human Race Have Failed. New Haven, CT: Yale University Press.
Scott, Peter Dale, and Jonathan Marshall. 1988. Cocaine Politics: Drugs, Armies,
and the CIA in Central America. Berkeley, CA: University of California Press.
Smith, Michael L., ed. 1992. Why People Grow Drugs: Narcotics and Development
in the Third World. London: The Panos Institute.
Snyder, Richard. 2006. “Does Lootable Wealth Breed Disorder? A Political Economy
of Extraction Framework.” Comparative Political Studies 39: 367–99.
Snyder, Richard, and Angélica Durán-Martínez. 2009. “Drugs, Violence, and
State-Sponsored Protection Rackets in Mexico and Colombia.” Colombia Inter-
nacional 70: 61–91.
Tarazona-Sevellano, Gabriela, with John B. Reuter. 1990. Sendero Luminoso and the
Threat of Narco-Terrorism. Washington, DC: Center for Strategic and Interna-
tional Studies.
Thoumi, Francisco E. 1992. “Why the Illegal Psychoactive Drugs Industry Grew in
Colombia.” Journal of InterAmerican Studies and World Affairs 34(3): 37–63.
UNODC (United Nations Office on Drugs and Crime). 2016. World Drug Report
2016. New York: The United Nations.
Weimer, Daniel. 2011. Seeing Drugs: Modernization, Counterinsurgency, and
U.S. Narcotics Control in the Third World, 1969–1976. Kent, OH: Kent State
­University Press.
Werlich, David P. 1968. The Conquest and Settlement of the Peruvian Montaña.
Unpublished Ph.D. diss., University of Minnesota.
Wolf, Eric R. 1982. Europe and the People without History. Berkeley, CA: University
of California Press.
2 The ghosts of development
past
Deforestation and coca in
western Amazonia
Liliana M. Dávalos1

Introduction
Coca growers and their illicit crops have been described as a critical factor
in understanding the extent and location of deforestation in the Andean
fringe of the Amazon for decades (e.g., Dávalos et al. (2011); Young and
León (1999). From a purely economic perspective, however, disproportion-
ate deforestation by coca growers is puzzling. Quite the opposite, growing
a highly lucrative crop should lead to decreases in both the cultivated area
and the rate at which growers have to bring forested land into production
(Kaimowitz 1997). As most coca since the 1970s has been produced for the
illegal cocaine market, at least two other explanations have been proposed
for coca deforestation. First, coca cultivation and harvesting might attract
growers who would otherwise intensify production of other crops at already
developed sites to new, forested sites. Second, aggressive efforts to suppress
the crops force growers into remote sites that would otherwise remain
untouched. In both cases, the resulting deforestation increases because of
the illegal nature of coca. Hence, deforestation and environmental damages
in western Amazonia would arise from coca prohibition and not expansion
of agriculture, or not primarily because of this expansion.
There are high stakes for discovering and addressing the dominant factors
driving growers to both adopt coca and contribute to deforestation in the
western Amazon. Just the last twenty years, as the decades-long war against
coca in the region has intensified, have seen most coca cultivation shift from
the edges of the Amazon of Bolivia, Peru, and Colombia, first to every eco-
system of Colombia, then to Peru (Dávalos et al. 2009), then back to forest
frontiers of Colombia again. All the while, foci of production in the Amazon
persisted even as eradication investments boomed, and alternative devel-
opment programs to persuade growers to switch away from coca to other
crops multiplied. But even as the dynamics of coca cultivation shifted across
the Andes, deforestation in the western Amazon continued apace, sometimes
worsening in parallel with programs for eradication and alternative devel-
opment (Bradley and Millington 2008b).
Here, I review current research on the quantity and location of both
deforestation and coca cultivation in western Amazonia, finding illegal
20  Liliana M. Dávalos
crops explain little deforestation. Instead, the wedges of deforestation into
the Amazon lowlands have tracked roads and sites targeted for coloniza-
tion and development decades ago. While this approach cannot address the
history and idiosyncratic trajectory of particular sites, it outlines features
common to the process of agricultural expansion into the Amazon across
the northern Andean countries. Grounded on analyses of land use change,
the synthesis presented here focuses on the forests, their fragmentation and
loss across the region, complementing the localized case studies of the rest
of the book.

Background: tropical deforestation, the problem and models

Since the 1980s, international attention has focused on tropical forests as
rapidly disappearing ecosystems harboring the greatest biological diversity
on Earth (Myers et al. 2000). While the task of finding optimal areas for
conservation has produced a large literature on the ecology of these complex
systems (Brienen et al. 2015), a parallel quest for solutions has emerged,
strongly linked to the history, economics, and human geography of the Ama-
zon and similar regions (Barber et al. 2014; Dávalos et al. 2014; Fearnside
1993). Clearly, the accelerated change from vast, continuous, old-growth
tropical forests to agricultural uses since the 1950s is one of the main current
threats to global biodiversity (Laurance 1999; Laurance et al. 2012). As trop-
ical forests provide critical ecosystem functions through carbon sequestra-
tion, regulation of water and sediment flows, and soil nutrient cycling (Asner
et al. 2009; Hedin et al. 2003), the prospect of an Amazonia dominated by
agriculture raises concern about the stability of both global climate and the
water cycle. Currently, tropical deforestation and tropical fires contribute
substantial and increasing proportions of global carbon emissions (DeFries
et al. 2002; Harris et al. 2012), but a substantially denuded and drier Amazon
could even switch from a carbon sink that absorbs emissions to a net source
that accelerates climate change (Brienen et al. 2015; Nepstad et al. 2008).
If the goal is to maintain productive agricultural systems based on a stable
climate into the future, the continued stability of Amazonian forests is a
global priority.
As remote sensing data and computing power have become increasingly
available, analyses of satellite imagery have confirmed agriculture—and not
logging or similar extractive activities—as the main direct cause of Amazon
deforestation over the last fifteen years (Graesser et al. 2015; Gutiérrez-Vélez
et al. 2011). Still, identifying agriculture, and in particular the expansion
of pastures (e.g., (Chadid et al. 2015; McAlpine et al. 2009)), as the main
contributor to forest loss is unhelpful when designing long-term policies to
address deforestation, as both factors enabling change and structural drivers
of change remain intact. Instead it is more helpful to distinguish between
proximate causes and underlying drivers when considering the human activ-
ities influencing tropical deforestation (Geist and Lambin 2002). Proximate
 The ghosts of development past  21
causes directly change land use from forest to human uses, including mining,
logging, roads, and particular forms of agriculture (Fearnside 2005; Laura-
nce et al. 2002).
In contrast to the clear and large role of agriculture and agriculturalists
as proximate causes of deforestation, the underlying drivers of deforestation
are subject to much debate (Lambin et al. 2001). Identifying and under-
standing these drivers is vital, as national and global development policies
aim to reduce or at least not unduly increase deforestation (Angelsen and
Kaimowitz 1999). The debate on the underlying or ultimate causes of defor-
estation pits proponents of population growth as the ultimate driver of all
environmental degradation (e.g., Ehrlich and Ehrlich (2002)), against social
scientists who argue human ingenuity and adaptation tend to avert envi-
ronmental catastrophe as populations grow (e.g., Bhattarai and Hammig
(2001); Boserup (1965)). The Malthusian view of population invariably
expanding to match productivity thereby undermining any gains in wellbe-
ing (Malthus 1798), is the basis of demography and poverty as explanations
for tropical deforestation (Geist and Lambin 2001; Rudel and Roper 1997).
The evidence, however, suggests population and poverty result in tropical
deforestation only when accompanied by specific economic development
policies, social arrangements, cultural practices, and even beliefs (Geist and
Lambin 2001).
The evidence for a strong institutional influence—and against a strictly
Malthusian view—on how people respond to economic opportunities and
the consequences for land use and deforestation has accumulated only
recently (Geist and Lambin 2001; Lambin et al. 2001; Rudel and Roper
1997). For example, even in the pre-industrial era, the rate of conversion of
natural habitats to human use was lower than expected given population
growth, according to historical reconstructions of land use (Ellis et al. 2013).
These findings are replicated with data from tropical countries, collected
recently using remote sensing and comprising four decades of surveillance.
Those studies confirm a decoupling of agricultural productivity and habitat
change. While developing countries have increased agricultural production
~3.3–3.4 percent annually, deforestation has increased agricultural area by
only 0.3 percent each year, suggesting forest conversion plays a minor role
in productivity gains (Angelsen 2010). At the same time, if deforestation has
expanded agricultural area by only ~0.3 percent annually, higher deforesta-
tion rates such as those recorded for the Amazon require additional expla-
nations (Table 2.1).
Divergent conceptual models of deforestation can help explain the appar-
ent contradiction between high rates of land use change to agriculture in
the short term, and lower rates of long-term growth in agricultural extent.
While the models elide much variation from one country to the next (e.g.,
the oil-centered economies of Ecuador and Venezuela, or the history of
armed conflict and its relationship with urbanization for Colombia and
Peru), they provide a framework for relating the geography of deforestation
22  Liliana M. Dávalos
Table 2.1  Deforestation rates in countries with coca cultivation recorded since
2000 and neighbors without records over same period

Country Annual loss Deforestation Coca

km2 rate (percent) cultivation

Bolivia 2339 0.407 Yes

Brazil 25480 0.536 –
Colombia 2022 0.258 Yes
Ecuador1 422 0.235 –
Peru 1259 0.164 Yes
Venezuela 939 0.175 –

Source: Deforestation data from Hansen et al. (2013), coca production data from UNODC
(2015). Only areas with >50 percent tree cover were included in calculating the rate. Defor-
estation rates were calculated following the compound interest formula of Fearnside (1993).
Positive rates indicate forest loss.
Note:
1
Although both opium poppy and coca cultivation have been detected in Ecuador (UNODC
and Gobierno Nacional de la República del Ecuador 2015), the area detected during the study
period is negligible compared to its Andean neighbors.

to the ­history of economic development. These models, first outlined in ref-

erence to global tropical deforestation (Rudel and Roper 1997), relate forest
loss to economic development and poverty in distinct ways.

The immiseration model of deforestation

The engine of deforestation in the immiseration model is a growing popu-
lation of small farmers with limited access to the means of intensification,
who then expand agriculture into marginal lands at the expense of forests
(Rudel and Roper 1997). Although sometimes linked to slash-and-burn
agriculture (Myers 1993), slash and burn can be sustainable when tropical
forests are used in an impermanent manner for fewer years than the land
is fallowed (Harris 1971). Instead, the immiseration model requires pov-
erty both among agriculturalists and more broadly in the national economy
which fails to absorb workers (Walker 1993). For this model to explain the
gap between growth in agricultural land and deforestation requires the ulti-
mate collapse of marginal lands brought into production and their failure
to lead to long-term permanent agriculture. In contrast to traditional slash-
and-burn agriculture, which requires abundant forests to be sustainable,
the long-term footprint of deforestation from immiseration is degraded and
unproductive land where forests used to be. No capital or investment is nec-
essary for deforestation to take place, just an abundance of poor growers.

The frontier model of deforestation

When forested land is abundant (e.g., Figure 2.1), entrepreneurs, small
farmers, and companies work together or separately to develop a region
 The ghosts of development past  23

Figure 2.1 Stylized trajectory of forest fragmentation at agricultural frontiers, from

old-growth forests to rural fields

and open a frontier (Rudel and Roper 1997). Although population growth
and poverty do contribute to deforesting the frontier, insofar as landless
rural laborers and smallholders help exploit and settle the newly opened
lands (Rudel and Roper 1997), poor growers are not the driving factors of
forest loss (Fearnside 1993; Lambin et al. 2001). Instead, both state assis-
tance (e.g., through road construction) and private capital are necessary to
open predominantly forested lands to exploitation (Hecht 1985). Lacking
infrastructure to adequately enforce property rights (Angelsen 1999), the
leading edge of the frontier invites conflict over the land and its resources.
Often, the forests are quickly cleared to establish ownership, extract as
much of its natural resources as possible, or both (Fearnside 2005; South-
gate 1990). In contrast with the framework of immiserization, which would
predict a decrease in forest loss with investment (e.g., for intensification),
public or private investment at the frontier increases deforestation (Rudel
and Roper 1997).
If immiserization were the better explanation for western Amazon defor-
estation, then investment into the frontier would not be a necessary condi-
tion for deforestation and only the presence of large campesino populations
would be enough. In contrast, if deforestation in the region arose through
the opening of the frontier, then development plans and in particular road
construction would be indispensable for deforestation. In both cases, coca
deforestation would concentrate among poor growers and regions. Here, I
review the different studies on deforestation to systematically evaluate these
models in light of deforestation data.

Coca in the deforestation literature: coca cultivation as a

special force for deforestation
Without the advantage of detailed remote sensing analyses, early studies on
coca and deforestation highlighted its uniquely destructive potential. For
example Álvarez (2002) used back-of-the envelope calculations to estimate
roughly 50 percent of 1990s deforestation in Colombia could be attributed
to coca growers and their crops. In another example, an estimate of “several
million hectares of tropical forest” cleared by coca growers in the Andean
countries (Young 2004b) was accompanied by an urgent call for ­collecting
24  Liliana M. Dávalos
systematic data on deforestation in Peru (Young 2004a). Today, the data
needed to assess the extent of forest transformed into coca cultivation
have become available both through analyses conducted by the UNODC
(UNODC 2008; UNODC and Peru Ministerio del Medio Ambiente 2011),
and through studies undertaken by independent research groups including
illicit crop cultivation as one of multiple human land uses (Armenteras et al.
2013b; Chadid et al. 2015; Dávalos et al. 2011). These studies have focused
on direct and indirect deforestation from coca.

Direct deforestation
The surface area devoted to coca is small compared to other land uses
(Dourojeanni 1992), but this small area is viewed as an underestimate of
the deforestation resulting from cultivation (Young 1996). This is because
coca is seen as the cash crop of pioneering transformation, taking agricul-
ture to remote locales where cultivation would not occur otherwise (Álva-
rez 2001; Young 2004b; Young and León 2000). This encroachment into
old-growth forests is believed to then lead to further forest loss, as other
forms of agriculture expand next to the illicit crops. Ancillary uses leading
to deforestation include other (subsistence) crops, pastures, airstrips, roads
and dwellings (Álvarez 2002).
A systematic search for remote sensing analyses providing sufficient infor-
mation to estimate deforestation rates in coca-growing areas is summarized
in Figure 2.2, and reviewed chronologically. The study reaching the longest

Figure 2.2  Annual deforestation rates from legal crops or from coca cultivation
Sources: 1 Dávalos et al. (2011), 2 UNODC and Peru Ministerio del Medio Ambiente (2011),
3 Armenteras et al. (2013b), and 4 Chadid et al. (2015). All analyses correspond to the
­Amazon frontier except 4
 The ghosts of development past  25
into the past was the analysis of Landsat5 coverage for 1986, 1993, and
2007 for Pichis-Palcazú in the Amazonia of Peru (UNODC and Peru Minis-
terio del Medio Ambiente 2011). The express purpose of the study was not
only to quantify deforestation, but to determine the economic contribution
of different activities and their opportunity costs. Therefore, estimates of
the value of frontier agriculture included both coca cultivation and raising
cattle in pastures cleared from the forest. Additionally, illegal logging was
also mentioned as contributing to the deforestation by enabling traffickers
to launder illegal revenue. Compared to pastures, which made up 57 percent
of the area cleared of forest for human uses, coca cultivation was a minor
use at 0.39 percent of total. The estimated net value from coca cultivation
and cattle ranching were estimated to sum about US $4.6 million, while the
partial value of the standing value of woods from the deforested area was
estimated at over US $19 million. Hence the opportunity costs of alterna-
tive exploitation were roughly four times the revenue generated from fron-
tier agriculture, making frontier exploitation both economically wasteful
and environmentally unsustainable. Although the study does not directly
address it, the contradiction between great economic potential from carbon
sequestration or careful logging and the reality of encroaching agriculture
highlights the tension between the formal value of the forest and clearing
dynamics in the western Amazon frontier.
The second study compared Landsat5 from 2001 and 2009 to quantify
land use change for Guaviare in Colombia, and relate changes to the fires
(Armenteras et al. 2013b). Although deforestation associated with coca cul-
tivation in Guaviare has been documented since 1990, it is at least a decade
older (Arcila et al. 1999; UNODC 2010). Coca cultivation declined steadily
throughout the period of analysis, but mosaics with illicit crops had signifi-
cantly lower probability of reverting to forest (6.8 percent) than mosaics
dominated by pastures (13.5 percent). This was further corroborated by the
finding that coca-dominated mosaics had the lowest probability of illicit
crop plots reverting to forest at 14.3 percent, compared to 26.4 percent
probability for coca plots in forest-dominated mosaics. In short, coca does
not need to occupy much area to signal a transformation of the landscape
toward forest loss. Subsequent analyses of these data confirmed both the
decline of coca and the contribution of other land uses, particularly pas-
tures, to high deforestation rates in the most rapidly developing section of
Guaviare (Dávalos et al. 2014), as discussed on pp. 35–36.
The third study relied on land use data for Colombia generated by the
UNODC from 2002 to 2007 to estimate the influence of coca cultivation
as a catalyst of deforestation beyond its immediate surface area (Dávalos
et al. 2011), discussed on pp. 28–30. Three regions of the country were
analyzed, none of which correspond to a single biogeographic region. The
northern region comprised primarily Andean forest remnants of the Sierra
Nevada of Santa Marta and the Serranía del Perijá (Álvarez 2002). The cen-
tral region included Andean forests of the three Colombian cordilleras, and
26  Liliana M. Dávalos
in particular the Cordillera Central in San Lucas (Dávalos 2001), as well as
remnant lowland forests of the corresponding inter-Andean valleys and the
Chocó biogeographic region (moist tropical forests of the western slopes
of the Cordillera Occidental and lowlands abutting the Pacific Ocean from
southern Panama to northern Ecuador). The southern region included rem-
nants of Chocó and Andean forests and overwhelmingly comprised Amazo-
nian forests, especially at the colonization frontier. Along with the common
pattern of much higher deforestation rates from uses other than coca cul-
tivation, this study showed 4- to 20-fold higher deforestation rates in sec-
ondary forests than in old-growth stands. This is roughly consistent with
the 10-fold increase in deforestation rate for secondary forests compared to
old-growth stands found in the Guaviare study (Armenteras et al. 2013b).
The final study also used the UNODC layers for Colombia with a much
narrower focus on modeling forest loss in the Andean and sub-Andean for-
ests of San Lucas (Chadid et al. 2015). Beyond the 2002–2007 period, anal-
yses expanded to 2007–2010. Coca cultivation tended to expand in San
Lucas from 941 hectares recorded in 2002 to 6,013 hectares in 2010, and
this makes the region unlike other locations analyzed. Despite this differ-
ence, coca cultivation was still a minor land use, with 0.3 percent of land
use even at its maximum in 2010. This contrasts sharply to pastures going
from 9 percent of land use to almost 24 percent of land use in less than
one decade. For comparison, the Guaviare study also found “considerable”
pasture, from 8 percent to 10.3 percent in the 2001–2009 period (Armen-
teras et al. 2013b). The deforestation models generated for San Lucas also
provide some insights on key differences between coca cultivation and pas-
tures, including optimal intermediate distance to other crops, high distance
to settlements, cultivation on slopes, and proximity to rivers (Chadid et al.
2015). This is the first quantitative confirmation of the observation of coca
cultivation taking place in slopes growers would not use for other agricul-
ture (López Rodríguez and Blanco-Libreros 2008; Young 2004a, 2004b;
Young and León 1999), and to systematically compare coca and pasture
deforestation.
Comparisons of deforestation rates across studies show two clear pat-
terns. First, deforestation rates for agricultural uses other than illicit crops
are higher by one order of magnitude or more (Figure 2.2). The small direct
footprint of coca is highlighted in all source studies (Armenteras et al. 2013b;
Chadid et al. 2015; Dávalos et al. 2011; UNODC and Peru Ministerio del
Medio Ambiente 2011). This is also expected because during the period of
analyses, illicit crops have been monitored through remote s­ ensing, resulting
in smaller coca plots (UNODC 2008; UNODC and Gobierno de Colom-
bia 2013). Labor availability for harvesting leaves is thought to constrain
plot size on these productive systems (Kaimowitz 1997). The data reviewed
here are insufficient to test this potential explanation, although a study pur-
porting to test this effect found mixed results in Chapare, Bolivia (Bradley
and Millington 2008b). Regardless of the mechanism, coca replaces only a
 The ghosts of development past  27
small fraction of the forest. The conversion of forests for other land uses is
what produces high deforestation rates in each of these agricultural fron-
tiers. These uses are thought to be associated with coca cultivation through
the activities of coca growers as agents of deforestation (contrasting with
commercial logging, for example).
Second, deforestation rates are higher for secondary forests than for old-
growth forests (Figure 2.2). Overall deforestation rates up to 6 percent have
been observed at sites in Santa Cruz, Bolivia (Steininger et al. 2001), with
rates from 1.2–4.5 percent historically more common at now-denuded low-
lands of Colombia, and peaking at 7.8 percent when secondary forests are
included (Etter et al. 2006b). By separating rates for old-growth and sec-
ondary forests, a pattern of high turnover for fallowed regeneration plots
becomes evident (Figure 2.2). In the agricultural frontier, secondary forests
are the result of previous human intervention, and their presence implies
an earlier process of land use change (Guariguata and Ostertag 2001). This
process of regrowth is therefore concentrated at the forest frontier, as almost
all the analyses highlight (Armenteras et al. 2013b; UNODC and Peru Min-
isterio del Medio Ambiente 2011), including by estimating the probability
of regrowth to be highest at the forest frontier (Dávalos et al. 2011). Even
when a secondary forest has regrown for several years, its physical character-
istics differ from old-growth forests. These differences include lower canopy
heights, lower biomass, and lower biodiversity (Guariguata et al. 1997; Lau-
rance 2015). The process of forest fragmentation in the agricultural frontier
proceeds more frequently toward greater fragmentation and physical sepa-
ration between patches of forest than toward regeneration (Figure 2.1). It
is also easier to access, light fires, and further fragment these fragmented
landscapes than large, unbroken stands of old-growth forests (Armenteras
et al. 2013b; Dávalos et al. 2014; Etter et al. 2006b; Fahrig 2003). In line
with these historical and physical considerations, deforestation rates from
coca cultivation were higher for secondary forests as well, with the single
exception of the Central Colombia region analyzed by Dávalos et al. (2011)
(Figure 2.2).
In conclusion, and contrary to some news headlines, coca causes little
direct deforestation. Measurements reveal coca replaces a minimum of for-
est along the agricultural frontier, amounting to one-tenth or more often
much less of the total transformation. These areas are not losing forest only
or mainly because they have coca. Instead, the high rates of loss of second-
ary forests suggest these sites correspond to the agricultural frontier where
colonization and migration only began over the last few decades (Dávalos
et al. 2011; Etter et al. 2006b, 2008; Young and León 1999). The use of
higher-slope terrain for coca cultivation where other crops are not grown
(Chadid et al. 2015), confirms decades-old claims using statistical analyses
(López Rodríguez and Blanco-Libreros 2008; Young 2004a, 2004b; Young
and León 1999), and indicates one uniquely unsustainable characteristic of
coca. Coca is grown on slopes where growers choose to plant nothing else.
28  Liliana M. Dávalos
The hypothesis that coca itself attracts growers to these sites and drives oth-
erwise nonexistent deforestation is discussed below.

Indirect deforestation
Analyses of direct deforestation from coca cultivation show this crop leads
to relatively small clearings. Besides the high economic return per hect-
are (Bradley 2005; Kaimowitz 1997), the multi-year productive cycle of
coca and its productivity despite replanting at the same site can reduce the
fragmentation and deforestation effects of this crop (Salisbury and Fagan
2011; Salisbury 2007). This last finding does not address the hypothesis of
coca as particularly destructive because of its unique tendency to promote
land use change in remote areas, or to attract colonists (Álvarez 2001,
2003; Young 2004b; Young and León 2000). There are two ways to eval-
uate this argument. The first is by demonstrating deforestation observed in
with areas influenced by coca is somehow related to coca, and not just the
result of pioneering or colonist agriculture in general. This is difficult to
document because coca cultivation concentrates along existing coloniza-
tion fronts in all three Andean countries (Andrade 2004; Etter et al. 2005,
2006a; Fajardo 2004; UNODC 2010, 2014; UNODC and Peru Ministe-
rio del Medio Ambiente 2011). The second way of evaluating this claim
requires comparing deforestation rates from sites influenced by coca to
those where coca is minimal or absent. If coca cultivation is uniquely dam-
aging, then deforestation rates in affected regions should exceed those of
unaffected regions (other things being equal, meaning along agricultural
frontiers).
There was a single study attempting to isolate the unique effect of illicit
crops as catalysts of forest loss throughout the landscape (Dávalos et al.
2011). The effect of coca cultivation was measured in two ways: as the
distance to the nearest coca plot, and as coca cultivation present per kilo-
meter square. If coca cultivation were a unique catalyst of land use change,
then the probability of a forest pixel converting to any human use should
decrease with distance to coca and increase with the quantity of cultivation
in the larger area. A series of landscape variables usually associated with
the probability of deforestation were also included: the proportion of for-
est remaining (Ewers 2006), distances to roads and rivers (Laurance et al.
2009; Mahecha et al. 2002; Viña et al. 2004), biophysical characteristics
related to agriculture in general such as climate, slope, and aspect (Etter et
al. 2006c), and the protection status of the land (Barber et al. 2014). The
results of models accounting for spatial autocorrelation inherent to the
landscape data showed the expected effect of coca cultivation in southern
Colombia, but not in the northern or central region (Dávalos et al. 2011).
Those results show for every two pixels of forest of any type converting
to human use in southern—mostly Amazonian—Colombia, 98 stay the
same during the 2002–2007 period. But when the quantity of coca in the
 The ghosts of development past  29
surrounding kilometer square increased by 2 hectares, only 84 stayed the
same. In contrast, when the distance to the nearest coca plot increased by
15 kilometers, 222 pixels stayed the same. The change in probability of
losing a forest pixel behaved as expected if coca was indeed a unique cat-
alyst of forest loss in the landscape.
Two additional results of Dávalos et al. (2011) merit discussion. First,
no similar landscape effect was demonstrated for northern and central
Colombia, despite the extent of coca cultivation and presumed association
with deforestation in both regions (Figure 2.3) (Chadid et al. 2015; Dávalos
2001; UNODC 2008). The large number of pixels sampled ensures this
result was not caused by low statistical power. Instead, this implies coca
did not behave as a special catalyst and instead was just one more crop
in agricultural colonization fronts. Second, analyses of deforestation rates
using municipalities found no evidence that the quantity of new coca cul-
tivation in 2002–2007 resulted in higher deforestation rates. This result
shows effects detectable across the landscape do not scale up to political
units for which socioeconomic data become available, and this will become
important when discussing analyses modeling deforestation at the subna-
tional scale.
Instead of finding coca cultivation (or eradication) as a factor explaining
deforestation rates, Dávalos et al. (2011) found gaining population density
increased rates in municipalities with new coca during the period. The 267
remaining municipalities in the sample showed no such pattern. This effect
could not be explained by coca attracting colonists: new coca cultivation
was unrelated to changes in population density. The authors interpreted

Figure 2.3 Summary of deforestation rates in areas or at times without illicit crops,

or when coca cultivation was present
Source: See Table 2.2
30  Liliana M. Dávalos
these results as evidence that coca cultivation was a symptom, and not the
ultimate cause of deforestation (p. 1225):

[W]e hypothesize that what sets coca-growing municipalities apart is

poor rural development. Gains in rural population density relate to
higher deforestation rates because most or all economic activities that
absorb immigrants, or used to occupy emigrants, require forest clearing.
Municipalities without new coca would have a diverse suite of economic
activities to accommodate population growth, so that the relationship
between population and deforestation breaks down. […] The expansion
of coca itself is an indication that these municipalities constitute the
agricultural frontier, where settled land ends and new inroads begin.
[…] Coca is expanding in these municipalities because they are under-
developed, rather than the converse. Coca is therefore a symptom rather
than the ultimate cause of deforestation, and structural features such as
socioeconomic inequality, failed agricultural development policies, and
armed conflict are the large-scale drivers of deforestation.

Subsequent analyses of MODIS imagery from 2001 to 2010 support this last
interpretation (Sánchez-Cuervo et al. 2013). Those analyses modeled land use
change as a function of a suite of biophysical and socioeconomic variables,
including climate, accessibility by road or river, changes in human population
density, poverty, changes in coca cultivation, displacement, and the activities
of armed groups. While the activities of armed groups explained forest loss
in particular ecoregions, changes in coca cultivation did not explain changes
in land use change at any spatial scale (Sánchez-Cuervo et al. 2013).
Another study from Colombia focused exclusively on deforestation rates
at different stages of colonization in the Meta/Guaviare colonization front
(Table 2.2, (Rodríguez et al. 2012)). Satellite imagery data from early periods
are difficult to parse, but coca has influenced land use change in the region
since the 2000s, may date back to the 1980s (Molano 1989), and was defi-
nitely present by 1990 (UNODC 2010). The key finding was deforestation
rates increase along the gradient of human influence, from lowest to highest
for indigenous settlements, colonist frontier, transition zones and settlement
zones (as proposed in Figure 2.1). Settlement zones have deforestation rates
100-fold greater than indigenous settlement areas, and transition zones have
10-fold greater rates of deforestation than colonist frontiers (Table 2.2). Despite
lacking a quantitative assessment of the influence of coca cultivation, coca agri-
culture was proposed as influencing both stages with the highest rates of land
use change: transition zones and settlement zones (Rodríguez et al. 2012).
A series on Colombian deforestation from Landsat imagery in 1985 and
2005 (Armenteras et al. 2011, 2013a) further evaluated the relationship
between coca cultivation and deforestation rates. The relevant analyses
encompassed all ecoregions but the Andes, including Amazonia (Armen-
teras et al. 2013a). While the goal in each case was to identify the drivers
 The ghosts of development past  31
Table 2.2  Local and regional rates of forest loss with and without coca in ­Andean
countries, in chronological order

Coca Annual Deforestation Period Country Source

presence loss (ha) rate (percent)

No 1,877 1.727 1973–1985 Colombia (Viña et al. 2004)

No 1,764 1.099 1973–1985 Ecuador (Viña et al. 2004)
No 7 0.224 1975–1983 Bolivia (Bradley 2005)
Yes 39 3.854 1975–1986 Bolivia (Bradley 2005)
Yes 2,500 0.026 1976–1986 Bolivia (Killeen et al. 2007)
No 0 0.000 1976–1986 Bolivia (Killeen et al. 2007)
Yes 5,000 0.250 1976–1986 Bolivia (Killeen et al. 2007)
No 6,700 0.110 1976–1986 Bolivia (Killeen et al. 2007)
Yes 1,200 0.021 1976–1986 Bolivia (Killeen et al. 2007)
Yes 32,900 0.175 1976–1986 Bolivia (Killeen et al. 2007)
Yes 2,800 0.213 1976–1986 Bolivia (Killeen et al. 2007)
No 59 4.836 1983–1986 Bolivia (Bradley 2005)
No 151 4.907 1983–1986 Bolivia (Bradley 2005)
Yes 2212 2.566 1985–1996 Colombia (Viña et al. 2004)
No 1,356 0.974 1985–1996 Ecuador (Viña et al. 2004)
Yes 0.041 1985–2002 Colombia (Rodríguez et al. 2012)
Yes 0.172 1985–2002 Colombia (Rodríguez et al. 2012)
Yes 1.993 1985–2002 Colombia (Rodríguez et al. 2012)
Yes 3.684 1985–2002 Colombia (Rodríguez et al. 2012)
Yes 27,420 0.747 1985–2005 Colombia (Armenteras et al. 2011)
Yes 46,477 0.634 1985–2005 Colombia (Armenteras et al. 2011)
Yes 108 4.117 1986–1992 Bolivia (Viña et al. 2004)
Yes −18 −3.061 1986–1993 Bolivia (Viña et al. 2004)
Yes 8 0.766 1986–1993 Bolivia (Viña et al. 2004)
Yes 13,400 0.137 1987–1991 Bolivia (Killeen et al. 2007)
No 0 0.000 1987–1991 Bolivia (Killeen et al. 2007)
Yes 21,800 1.118 1987–1991 Bolivia (Killeen et al. 2007)
No 3,800 0.063 1987–1991 Bolivia (Killeen et al. 2007)
Yes 9,600 0.166 1987–1991 Bolivia (Killeen et al. 2007)
Yes 87,000 0.472 1987–1991 Bolivia (Killeen et al. 2007)
Yes 2,300 0.179 1987–1991 Bolivia (Killeen et al. 2007)
Yes 46,346 0.608 1990–2002 Colombia (UNODC 2010)
Yes 31,524 0.421 1990–2005 Colombia (Armenteras, et al. 2013a)
Yes 33,822 0.216 1990–2005 Colombia (Armenteras et al. 2013a)
Yes 50,260 0.333 1990–2005 Colombia (Armenteras et al. 2013a)
Yes 125,785 0.260 1990–2005 Colombia (Armenteras et al. 2013a)
Yes 112 5.677 1992–1996 Bolivia (Bradley 2005)
Yes 8,000 0.083 1992–2000 Bolivia (Killeen et al. 2007)
No 900 0.059 1992–2000 Bolivia (Killeen et al. 2007)
Yes 10,100 0.542 1992–2000 Bolivia (Killeen et al. 2007)
No 5,100 0.085 1992–2000 Bolivia (Killeen et al. 2007)
Yes 3,000 0.052 1992–2000 Bolivia (Killeen et al. 2007)
Yes 122,900 0.680 1992–2000 Bolivia (Killeen et al. 2007)
Yes 700 0.055 1992–2000 Bolivia (Killeen et al. 2007)
Yes 84,250 0.211 1992–2004 Bolivia (Müller et al. 2012)
Yes 29,667 0.074 1992–2004 Bolivia (Müller et al. 2012)
Yes 43,083 0.108 1992–2004 Bolivia (Müller et al. 2012)

(Continued)
32  Liliana M. Dávalos
Table 2.2  (Continued)

Coca Annual Deforestation Period Country Source

presence loss (ha) rate (percent)

Yes 35 3.55 1993–1996 Bolivia

(Bradley 2005)
Yes 45 6.276 1993–1996 Bolivia
(Bradley 2005)
Yes 40,000 4.1 1996–1999 Colombia
(Etter et al. 2006a)
Yes −50 −8.591 1996–2000 Bolivia
(Bradley 2005)
No 40 2.621 1996–2000 Bolivia
(Bradley 2005)
Yes 57 6.463 1996–2000 Bolivia
(Bradley 2005)
No 1,071 2.032 2000–2005 Peru (UNODC 2014)
No 1,382 0.308 2000–2005 Peru (UNODC 2014)
No 1,517 0.273 2000–2005 Peru (UNODC 2014)
No 1,554 1.250 2000–2005 Peru (UNODC 2014)
Yes 1,771 0.514 2000–2005 Peru (UNODC 2014)
No 2,167 0.687 2000–2005 Peru (UNODC 2014)
No 2,246 0.892 2000–2005 Peru (UNODC 2014)
No 2,813 2.430 2000–2005 Peru (UNODC 2014)
Yes 4,014 0.607 2000–2005 Peru (UNODC 2014)
Yes 4,464 0.673 2000–2005 Peru (UNODC 2014)
Yes 20,800 0.216 2001–2004 Bolivia
(Killeen et al. 2007)
No 400 0.027 2001–2004 Bolivia
(Killeen et al. 2007)
Yes 23,800 1.335 2001–2004 Bolivia
(Killeen et al. 2007)
No 4,100 0.069 2001–2004 Bolivia
(Killeen et al. 2007)
Yes 8,800 0.153 2001–2004 Bolivia
(Killeen, et al. 2007)
Yes 160,800 0.940 2001–2004 Bolivia
(Killeen, et al. 2007)
Yes 5,900 0.464 2001–2004 Bolivia
(Killeen et al. 2007)
Yes 10,867 0.154 2002–2009 Colombia
(UNODC 2010)
No 863 1.637 2005–2010 Peru (UNODC 2014)
No 1,236 0.994 2005–2010 Peru (UNODC 2014)
Yes 1,587 0.460 2005–2010 Peru (UNODC 2014)
No 1,944 0.617 2005–2010 Peru (UNODC 2014)
No 1,977 0.356 2005–2010 Peru (UNODC 2014)
No 1,977 0.440 2005–2010 Peru (UNODC 2014)
No 2,064 1.783 2005–2010 Peru (UNODC 2014)
No 3,042 1.208 2005–2010 Peru (UNODC 2014)
Yes 3,979 0.601 2005–2010 Peru (UNODC 2014)
Yes 4,878 0.736 2005–2010 Peru (UNODC 2014)
Yes −18,8478 −0.3255 2011–2010 Colombia
(Sánchez-Cuervo
et al. 2012)
Yes 325,356 2.1416 2011–2010 Colombia (Sánchez-Cuervo
et al. 2012)

Note: Deforestation rates were calculated following the formula of Fearnside (1993). Positive
rates indicate forest loss.
1
Rate for indigenous settlement zones (Rodríguez et al. 2012).
2
Rate for colonist frontier zones (Rodríguez et al. 2012).
3
Rate for transition zones (Rodríguez et al. 2012).
4
Rate for settlement zones (Rodríguez et al. 2012).
5
This is the rate for the woody vegetation category (Sánchez-Cuervo et al. 2012).
6
This is the rate for the mixed woody/non woody vegetation category, probably comprising
secondary growth (Sánchez-Cuervo et al. 2012).
 The ghosts of development past  33
of deforestation, the quantity of coca cultivation and its variation between
regions was a special focus of the discussion. Additionally, those analyses
controlled for multiple factors known to influence deforestation such as
the extent of pastures and legal crops, temperature and precipitation, road
density, and rural and urban population. Surprisingly, the amount of coca
cultivation was a positive covariate of deforestation in the Caribbean and
Orinoco regions, but not in Amazonia (Armenteras et al. 2013a). To explain
the finding that coca cultivation was not a covariate of deforestation rates in
Amazonia, the authors discuss two alternatives (p. 1191):

One possible explanation is that the effect of illicit crops in Amazonia was
not captured by our model or either they have smaller impact on total
deforestation than previously expected […]. Alternatively, there might
be a link between rural population density, deforestation, and an illegal
economy such that, as Dávalos et al. (2011) suggest coca growing might
be a consequence (attractor) of poverty and not a cause of deforestation.

Two important conclusions follow from these subnational analyses. First,

and in line with the municipality analyses of Dávalos et al. (2011) and Sán-
chez-Cuervo et al. (2013), the effect of coca on the landscape in Amazonia
does not scale up to the levels of political units or disappear when demo-
graphic and socioeconomic factors are included. Second, coca cultivation
was a covariate of deforestation in two regions: the Caribbean and Ori-
noco. Importantly, the Orinoco region excluded the vast frontier along the
east Andes surrounding the Picachos, Macarena, and Tinigua parks—which
corresponds roughly to the Ariari colonization front studied in Chapter 5—
which are analyzed instead as part of Amazonia (Armenteras et al. 2013a).
In other words, coca cultivation was not correlated to deforestation rates in
the regions with the most coca.
Another way of demonstrating the unique properties of coca in causing
deforestation would be to show that deforestation rates influenced by coca
are higher than what they would be without coca cultivation. Studies of defor-
estation from Andean countries with sufficient data to determine whether
or not coca influences deforestation rates are summarized in Table 2.2.
Data were available for Bolivia, Colombia, Ecuador, and Peru.
The earliest study was the analysis of Landsat coverage for 1973, 1985,
and 1996 along the Colombia–Ecuador border, or the Putumayo coloniza-
tion wedge (Viña et al. 2004). Although the deforestation rate in Colombia
almost doubled the rate of Ecuador for the first period and almost trebled
it during the later period (Table 2.2), the causes of these differences are
unclear. For the first period, the higher Colombian rates were attributed
to higher colonization pressures from oil exploitation (Wesche 1968), and
coca cultivation during the second period. But there was no quantitative
evidence to support this last explanation. The patterns of deforestation,
however, are indicative. In Ecuador deforestation followed roads, forming a
34  Liliana M. Dávalos
clear ­“herring-bone” spatial pattern, while in Colombia a wedge formed at
the road’s terminus early on, resulting in concentric and outwardly expand-
ing agricultural plots. For the purpose of comparing results to other studies
(Figure 2.3), the data from Ecuador were coded as having no influence from
coca although the region is tightly interconnected.
The second earliest study analyzed land use change using Landsat images
from 1975, 1985, 1992 or 1993 and 2000 at three sites in Chapare, Bolivia
(Bradley 2005). Uniquely among all studies analyzed, Bradley (2005) con-
ducted interviews to ascertain the decision-making process of colonists
regarding land use. As a result, published sections of this dissertation are
among the few comparisons of deforestation rates during different anti-coca
regimes (Bradley and Millington 2008a, 2008b). At each of the three sites,
three periods were demarcated: pre-coca, coca-dominant, and post-coca dom-
inant. The last two periods are designated as influenced by coca in Table 2.2
and Figure 2.3. The interviews also helped determine the immediate factors
motivating deforestation agents, including sale prices of local agricultural
commodities (cattle and milk, coca, bananas, pineapple, oranges, and heart
of palm), and government enforcement of anti-coca policies. Although the
general conclusion is that laissez-faire approaches to coca cultivation gener-
ated less deforestation than alternative development projects, there was high
variance in deforestation rates, as illustrated in Figure 2.3.
In contrast with this result, another study from Bolivia found coca per-
missiveness increased deforestation rates. Killeen et al. (2007) examined
deforestation for 7 departamentos (departments) using Landsat imagery
from 1975/1976, 1986/1987, 1991/1992, 2000/2001 and 2004/2005. Addi-
tionally, subsequent analysis disaggregated potential agents of land use
change by agricultural sector (Killeen et al. 2008). During the entire period
mechanized agriculture and cattle ranching were identified as key drivers
of rapid rise in deforestation rates in Santa Cruz (Killeen et al. 2007). By
cross-referencing coca cultivation reports, all departamentos except Pando
and Chuquisaca were influenced by the illicit crops category in Table 2.2
and Figure 2.3 (UNODC and Estado Plurinacional de Bolivia 2011). The
relationship between rates and agriculture in general is analyzed in detail
below.
One additional analysis from eastern Bolivia between 1992 and 2004 dis-
criminated between direct causes of deforestation by mechanized agricul-
ture, smallholder production, or pastures for cattle (Müller et al. 2012). The
study noted the regional importance of coca cultivation, particularly among
local smallholders in the Chapare (Tejada et al. 2016). Based on the timing
and spatial location of those analyses, all estimates were assigned to the
influenced by coca category of Table 2.2 and Figure 2.3.
A few additional studies provided sufficient information to estimate
deforestation rates in Colombia when coca might influence these measure-
ments (Etter et al. 2006a; Sánchez-Cuervo et al. 2012). The earliest of these
estimates examined waves of unplanned deforestation in Caquetá for the
 The ghosts of development past  35
1982–2002 period, with the peak deforestation rate of 4.1 percent (Table 2.2)
reported for the 1996–1999 period (Etter et al. 2006a). This peak deforesta-
tion rate was attributed in part to coca cultivation, although without quan-
titative evidence. Finally, Sánchez-Cuervo et al. (2012) analyzed MODIS
imagery for the 2001–2010 period, finding overall gain in woody vegetation
(or forest-rich landscapes) and loss of mixed woody and non-woody vegeta-
tion, which would be roughly equivalent to secondary growth. Eight of the
10 municipalities recording the greatest loss of wood vegetation correspond
to areas in the Orinoco basin, and oil exploration and exploitation is one of
the explanations proposed for these deforestation outliers (Sánchez-Cuervo
et al. 2012).
Two analyses conducted by the UNODC, one in Colombia (UNODC
2010), and another in Peru (UNODC 2014), provided extensive informa-
tion on deforestation along coca-producing colonization fronts. The goal
of the first study was to give policymakers better information on the bio-
physical, socioeconomic, and security aspects in regions affected by coca
cultivation. Those analyses focused on the ecological transition between the
Orinoco basin and Amazonian forests in the departments of Meta/Guaviare,
a region experiencing the rapid conversion of forests to pastures and, to a
much lesser extent, coca cultivation (Table 2.2) (Armenteras et al. 2013b;
Dávalos et al. 2014). The UNODC analyses found a 4-fold higher deforesta-
tion rate for 1990–2002 than for 2002–2009 (Table 2.2). Coca was initially
concentrated along the Andean slopes at the western end of the region, and
by the end of the study period, cultivation had shifted to the more isolated
moist savannas in the easternmost flank of Meta. The study highlights two
parallel and seemingly contradictory dynamics:

[C]oca cultivation presents two simultaneous and antagonistic pro-

cesses. At one end, rural consolidation spreading from population cen-
ters occupying a zone of 1,045,000 ha; at the other, a colonizing front
progressively taking over the Amazonian that currently ecosystem occu-
pies 1,500,000 ha. That zone requires action to limit effects on strategic
ecosystems.

The first dynamic corresponds to the integration of former forests (and for-
mer coca cultivation) into the urban land markets of emergent, regionally
important cities (Dávalos et al. 2014). These areas were identified by the
UNODC as low risk for coca cultivation (UNODC 2010), despite corre-
sponding to centers fringed by large clusters of cultivation in the 1990s (Lee
and Clawson 1993). The declining trends in coca cultivation in those zones
strongly relate to the growing fraction of the local population living in the
core cities, signaling intensifying urbanization and greater importance to
the regional economy (Dávalos et al. 2014). These changes signal a com-
plete rearrangement of the landscape from one almost 50 percent forested
in 2000, to one comprising vast open areas of low-productivity pastures
36  Liliana M. Dávalos
in 2010 (Dávalos et al. 2014). These would also constitute settlement
zones with the greatest rates of deforestation of Rodríguez et al. (2012).
Although Rodríguez et al. (2012) mentioned coca cultivation—­ without
­quantification—as an important component of the landscape in settlement
zones, the decade-long trend in the San José-Calamar axis of Guaviare
instead suggests coca cultivation is declining (Dávalos et al. 2014). There is
no contradiction between Rodríguez et al. (2012) and Dávalos et al. (2014),
as coca cultivation is indeed present but declining. At the forested edges of
the newly consolidated rural spaces lies the forest frontier, from which new
waves of frontier agriculture depart along the large and navigable rivers.
These zones correspond to the second dynamics, harboring more than 70
percent of the total coca cultivation (UNODC 2010). Using the classifica-
tion of Rodríguez et al. (2012), these are transition zones between new or
early colonist territories and settlement areas.
The UNODC study of San Martín, Peru, aimed to analyze economic
effects from alternative development programs and other productive ini-
tiatives relative to deforestation and coca cultivation (UNODC 2014). In
contrast with the Meta/Guaviare study, a key feature of San Martín is the
decline in coca production since its peak in the 1980s, despite sudden jumps
in production recorded in 2004 and 2010. San Martin comprises the middle
of the Huallaga River Valley, abutting the Andes to the west and extending
into Amazonian lowlands to the east. The natural vegetation encompasses
a gradient from subtropical montane forests along the eastern flank of the
Andes to Amazonian lowland forests to the east and south. Based on prelim-
inary analyses conducted by Conservation International, the report included
forest cover for each of the provinces of the region for the 2000–2005 and
2005–2010 periods. By revenue, the top licit products were rice, concen-
trated in the central Huallaga Valley, coffee in the mid elevations of the val-
ley to the north, and plantain, presumably in the lowlands. Coca cultivation
and deforestation were tallied as losses in economic analyses. The net bal-
ance was negative, with losses exceeding revenue between 2002 and 2011
by almost a factor of 2 (total revenue of US $2,900 million, losses of US
$5,300). There were no quantitative analyses on the relationships between
agricultural uses (including coca cultivation) and deforestation. Neverthe-
less, low-productivity cattle ranching (1 head per hectare) was deemed a
key driver of deforestation during the last decade. Two large deforestation
fronts were evident based on the data: an inter-Andean valley front predom-
inantly associated with licit agriculture and the road network in northern
San Martin, and another pushing northward from the south along the upper
Huallaga River. This last front overlaps with areas of varying density of coca
cultivation.
Plotting the different deforestation rates highlights three patterns (Table 2.2,
Figure 2.3). First, the greatest variance in deforestation rates as well as high-
est rates of forest loss were recorded for smaller areas in Bolivia (Bradley
2005). This is likely related to the history of fragmentation of Chapare gen-
erating small forest patches from deforestation spreading outward along
 The ghosts of development past  37
the Cochabamba–Santa Cruz road (Millington et al. 2003). This landscape
configuration makes further land cover change easier and faster than larger
contiguous forest blocs. As discussed before, such a dynamic landscape will
also experience regrowth in abandoned coca plots and fallows (Dávalos
et al. 2011). The highly dynamic landscape then explains the variance cor-
responding to forest gains in places and at times in Bolivia. Another pattern
is the great majority of records corresponding to deforestation instead of
regrowth. Only a few records from Chapare in Bolivia, and the MODIS
analysis of woody vegetation in Colombia show regrowth (Table 2.2). The
final pattern is lower deforestation rates during periods influenced by coca
in Colombia and Peru, but not Bolivia (Figure 2.3). If coca were a unique
catalyst of deforestation, then coca-influenced records should correspond to
high deforestation rates, but the opposite trend is evident in Colombia and
Peru. This is consistent with models by several authors who proposed coca
generates less deforestation than expected from other crops (Kaimowitz
1997). The only country fitting the prediction of higher deforestation rates
when coca is part of the agricultural frontier is Bolivia (Figure 2.3).
To summarize: claims of coca as a promoter of deforestation beyond that
expected at the forest frontier in western Amazonia region are at odds with
almost all the data. A single study detected the effect of coca, only in south-
ern Colombia, and likely because relevant socioeconomic variables such
as changes in population density were unavailable at the relevant spatial
scale (Dávalos et al. 2011). Once socioeconomic characteristics—including
variables related to economic development, roads, and armed conflict—are
included, analyses of independently collected data show coca cultivation
fails to explain variation in deforestation rates in the Amazonian region
most affected by this type of agriculture (Armenteras et al. 2011, 2013a;
Dávalos et al. 2011; Sánchez-Cuervo et al. 2013). Instead, most analyses
bolster the interpretation of Dávalos et al. (2011): the presence of coca
is an indicator or symptom of the conditions of the agricultural frontier.
These conditions, and the model of extractive development they embody,
both drive deforestation rates associated with immigration and provide the
medium for coca cultivation.
Analyses showing that Bolivian departamentos without coca cultiva-
tion have lower deforestation rates are also potentially consistent with this
hypothesis (Killeen et al. 2007, 2008). Both Bolivian studies lacked demo-
graphic and economic covariates, or the amount of coca cultivation as a fac-
tor on deforestation rates: the pattern of Figure 2.3 may correspond to the
forest frontier actively attracting migration for extractive activities at times
when coca and/or another factor fuels the regional economy.

Discussion: frontier deforestation dynamics

in the Andean region
Although impoverished farmers are often cited as a key factor in ­Amazon
deforestation (Myers 1993), deforestation and coca cultivation in the
38  Liliana M. Dávalos
Andean region cannot be explained without reference to the agricultural
frontier, and its ecological and socioeconomic conditions. The resource fron-
tier model helps explain several vexing features of Amazonian deforestation
in the Andean region, and its coca-related manifestation. First, it provides
a geographic focus. Even in Colombia, the country with the most ecologi-
cally diverse distribution of coca, cultivation is mostly restricted to the last
remnants of mostly forested natural habitats at the agricultural frontiers.
The exceptions (e.g., in remote outposts in Guaínia; small concentration
of coca in Pando; Bolivia in the 2000s (UNODC and Estado Plurinacional
de Bolivia 2011)) are, without migration, short-lived and easy to eradicate
compared to the large clusters at the ecotone of the Andes and the Amazo-
nian lowlands. The coca clusters at this transition in the Ariari of Colom-
bia, Huallaga and Apurímac in Peru, and Yungas de la Paz and Chapare in
Bolivia persist to this day.
Second, it helps explain why, though surrounded by a seemingly extraor-
dinary bounty of forest products with the potential to yield great benefits if
managed sustainably (e.g., UNODC and Peru Ministerio del Medio Ambiente
(2011)), the final land use tends to be pastures for extensive cattle ranching
(Hecht 1993). Opening the frontier to agriculture requires investment (often
public, as outlined above), and always involves personal risks for smallhold-
ers. Once at the frontier, and as long as territorial control is weak, managing
the productivity of agricultural lands is more expensive than opening a new
frontier. This creates progressive encroachment into Amazonian lowlands,
sometimes perceived as the result of illegal drug prohibition (McSweeney
2015). The advancement of the frontier, however, continues to take place in
countries entirely lacking coca cultivation, as in Ecuador or Brazil (Graesser
et al. 2015; Rodrigues et al. 2009; Rudel et al. 2002). In short: the frontier
continues to advance as the older deforested areas either become commer-
cial and population centers in their own right—as in San José del Guaviare
and El Retorno in Colombia and Santa Cruz in Bolivia—or decline as their
population migrates to cities or further afield (Carr 2009; Hecht et al. 2015).
Finally, the frontier model helps explain why the closing of the forest
frontier, when the near-complete transformation of the landscape has played
out in a region, also signals the decline of coca cultivation. This is some-
times incorrectly interpreted as the result of anti-coca policies (Dávalos
et al. 2014). Instead, it relates to smallholders dependent on coca migrating
(to cities or other frontiers), while formerly forested lands become proper-
ties for investment in an emergent, now better-connected region (Dávalos et
al. 2014; Rudel et al. 2002). This process may involve land grabs and a great
deal of violence (Fergusson et al. 2014; Salisbury and Fagan 2011). Newly
settled agricultural lands where state control and property rights remain
fluid provide opportunities to forcibly take the land, a scarce and increas-
ingly valuable resource (Borras et al. 2012; Lambin and Meyfroidt 2011).
These are generalizations and many objections can be raised, but the
frontier model has distinct advantages over its alternative for ­understanding
 The ghosts of development past  39
deforestation in western Amazonia. To understand these advantages requires
first reviewing the role of coca in discussions on deforestation in the Andean
countries, and examining the evidence on this purported role. These dis-
cussions on coca and its role in deforestation provide the background to
the central thesis of this chapter: that government investment in opening
the western frontier of Amazonia played a decisive role in the subsequent
onslaught of the deforestation as well as the establishment of coca.

The human geography of the Amazonian frontier of the Andes

If not coca, then what factors explain the location and rates of deforestation
in western Amazonia? At the center of this book is the history of Andean
colonists at the Amazonian forest frontier, a topic of longstanding inter-
est in the social sciences (e.g., Crist and Nissly (1973). This crucial history,
however, tends to be overlooked by studies of land use change (e.g., Etter
et al. (2008), but see Young and León (1999)) even though it is indispensable
to understand both the location and extent of transformation of western
Amazonia.
To summarize a vast literature: the Andean nations of Colombia, Ecua-
dor, Peru and Bolivia coordinated efforts to develop road infrastructure into
their Amazonian lowlands with the ultimate—and still unachieved—goal
of interconnecting the Andean section of the Amazon basin from Venezuela
to Bolivia (Denevan 1966). The hemispheric Declaration of the Presidents
of America in Punta del Este in 1967 crystalized the scope and ambition of
this massive development plan (Meeting of American Chiefs of State 1967).
The goals of laying the foundation for economic integration by complet-
ing the Carretera Marginal de la Selva and modernizing agricultural food
production through development, agrarian reform, and land settlement are
the most relevant to land use change in the declaration (Meeting of Ameri-
can Chiefs of State 1967). These goals were soon bolstered by international
support and, indeed, development funds and multilateral loans became
conditioned on reforms to achieve a solution to the political challenge of
landless campesinos (INCORA 1974a). The goal of developing and settling
Amazonia, however, did not begin with this declaration. Instead, the vision
embodied by the Carretera Marginal de la Selva was in itself a culmination
of processes begun decades earlier within Andean nations.
The Carretera Marginal de la Selva was a long-held goal of presidential
candidate Fernando Belaúnde Terry of Peru, an important factor in his 1963
election (Denevan 1966). By October 1963 the government of Peru started
spearheading the Marginal de la Selva as an Andean initiative in meetings
with ministers from Colombia, Ecuador, and Bolivia. Construction of Peru’s
Marginal de la Selva started in 1966, and was supposed to complement a
network of 28 planned access roads from the Andes into the lowlands. The
top priority for construction was the Tarapoto highway linking Tarapoto up
the Huallaga River Valley to Juanjui and connecting to the road branching
40  Liliana M. Dávalos
south from Tingo María, thereby opening the entire Huallaga Valley to agri-
culture with the expectation of allocating up to 7 million hectares of land and
accommodating no fewer than 1.5 million settlers (Denevan 1966; Young
and León 1999). At the time, this road was seen as the Peruvian answer to
Bolivia’s success in building the first Cochabamba–Santa Cruz road in the
1950s, which ushered migration and a series of agricultural booms—very
much including illegal coca—that persist even today (Gallup et al. 2003).
In contrast with these projects led by the central government, efforts to
improve access into the Amazon frontier in Colombia and Ecuador were,
at least at first, private or undertaken by local governments. As early as
the 1940s, rubber companies and local governments developed often-failing
roads along the Ariari River from Villavicencio to Calamar through San
José, Guaviare (Molano 1989). In southern Colombia, oil companies helped
improve the trail from Nariño to the upper Caquetá and upper Putumayo
rivers (Wesche 1968). It was only until the 1960s, and directly connected to
settlement programs, that national resources were deployed to improve the
roads (INCORA 1974a). As in southern Colombia, Ecuador’s road into the
Amazon was built with support from oil companies, but in contrast with the
Colombian example, Ecuador’s colonization projects focused on the Pacific
and north of the Andes, not the east (Schuurman 1979). In sum, by the time
the Carretera Marginal de la Selva was proposed as a hemispheric project,
all Andean countries had already built some infrastructure to reach their
Amazonian foothills and to eventually reach the lowlands, even as the scope
of the infrastructure and its economic targets varied (Figure 2.4).
Both Peru’s Marginal de la Selva and Bolivia’s Cochabamba-Santa Cruz
roads transformed local landscapes by attracting settlers who fragmented
the forest for agriculture and whose products were now more accessible
to the Andean core (Denevan 1966; Young and León 1999). Even the less
ambitious Colombian access roads had similar effects, accelerating the for-
merly slow process of clearing and colonization (Brücher 1968; Etter et al.
2006b, 2008). Transforming the entire region into a wedge of colonization,
however, required the hemispheric goal of building the Marginal de la Selva,
which released hitherto unavailable international financing, and provided a
focus for agrarian reform programs just as actions to expand the agricultural
land base became urgent (Crist and Nissly 1973). Despite the then known
shortcomings of Amazonian soils for continued cultivation (Denevan 1966),
opening the vast Amazonian forests to campesino cultivation was seen as
one of the keys to meeting the political clamor for land reform, increasing
agricultural productivity, and relieving pressure from mass migration into
cities. Agricultural development based not on intensification but on exten-
sive clearing followed by steady production promised to also secure domestic
food supplies for each expanding nation (Schuurman 1979; Wesche 1968).
The access roads facilitated migration into the already existing towns at
the foothills of the Andes, where most of the agricultural development was
expected to concentrate (Denevan 1966). But the same forces that made
 The ghosts of development past  41

Figure 2.4 Left: planned Carretera Marginal de La Selva connecting the Colom-

bia–Venezuela border to Santa Cruz. Major centers of illegal coca pro-
duction. Right: illegal coca cultivation in the Andean countries for 2014
Sources: Brücher (1977), Lee and Clawson (1993), UNODC (2015)

campesinos migrate from the Andes—demographic growth, vastly inequita-

ble land distribution in the Andes, and pervasive lack of capital and credit—
pushed farmers farther into the Amazonian frontier. As early as the beginning
of the 1970s, land grabs for the most fertile plots, along with soil erosion
following deforestation had emerged as crucial challenges (INCORA 1974b).
Colonization projects to complement the roads had been planned from the
beginning, but with the massive influx of colonists and aided by state-spon-
sored calls for migration into the internal frontier, setting up the projects
gained new urgency. Although originally the projects aimed to direct coloni-
zation, most often they followed the migration flows of Andean farmers seek-
ing land, a new start, or fleeing violence (Brücher 1968; Jülich 1975; Maass
1969; Schoop 1970; Wesche 1968). Table 2.3 summarizes the major coloni-
zation projects in the Amazon frontier of the Andean countries, including the
smaller projects undertaken in Ecuador (Schuurman 1979). The major proj-
ects were concentrated in Colombia, Peru, and Bolivia. Colombia opened
projects in Meta along the Ariari, at El Retorno in Guaviare, near Floren-
cia Caquetá, and Puerto Asís, Putumayo (Brücher 1968, 1977; Schuurman
1978, 1979). Peru opened projects in Alto Marañon, Tingo María-Tocache
in the Huallaga Valley, and Apurímac, in 1978 at Pichis-­Palcazú, as well
as smaller research-oriented projects in Jenaro ­Herrera and ­Caballococha
42  Liliana M. Dávalos
Table 2.3  Government-sponsored colonization projects in western Amazonia,
1960s–1970s

Project Country Location Name Brücher Name

(1977) Schuurman
(1978)

Ariari-Güéjar Colombia Ariari, Meta — Meta

Guaviare Colombia San José, S. José El Retorno
Guaviare
Caquetá Colombia Florencia, Caquetá Caquetá
Caquetá
Putumayo Colombia Puerto Asís, Puerto —
Putumayo Leguízamo
Lago Agrio Ecuador not mapped — Shushufindi
Payamino Ecuador not mapped — Payamino
Palora-Pastaza Ecuador not mapped — Palora Pastaza
Upano Ecuador not mapped — Upano vallei
Morona Ecuador not mapped — San José de
Morona
Alto Marañón Peru Alto Marañón, Alto Marañón Alto Marañon
Marañón
Jenaro Herrera Peru Jenaro Herrera, Genaro Herrera Jenaro Herrera
Loreto
Middle Peru South of Tingo Mittl. Huallaga Tingo María-
Huallaga María, Tocache
Leoncio Prado
Pichis-Palcazú Peru Puerto Bermudez, — —
Oxapampa
Apurímac river Peru San Francisco, Apurímac Apurímac
Ayacucho
Alto Beni Bolivia Yungas de la Paz Alto Beni Alto Beni
Chapare Bolivia Chapare, Chapare Chimoré
Cochabamba
Santa Cruz Bolivia West of Santa Sta. Cruz Yapacaní
Cruz de la
Sierra

(Reategui and Taminche 1980; Schuurman 1978, 1979; UNODC and Peru
Ministerio del Medio Ambiente 2011). Bolivia opened projects in Santa
Cruz, Chapare, and Alto Beni (Schoop 1970).
Based on contemporary accounts of the colonization projects most
directly associated with the Marginal de la Selva road (locations mapped in
Table 2.3), Dávalos et al. (2016) tested the spatial relationship between coca
cultivation in 2014 and the projects of the 1970s (Figure 2.4). Despite more
than four decades separating the projects from the contemporary distribu-
tion of illegal coca in the Amazon, spatial models using only the distance
from the projects can accurately predict the location of coca cultivation.
This demonstrates persistent spatial clustering in spite of many multi-lateral
efforts to eradicate coca. Coca cultivation to date still clusters around the
 The ghosts of development past  43
colonization projects, which can be traced to at least the 1980s (Figure 2.4).
The association between the colonization centers and coca can help explain
why this crop is invariably part of the western Amazon deforestation fron-
tier: both large-scale deforestation and coca share a common origin with
the mass immigration from the Andes facilitated by the roads and at least
partially supported by the projects.
The migrant waves of Andean campesinos associated with the coloni-
zation projects encountered poor and incomplete roads, sparse—if any—­
infrastructure, and greater challenges to agriculture than in the Andes
(Clawson 1982; Schuurman 1979). Although some colonization projects
succeeded in directing colonization and legitimating land claims, agricultural
credit was scarce and in any case most farmers were unfamiliar with it, or
with lowland tropical agriculture. In the face of these challenges, colonists
traded labor for access to cleared land, and moved on as fertility declined
thus establishing a cycle of frontier clearing that continues to this day. Even
very early on, it became clear colonization projects were not delivering on the
promises of more equitable land distribution or even food security (INCORA
1974a). With high transport costs, few products of the Amazon frontier
could compete with the yields from Andean farms. Not coincidentally, these
beachheads of Amazonian colonization from the Andes in Colombia, Peru
and Bolivia, also became centers of coca cultivation (Figure 2.4).
Perhaps it could not have been any other way and the coca/deforestation
frontier would have emerged with or without the development vision that
built the roads and the colonization projects. After all, the coca leaf to supply
the cocaine that became a consumption trend in the 1970s and 1980s had
to be grown somewhere. There is, however, and important counterfactual in
the trajectory of colonization, deforestation, and agriculture of the region:
Ecuador. Both Brücher (1977) and Schuurman (1979) discounted the Ecua-
dorian projects east of the Andes—as opposed to those along the Pacific—as
being too small and disconnected from Andean markets to accomplish their
goal of attracting settlers. In the Ecuadorian Oriente, only the Upano Valley
project was linked to the Marginal de la Selva, and it was hindered by the
poor state of the road from Cuenca to Limón (Schuurman 1979). Other
projects in the Oriente, such as the Lago Agrio, focused on providing sup-
port for colonists along the single and oil extraction road (Schuurman 1978,
1979). The road itself was completed fairly late, by 1971, in contrast with
the earlier completion of state-sponsored access roads in both Bolivia and
Peru. Deforestation radiates out of this road and its later tributaries, and
contrasts with the wedge pattern of nearby Putumayo in Colombia (Viña
et al. 2004; Wesche 1968). The colonization projects in eastern Ecuador had
neither the scope nor the agricultural focus of those in the other Andean
countries linked through the Marginal del la Selva, and the resulting defor-
estation also differs in pattern and extent. This example suggests the massive
influx of Andean campesinos enabled by the roads and colonization projects
was a necessary condition for the creation of the coca frontier.
44  Liliana M. Dávalos
Conclusion
The forest frontier of western Amazonia opened through the development
of access roads allowing mass migration into the foothills of the northern
Andes. Although local and private efforts to build roads into the Ama-
zon had been underway for decades, the Carretera Marginal de la Selva
unleashed international financing specifically for projects to direct and aid
colonists, further boosting the attraction of the region to upland Andean
campesinos. In time, neither the social benefits of more equitable access to
land and credit, nor the steady production for Andean markets were realized
on the scale needed for the frontier to become prosperous or self-sustain-
ing. Instead, new waves of frontier colonization launched from the foothills
deeper into the Amazon, as land tenure concentrated near towns swelled
from migration. The combination of high transport costs and low produc-
tivity in the challenging tropical environment led to uncompetitive agricul-
ture. By this time, investment in both the roads and support for the colonists
dwindled, while campesinos adopted coca cultivation for the burgeoning
illegal market. Hence the apparently puzzling finding that coca cultivation is
a poor predictor of deforestation rates and yet seems to be present at almost
every forest frontier in the western Amazon has a simple explanation. Both
coca and deforestation are the result of a grand twentieth-century modern-
izing effort to develop this vast region whose consequences are visible today,
even from space.

Note
1 This work was partially supported by a commission from the Gesellschaft für
Internationale Zusammenarbeit (GIZ) to L.M.D. I thank A. Corthals for Ger-
man-language translations; A. Corthals, E. Dávalos, E. Lauterbur, M. Lim, J.
Retana, and L.R. Yohe for comments; and D. Armenteras, A. Bejarano, and L.
Correa for discussions on ruralization. For data, I thank C. Bussink, L. Correa,
M. Gualdrón, A. Korenblik, L. Vallejos and A. Vella from the United Nations
Office on Drug and Crime (UNODC). All opinions are exclusively the author’s
and do not reflect GIZ or UNODC positions.

References
Álvarez, María D. 2001. Forests under fire. NACLA Report on the Americas
35:29–34.
———. 2002. Illicit crops and bird conservation priorities in Colombia. Conserva-
tion Biology 16(4):1086–1096.
———. 2003. Forests in the time of violence: conservation implications of the
Colombian war. Journal of Sustainable Forestry 16(3–4):49–70.
Andrade, Germán I. 2004. Selvas sin Ley. Conflicto, drogas y globalización de
la deforestación de Colombia. In Guerra, Sociedad y Medio Ambiente. M.
Cárdenas and M. Rodríguez, eds. 106–173. Bogotá, Colombia: Foro Nacional
Ambiental.
 The ghosts of development past  45
Angelsen, Arild. 1999. Agricultural expansion and deforestation: modelling the
impact of population, market forces and property rights. Journal of Development
Economics 58(1):185–218.
———. 2010. Policies for reduced deforestation and their impact on agricultural produc-
tion. Proceedings of the National Academy of Sciences, USA 107(46):19639–19644.
Angelsen, Arild and David Kaimowitz. 1999. Rethinking causes of deforestation:
lessons from economic models. World Bank Research Observer 14:73–98.
Arcila, Oscar Hernando, Gloria Ines Gonzalez, and Carlos Ariel Salazar. 1999. Gua-
viare: población y territorio. Bogotá: Tercer Mundo Editores.
Armenteras, Dolors, N. Rodriguez, and J. Retana. 2011. Understanding deforesta-
tion in montane and lowland forests of the Colombian Andes. Regional Environ-
mental Change 11(3):693–705.
———. 2013a. National and regional determinants of tropical deforestation in
Colombia. Regional Environmental Change 13(6):1181–1193.
———. 2013b. Landscape dynamics in northwestern Amazonia: an assessment of
pastures, fire and illicit crops as drivers of tropical deforestation. PLoS ONE
8(1):e54310.
Asner, Gregory P., Thomas K. Rudel, T. Mitchell Aide, and Ruth Defries. 2009.
A contemporary assessment of change in humid tropical forests. Conservation
Biology 23(6):1386–1395.
Barber, Christopher P., M. A. Cochrane, C. M. Souza, and W. F. Laurance. 2014.
Roads, deforestation, and the mitigating effect of protected areas in the Amazon.
Biological Conservation 177(0):203–209.
Bhattarai, Madhusudan, and Michael Hammig. 2001. Institutions and the environ-
mental kuznets curve for deforestation: a crosscountry analysis for Latin Amer-
ica, Africa and Asia. World Development 29(6):995–1010.
Borras, Saturnino M., J. Franco, S. Gómez, C. Kay, and M. Spoor. 2012. Land
grabbing in Latin America and the Caribbean. The Journal of Peasant Studies
39(3–4):845–872.
Boserup, Ester. 1965. The Conditions of Agricultural Growth: The Economics of
Agrarian Change Under Population Pressure. London: Earthscan Publications.
Bradley, Andrew V. 2005. Land-use and land-cover change in the Chapare region of
Bolivia. PhD thesis, University of Leicester.
Bradley, Andrew V. and Andrew Millington. 2008a. Agricultural land-use trajecto-
ries in a cocaine source region: Chapare, Bolivia. In Land Change Science in the
Tropics. W. Jepson and A. Millington, eds., pp. 231–250. New York: Springer.
Bradley, Andrew V. and Andrew C. Millington. 2008b. Coca and colonists: quantify-
ing and explaining forest clearance under coca and anti-narcotics policy regimes.
Ecology and Society 13(1):31.
Brienen, Roel J. W., O. L. Phillips, T. R. Feldpausch, E. Gloor, T. R. Baker, J. Lloyd,
G. Lopez-Gonzalez, A. Monteagudo-Mendoza, Y. Malhi, S. L. Lewis, R. Vasquez
Martinez, M. Alexiades, E. Alvarez Davila, P. Alvarez-Loayza, A. Andrade, L. E.
O. C. Aragao, A. Araujo-Murakami, E. J. M. M. Arets, L. Arroyo, G. A. Aymard C,
O. S. Banki, C. Baraloto, J. Barroso, D. Bonal, R. G. A. Boot, J. L. C. Camargo, C.
V. Castilho, V. Chama, K. J. Chao, J. Chave, J. A. Comiskey, F. Cornejo Valverde,
L. da Costa, E. A. de Oliveira, A. Di Fiore, T. L. Erwin, S. Fauset, M. Forsthofer,
D. R. Galbraith, E. S. Grahame, N. Groot, B. Herault, N. Higuchi, E. N. Honorio
Coronado, H. Keeling, T. J. Killeen, W. F. Laurance, S. Laurance, J. Licona, W. E.
46  Liliana M. Dávalos
Magnussen, B. S. Marimon, B. H. Marimon-Junior, C. Mendoza, D. A. Neill, E. M.
Nogueira, P. Nunez, N. C. Pallqui Camacho, A. Parada, G. Pardo-Molina, J. Pea-
cock, M. Pena-Claros, G. C. Pickavance, N. C. A. Pitman, L. Poorter, A. Prieto, C.
A. Quesada, F. Ramirez, H. Ramirez-Angulo, Z. Restrepo, A. Roopsind, A. Rudas,
R. P. Salomao, M. Schwarz, N. Silva, J. E. Silva-Espejo, M. Silveira, J. Stropp, J.
Talbot, H. ter Steege, J. Teran-Aguilar, J. Terborgh, R. Thomas-Caesar, M. Toledo,
M. Torello-Raventos, R. K. Umetsu, G. M. F. van der Heijden, P. van der Hout,
I. C. Guimaraes Vieira, S. A. Vieira, E. Vilanova, V. A. Vos, and R. J. Zagt. 2015.
Long-term decline of the Amazon carbon sink. Nature 519(7543):344–348.
Brücher, Wolfgang. 1968. Die Erschliessung des tropischen Regenwaldes am Ostrand
der kolumbianischen Anden: der Raum zwischen Río Ariari und Ecuador. Tübin-
gen: Im Selbstverlag des Geographischen Instituts del Universität Tübingen.
———. 1977. Formen und effizienz staatlicher Agrarkolonisation in den Östlichen Reg-
enwaldgebieten der tropischen Andenländer. Geographische Zeitschrift 65(1):3–22.
Carr, David. 2009. Population and deforestation: why rural migration matters. Prog-
ress in Human Geography 33(3):355–378.
Chadid, Maria, L. M. Dávalos, J. Molina, and D. Armenteras. 2015. A Bayesian spatial
model highlights distinct dynamics in deforestation from coca and pastures in an
Andean biodiversity hotspot. Forests 6(11):3828.
Clawson, David L. 1982. Obstacles to successful highlander colonization of the Ama-
zon and Orinoco basins. American Journal of Economics & Sociology 41(4):351.
Crist, Raymond E. and Charles M. Nissly. 1973. East from the Andes. Gainesville:
University of Florida Press.
Dávalos, Liliana M. 2001. The San Lucas mountain range in Colombia: how much
conservation is owed to the violence? Biodiversity and Conservation 10(1):69–78.
———. Adriana C. Bejarano, and H. Leonardo Correa. 2009. Disabusing cocaine:
pervasive myths and enduring realities of a globalised commodity. International
Journal of Drug Policy 20(5):381–386.
———. A. C. Bejarano, M. A. Hall, H. L. Correa, A. P. Corthals, and O. J. Espejo.
2011. Forests and drugs: coca-driven deforestation in tropical biodiversity
hotspots. Environmental Science & Technology 45(4):1219–1227.
———., J. S. Holmes, N. Rodríguez, and D. Armenteras. 2014. Land, not beef mar-
kets, drives deforestation and fragmentation in northwestern Amazonia. Biolog-
ical Conservation 170:64–73.
———. Karina M. Sanchez, and Dolors Armenteras. 2016. Deforestation and coca culti-
vation rooted in twentieth-century development projects. BioScience 66(11):974–982.
DeFries, Ruth S., Richard A. Houghton, Matthew C. Hansen, Christopher B. Field,
David Skole, and John Townshend. 2002. Carbon emissions from tropical defor-
estation and regrowth based on satellite observations for the 1980s and 1990s.
Proceedings of the National Academy of Sciences, USA 99(22):14256–14261.
Denevan, William M. 1966. Carretera marginal de la selva and the central Huallaga
region of Peru. Geographical Review 56:440–443.
Dourojeanni, M. J. 1992. Environmental impact of coca cultivation and cocaine pro-
duction in the Amazon region of Peru. UN-ODCCP Bulletin on Narcotics 2:37–53.
Ehrlich, Paul R. and Anne H. Ehrlich. 2002. Population, development, and human
natures. Environment and Development Economics 7(01):158–170.
Ellis, Erle C., Jed O. Kaplan, Dorian Q. Fuller, Steve Vavrus, Kees Klein Goldewijk,
and Peter H. Verburg. 2013. Used planet: a global history. Proceedings of the
National Academy of Sciences, USA 110(20):7978–7985.
 The ghosts of development past  47
Etter, Andrés, C. McAlpine, D. Pullar, and H. Possingham. 2005. Modeling the age of
tropical moist forest fragments in heavily-cleared lowland landscapes of Colom-
bia. Forest Ecology and Management 208(1–3):249–260.
Etter, Andrés, C. McAlpine, S. Phinn, D. Pullar, and H. Possingham. 2006a. Unplanned
land clearing of Colombian rainforests: spreading like disease? Landscape and
Urban Planning 77(3):240–254.
———. 2006b. Modelling the conversion of Colombian lowland ecosystems since 1940:
drivers, patterns and rates. Journal of Environmental Management 79(1):74–87.
———. 2006c. Regional patterns of agricultural land use and deforestation in
Colombia. Agriculture, Ecosystems & Environment 114(2–4):369–386.
Etter, Andrés, Clive McAlpine, and Hugh Possingham. 2008. Historical patterns
and drivers of landscape change in Colombia since 1500: a regionalized spatial
approach. Annals of the Association of American Geographers 98(1):2–23.
Ewers, Robert M. 2006. Interaction effects between economic development and
forest cover determine deforestation rates. Global Environmental Change
16(2):161–169.
Fahrig, Lenore. 2003. Efects of habitat fragmentation on biodiversity. Annual Review
of Ecology, Evolution, and Systematics 34(1):487–515.
Fajardo M., Dario. 2004. El conflicto armado y su proyección en el campo. In
Guerra, Sociedad y Medio Ambiente. M. Cárdenas and M. Rodríguez, eds.
67–105. Bogotá, Colombia: Foro Nacional Ambiental.
Fearnside, Philip M. 2005. Deforestation in Brazilian Amazonia: history, rates, and
consequences. Conservation Biology 19(3):680–688.
——. 1993. Deforestation in Brazilian Amazonia: the effect of population and land
tenure. Ambio 22(8):537–545.
Fergusson, Leopoldo, Dario Romero, and Juan F. Vargas. 2014. The environmen-
tal impact of civil conflict: the deforestation effect of paramilitary expansion in
Colombia. Documento CEDE 2014(36).
Gallup, John Luke, Alejandro Gaviria, and Eduardo Lora. 2003. Is Geography Des-
tiny?: Lessons from Latin America: Stanford University Press.
Geist, Helmut J., and Eric F. Lambin. 2001. What Drives Tropical Deforestation?
0959–3780. Barcelona, Spain: LUCC International Project Office.
———. 2002. Proximate causes and underlying driving forces of tropical deforesta-
tion. Bioscience 52(2):143–150.
Graesser, Jordan, T. M. Aide, H. R. Grau, and N. Ramankutty. 2015. Cropland/pas-
tureland dynamics and the slowdown of deforestation in Latin America. Envi-
ronmental Research Letters 10(3). DOI: 034017.
Guariguata, Manuel R., and Rebecca Ostertag. 2001. Neotropical secondary forest
succession: changes in structural and functional characteristics. Forest Ecology
and Management 148(1–3):185–206.
——., R. L. Chazdon, J. S. Denslow, J. M. Dupuy, and L. Anderson. 1997. Structure
and floristics of secondary and old-growth forest stands in lowland Costa Rica.
Plant Ecology 132(1):107–120.
Gutiérrez-Vélez, Víctor H., R. DeFries R, M. Pinedo-Vásquez, M. Uriarte, C. Padoch,
W. Baethgen, K. Fernandes, and Y. Lim. 2011. High-yield oil palm expansion
spares land at the expense of forests in the Peruvian Amazon. Environmental
Research Letters 6(4):044029.
Hansen, M. C., P. V. Potapov, R. Moore, M. Hancher, S. A. Turubanova, A. Tyu-
kavina, D. Thau, S. V. Stehman, S. J. Goetz, T. R. Loveland, A. Kommareddy,
48  Liliana M. Dávalos
A. Egorov, L. Chini, C. O. Justice, and J. R. G. Townshend. 2013. “High-­
Resolution Global Maps of 21st-Century Forest Cover Change.” Science 342
(6160):850-853. doi: 10.1126/science.1244693.
Harris, David R. 1971. The ecology of swidden cultivation in the Upper Orinoco
Rain Forest, Venezuela. Geographical Review 61(4):475–495.
Harris, Nancy L., S. Brown, S. C. Hagen, S. S. Saatchi, S. Petrova, W. Salas, M. C.
Hansen, P. V. Potapov, and A. Lotsch. 2012. Baseline map of carbon emissions
from deforestation in tropical regions. Science 336(6088):1573–1576.
Hecht, Susanna B. 1985. Environment, development and politics: capital accu-
mulation and the livestock sector in Eastern Amazonia. World Development
13(6):663–684.
———. 1993. The logic of livestock and deforestation in Amazonia. BioScience
43(10):687–695.
——., Anastasia Lucy Yang, B Sijapati Basnett, Christine Padoch, and NL Peluso.
2015. People in Motion, Forests in Transition: Trends in Migration, Urbaniza-
tion, and Remittances and Their Effects on Tropical Forests. 6023870139. Bogor,
Indonesia: Center for International Forestry Research (CIFOR).
Hedin, Lars O., Peter M. Vitousek, and Pamela A. Matson. 2003. Nutrient losses
over four million years of tropical forest development. Ecology 84:2231–2255.
INCORA. 1974a. La Colonización en Colombia: una evaluación del proceso. San
José, Costa Rica: Instituto Interamericano de Ciencias Agricolas, IICA.
———. 1974b. La Colonización en Colombia: una evaluación del proceso. San José,
Costa Rica: Instituto Interamericano de Ciencias Agricolas, IICA.
Jülich, Volker. 1975. Die Agrarkolonisation im Regenwald des mittleren Río Huall-
aga (Peru). Marburg: Im Selbstverlag des Geographischen Institutes der Univer-
sität Marburg.
Kaimowitz, David. 1997. Factors determining low deforestation: the Bolivian Ama-
zon. AMBIO: A Journal of the Human Environment 26(8):536–540.
Killeen, Timothy J., V. Calderon, L. Soria, B. Quezada, M. K. Steininger, G. Harper,
L. A. Solórzano, and C. J. Tucker. 2007. Thirty years of land-cover change in
Bolivia. AMBIO: A Journal of the Human Environment 36(7):600–606.
———., A. Guerra, M. Calzada, L. Correa, V. Calderon, L. Soria, B. Quezada, and
M. K. Steininger. 2008. Total historical land-use change in eastern Bolivia: who,
where, when, and how much. Ecology and Society 13(1):36.
Lambin, Eric F. and Patrick Meyfroidt. 2011. Global land use change, economic glo-
balization, and the looming land scarcity. Proceedings of the National Academy
of Sciences, USA 108(9):3465–3472.
———., Eric F. Lambin, B. L. Turner, Helmut J. Geist, Samuel B. Agbola, Arild
Angelsen, John W. Bruce, Oliver T. Coomes, Rodolfo Dirzo, Günther Fischer,
Carl Folke, P. S. George, Katherine Homewood, Jacques Imbernon, Rik Lee-
mans, Xiubin Li, Emilio F. Moran, Michael Mortimore, P. S. Ramakrishnan,
John F. Richards, Helle Skånes, Will Steffen, Glenn D. Stone, Uno Svedin, Tom
A. Veldkamp, Coleen Vogel, and Jianchu Xu. 2001. The causes of land-use and
land-cover change: moving beyond the myths. Global Environmental Change
11(4):261–269.
Laurance, William F. 1999. Reflections on the tropical deforestation crisis. Biological
Conservation 91(2–3):109–117.
———. 2015. Emerging threats to tropical forests. Annals of the Missouri Botanical
Garden 100(3):159–169.
 The ghosts of development past  49
Laurance, William F., A. K. M. Albernaz, G. Schroth, P. M. Fearnside, S. Bergen, E.
M. Venticinque, and C. Da Costa, C. 2002. Predictors of deforestation in the
Brazilian Amazon. Journal of Biogeography 29(5–6):737–748.
——— D. Carolina Useche, Julio Rendeiro, Margareta Kalka, Corey J. A. Bradshaw,
Sean P. Sloan, Susan G. Laurance, Mason Campbell, Kate Abernethy, Patricia
Alvarez, Victor Arroyo-Rodriguez, Peter Ashton, Julieta Benitez-Malvido, Allard
Blom, Kadiri S. Bobo, Charles H. Cannon, Min Cao, Richard Carroll, Colin
Chapman, Rosamond Coates, Marina Cords, Finn Danielsen, Bart De Dijn, Eric
Dinerstein, Maureen A. Donnelly, David Edwards, Felicity Edwards, Nina Farwig,
Peter Fashing, Pierre-Michel Forget, Mercedes Foster, George Gale, David Har-
ris, Rhett Harrison, John Hart, Sarah Karpanty, W. John Kress, Jagdish Krishnas-
wamy, Willis Logsdon, Jon Lovett, William Magnusson, Fiona Maisels, Andrew
R. Marshall, Deedra McClearn, Divya Mudappa, Martin R. Nielsen, Richard
Pearson, Nigel Pitman, Jan van der Ploeg, Andrew Plumptre, John Poulsen, Mau-
ricio Quesada, Hugo Rainey, Douglas Robinson, Christiane Roetgers, Francesco
Rovero, Frederick Scatena, Christian Schulze, Douglas Sheil, Thomas Struhsa-
ker, John Terborgh, Duncan Thomas, Robert Timm, J. Nicolas Urbina-Cardona,
Karthikeyan Vasudevan, S. Joseph Wright, Juan Carlos Arias-G, Luzmila Arroyo,
Mark Ashton, Philippe Auzel, Dennis Babaasa, Fred Babweteera, Patrick Baker,
Olaf Banki, Margot Bass, Inogwabini Bila-Isia, Stephen Blake, Warren Brockel-
man, Nicholas Brokaw, Carsten A. Bruhl, Sarayudh Bunyavejchewin, Jung-Tai
Chao, Jerome Chave, Ravi Chellam, Connie J. Clark, Jose Clavijo, Robert Cong-
don, Richard Corlett, H. S. Dattaraja, Chittaranjan Dave, Glyn Davies, Beatriz de
Mello Beisiegel, Rosa de Nazare Paes da Silva, Anthony Di Fiore, Arvin Diesmos,
Rodolfo Dirzo, Diane Doran-Sheehy, Mitchell Eaton, Louise Emmons, Alejan-
dro Estrada, Corneille Ewango, Linda Fedigan, Francois Feer, Barbara Fruth,
Jacalyn Giacalone Willis, Uromi Goodale, Steven Goodman, Juan C. Guix, Paul
Guthiga, William Haber, Keith Hamer, Ilka Herbinger, Jane Hill, Zhongliang
Huang, I. Fang Sun, Kalan Ickes, Akira Itoh, Natalia Ivanauskas, Betsy Jackes,
John Janovec, Daniel Janzen, Mo Jiangming, Chen Jin, Trevor Jones, Hermes Jus-
tiniano, Elisabeth K. V. Kalko, Aventino Kasangaki, Timothy Killeen, Hen-biau
King, Erik Klop, Cheryl Knott, Inza Kone, Enoka Kudavidanage, Jose Lahoz da
Silva Ribeiro, John Lattke, Richard Laval, Robert Lawton, Miguel Leal, Mark
Leighton, Miguel Lentino, Cristiane Leonel, Jeremy Lindsell, Lee Ling-Ling, K.
Eduard Linsenmair, Elizabeth Losos, Ariel Lugo, Jeremiah Lwanga, Andrew L.
Mack, Marlucia Martins, W. Scott McGraw, Roan McNab, Luciano Montag, Jo
Myers Thompson, Jacob Nabe-Nielsen, Michiko Nakagawa, Sanjay Nepal, Mar-
ilyn Norconk, Vojtech Novotny, Sean O’Donnell, Muse Opiang, Paul Ouboter,
Kenneth Parker, N. Parthasarathy, Katia Pisciotta, Dewi Prawiradilaga, Cath-
erine Pringle, Subaraj Rajathurai, Ulrich Reichard, Gay Reinartz, Katherine
Renton, Glen Reynolds, Vernon Reynolds, Erin Riley, Mark-Oliver Rodel, Jessica
Rothman, Philip Round, Shoko Sakai, Tania Sanaiotti, Tommaso Savini, Gertrud
Schaab, John Seidensticker, Alhaji Siaka, Miles R. Silman, Thomas B. Smith, Sam-
uel Soares de Almeida, Navjot Sodhi, Craig Stanford, Kristine Stewart, Emma
Stokes, Kathryn E. Stoner, Raman Sukumar, Martin Surbeck, Mathias Tobler,
Teja Tscharntke, Andrea Turkalo, Govindaswamy Umapathy, Merlijn van Weerd,
Jorge Vega Rivera, Meena Venkataraman, Linda Venn, Carlos Verea, Carolina
Volkmer de Castilho, Matthias Waltert, Benjamin Wang, David Watts, William
Weber, Paige West, David Whitacre, Ken Whitney, David Wilkie, Stephen Williams,
50  Liliana M. Dávalos
Debra D. Wright, Patricia Wright, Lu Xiankai, Pralad Yonzon, and Franky
Zamzani. 2012. Averting biodiversity collapse in tropical forest protected areas.
Nature 489:290–294.
———., Miriam Goosem, and Susan G. W. Laurance. 2009. Impacts of roads and
linear clearings on tropical forests. Trends in Ecology & Evolution 24(12):
659–669.
Lee, Rensselaer, and Patrick Clawson. 1993. Crop Substitution in the Andes. Wash-
ington, DC: Office of National Drug Control Policy.
López Rodríguez, Sara R., and Juan F. Blanco-Libreros. 2008. Illicit crops in tropical
America: deforestation, landslides, and the terrestrial carbon stocks. AMBIO: A
Journal of the Human Environment 37(2):141–143.
Maass, Alfredo. 1969. Entwicklung und Perspektiven der wirtschaftlichen
Erschließung des tropischen Waldlandes von Peru, unter besonderer Berücksich-
tigung der verkehrsgeographischen Problematik. Tübingen: Im Selbstverlag des
Geographische Instituts del Universität Tübingen.
McAlpine, Clive, A. Etter, P. M. Fearnside, L. Seabrook, and W. F. Laurance. 2009.
Increasing world consumption of beef as a driver of regional and global change:
a call for policy action based on evidence from Queensland (Australia), Colom-
bia and Brazil. Global Environmental Change 19(1):21–33.
McSweeney, Kendra. 2015. The Impact of Drug Policy on the Environment. New
York: Open Society Foundations.
Mahecha, Liliana, Luis A Gallego, and Francisco J Peláez. 2002. Situación actual
de la ganadería de carne en Colombia y alternativas para impulsar su com-
petitividad y sostenibilidad. Revista Colombiana de Ciencias Pecuarias 15(2):
213–225.
Malthus, Thoma Robert. 1798. An Essay on the Principle of Population as It Affects
the Future Improvement of Society With Remarks on the Speculations of Mr.
Godwin, M. Condorcet, and Other Writers. London: J. Johnson.
Meeting of American Chiefs of State. 1967. Declaration of the Presidents of America
signed at the meeting of American Chiefs of State, Punta del Este, Uruguay, April
14, 1967. Washington, DC: General Secretariat of the Organization of American
States, Department of Information and Public Affairs.
Millington, Andrew C., Ximena M. Velez-Liendo, and Andrew V. Bradley. 2003. Scale
dependence in multitemporal mapping of forest fragmentation in Bolivia: impli-
cations for explaining temporal trends in landscape ecology and applications to
biodiversity conservation. ISPRS Journal of Photogrammetry and Remote Sens-
ing 57(4):289–299.
Molano, Alfredo. 1989. Aproximación al proceso de colonización de la región del
Ariari–Güejar–Guayabero. In La Macarena Reserva biológica de la humanidad
Territorio de conflictos. Cubides, Fernando, Antanas Mockus Sivickas, Mario
Avellaneda, Henry González, Óscar Arcila Niño, Alfredo Molano, Hugo Acero,
Juan Carlos Pacheco, and Ricardo Mosquera Mesa, eds. 279–304. Bogotá:
Universidad Nacional de Colombia.
Müller, Robert, Daniel  Müller, Florian Schierhorn, Gerhard Gerold, and Pablo
Pacheco. 2012. Proximate causes of deforestation in the Bolivian lowlands: an
analysis of spatial dynamics. Regional Environmental Change 12(3):445–459.
Myers, Norman. 1993. Tropical forests: the main deforestation fronts. Environmen-
tal Conservation 20(01):9–16.
 The ghosts of development past  51
———, Russell A. Mittermeier, Cristina G. Mittermeier, Gustavo A. B. da Fonseca,
and Jennifer Kent. 2000. Biodiversity hotspots for conservation priorities. Nature
403(6772):853–858.
Nepstad, Daniel C., C. M. Stickler, B. Soares-Filho, and F. dan Merry. 2008. Inter-
actions among Amazon land use, forests and climate: prospects for a near-term
forest tipping point. Philosophical Transactions of the Royal Society B: Biologi-
cal Sciences 363(1498):1737–1746.
Reategui A., Ricardo, and Pascual Taminche T. 1980. Proyecto de Asentamiento
Rural Iitergrado (PARI) en Jenaro Herrera, Peru. Proyectos forestales para el
desarrollo rural de America tropical, Turrialba, 1980. Vol. 6, pp. 45–55. CATIE.
Rodrigues, Ana S. L., Robert M. Ewers, Luke Parry, Carlos Souza, Jr., Adalberto
Veríssimo, and Andrew Balmford. 2009. Boom-and-bust development patterns
across the Amazon deforestation frontier. Science 324(5933):1435–1437.
Rodríguez, Nelly, D. Armenteras, R. Molowny-Horas, and J. Retana. 2012. Patterns
and trends of forest loss in the Colombian guyana. Biotropica 44(1):123–132.
Rudel, Thomas K., Diane Bates, and Rafael Machinguiashi. 2002. A tropical for-
est transition? Agricultural change, out-migration, and secondary forests in
the Ecuadorian Amazon. Annals of the Association of American Geographers
92(1):87–102.
———. and Jill Roper. 1997. The paths to rain forest destruction: crossnational
patterns of tropical deforestation, 1975–1990. World Development 25(1):53–65.
Salisbury, David S. 2007. Overcoming Marginality on the Margins: Mapping, Log-
ging, and Coca in the Amazon Borderlands. Austin: University of Texas.
———. and C. Fagan. 2011. Coca and conservation: cultivation, eradication, and
trafficking in the Amazon borderlands. GeoJournal 78 (1):41–60.
Sánchez-Cuervo, Ana María, and T. Mitchell Aide. 2013. Consequences of the armed
conflict, forced human displacement, and land abandonment on forest cover
change in Colombia: a multi-scaled analysis. Ecosystems 16(5):1052–1070.
———. T. M. Aide, M. L. Clark, and A. Etter. 2012. Land cover change in Colom-
bia: surprising forest recovery trends between 2001 and 2010. PLoS ONE
7(8):e43943.
Schoop, Wolfgang. 1970. Vergleichende untersuchungen zur Agrarkolonisation der
Hochlandindianer am Andenabfall und im Tiefland Ostboliviens. Wiesbaden:
F. Steiner Verlag.
Schuurman, Frans J. 1978. From resource frontier to periphery: agricultural colo-
nization east of the Andes. Tijdschrift voor Economische en Sociale Geografie
69(1–2):95–104.
———. 1979. Colonization policy and peasant economy in the Amazon basin.
Boletín de Estudios Latinoamericanos y del Caribe (27):29–41.
Southgate, Douglas. 1990. The causes of land degradation along “sponta-
neously” expanding agricultural frontiers in the third world. Land Economics
66(1):93–101.
Steininger, M. K., C. J. Tucker, P. Ersts, T. J. Killeen, Z. Villegas, and S. B. Hecht. 2001.
Clearance and fragmentation of tropical deciduous forest in the Tierras Bajas,
Santa Cruz, Bolivia. Conservation Biology 15(4):856–866.
Tejada, Graciela, Eloi Dalla-Nora, Diana Cordoba, Raffaele Lafortezza, Alex
Ovando, Talita Assis, and Ana Paula Aguiar. 2016. Deforestation scenarios for
the Bolivian lowlands. Environmental Research 144, Part B:49–63.
52  Liliana M. Dávalos
UNODC. 2008. Análisis multitemporal de cultivos de coca, diciembre 31 2001 –
­diciembre 31 2006. Naciones Unidas, Oficina contra la Droga y el Delito: Bogotá.
———. 2010. Transformación socioeconómica y biofísica asociadas con los cultivos
ilícitos en la region Sur del Meta-Guaviare. UNODC, Embajada de la Republica
Federal Alemana, and Ministerio de Defensa Republica de Colombia: Bogotá.
———. 2014. San Martín. Análisis Económico del Impacto del Desarrollo Alternativo,
en relación a la Deforestación y la Actividad Cocalera. UNODC and Ministerio del
Medio Ambiente, Peru: Lima.
———. 2015. World Drug Report 2015. In World Drug Report. New York: United
Nations Office on Drugs and Crime.
———. and Estado Plurinacional de Bolivia. 2011. Estado Plurinacional de Bolivia:
Monitoreo de Cultivos de Coca 2010. UNODC: La Paz.
———. and Gobierno de Colombia. 2013. Monitoreo de Cultivos de Coca 2012.
UNODC.
———. and Peru Ministerio del Medio Ambiente. 2011. Análisis económico de las
actividades causantes de la deforestación en Pichis-Palcazú. UNODC and Minis-
terio del Medio Ambiente: Lima.
———. and Gobierno Nacional de la República del Ecuador. 2015. Indicadores de
Cultivos Ilícitos en el Ecuador 2014. In Monitoreo de Cultivos de Coca. Quito:
UNODC.
Viña, Andres, Fernando R. Echavarria, and Donald C. Rundquist. 2004. Satellite
change detection analysis of deforestation rates and patterns along the Colombia–
Ecuador border. AMBIO: A Journal of the Human Environment 33(3):118–125.
Walker, Robert. 1993. Deforestation and economic development. Canadian Journal
of Regional Science 16:481–497.
Wesche, Rolf Jürgen. 1968. The Settler Wedge of the Upper Putumayo River. Geog-
raphy. Gainesville: University of Florida.
Young, Kenneth R. 1996. Threats to biological diversity caused by coca/cocaine
deforestation. Environmental Conservation 23(1):7–15.
———. 2004a. Environmental and social consequences of coca/cocaine in Peru. In
Dangerous Harvest: Drug Plants and the Transformation of Indigenous Land-
scapes. M. K. Steinberg, J. J. Hobbs, and K. Mathewson, eds., 249–273. Oxford:
Oxford University Press.
———. 2004b. A Geographical perspective on coca/cocaine impacts in South
­America. In WorldMinds: Geographical Perspectives on 100 Problems. D. Janelle,
B. Warf, and K. Hansen, eds., 363–367: The Netherlands: Springer.
———. and Blanca León. 1999. Peru’s humid eastern montane forests. Centre for
Research on the Cultural and Biological Diversity of Andean Rainforests (DIVA).
Reporte Técnico (5).
———. 2000. Biodiversity conservation in Peru’s Eastern Montane forests. Moun-
tain Research and Development 20(3):208–211.
3 Ideas of modernization and
territorial transformation
The case of the Upper Huallaga
Valley of Peru
Maritza Paredes and Hernán Manrique1

Introduction
The sparse research on coca production in Peru has primarily focused on the
consequences of the illicit cocaine trade, such as crime, violence, and corrup-
tion (Dreyfus 1999; Felbab-Brown 2010; Gonzales 1992; Kay 1999; Marcy
2010; Morales 1989; Páucar 2006). However, less attention has focused
on how specific territories of Peru came to be integrated in this illicit econ-
omy, thus producing spaces not just of violence but also persistently fragile
and unstable livelihoods for their long-term inhabitants (Aramburú 1982;
Rumrrill 1992; Manrique 2016). To better understand what we now call
“los valles cocaleros” [coca-producing valleys] and their internal dynamics
in Peru requires understanding the historical transformation experienced in
Amazonian territories such as the Alto Huallaga and the impact of this pro-
cess in the subsequent articulation of these territories by illicit economies
spurred by drug trafficking.
Through a case study of the Valle del Alto Huallaga in Peru (Upper Hual-
laga Valley or VAH), this chapter seeks to explain how the productive char-
acteristics of the valley unexpectedly shifted during the 1960s and 1970s,
facilitating the expansion of the illegal coca economy in the region. It also
explores how the ideas of modernization of Peruvian elites and international
development agencies diffused into blueprints for agricultural modernization
and tropical colonization that shaped policies for the valley’s transforma-
tion. By the end of the 1970s, the failure of these official colonization policies
made it possible for the valley’s changed social structure to be incorporated
into the massive, illegal cultivation of coca. Structural factors such as the val-
ley’s geographic remoteness from coastal Peru and distance from the central
state may have contributed to the development of illicit coca cultivation. But
this chapter traces the origins of the expansion of the illegal coca trade to
newly-facilitated access to frontier lands suited to coca production leading
to revolutionary changes in the valley between the 1960s and 1970s. This led
in time to the formation of an enclave of poor peasants left without sustain-
able livelihoods in the region’s increasingly degraded acidic soil.
Such changes were driven by a national elite that sought to use policies
for colonizing Peru’s Selva Alta or upper Amazon jungle as an inexpensive
54  Maritza Paredes and Hernán Manrique
and politically expedient way out of the problem of land distribution in
the highlands. Tropical colonization policies took on particular importance
in the polarized 1960s when they were viewed by Peruvian landed elites
as a less costly alternative to the radicalism of agrarian reform that might
force them to relinquish their hacienda lands (Albertus 2015a; Aramburú
1979). Intensifying peasant land grabs during the 1960s in the highlands
and the political tensions of the Cold War drove Peruvian elites to view
frontier expansion as an alternative to reform that could bolster political
stability without undermining their economic interests. These elites directly
benefited from a wave of international and U.S. developmentalist interests
that sought to implement international socio-economic formulas to solve
the local problems of Latin America. Developmentalism was the broad idea,
which peaked in the 1960s, that a combination of enhanced investment,
government intervention and planning, industrialization, and structural
reforms would overcome Third-World “under-development” with acceler-
ated economic growth. Both Latin American economists at the Economic
Commission for Latin America and the Caribbean (ECLAC) and first world
institutions such as the World Bank shared these developmentalist concepts
and objectives. The interests of the Peruvian political and landed elites came
to overlap with the efforts of international development organizations on
the shape and form of Peru’s agrarian reform, affecting the territorial trans-
formation of the Huallaga Valley.
The colonization programs ultimately failed to achieve their initial
objective of transforming the Peruvian upper jungle into a rich agricul-
tural resource, instead producing an enclave of tens of thousands of poor
peasants who struggled to adapt to unknown territory or produce sustain-
able livelihoods. Andean migrants to these tropical lands were unfamiliar
with the soil and climate of the valley, and their adaptation was slow. Since
Andean migrants to this tropical region had little access to markets or com-
mercial networks, they became heavily dependent on the state for initial
support (Aramburú 1989, 53). However, disillusion with these colonization
policies coincided with both the mid-1970s national economic crisis that
brought the prices of the principal products of Amazonia tumbling (Thorp
and Paredes 2010), and a new surge in international demand for coca from
the Huallaga, due to the intensifying consumption of cocaine in the 1970s
U.S. and the building up of the new Colombian drug trafficking organiza-
tions connected to the valley (Gootenberg 2008, 295–306). Coca became
their most dependable and profitable crop, although it was no panacea since
regional political violence and the coming repression of the war on drugs
(decreed in Peru by 1978) also adversely affected coca-producing peasants
by the 1980s.
This chapter is based on a close case study of the colonization of the Valle de
Alto Huallaga (henceforth VAH). Until the mid-1980s, this valley was com-
prised of the Leoncio Prado and Mariscal Cáceres provinces, which straddle
the departments of Huánuco and San Martín respectively (see F ­ igure 3.1).2
During the 1970s and 1980s, the VAH became the first and largest global
 The Upper Huallaga Valley, Peru  55
illegal supply shed for coca and cocaine. In the heyday of this boom, half of
the international market’s coca and locally made cocaine paste (pasta básica
de cocaína, or PBC) used in cocaine production was sourced from the VAH
(Gootenberg 2008, 292). These characteristics have made the VAH a case
study in drug trafficking research (Aramburú 1989; Dreyfus 1999; Gonzales
1992; Kay 1999; Kernaghan 2009; Manrique 2016). However, Gootenberg’s
suggestions aside (2008), there is little specific work studying the origins of
the valley itself as a center of drug production and its longer links to state
colonization policies and the agrarian reform process of the 1960s.3 The
methodological approach is two-fold. First, we conduct a literature review of
specialized texts and primary documents from outside development agencies
like the ECLAC and the World Bank.4 Second, we review Peruvian sources,
chief among them key pieces of Peruvian legislation, operational plans, Presi-
dential addresses, congressional debate records, and evaluation reports from
the varied institutions created to implement tropical colonization policies.
The chapter traces the evolution of the Valle de Alto Huallaga over a
two-decade period 1960–1980 in which it was transformed from a state-
driven internal colonization project to a highly contested coca-producing
valley that housed not only migrant colonists but new drug traffickers and
armed groups. The first section explores the motivations of landholding
Peruvian elites who pushed for the expansion of the agricultural frontier
into the upper Amazon to highlight the historical connections between
calls for land reform and subsequent internal colonization policies. The
next section details how development agency blueprints for international
development and agricultural modernization coincided with elite interests,
marrying internal colonization policies with Peru’s comprehensive agrarian
development plan. This section reveals the failure of these colonization pol-
icies in terms of sustainable livelihoods and the ensuing dependence of new
migrants on both the state and international financial and technical support
for survival. The final section details how demographic, environmental, and
economic conditions forced colonists to shift their reliance on the state to
illegal coca production, making the VAH the cocaine capital of Peru. The
shift from colono to cocalero did not alleviate the insecure, unstable condi-
tions of peasant life as the inhabitants of the VAH were often caught in the
crossfire of the state’s new war on drugs and its later fight against terror-
ism. The chapter concludes with reflections on the new forms of governance
employed by the state, the historical preservation of elite interests in the
Andes, and the international push for agrarian reform across Latin America
which together transformed the upper Amazon from an imagined agricul-
tural “paradise” to the home of the thriving illegal drug trade it still is today.

Peruvian elites and the impetus for Amazonian colonization

In the Peruvian elite imagination, Amazonia was long viewed as “El Dorado,”
a source of inexhaustible wealth (Cotler 1999, 112). This section shows
how during the 1960s this Peruvian elite imaginary led them to push for
56  Maritza Paredes and Hernán Manrique
the ­colonization of Amazonia as a solution to the growing problem of land
conflicts. The expansion of the agricultural frontier towards Amazonia rep-
resented the renewal of an old promise, not of precious gold but of “empty”
land, a “land without men for men without land” in the words of soon-to-be
President Fernando Belaúnde Terry (Belaúnde 1959, 105).
The Peruvian case is marked by a close link between agrarian reform and
internal colonization policies. Both policies were carried out simultaneously
and complemented each other even though they had geographically distinct
objectives. While the primary objective of agrarian reform was to resolve
conflicts over the distribution of land and land tenure in the Andean high-
lands and Peru’s northern coast, Amazonian colonization policies sought
to expand the agricultural frontier by transforming Amazonian headwaters
into productive farmland. When the agrarian reform expanded under Gen-
eral Velasco’s presidency, with the unexpected expropriation of hacienda
lands in 1969, colonization policies maintained this geographic focus on the
high-elevation tropics or ceja de la montaña (eyebrow of the mountains or
Amazonian piedmont) as they are known in Peru.5
The debate over land conflicts and agrarian reform in Peru dates to the
second decade of the twentieth century (Lastarria-Cornhiel 1989, 127). At
the time, Peru had without doubt one of the region’s worst extremes of land
inequality (CIDA 1966); highland peasants unleashed a wave of protests,
and the debate over agrarian reform filtered into national politics. This sec-
tion addresses how economic elites and their policies responded to this crisis
with a vision of expanding the agricultural frontier to the upper Amazon in
order to appease agitating peasants. For landed elites, the proposed coloni-
zation of Amazonia represented a less radical solution to land conflicts than
did agrarian reform, which would have meant an overhaul of the lopsided
land tenure structure that benefited them.
The notion of colonizing the jungle was hardly new. Various European
colonies, such as Pozuzo, were tried in the nineteenth century, and during
the first half of the twentieth century the “conquest of the Amazon” was
adopted by different Peruvian governments and promoted by different
Peruvian elites.6 Modern attempts to actually colonize the ceja de montaña
can be traced back to the end of the 1930s. In 1938, the passage of Law
8621 permitted the state to expropriate uncultivated mountain lands run-
ning alongside both current and planned highways to the west of the Andes
mountain range. This law would later allow the state to freely sell, award, or
auction (as a concession) such lands if it so desired. Various other laws from
this period further confirmed this interest in colonizing the upper jungle,
particularly the Tingo María area along the Huallaga River.7
In the same decade, the construction of so-called “highways of penetra-
tion” reflected Peruvian elites’ growing interest in colonizing the ceja de
montaña. These interests viewed the Amazonian region “not only as a food
basket for the country, but [for] the world” (Cotler 1999, 112). The fruit of
these challenging, but persistent efforts to construct these first roads east
 The Upper Huallaga Valley, Peru  57
was the interconnection of the Huánuco-Tingo María highway with the
recently built Tingo María-Aguyatía-Pucallpa highway in 1943 (Aramburú
1981, 50; 1982, 2; Crist and Nissly 1973, 96). By connecting Lima with
the Ucayali River in Pucallpa for the first time, the city of Tingo María was
converted into northwest Peru’s gateway to the upper jungle.
However, these laws did not encompass a significant range of the agri-
cultural frontier, nor did they succeed in actually increasing the popula-
tion of the Peruvian Amazon.8 In spite of these incipient transformations
into the ceja de montaña, it was only towards the end of the 1950s that the
discussion over agrarian reform would catapult the colonizing impulses
of Peruvian elites to the forefront. The challenge posed by land conflicts
in the Andes and the need for agrarian reform revived elite interest in
expanding the agricultural frontier to include the highlands of the Ama-
zon. In this way, the agrarian reform debate acquired unprecedented
visibility in the country at the end of the 1950s. After the demise of Gen-
eral Odria’s conservative dictatorship (1948–1956), a possible agrarian
reform became one of the most discussed topics in Peru’s general elections
(Chirinos 1975, 47).

Peasant mobilization and agrarian reform

In the 1956 elections, the winner was a representative of the oligarchy and
hacienda-owners: Manuel Prado Ugarteche. In spite of his frankly conserva-
tive character, his government took the opportunity to seek and obtain inter-
national funds for agrarian reform. Thus, in 1956, the government formed
the Commission for Agrarian Reform and Housing (CRAV in Spanish) to
design and propose a law of agrarian reform. Paradoxically, five of the com-
mission’s eight members were linked to haciendas, including its well-known
new director, a hacienda owner and president of the National Agrarian Soci-
ety (SNA), Pedro Beltrán (Lowenthal 1975, 28).
As a part of his work, Beltrán obtained a loan of $25 million from the U.S.
Export-Import Bank to finance agrarian reform, colonization programs,
and infrastructure construction. There was also an agricultural develop-
ment loan of $5 million from the World Bank to make related bureaucratic
reforms (Kofas 1996, 189). However, after four years of work, the CRAV
proposed a limited agrarian reform law more focused on colonization and
irrigation than on the redistribution of land. The lack of state interest in
the redistribution of land not only disillusioned various groups on the left,
whose own agrarian reform proposal presented by the Revolutionary Insur-
gent Movement (MIR) had been rejected, but also alienated diverse regional
elites. For example, the landholders of the Agro-Livestock Society of Cusco,
an organization affiliated with SNA, demanded that President Prado request
state support for agricultural modernization (Ramírez 2011), adding yet
another voice to the mounting peasant mobilization taking place around
highland and lowland Cusco.
58  Maritza Paredes and Hernán Manrique
During the last years of the Prado Ugarteche administration, peasant
movements and land grabs became more intense and radical. An important
precedent for this was the growing association between leftist political par-
ties and indigenous and peasant organizations that started in the mid-1950s
(Paredes 2011, 135). While towards the end of the decade, many peasant
movements were led by traditional communities, peasants, colonists, and
peons, by the middle of the 1960s, these movements were led by unions and
regional federations (Guzmán and Vargas 1981). This was not just about
changes in leadership but rather a shift in the types of organizations, forms
of contention, and their results. Just between August and December of 1963
alone, a wave of 77 hacienda invasions occurred across the entire country
(Cotler and Portocarrero 1976, 292). The most important context to this
movement was Cerro de Pasco, where the multinational U.S. mining firm
Cerro de Pasco Corporation controlled the majority of the land in the Valle
de La Convención y Lares, a colonizing area in Cusco ceja de montaña.
Here, a peasant union struggle led by the Marxist Hugo Blanco checkmated
traditional power structures, serving as an example to similar organizations
in the country. Under the radical rallying cry Tierra o Muerte (“Land or
Death”), these now politicized movements undermined the foundations of
the latifundia system, generating a sense of instability in the governing class.
In the face of the social and political crisis produced by grassroots land
takeovers and political tensions arising from the contentious 1962 elec-
tions, Prado’s administration was cut short. The Armed Forces executed an
unprecedented kind of institutional coup d’état (Manrique 1995, 298). In
opposition to national and regional elites, the military junta implemented an
agrarian reform in the tumultuous regions of La Convención y Lares, Cusco.
Thereafter, agrarian reform was impossible to ignore in Peruvian politics.
A year later, the military junta retreated from the Government Palace and
returned to calling for general elections.
Agrarian reform, with greater intensity than the 1956 elections that
brought Prado Ugarteche to power, was the central theme of national debate
in the new elections of 1962. A special official press release during the elec-
tions noted:

one of the most contested issues currently being debated is, without a
question, the Agrarian Reform. Not one party or parliamentary candi-
date denies the urgency with which this Reform must take place; rather,
they fundamentally disagree on the manner in which it should be car-
ried out.
(El Comercio, May 1962, in Ramírez 2011)

In reality, the elites represented in Congress had similar ideas regarding

agrarian reform. For the senator from Cusco representing the government
party, Acción Popular, the recipe for agrarian reform was simple: it included
technical modernization of agriculture to boost production, a more just
 The Upper Huallaga Valley, Peru  59

Figure 3.1  Inter-Andean valleys and colonization zones

Source: Belaúnde (1959, 102–104)

redistribution of land leaning towards elimination of the latifundia system,

and the conquest of new lands to be colonized, connected, irrigated, and
developed.

The architect of internal colonization and the Cold War

Fernando Belaúnde was, without a doubt, one of the candidates most aware
of the dire need for agrarian reform. However, his plan to modernize Peru
60  Maritza Paredes and Hernán Manrique
went much further than the redistribution of land tenure in the Andes. From
his architecturally inspired perspective, expressed since the late 1950s, Peru
would not be able to become fully developed if it did not incorporate the
vast Amazon region into its nation-building project. Inspired as well by
the development lessons of the U.S. Tennessee Valley Authority (TVA) of
the 1930s, Belaúnde (1965, 26–31) proposed the large-scale incorporation
of arable lands and the wholesale exploitation of hydraulic power in Peru.
His most ambitious project in this respect was the Carretera Marginal de
la Selva [Marginal Highway of the Jungle] designed to connect “the best
valleys of the eyebrow of the mountains that cross many already-formed
or towns in-formation, giving birth to new towns where the excess of our
population will find the ways of life that they so need” (Belaúnde 1967, 43).
For him, the Marginal Highway would unite “the Mayo, Huallaga, Pozuzo,
Perené, Ene, Apurímac, Urubamba, Camisea, and Alto Madre de Dios val-
leys, touching the Ucayali [River] at the port of Atalaya and the Huallaga
[River] at the port of Yurimaguas [with] a coastal exit through the Olmos
route” (Belaúnde 1967, 31). To carry out the highway’s feasibility studies,
the state received grants from the Inter-American Development Bank (IDB).
There were a variety of reasons why Belaúnde privileged the upper jungle
over the lower jungle for colonization. In addition to its proximity to Lima,
the geographic conditions of the upper jungle were considered more desir-
able. Belaúnde maintained that: “unlike the flooded lower jungle, the eye-
brow of the mountains offers various advantages with its natural drainage
and facilitates an abundance of materials for establishing routes” (Belaúnde
1959, 101). The main reason for prioritizing the upper jungle was the pres-
ence of “hotbeds of men” or “colonist nuclei” in the inter-Andean valleys
adjacent to the upper jungle (Figure 3.1). For Belaúnde, the people inhab-
iting lands directly bordering the Andean mountain range would ideally
migrate horizontally towards the edge of the adjacent jungle and swiftly col-
onize those areas. Among these were, principally, the populations of Ancash
and Huánuco and Pasco that were physically closest to the Huallaga system.
To make this happen, all these communities needed to do was use Belaúnde’s
proposed road infrastructure, represented by the thick black line.
With his extensive geographical knowledge, Belaúnde galvanized the
Peruvian electorate and won the 1963 presidential elections. More than any
of its predecessors, the new administration had the opportunity to conduct
the structural reforms now expected by large sectors of the population.
The Belaúnde government was not alone in its efforts to modernize Peru’s
agrarian sector and the drive for colonization of the Amazonian highlands
received support from the U.S. through various initiatives. Most impor-
tantly, the ascendance of Belaúnde’s government coincided with the recently
initiated “Alliance for Progress” (1961–1964) agenda, propelling the coloni-
zation campaign as never before.
The Kennedy administration marked a watershed in the collaboration
between the U.S. and Latin America on development initiatives. In the
 The Upper Huallaga Valley, Peru  61
­ revious decade, at the 1954 meeting of the Inter-American Economic and
p
Social Commission of the Organization of American States (OAS) in Quin-
tadinha, Brazil, a group of Latin American development experts raised the
urgent necessity of establishing an Inter-American Development Bank and
of receiving $1 billion in foreign assistance (Griffith-Jones 1979, 424). This
scale project would mature once Kennedy created the Alliance for Progress
in 1961 at a conference of the Inter-American Economic and Social Com-
mission of the OAS in Punta del Este. All the Latin American states apart
from Cuba signed the treaty, which was executed in large part as a positive
North American response to the Communist threat of Cuba. The Alliance
for Progress program was envisioned to provide at least $20 billion of finan-
cial assistance to Latin America for a period of ten years—the equivalent of
$100 billion in 2004 dollars (Taffet 2007, 5)—an unprecedented investment
for Latin American governments. In the words of Teodoro Moscoso, then
director of Alliance for Progress, this was “an agreement for peaceful revo-
lution at a hemispheric scale” (Moscoso 1962, 4).
With this international support, Andean countries quickly began a series
of colonizing projects and extensive highway construction. In Alto Huall-
aga, the Peruvian state built highways through the forest, invested in credits
to finance small farmers, and designed and supported programs of formal
colonization. Belaúnde was able to use an IBD loan of $15 million to estab-
lish the first of the state-directed zones of colonization, strategically located
in the Alto Huallaga: the Colonization Project Tingo María-Tocache-­
Campanilla. A decade later, these colonized sites would become some of the
largest producers of coca and cocaine in the country.
In the context of the Alliance for Progress, and severe poverty throughout
Latin America, Cuba was feared as a possible trigger for the appearance of
more pro-Communist groups in the Western Hemisphere (Taffet 2007, 36).
This quickly made agrarian reform a central plank in international Cold
War policy, as reform could stem the escalating land conflicts that had the
potential to radicalize the landless peasantry. Since the early 1950s, U.S.
officials highlighted the need for agrarian reform in a variety of meetings
with the UN and the FAO. As the U.S. Secretary of State even declared in
1953 “Land reform is absolutely foremost in our whole international pol-
icy” (Kapstein 2014, 110). In this way, the Alliance for Progress imbued
agrarian reform with a double objective: to modernize agriculture and to
counter the perceived threat of Communism (Albertus and Kaplan 2013, 2;
Kapstein 2014, 110).
In Peru, Belaúnde was committed to developing his own program of land
reform. However, for national political reasons, land reform would fail to
materialize. This only intensified the stress on Amazonian colonization.
Landed elites proved more supportive of colonization as a policy since the
land to be occupied in the tropics was still imagined as “empty” and thus
a win–win solution to the conflicts taking place over elite landholdings in
Peru’s highlands.
62  Maritza Paredes and Hernán Manrique
During Belaúnde’s tenure, agrarian reform in Peru continued to evoke
controversy and no consensus emerged on the best way to carry out reform
(Barraclough and Domike 1966). The problem of highly concentrated land
holdings in Peru appeared to have no peaceful solution. On the one hand,
a mid-1960s report from the Inter-American Committee for Agricultural
Development (CIDA) indicated that Peru clearly had the region’s highest
disparity in landholding concentration, for example, twice that of Argentina
and Guatemala (Barraclough and Domike 1966, 239). On the other hand,
this structure of landholding could not be changed given the implacable
opposition of elites and maintaining political stability.
Belaúnde’s agrarian reform law struggled from the beginning. On August
12, 1963, Enrique Torres Llona, the Minister of Agriculture, presented the
proposed agrarian reform law. He noted that

agrarian reform is not a problem specific to Peru; it is a problem that

affects all insufficiently developed countries (...) It cannot be forgot-
ten that [land reform] comes up against powerful interests, interests of
those who, precisely because of the high concentration of property, are
not farmers but landholders and who oftentimes hold great political
power.
(Diario de los Debates de la Cámara de Diputados – henceforth
DDCD 1963, vol. I, 241–264)

However, the government’s proposed law agrarian reform law was not the
one debated. Opposing political caucuses such as the Alianza Popular Revo-
lucionaria Americana (APRA party) and the Unión Nacional Odrísta (UNO
party) also presented options. Among these, for instance, was a new national
rural code or the creation of a National Corporation for Agrarian Reform,
proposed by UNO. UNO attacked Belaúnde’s proposed law on the class-
based grounds that it threatened “to substitute the existing capitalist agrar-
ian structure” (DDCD 1963, vol. III, 404). For the UNO representative in
Lima, Víctor Freundt Rosell, Belaúnde’s law meant that while the:

change in rural landholding from one to many hands will clearly result
in the social justice [via] greater distribution of land [it] does not mean
greater production in the future (...) Because new possessors of land will
effectively have no technical experience or economic capacity for inten-
sive exploitation (...) Given these land, wealth, and water limitations,
Agrarian Reform in Peru would be a resounding failure in little time,
we would have little to eat, [there will be] a tremendous increase in the
costs of living and abandonment of lands
(DDCD 1963, vol. III, 409)

Freundt Rosell’s claims do not constitute a rejection of agrarian reform as

such, they suggest a more moderate reform and expand the discussions to
 The Upper Huallaga Valley, Peru  63
include issues of irrigation and colonization. UNO’s reasons for its proposal
to create a National Economic Agrarian Reform Corporation asserted that:
“the expropriation and splitting of estates currently being projected is insuf-
ficient in that these lands are not creating a noticeable increase in produc-
tion, which is indispensable to obtain through more farmable area through
irrigation and colonization projects” (DDCD 1963, vol. III, 409). Obviously,
irrigation and colonization were perceived by some parliamentarians from
opposing parties as a better alternative to land reform.
This was not Belaúnde’s vision, as he viewed irrigation and colonization
as complementary. After many months of prolonged and clamorous debates
in the Chamber of Deputies and Senate, Law 15037—the Agrarian Reform
Law—was finally passed on May 21, 1964. The spirit of this law, bore an
uncanny resemblance to the proposals and recommendations of the inter-
national development agencies (Art. 1, Ley 15037). The Law of Agrarian
Reform created a series of institutions entrusted with bringing about the
reform. Chief among these were the Institute for Agrarian Reform and
Promotion (IRPA in Spanish) (Art. 191), whose governing body was the
National Agrarian Council (CNA) comprised of a multi-sectoral commis-
sion (Art. 195). Among the executive bodies of the IRPA were the National
Office for Agrarian Reform (ONRA in Spanish) and the Agrarian Research
and Promotion Service (SIPA) (Art. 194).
However, parliamentary opposition successfully blunted the Agrarian
Reform Law, with arbitrary interventions on more than one occasion. Thus,
the final law contained a myriad of costly and bureaucratic exceptions.
When it was finally possible to implement the law, the resulting process
was difficult and expensive to carry through. In the end, Belaúnde’s agrar-
ian reform attempt only covered 380,000 hectares and reached fewer than
15,000 peasants (Albertus 2015b, 215). Belaúnde’s aspiration to carry out
a national agrarian reform in democratic fashion were dashed as his term
came to an abrupt end, in part because of the infamously unresolved inter-
national tax conflict with the multinational International Petroleum Corpo-
ration. Belaúnde was deposed on October 3, 1968 through a coup d’état led
by General Velasco Alvarado and his new-style leftist Revolutionary Gov-
ernment of the Armed Forces (GRFA in Spanish). But while Belaúnde’s gov-
ernment had failed in its efforts to implement a national agrarian reform,
it had succeeded in starting and promoting colonization programs. Unlike
Belaúnde, the Velasco regime implemented agrarian reform far more deci-
sively and with greater intensity. These military-like moves also reinforced
the role of the expanding state in colonization sites near international
borders, including the transformation of the VAH (Manrique 2016, 67).

The colonization of Alto Huallaga

As previously noted, colonization projects were not entirely new to the
VAH. These efforts began decades before through varied public and private
64  Maritza Paredes and Hernán Manrique
endeavors. Tingo María had been designated an official zone for coloniza-
tion since the 1930s. Beginning in 1949, its official territory of coloniza-
tion encompassed more than 3,000 hectares. In private initiatives, around
15,000 hectares of the Alto Huallaga pertained to a farming colonization
project by Italian entrepreneurs. Many private companies in the valley
used traditional enganche labor contracts to obtain labor (Bedoya 1982).
Enganche indirectly promoted colonization because it gave the seasonal
workers who migrated to work as day laborers for companies like Saipai
or Tea Gardens an opportunity to see the expanses of unoccupied land and
they became motivated to take up individual farming over continuing on as
precarious seasonal workers. However, only in the 1960s with Belaúnde and
the Alliance for Progress did large-scale social transformation hit the VAH,
as expansion of the agricultural frontier and colonization became an imper-
ative for national progress. In contrast to the Peruvian state’s early attempts
to populate the region in the first part of the twentieth century, the “integra-
tion” of the Peruvian Amazonia in the following decades relied on a more
developed system of technical support and the deep pockets of international
development organizations.
The colonization of the VAH occurred through a combination of direct
and indirect incentives. Along with the resources to grant more agricultural
credits, the Belaúnde government put in place an aggressive investment
plan for highway construction using its loans from development agencies.
Public investment in highways and ports grew from 327 million soles in
1960 to 1.07 billion soles in 1964 to 3.15 billion in 1967, by the end of his
term (Parodi 2014, 443). Of these highway projects, the most ambitious of
them all was undoubtedly the construction of the Marginal Highway, which
proved fundamental to the colonization of the Peruvian Amazonia since it
connected Tingo María with towns like Tocache, Juanjuí, and Tarapoto.
As stressed by Kernaghan (2009, 8), the modern history of the Huallaga
began with this one highway. Its construction incorporated the VAH, from
northern part of Huánuco to southern San Martín, into the transport infra-
structure of the rest of the country, opening the door to large-scale internal
migration and enhanced agricultural production.
Beyond these indirect infrastructure incentives, the state’s campaign to
colonize in the 1960s involved new and ambitious direct population reset-
tlement policies. Four main colonization programs were implemented in this
decade.9
Of the colonization projects presented in Table 3.1, the state’s largest
colonization project was that of Tingo María-Tocache-Campanilla, which
began operations in 1966 (CENCIRA 1974, 13). The feasibility of the proj-
ect was preceded by extensive scientific and technical studies carried out by
the National Office for the Evaluation of Natural Resources (ONERN). The
results of these evaluations established the valley as an ideal setting for agri-
cultural production. Soil testing suggested a potential use of 35.64 percent
in annual crops, 15.65 percent in permanent crops (like tea or fruit), and
 The Upper Huallaga Valley, Peru  65
Table 3.1  Major colonization projects in the Peruvian jungle from 1961 to 1972

Colonization project Location Year established

Pichari Upper jungle of Cusco 1961
Jenaro Herrera Lower jungle of Loreto 1965
Tingo María- Upper jungle of Huánuco 1966
Tocache-Campanilla and San Martín
Marichín-Río Yavari’ Lower jungle of Loreto 1971
Caballococha
Saispampa Lower jungle of Ucayali 1972
Source: Compiled by authors based on data from OEA (1987).

37.72 percent for grazing (Bedoya 1985, 84). In addition to these studies,
research conducted at the Experimental Station in Tingo María (the hemi-
sphere’s largest U.S. sponsored tropical research center, dating to war-era
strategic commodities), confirmed the zone’s productive possibilities, pro-
viding further support for moving ahead into the colonization program
(Lesevic 1984, 28).
Backed by agricultural research, a $15 million loan from the IDB plus
Peruvian public funds, the colonization project Tingo María-Tocache-­
Campanilla started in 1966. This project was Peru’s largest colonization
experiment of the 1960s, enhanced by the simultaneous construction of
the Marginal Highway of the Jungle (Nelson 1973, 103). The subsequent
modern history of VAH was inaugurated by this heady confluence of state
colonization projects, transport infrastructure expansion, and a large
spontaneous migration flow (Kernaghan 2009, 7).10 The 1966 project was
granted 456,800 hectares with a potential total of 130,000 hectares for
direct agricultural use. Officials planned to settle 4,227 families in the des-
ignated area between San Martín and Huánuco. The project’s territories
encompassed a large part of the districts of José Crespo and Castillo in
the Leonicio Prado province, the district of Cholón in Marañón province
(both in the department Huánuco) as well as the districts of Uchiza and
Tocache in Mariscal Cáceres province in the department of San Martín
(Bartra 1978, 71). The Tingo María-Tocache-Campanilla, project was, in
effect, in the heart of the VAH.
Just two years into the execution of the colonization project, the Rev-
olutionary Government of the Armed Forces removed Belaúnde from the
presidency. The new military government, which for various reasons redou-
bled colonization of the VAH, made two primary changes. First, the strat-
egy of individual and family settlement was replaced with the formation
of cooperatives like the Agrarian Cooperatives of Production (CAP) and
Agrarian Cooperatives of Service (CAS) (Bedoya 1981, 7). These were
medium- and large-size cooperatives administered by the state related to
66  Maritza Paredes and Hernán Manrique
the larger “socialist” character of Velasco-era agrarian reform. Second, the
military government reoriented the region’s agricultural projects towards
livestock and meat production (Bedoya 1981, 8), offering small loans for
grazers adopting mechanized livestock methods, with the hope of providing
greater meat supplies to the Peruvian population.

Colonization: blueprint for the modernization of agriculture

During the 1960s and 1970s, the colonization program launched by
Belaúnde’s administration and later by Velasco relied on international allies
for both financing and technical support. In contrast to other U.S. presi-
dents, Kennedy had the benefit of the presence of well-developed financial
institutions granting multilateral credit (such as the World Bank) as well
as a pre-existing infrastructure for the promotion of development world-
wide (Rist 2002). This infrastructure, in addition to channeling of financial
resources, offered technocratic expertise, knowledge and tools that prom-
ised to put scientific knowledge to work in the service of the newly-defined,
under-developed “Third World” (Ferguson 1994). The Kennedy administra-
tion set out to incorporate this developmentalist apparatus and its advo-
cates. Among them was W.W. Rostow, whose influential 1960 book The
Stages of Economic Growth: A Non-Communist Manifesto helped advance
a financing program to propel the so-called “takeoff” of the Latin American
economies (Taffet 2007, 21). These ideas resonated deeply with Fernando
Belaúnde, whose 1965 presidential address signaled that in order to “posi-
tion the country for the Rostow’s conditions for [economic] takeoff,” his
government would prioritize spending in the agrarian sector (Message to
Congress 1965).
The spread of agricultural developmentalism in Latin America was part
of a global agenda that supported Belaúnde’s government and animated its
plan to colonize Amazonia. This agenda promoted agricultural moderniza-
tion through the application of specialized knowledge of Peru’s agrarian
and rural realities. Development agencies propagated a new awareness of
potential and available strategies and a justification for their progressive
exploitation that, combined with structural reforms, would jumpstart the
modernization of agriculture.
During the 1960s, the United States supported a great number of agricul-
tural modernization projects through the Alliance for Progress and interna-
tional development institutions. A key part of the plan to funnel U.S. funds
to these initiatives was the creation of the Inter-American Development
Bank (IDB). Many of the expansion and modernization initiatives taking
place across the Andean countries were substantially funded by the IDB
and the World Bank, and received technical support from the Economic
Commission for Latin America and the Caribbean (ECLAC) (Barclay 1999,
61–62). With this international support, Andean countries began a series of
highway construction drives and related colonization projects.
 The Upper Huallaga Valley, Peru  67
With the signing of the Alliance for Progress treaty in Punta del Este, the
group of actors that supported agrarian reform and agricultural moderniza-
tion in Latin America became larger than ever. Organizations like the IDB,
the OAS, the FAO, ECLAC, and the Inter-American Institute for Cooper-
ation on Agriculture (IICA), among others, set out to directly tackle these
issues. Agricultural research in the region acquired relevance by guiding the
decisions of the agrarian reform and influencing agricultural planning. To
this end, new bodies like the Inter-American Committee for Agricultural
Development (CIDA)—comprised of specialists of the OAS, IDB, FAO,
ECLAC, and the IICA—formed with the objective of undertaking studies
to be used as the basis for an agrarian reform with a regional perspective.
These scientific studies turned out to be fundamental for understanding the
underlining causes of agricultural under-development and for providing
possible solutions. As noted by a CIDA evaluation document,

From a rural point of view, the latifundio system does not adapt to
growing demand for employment nor to growing demand for food.
Only a radical change in the distribution of land and land tenure can
produce an intensification in the use of arable lands, and, with it, an
increase in rural employment.
(CIDA 1966, 4)

The CIDA studies and FAO and ECLAC reports described the “striking par-
allels in the agrarian problem across all Latin American countries” (CEPAL
1978, 86). The belief that the land problem was similar across Latin Amer-
ica provided the basis for producing uniform development “blueprints,”
where international development agency recommendations, coordination
plans, and training would attain new levels of relevance. These blueprints
were shared and discussed in diverse settings and international meetings for
training Latin American land reform officials and experts. Organizations
involved in the promotion of agrarian reform like the IICA and the CIRA
carried out various International Executive Agrarian Reform Meetings with
the support of Project 206 “Training and Studies of Rural Development and
Agrarian Reform” in the Technical Cooperation Program of the OAS. These
development agencies hosted the first of these events in Lima and Bogotá in
1965 with the goal of cultivating a network for countries to exchange infor-
mation, regional ideas, and experiences, thereby fortifying the objectives of
agrarian reform in Latin America (IICA and CIRA 1966, 5). Similarly, the
IICA (with the support of Project 206) held international meetings on agrar-
ian reform at least a dozen times in varied countries of the region.
In the case of Peru, these development agency recommendations were
laden with developmentalist ideas and concepts that ended up playing a
decisive role in defining agrarian reform and the colonization of the Ama-
zon. ECLAC, which originally focused solely on themes of industrializa-
tion, moved during the 1960s toward agricultural development issues and,
68  Maritza Paredes and Hernán Manrique
in particular, toward agrarian reform policies (Bielchowsky 1998, 30). In
the words of Raúl Prébisch, the leading ECLAC structuralist thinker of the
era, “agrarian reform is imperative.” To bring about reform would require
a structural change that fostered “social mobility, satisfying the demands
of and improving the diets of the rapidly growing urban populations and
enhancing the quality of life for the rural masses” (Prébisch 1963, 47). With
this objective in mind, at the 9th ECLAC session in 1966, the commission
openly assumed the role of technical advising on design of national policies
for land reform (CEPAL 1961; Ortega 1988).
The agrarian reform blueprint ECLAC disseminated included legislation
that established concrete limits on the size of landholdings, downplayed
agrarian compensation in cash, and emphasized the technological modern-
ization of agriculture (CEPAL 1961). This last measure became a fundamen-
tal component of ECLAC’s work. In fact, with this particular objective in
mind, ECLAC, in conjunction with the FAO and the IDB, conducted a series
of diagnostic and future planning estimates about the physical inputs used
in agricultural activities such as fertilizers, pesticides, and farm machinery.
This Peruvian case study revealed inadequate technological mechanization
in the agricultural sector, insufficiently intensive exploitation of land, and
low agricultural productivity (CEPAL 1967, 2).
Throughout this decade, ECLAC also initiated a series of studies about the
production structures and trends in regional production. In its 1968 Peru-
vian case study, ECLAC highlighted the overall deficient state of national
agricultural production. The report attributed the failing condition of Peru-
vian agriculture to: (1) the poor use of natural resources like land and water
and the low availability of capital and qualified technical personnel in Peru;
(2) the deficient structure of landholding and land-owning and insufficient
development of irrigation technology; (3) the absence of a rational transport
system for the commercialization and industrialization of agricultural prod-
ucts; (4) inadequate conditions for primary agricultural materials destined
for export; and (5) scarce coverage and depth of developmental activity car-
ried out by the state (CEPAL 1968, 10).
The majority of international development agencies were actually
guarded in their recommendations regarding colonization policies, but they
still ended up funding them. ECLAC considered colonization as part of a
series of measures in a larger package of policies for agrarian reform, to
be a valid form of expanding the agricultural frontier. It explicitly warned
that “given the high costs [of colonization due to] direct investments and
infrastructure,” the execution of colonization policies was contingent on
the availability of capital (CEPAL 1961, 41). Moreover, in one evaluation
document, ECLAC notes that it “sometimes advocates the colonization
of new lands, even when there are poorly-exploited latifundios that may
be subdivided into many smaller investments” (CEPAL 1961, 41). Other
agencies, notably the World Bank, seemed less supportive of the restructur-
ing of land tenure and backed agricultural modernization in zones like the
 The Upper Huallaga Valley, Peru  69
lowland tropics as a more “promising” geographic region of Peru, what-
ever that meant. Once there, they suggested increasing efforts in mecha-
nizing agriculture, fertilizer use, agricultural credits, and scientific research
(van der Laar 1980).
The global spread of modernization policies and agrarian reform was
driven by international development agencies. However, the policies of the
Andes were distinct from those in the rest of Latin America in the 1960s
given the intensity with which elites opposed policies of land redistribution.
Instead, agrarian and political elites channeled international support for
land reform into the expansion of the agricultural frontier. These agencies’
financial and technical support would underwrite projects to colonize the
upper jungle through directed resettlements, road construction, and incen-
tives like credits, subsidies, and technical support to would-be colonists.
With an eye towards Amazonia, Peruvian elites equipped with international
support hoped “to construct a social and economic space not subject to the
hurdles or institutional and cultural distortions that are colonial in origin
which impede the modernization of other spaces, particularly in the Andean
area” (Barclay 1999, 48).
The spread of ideas, technical knowledge, and financing from international
development agencies for colonization of the Amazon is observed in the
expansion of agricultural credit for small producers. Along with the increase
in highways, agricultural credit was the single most important instrument
for promoting the colonization of the Amazon. This tool was advocated by
international development agencies in Latin America after CIDA discovered
the sparse use of agricultural credits in the region. According to statistics by
van der Laar (1980, 160), between 1960 and 1969 around half of the total
amount of foreign assistance provided by USAID, the IDB, and the World
Bank for agriculture in Latin America—or some $915 million ­dollars—was
designated for agricultural credits.
The World Bank led the agencies promoting agricultural credits. In Peru,
technical and financial support from the World Bank to the Banco de Fomento
Agropecuario [Bank for the Promotion of Agriculture] (BFA) began with
the creation of the BFA in the early 1960s as a side agreement signed with
Alliance for Progress. This support buttressed the crusade for colonization
in the 1960s and was justified by the CIDA report’s findings regarding the
lack of usable agricultural credit. Table 3.2 shows the World Bank loans des-
tined for agricultural credits granted to the Peruvian state between 1954 and
1983, a total of $45 million dollars. The first three loans, granted between
1954 and the mid-1960s, were short-term projects, two or three years each,
designed to “increase agricultural productivity in Peru through the import
of machinery, equipment, materials, and cattle and through the expansion of
available credit for agricultural producers” (World Bank 1954, 1957, 1960).
By 1965, the capital loaned for the BFA had increased considerably, largely
directed towards the mechanization of agriculture and the importation of
livestock for small producers (van der Laar 1980).
70  Maritza Paredes and Hernán Manrique
Table 3.2  List of World Bank loans to the Peruvian government through the Banco
de Fomento Agropecuario (BFA) to expand agricultural credits (in USD)

Year Loan name Loan identification Amount

number

1954 First Agricultural Credit Project 105 PE $5,000,000

1957 Second Agricultural Credit Project 162 PE $5,000,000
1960 Third Agricultural Credit Project 257 PE $5,000,000
1965 Fourth Agricultural Credit Project 415 PE $15,000,000
1973 Fifth Agricultural Credit Project 933 PE $25,000,000

Source: World Bank (1954, 1957, 1960, 1965, 1981).

Table 3.3  Percentage of BFA credits ­granted

in the jungle, 1941–1989

Period Percentage of total

credit

1941–1950 3%
1951–1960 5%
1961–1970 20%
1971–1980 22%
1981–1990 24%

Source: Barclay (1999), BFA and Banco Agrario del

Perú archives.

Initially, agricultural credit was fundamentally directed at coastal plant-

ers. Progressively, however, these credits expanded to agriculturalists in the
Amazonian regions under colonization in the 1960s. Loans for agricultural
and colonization projects multiplied in the region during this period, rising
from 5 percent of the loans in the 1950s to as high as 20 percent in the
1960s (Table 3.3).
As demonstrated in Table 3.2, after Velasco’s “revolutionary” military
government of the early 1970s, the sum of World Bank financing for agri-
cultural and farming credits notably increased. The World Bank and the
IDB opened lines of credit for large-scale animal husbandry and mechanized
clearing (i.e., deforestation) in Amazonian colonization sites. The military
regime established cooperative production units, the Agrarian Production
Cooperatives (CAP) and the Agrarian Service Cooperatives (CAS), in the
Project Tingo María-Tocache-Campanilla zone, which used these lines of
credit to purchase cattle and machinery. These were medium- and large-scale
peasant-run cooperatives administered by the state. From the mid-1970s,
loans granted for livestock represented more than 60 percent of the debt
of 15 CAP, CAS, and companies in the region owed to the Banco Agrario
(Bedoya 1981, 8).
 The Upper Huallaga Valley, Peru  71
By legitimating the justification for and technical operations of agricul-
tural credits, the World Bank and its kindred development agencies deeply
influenced the Peruvian state’s colonization programs. The intensive use of
agricultural credit was not particular to Peru. Under the assumption that
credits would effectively increase farmer productivity beyond daily subsis-
tence, development agencies such as the World Bank made this agricultural
instrument global in the 1970s. Loans to small agriculturalists would raise
production and agricultural productivity through mechanization. Financial
investments constituted the most economic use of both the agriculturalists
and national resources, developing nationwide agrarian credit institutions in
the process (van der Laar 1980). Thus, it was the structure of World Bank
loans in the 1970s, and not Peruvian government preferences, that priori-
tized for loans for machinery and livestock to small farmers.
Yet, these livestock investments yielded less than was hoped for, and var-
ious cooperatives wound up severely indebted. By 1976, for example, the
CAP and CAS in the Tingo Maria-Tocache-Campanilla project region were
more than 230 million soles in debt (Table 3.4). This enormous debt was
decisive for the later imbalances and bankruptcies of these cooperatives.
In sum, the diffusion of agricultural credits among small farmers as a
tool for farming modernization was a perfect example of the dissemina-
tion effect of international development agency ideas, technical methods,
and resources. It also demonstrates how powerful local actors employed
these disseminating ideals, methods, and financing for their own gain, by
channeling farming credits toward the colonization of the Amazon to push
poor peasants to move to “free lands.” Few considered the possible failure

Table 3.4  Debts of cooperatives affiliated with colonization project Tingo

­Maria-Tocache-Campanilla for clearing forest

Cooperative name Total debt Debt incurred Percentage

incurred (in for mechanical (of total debt)
soles) clearing (in soles)

CAP Perú Oriental 15,449,192 1,653,819 10.7

CAP Arequipa 20,781,789 2,230,453 10.73
CAP Nuevo 4,118,740 2,302,529 50.7
Horizonte
CAP Piura 18,423,771 2784607.82 15.11
CAS Anda Pacae 3,399,969 876,603 25.78
CAS La Marginal 11,511,153 213,098 1.85
CAS La Morada 12,017,242 6,013,976 50.04
CAS El Progreso 13,142,289 1,521,205 11.57
CAS San Martín 20,397,461 6,229,295 30.53
de Pucate
Total 126,241,603 23,825,587 18.87

Source: Bedoya, La destrucción del equilibrio ecológico en las cooperativas del Alto Huallaga
(1981, 37).
72  Maritza Paredes and Hernán Manrique
of credit policies to increase productivity, which peasants often used for
immediate subsistence instead. With different goals, the Peruvian state and
international development agencies pushed for colonization through dubi-
ous new technical instruments like agricultural credits.

Refugees in the Huallaga: between reliance on the state or

dependence on illicit coca
This section focuses on the radical unexpected transformation of the VAH
in this two-decade span and its unpredictable results. In particular, the VAH
saw the formation of an enclave of poor peasants dependent on the state in
an environment of growing illegality and violence. Through state coloniza-
tion policies, colonists and peasants came to be strongly dependent on the
intervention of Peruvian state institutions and allied international actors.
When the Peruvian state both retreated following the failure of colonization
practices in the region, and underwent its own financial bankruptcy in the
mid-1970s, local peasant reliance quickly shifted to coca cropping.
The use of new institutions and techniques that measured and evaluated
the economic potential of soil in the region had generated false expecta-
tions about tropical land productivity in the complex ecological context
of the Huallaga. Inexperience with the local ecology made it difficult to
execute directives to produce information and analyses on soil fertility
to serve the colonists in the VAH. In effect, the Huallaga’s wide range of
soil types and the rapid onset of spontaneous colonization into the valley
outstripped the capacities of technical studies and the ability to generate
land capacity maps. As a result, many of the new inhabitants settled in
less-fertile corners of the valley. Previous assumptions about the improv-
ability and potential of tropical productivity and crop yields were used to
justify the liberal agricultural credit policies of international development
agencies for extremely impoverished new families. Having never received
credit before, and little onsite advice, these loans were often used for their
immediate subsistence instead of the intended goal of improving produc-
tive capital and agricultural investments.
The VAH underwent two important transformations during the 1960s
and 1970s. The first was demographic. An unexpectedly large number of
poor peasants migrated to the VAH. The construction of roads and other
infrastructure and the Tingo María-Tocache-Campanilla colonization
project under Belaúnde brought an unplanned settlement of thousands of
colonists on both sides of the Huallaga River. Overall, the population of
VAH increased by more than 100 percent in less than a decade, rising from
45,187 inhabitants in 1961 to 92,237 inhabitants in 1972 (see Table 3.5).
These numbers clearly exceeded Belaúnde’s or anyone’s initial expectations.
Official settlement projects and road construction initiated by the gov-
ernment were likely intended to indirectly spur more spontaneous migra-
tion to the valley, but the scale of migration eclipsed anything planned. The
 The Upper Huallaga Valley, Peru  73
Table 3.5  Total population of Valle de Alto Huallaga (VAH), 1940s–1980s

Census year 1940 1961 1972 1981

Total population 11,623 45,187 92,237 134,600

Source: Aramburú (1989, 234)

r­ esulting land occupations and transitory agricultural practices such as slash

and burn of the jungle led to a serious problem in subsequent abandonment
of land parcels. Many inhabitants—above all those who had arrived after
the initial phases of the project—occupied abandoned parcels only to find
that these lands already had legal owners and now lay fallow (Aramburú
1981, 58). This threatened the existing colonization project, as the demand
for land soon outstripped its availability (Aramburú 1989, 248). Conse-
quently, in the face of land shortages, by the 1970s new colonists expanded
the agricultural frontier into the adjacent jungles and hillsides in search of
plots to exploit and settle.
At the start of the project, more than 5,400 families were already liv-
ing in the zone designated for colonization. But by the early 1970s, the
estimated absorption of this area was already 28 percent over capacity
(Aramburú 1989, 246). A 1974 diagnostic report by CENCIRA warned
of the glut of migrants to the colonization zone. CENCIRA expected the
settlement of 4,227 families by the end of the project—a total of 23,400
colonists assuming five members per family. However, according to the
1972 Census, the zone alone already had 29,998 inhabitants—6,500 over
target (CENCIRA 1974, 95). Moreover, both the spontaneously migrat-
ing colonists and those the state had intentionally settled in the region
established themselves in scattered zones apart from the large land area
officially designated by the colonization project. In part, this mass migra-
tion had to do with the rapid pace of Velasco’s agrarian reform in the
highlands, which along with providing peasants with new access to land
and employment was also expelling them (Meyer 2009). These same fac-
tors contributed to the explosive growth of Peru’s towns and cities in this
era, notably on the coast. With the opening of border commerce and the
construction of the Marginal Highway, colonists continued arriving and
forming new towns in VAH. Demographic pressure rose steeply, whether
through the spontaneous migrations that produced the towns of Pucayacu
and Santo Domingo de Anda, or the state-sponsored settlements of La
Morada and Pucate. While the VAH population was not the only one in
Peru experiencing exponential growth, from the 1960s onward growth
rates in the provinces of the VAH proved higher than in the rest of the
upper jungle (Table 3.6).
The second transformation in the VAH was the spread of a culture of state
dependence, which emerged in part from misguided institutional fantasies
about the productive capacity of the Amazonian frontier. Colonists faced
74  Maritza Paredes and Hernán Manrique
Table 3.6  Rural population growth rates in Alto Huallaga

Zone/Province Rural population

1940–1961 1961–1972 1972–1981

Selva Alta (Upper jungle zone) 3.8 4.69 3.67

Mariscal Cáceres (Province) 3.19 6.73 4.6
Leoncio Prado (Province) 7.61 6.01 2.96

Source: Rodríguez (1991, 149)

multiple challenges in maintaining sustainable economic a­ utonomy, and,

significantly, this crisis occurred in a zone without any traditional landed
class to offer campesinos direction, patronage, or authority. Anthropologist
David Nugent notes something similar in his study of the effort to modern-
ize post-1920s Chachapoyas, writing that:

The “population” that government officials assumed was [out] there

waiting to be put to work was not a thing of fact, but of fiction. It
was conjured into being, bit by bit, from the imaginations of individual
bureaucratic planners … unable to procure this (fictive) labor supply, a
crisis ensued.
(Krupa and Nugent 2015, 192)

The VAH was not in need of more labor per se, but faced a looming crisis
as a result of exaggerated institutional projections of productive capacity
for traditional commercial agricultural crops. The Tingo María-Tocache-­
Campanilla colonization project easily reached its settlement quotas, but
in so doing also created colonist dependence on the state and international
development agencies initially supporting the project (Aramburú 1981, 53).
But the state ran into serious difficulties keeping its promises and fulfill-
ing its objectives, as it did throughout Peru in this era. The ONERN, the
body responsible for undertaking and analyzing the social and economic
potential of colonization projects, significantly overestimated the economic
potential of the lands, providing an unwarranted legitimization of the VAH
colonization (Lizárraga 1968, 251). However, over the project’s lifetime, the
legendary productivity of upper jungle lands transformed into a nightmare
for colonists, who suffered constantly from problems stemming from unpro-
ductive crops. By the 1980s, ONERN published revised studies revealing the
substantially lower economic potential of VAH soils than initially suggested
(Bedoya 1985).
Having first prioritized farming incentives for the cooperatives, access
to such incentives—a primary source of state support—became greatly
restricted for individuals and families during the later Velasco regime.
During the Revolutionary Government of the Armed Forces, military forces
 The Upper Huallaga Valley, Peru  75
r­ eorganized the colonization of the valley in accordance with official guide-
lines. The individual, family, and cooperative settlements created in the orig-
inal project were replaced with a model of cooperatives—specifically, CAP
and CAS. The larger cooperatives quickly hoarded available credits, leaving
many other colonists empty-handed. In one interview, Alberto Chirif noted
the importance of these credits for colonists launching their own produc-
tive agricultural activities. With individual credits slashed, smaller colonists
became desperate. However, this was not the only change introduced by the
Revolutionary Government. Beginning in the 1970s, the military sought to
improve and expand not just large-scale agricultural production but farming
as well (Manrique 2016, 67). Bearing in mind soil problems and poor crop
yields in VAH, the military government prioritized deforestation for larger
farms and cattle ranching. This changed policy only aggravated the agri-
cultural crisis, as mechanized (by bulldozers or tractors) and manual defor-
estation seriously eroded already precarious topsoil. One observer described
the “destruction of ecological balance” (Bedoya 1981, 1–8) which left some
soils so depleted that it became costly to produce any crop other than coca.
At the start of the 1970s, CENCIRA’s report evaluating the colonization
project’s outcomes expressed serious concerns. According to its 1974 report
(CENCIRA 1974, 11), since 1973 parts of the project’s core suffered from
a high level of land abandonment as a result of the low agricultural per-
formance of the selva alta. This type of reconsideration was inconsistent
with the scientific evidence initially deployed to galvanize the project. Given
the gravity of these concerns, the same report recommended ending the
promotion of colonist migration, without reinforcing those farmers already
there, and further suggested establishing restrictions on further mechanized
deforestation (24, 38–9). Observers also reiterated the remoteness of colo-
nists from centers of state assistance, such as hospitals, clinics, police stations,
or schools, as well as normal networks of communication like telephones.

Crisis and the Coca Boom

In the mid-1970s Huallaga, the increase in U.S.-based demand for cocaine
transformed coca into a veritable “white gold” for the peasants of VAH
(Morales 1989). At its peak, one kilo of coca could fetch $3.60, far more than
any other legal crop (Briceño and Martínez 1989, 267). Coca and cocaine
were hardly new to the region. On the contrary, the department of Huánuco,
which partially housed the colonization project, had been the capital of legal
cocaine production since the start of the twentieth century (Gootenberg
2008). Concentrated in higher elevation Chinchao, cocaine was legal to pro-
cess from local coca in licensed local workshops until 1948. By the 1940s
and 1950s legal, sometimes large, commercial coca plantations had spread
down the valley into the areas surrounding Tingo María and Tulumayo, and
the frontiers of nearby Monzón. By the 1970s, this new demand for illicit
cocaine abroad and trafficking networks erupting in and to the VAH would
76  Maritza Paredes and Hernán Manrique
coincide with a growing recession that came to a head in the 1980s. Many
crop prices nosedived in large part due to the passage of Public Law 480 in
the U.S., ostensibly a form of foreign aid, which resulted in Peru’s importa-
tion of U.S.-grown surplus foodstuffs like potatoes, wheat, corn, and rice,
thus undermining most local crop production (Thorp and Paredes 2010,
196). In the VAH, this perfect storm meant few profitable alternatives to
coca cultivation. In conjunction with the region’s unexpectedly poor soils,
struggling colonists turned to coca for subsistence, and shifted their reliance
on vanishing state support to a new reliance on producing illicit coca. In just
a few years, the VAH became the cocaine capital of the world.
In the wake of coca’s rapid spread in the VAH, two conflicts material-
ized in the valley: the war on drug trafficking and the war on terrorism.
In the face of proliferating drug production and U.S. pressures against the
now internationally bankrupt Peruvian state, in 1978 the state passed Legal
Decree 22095 in 1978, known as the “Law to Repress the Illegal Traffic of
Drugs.” In 1979 and 1980, the government carried out two raids, Verde Mar
I and II, in which hundreds of police and military descended on the valley
to not only eradicate coca but also to confiscate lands and detain offending
peasants. This initiated a new hostile relationship between the state and the
once-favored colonists. In spite of this campaign, or perhaps because of it,
coca production continued its increase and the valley cultivated anywhere
between an estimated 120,000 to 195,000 hectares of coca leaf (Gootenberg
2008, 300; Ugarte 2014, 80). As for the war on terrorism, the valley’s coca
boom became a golden opportunity for Peru’s two insurgencies, Sendero
Luminoso and the MRTA, in the early 1980s. The coca boom could pro-
vide financing for their armed struggle (something Sendero at first resisted)
and also provided the possibility of cadres winning local political legitimacy
by providing the colonists-turned-cocaleros with armed protection against
state efforts to eradicate coca and confiscate drugs (Manrique 2015, 42).
During the most critical moments of Peru’s internal conflict of the next
decade, the inhabitants of the VAH found themselves caught in the crossfire
between terrorist groups and military forces, in addition to becoming victims
of abuse by drug traffickers. The escalating militarization of the valley created
unimaginable legal and survival challenges for ex-colonists, as once promised
economic and moral support became increasingly restricted and controlled.
In their effort to escape these dilemmas and restore social peace, many colo-
nists involved themselves in the fight against terrorism, forming self-defense
groups, Comités de autodefensa and rondas campesinas (Manrique 2016,
230). However, the military defeat of terrorist groups by the mid-1990s had
an unexpected secondary effect. In reality, defeating these groups (along with
short-lived aerial interception strategies and Colombia’s own massive coca
boom) destabilized the drug trade in the region, sharply depressing coca prices.
It threatened the population and hamlets of the VAH by eliminating coca as a
secure source of sustenance (Manrique 2015, 45). Again, this forced Huallaga
campesinos to return to the state for both subsistence and protection.
 The Upper Huallaga Valley, Peru  77
This section has briefly illustrated how the Valle del Alto Huallaga was
transformed into a key global cocaine production zone. In less than two
decades, from 1960 to 1980, state and development agency intervention in
the upper jungle of Peru unintentionally converted the valley into a region
where a large sector of the local population, brought by failed colonization
projects and poor agricultural prospects, became dependent on coca leaf for
their livelihoods by the late 1970s. By the 1980s, they fell under dire threat
in the fire-fight from the twin wars on drugs and terrorism. Only after years
of violent struggle, and the conclusion of Peru’s internal conflict, would new
initiatives emerge for productive restructuring of the zone and the alterna-
tive forms of development that its inhabitants had long requested from the
state. The following testimony provided to one of these authors, reflects as
much:

I believe that the violence also opened the eyes of the state to how Alto
Huallaga had been totally neglected and abandoned. From then on, the
state began to react and invest more in schools, infrastructure, more
health, education, and agricultural benefits
(Interview with Tocache farmer; Manrique 2016, 163)

It is a history filled with paradoxical twists and turns.

Conclusion
This chapter reveals the various radical and unexpected sides of the transfor-
mation that took place in VAH in the span of two decades after 1960, as well
as the unanticipated consequences in the process that enabled the cultivation
of an illicit coca market. Colonization policies motivated by elite political
self-interest and funded by international development agencies facilitated
the formation of colonies of poor peasants with little access to sustainable
livelihoods and an abject new dependence on state support. These policies
were also concomitant with calls for agrarian reform from both inside and
outside Peru; colonization efforts were the result both of elite efforts to cir-
cumvent the need to relinquish large landholdings and a general Cold War
effort to quash Communist or rebellious sympathies among the poor by
“developing” them. International agencies spearheaded or funded by Ken-
nedy’s Alliance for Progress designed agricultural development strategies for
all of Latin America, generating a new global perspective on land reform. It
combined the promised potential of a country’s resources with a justification
for structural reforms that would enable the modernization of agriculture
and of the peasantry, and like other U.S. Cold War policies had a political
or imperial content.
The foreign-funded state project to colonize Peru’s Amazonian interior
was, in turn, specifically catalyzed by new forms of governance, state con-
trol, and intervention pioneered by international development agencies
78  Maritza Paredes and Hernán Manrique
armed with an unwavering faith in the power of small agricultural credits
and science about the imagined productive promise of the highland jungle.
Yet when this infusion of credits was used in unanticipated ways, for subsis-
tence instead of investments in productive capital, and when the ecology of
this tropical region failed to support the kind of large-scale agriculture envi-
sioned by the state, the dream of breaking ground on an agriculturally rich
frontier began to quickly crumble. These failures collided with collapsing
domestic crop prices and rising international demand for cocaine, spurring
the rapid expansion of coca production in the VAH by the late 1970s. With
this turn to illegal coca cultivation, the peasants promised an agricultural
paradise found themselves mired instead in the valley’s deepening violence
as state forces cracked down on illegal coca production and terrorist groups
entering the region. In this way, from the 1960s to the 1980s, the valley
transformed from a “win–win” political fantasy, first dreamt up by national
landholding elites and the modernizing Peruvian state, to the unforeseen
scenario of colonists’ dependence on illegal coca for survival.

Notes
1 Both authors want to acknowledge the translation assistance of Adele Zhang and
comments of Paul Gootenberg and Liliana M. Dávalos on previous versions of
this chapter.
2 In 1984, the province of Tocache was created by partitioning the province of
Mariscal Cáceres in the department San Martín.
3 While the central focus of Gootenberg’s 2008 work Andean Cocaine is not the
VAH so much as the history of cocaine as a market good, it provides substantial
background on the origins of coca and cocaine in the valley, including some on
migration-colonization patterns and programs dating to the 1930s.
4 This information may be found in the IDB’s repository, although digital access is
restricted.
5 General Velasco executed a military coup on October 3, 1968, deposing the dem-
ocratically elected Fernando Belaúnde Terry. His administration was called the
Revolutionary Government of the Armed Forces and proposed anti-oligarchic
and nationalistic policies for the country that included one of the most radical
land reforms in Latin America. The law passed stipulated that all lands of greater
than 150 hectares on the coast and between 15 to 55 hectares in the highlands
(depending on the location) would be subject to expropriation, with no excep-
tions granted.
6 A variety of laws sought to promote internal colonization and the expansion of
the agricultural frontier into Amazonia, particularly the highland jungle, in Peru
called called “the montaña” (crudely, the “Mountain”). For this reason, the Law
of Immigration and Colonization passed in 1893 to permit foreign colonists to
settle Amazonia, and the Organic Law of Montaña Lands of 1898 to promote
internal colonization and land acquisition by foreign companies (Monterroso
2017, 4). The most important law, however, was the Legislation of Lands (Law
1220) passed in 1909—with a term of 65 years—that converted all undistributed
land into state property. This enabled the state to sell, adjudicate, or freely offer
concessions (Law 1220, Art.2).
7 Law 8687 passed in the same year, promoting the colonization of “zones that
cross the roadway that goes from the capital of Huánuco to the Bajo Ucayali
 The Upper Huallaga Valley, Peru  79
River for a length of twenty kilometers on both sides of the river” (Law 8687).
That same year, the Regulation of the Tingo María Colonization Center ordered
the execution of Law 8687. This regulation privileged Peruvian colonists and
offered free: (1) transport to the zone, (2) houses or shelter, (3) adequate fur-
nishings and (4) tools and seeds (Gazzolo 1966, 86). Some years later, in 1950,
Law 114361 was passed to fortify these efforts, extending colonization along the
Huallaga River in the Tingo María-Yurimaguas region.
8 According to the 1940 Census, the Amazonian departments of San Martín,
Madre de Dios, Loreto, and Amazonas represented a mere 5 percent of the Peru-
vian population.
9 These were not Peru’s only colonization projects. Outside of Amazonia, other
regions like the right bank of the Apurímac River, the port of Chimbote, and La
Joya in Arequipa were selected for colonization (CIDA 1966, 452–454).
10 While this claim may be disputed, Gootenberg 2008 describes how part of the
VAH had already been shaped by the legal cocaine industry starting in the late
nineteenth century, although modest in comparison to the massive boom that
took place from the 1960s onward.

Bibliography
Alacevich, Michele. 2009. The Political Economy of the World Bank: The Early
Years. Stanford: Stanford University Press and the World Bank.
Albertus, Michael. 2015a. “Explaining Patterns of Redistribution Under Autocracy:
The Case of Peru’s Revolution from Above.” Latin American Research Review,
50 (2): 107–134.
Albertus, Michael. 2015b. Autocracy and Redistribution: The Politics of Land
Reform. Cambridge: Cambridge University Press.
Albertus, Michael and Oliver Kaplan. 2013. “Land Reform as a Counterinsurgency
Policy: Evidence from Colombia.” Journal of Conflict Resolution, 57 (2): 198–231.
Altman, David and Juan Pablo Luna. 2012. “Introducción: El Estado latinoameri-
cano en su laberinto.” Revista de Ciencia Política, 32 (3): 521–543.
Aramburú, Carlos Eduardo. 1979. “Las migraciones a las zonas de colonización en
la selva peruana: perspectivas y avances.” Debates en Sociología, 4: 81–94.
Aramburú, Carlos Eduardo. 1981. “Problemas del desarrollo rural y la colonización
en la Amazonía peruana.” Debates en Sociología, 6: 41–70.
Aramburú, Carlos Eduardo. 1982. “Expansión de la frontera agraria y demográfica
en la selva alta peruana.” In Centro de Investigación y Promoción Amazónica
(CIPA) (ed.), Colonización en la Amazonía. Lima: CIPA, 1–39.
Aramburú, Carlos Eduardo. 1989. “La economía parcelaria y el cultivo de la coca: el
caso del Alto Huallaga.” In León, F. and Castro de la Mata, R. (eds.), Pasta Básica
de Cocaína: un estudio multidisciplinario. Lima: CEDRO, 231–260.
Banco Mundial. 1975. Land Reform. Sector Policy Paper. Washington, DC: World Bank.
Barclay, Frederica. 1999. “Protagonismo del Estado en el proceso de incorporación
de la amazonía.” In Barclay, F. (ed.), Amazonía 1940–1990, el extravío de una
ilusión. Lima: Terra Nova y PUCP, 43–100.
Barraclough, Solon and Arthur Domike. 1966. “Agrarian Structure in Seven Latin
American Countries.” Land Economics, 42 (4): 391–424.
Bartra, Werner. 1978. “Situación de los asentamiento rurales en la colonización
Tingo María-Tocache-Campanilla.” In IICA (ed.), Seminario sobre Asentamien-
tos Rurales Integrales. Tingo María: UNAS.
80  Maritza Paredes and Hernán Manrique
Bedoya, Eduardo. 1981. La destrucción del equilibrio ecológico en las cooperativas
del Alto Huallaga. Lima: CIPA.
Bedoya, Eduardo. 1982. “Expansión de la frontera agraria y demográfica en la selva
alta Peruana.” In Aramburú, Carlos Bedoya, C. and Bedoya Eduardo (eds.) Col-
onización en la amazonía peruana. Lima: Edición CIPA.
Bedoya, Eduardo. 1985. “Intensificación y Degradación en los Sistemas Agrícolas de
la Selva Alta: El caso del Alto Huallaga.” In Bedoya, E. (ed.), Estrategias produc-
tivas y recursos naturales en la Amazonia. Lima: CIPA, 48–98.
Belaúnde, Fernando. 1959. La conquista del Perú por los peruanos. Lima: Ediciones
Tawantinsuyu.
Belaúnde, Fernando. 1967. Carretera Bolivariana Marginal de la Selva. Lima: n/d.
Belaúnde, Fernando. 1965. Mensaje del Presidente Constitucional del Perú, Arqui-
tecto Fernando Belaunde Terry, ante el Congreso Nacional, el 28 de julio de
1965. Available at: www.congreso.gob.pe/Docs/participacion/museo/congreso/
files/mensajes/1961-1980/files/mensaje-1965.pdf
Bielchowsky, Ricardo. 1998. “La evolución de las ideas de la CEPAL.” Revista de la
CEPAL, número extraordinario, Oct 1998: 21–45.
Briceño, Juan and Jorge Martínez. 1989. “El ciclo operativo del tráfico ilícito de la
coca y sus derivados: implicancias en la liquidez del sistema financiero.” In León,
F. and Castro de la Mata, R. (eds.), Pasta Básica de Cocaína: un estudio multidis-
ciplinario. Lima: CEDRO, 261–279.
CENCIRA. 1974. Diagnóstico Socio-Económico del Proyecto de Colonización
Tingo María-Tocache-Campanilla. Lima: CENCIRA.
CEPAL. 1961. Economic Development, Planning and International Co-operation.
Santiago de Chile: CEPAL.
CEPAL. 1963. Programas y tendencias sociales en América Latina. Mar de Plata:
CEPAL.
CEPAL. 1967. El uso de fertilizantes en el Perú. Santiago de Chile: CEPAL.
CEPAL. 1968. Desarrollo Económico e Integración de América Latina: el caso del
Perú. Santiago de Chile: CEPAL.
CEPAL. 1978. 25 años en la agricultura de América Latina: Rasgos principales
(1950–1975). Santiago de Chile: CEPAL.
Chirinos, Alfonso. 1975. “La Reforma Agraria peruana.” Nueva Sociedad, 21: 47–64.
Cleaves, Peter and Martin Scurrah. 1980. Agriculture, Bureaucracy, and Military
Government in Peru. Ithaca: Cornell University Press.
Comité Interamericano de Desarrollo Agrícola. 1966. Tenencia de la tierra y desar-
rollo, socio-económico del sector agrícola, Perú. Washington, D.C.: Unión
Panamericana.
Cotler, Julio. 1999. Drogas y política en el Perú: la conexión norteamericana. Lima:
Instituto de Estudios Peruanos.
Cotler, Julio and Felipe Portocarrero. 1976. “Organizaciones campesinas, en el Perú.”
In Mar, M. (ed.), Hacienda, comunidad y campesinado en el Perú. Lima: Instituto
de Estudios Peruanos: 267–310.
Crist, Raymond and Charles Nissly. 1973. East from the Andes. Pioneer Settlements
in the South American Heartland. Gainesville: University of Florida Press.
DDCD. 1963. Official documents. Congreso de la República del Perú, Diario de los
Debates de la Cámara de Diputados, 1963, vol. I.
DEVIDA. 2015. Compendio Normativo sobre Tráfico Ilícito de Drogas y Desarrollo
Alternativo. Lima: DEVIDA.
 The Upper Huallaga Valley, Peru  81
Dreyfus, Pablo. 1999. “When All the Evils Come Together.” Journal of Contempo-
rary Criminal Justice, 15 (4): 370–396.
Eidt, Robert. 1962. “Pioneer Settlement in Eastern Peru.” Annals of the Association
of American Geographers, 52 (3): 255–278.
Felbab-Brown, Vanda. 2010. Shooting up: Counterinsurgency and the War on Drugs.
Washington, DC: The Brookings Institution.
Ferguson, James. 1994. The Anti-Politics Machine: Development, Depoliticization
and Bureaucratic Power in Lesotho. Minneapolis: University of Minnesota Press.
Fernández-Shaw, Félix. 1962. “La integración económica iberoamericana.” Revista
de Estudios Políticos, 121: 147–174.
Garcé, Adolfo. 2002. Ideas y competencia política en Uruguay (1960–1973). Revis-
ando el fracaso de la CIDE. Montevideo: Ediciones Trilce.
Gazzolo, Luis. 1966. “El régimen legal de tierras de montaña.” Derecho Rural, Part
I: 86–98.
Gonzales, José. 1992. “Guerrillas and coca in the Upper Huallaga Valley.” In Scott
Palmer, D. (ed.), Shining Path of Peru. New York: St. Martin’s Press, 105–126.
Gootenberg, Paul. 2008. Andean Cocaine: The Making of a Global Drug. Chapel
Hill: University of North Carolina Press.
Griffith-Jones, Stephany. 1979. “The Alliance for Progress: An Attempt at Interpreta-
tion.” Development and Change, 10 (3): 423–443.
Guzmán, Virginia and Virginia Vargas. 1981. Cronología de los movimientos cam-
pesinos, 1956–1964. Lima: Investigación, Documentación, Educación, Asesoría,
Servicios.
Huntington, Samuel. 1968. Political Order in Changing Societies. New Haven: Yale
University Press.
IICA and CIRA. 1966. Informe de la Reunión Internacinal de Ejecutivos de la
Reforma Agraria y de la Reunión de Evaluación y Planeamiento del Proyecto
206. Bogotá: IICA and CIRA.
Kapstein, Ethan. 2014. “Land and Rebellion : Lessons for Counter-insurgency.”
Global Politics and Strategy, 56 (2): 109–128.
Kay, Bruce. 1999. “Violent Opportunities: The Rise and Fall of ‘King Coca’ and Shin-
ing Path.” Journal of Interamerican Studies and World Affairs, 41 (3): 97–127.
Kernaghan, Richard. 2009. Coca’s Gone: Of Might and Right in the Upper Huallaga
Valley Post-boom. Stanford: Stanford University Press.
Krupa, C., and D. Nugent. 2015. State Theory and Andean Politics: New Approaches
to the Study of Rule. University of Pennsylvania Press, Incorporated.
Kofas, Jon. 1996. Foreign Debt and Underdevelopment: U.S.-Peru Economic Rela-
tions, 1930–1970. Lanham and London: University Press of America.
Lastarria-Cornhiel, Susana. 1989. “Agrarian Reforms of the 1960s and 1970s in
Peru.” In Thiesenhusen, W. (ed.). Searching for Agrarian Reform in Latin Amer-
ica. Boston: Unwin Hyman, 127–156.
Lesevic, Bruno. 1984. “Dinámica demográfica y colonización en la selva alta peru-
ana: 1940–1981.” In Consejo Nacional de Población y Centro de Investigación y
Promoción Amazónica (ed.), Población y colonización en la alta amazonía peru-
ana. Lima: CNP y CIPA, 11–50.
Lizárraga, José. 1968. “Oficina Nacional de Evaluación de Recursos Naturales.” In
National Academy of Sciences (ed.), Second Peru-U.S. Workshop on Science and
Technology in Economic Development, Vol. 2. Washington, DC: National Acad-
emy of Sciences, 248–254.
82  Maritza Paredes and Hernán Manrique
Lowenthal, Abraham. 1975. “Peru’s Ambiguous Revolution.” In Lowenthal, A. (ed.),
The Peruvian Experiment: Continuity and Change Under Military Rule. Prince-
ton: Princeton University Press, 3–43.
Manrique, Hernán. 2015. “Las bases históricas del ‘milagro de San Martín’: con-
trol territorial y estrategias estatales contra el narcotráfico y subversión (1980–
1995).” Politai: Revista de Ciencia Política, 6 (11): 33–51.
Manrique, Hernán. 2016. El fin de la Guerra de la cocaína: construcción del Estado
y desarrollo alternativo en la región San Martín. Tesis de Licenciatura PUCP.
Manrique, Nelson. 1995. Historia de la Republica. Lima: COFIDE.
Marcy, William. 2010. The Politics of Cocaine. How U.S. Foreign Policy Has Created
a Thriving Drug Industry in Central and South America. Chicago: Lawrence Hill
Books.
Meyer, Enrique. 2009. Ugly Stories of the Peruvian Agrarian Reform. Durham: Duke
University Press.
Monterroso, Iliana. 2017. Reivindicación de derechos colectivos. Reformas de tenen-
cia de tierras y bosques en el Perú (1960–2016). Documento de trabajo 227.
Bogor: CIFOR.
Morales, Edmundo. 1989. Cocaine: White Gold Rush in Peru. Tucson: University of
Arizona Press.
Moscoso, Teodoro. 1962. The Alliance for Progress: Its Program and Goals. Wash-
ington, DC: USAID.
Nelson, Michael. 1973. The Development of Tropical Lands: Policy Issues in Latin
America. Baltimore: John Hopkins University Press.
OEA. 1987. Estudio de Casos de Manejo Ambiental: Desarrollo Integrado de un
área de los Trópicos Húmedos – Selva Central del Perú. Washington, DC: OEA.
Ortega, Emiliano. 1988. “La agricultura en la óptica de la CEPAL.” Revista de la
CEPAL, 35: 13–37.
Paredes, Maritza. 2011. “Indigenous Politics and the Legacy of the Left.” In Crab-
tree, J. (ed.), Fractured Politics: Peruvian Democracy, Past and Present. London:
Institute for the Study of the Americas, 129–157.
Parodi, Carlos. 2014. “El gasto público y la inversión para el desarrollo social, 1930–
1980.” In Zegarra, Luis et al. (eds.), Compendio de Historia Económica del Perú.
La economía peruana entre la gran depresión y el reformismo militar (Tomo V).
Lima: IEP y BCR, 415–464.
Páucar, Felipe. 2006. La Guerra Oculta en el Huallaga, Monzón y Aguaytía. La
política y el poder corruptor del narcotráfico: la historia de 30 años de violencia
y desarrollo en Huánuco, San Martín y Ucayali. Tingo María: CEPDA.
Prébisch, Raúl. 1963. Hacia una dinámica del desarrollo latinoamericano. México,
D.F.: Fondo de Cultura Económica.
Ramírez, María Clemencia. 2011. Between the Guerrillas and the State: The Cocalero
Movement, Citizenship, and Identity in the Colombian Amazon. Durham, NC:
Duke University Press.
Rist, Gilbert. 2002. The History of Development: From Western Origins to Global
Faith. London & New York: Zed Books.
Rodríguez Achung, F. 1995. El recurso del suelo en la amazonía peruana, diagnóstico
para su investigación (Segunda Aproximación). Iquitos: IIAP.
Rodriguez , Martha. 1991. “Proceso de ocupación y construcción social del espacio
amazónico.” En: Amazonía 1940-1990. El Estravío de una Ilusión. Barclay, F;
Rodriguez, M; Santos, F y M. Valcarcel (eds). Terra Nuova/CISEPA.
 The Upper Huallaga Valley, Peru  83
Rumrrill, Roger. 1992. “The Highs and Lows of a Cocaine Economy.” In Smith, M.
et al. (eds.), Why People Grow Drugs. Narcotics and Development in the Third
World. London: PANOS, 79–90.
Taffet, Jeffrey. 2007. Foreign Aid as Foreign Policy. The Alliance for Progress in Latin
America. New York and London: Routledge.
Thorp, Rosemary and Maritza Paredes. 2010. Ethnicity and the Persistence of Inequal-
ity: The Case of Peru Conflict, Inequality and Ethnicity. Palgrave Macmillan.
Tussie, Diana. 1995. The Multilateral Development Banks. Vol. 4: The Inter-Ameri-
can Development Bank. Boulder and London: Lynne Rienner Publishers.
Ugarte, Daniela. 2014. Después de la Coca: La persistencia del cambio institucional
en Tocache. Unpublished thesis. Pontificia Universidad Católica del Perú.
van der Laar, Aart. 1980. The World Bank and the Poor. Boston: M. Nijhoff.
Wainwright, Tom. 2016. Narconomics: How to Run a Drug Cartel. New York: Pub-
lic Affairs.
World Bank. 1954. Loan Agreement (Second Agricultural Machinery Project)
between Republic of Peru and International Bank for reconstruction and Devel-
opment, 1954 (Loan Number 98 PE).
World Bank.1957. Loan Agreement (Second Agricultural Credit Project) between
International Bank for reconstruction and Development and Banco de Fomento
Agropecuario del Peru, 1957 (Loan Number 162 PE).
World Bank. 1960. Guarantee Agreement (Third Agricultural Credit Project).
Republic of Peru and International Bank for reconstruction and Development,
1960 (Loan Number 257 PE).
World Bank. 1965. Loan Agreement (Fourth Agricultural Credit Project) between
International Bank for reconstruction and Development and Banco de Fomento
Agropecuario del Peru, 1965 (Loan Number 415 PE).
World Bank. 1973. Project performance audit report. Peru Fifth Agricultural Credit
Project (Loan 933 PE).
4 Creating coca frontiers and
cocaleros in Chapare
Bolivia, 1940 to 1990
Andrew C. Millington1

Introduction
This chapter re-analyzes the establishment and growth of the most noto-
rious coca frontier in Bolivia—Chapare—between 1940 and 1990. It syn-
thesizes previously published material with new information from archives
and libraries in Bolivia, Germany, and the U.S. that have not been analyzed
previously. It also introduces a new source of information that has not been
used before in studying the historical geography of colonization frontiers in
South America before: declassified intelligence community space photogra-
phy from the U.S. government’s Cold War-era Corona missions.
The analysis reveals the following historical sequence of colonization.
First, an embryonic frontier emerged in the late nineteenth and early twen-
tieth centuries. This emergent frontier had two geographical foci around
the present-day town of Villa Tunari: the first, in Alto Chipiriri, and the
second in the lowlands of the Espíritu Santo and San Mateo rivers. Second,
an establishment phase from the late 1930s to the late 1950s during which
poles of lowland agrarian colonization were established—mainly by internal
migration, though there were efforts to attract migrants from overseas. The
former can be linked to high levels of discontent with working conditions
among tied labor on haciendas and the mining sector. The 1952 Agrarian
Revolution and left-wing agitation in the mines were key agents in the estab-
lishment phase. Simultaneously, Cochabamba came under the influence of
the Cuban and U.S. East Coast mafia, and is identified as the main source
of cocaine supplied to New York in the 1950s. Finally, there was an element
of persistence before and after the 1970s “cocaine boom,” in which coca
cultivation expanded throughout Chapare as an essential agent of growth in
agricultural settlements.

Chapare as a borderland
Chapare is in the cultural, political, and ecological borderland that occupies
the actual and perceived spaces between highland and lowland Bolivia. His-
torically, these borderlands have been the boundary between highland and
 Coca frontiers and cocaleros in Chapare  85
lowland peoples. They formed the de facto border between the Viceroyalty
of Alto Peru in the highlands and the tropical lowland landscapes that were
incorporated into the Viceroyalty of Río de la Plata after 1776. In 1825,
newly independent Bolivia inherited these borderlands, which separated the
commercial and political powerbases in the highland cities of Cochabamba,
La Paz, Oruro, and Sucre and the extensive, inadequately connected, poorly
known, but commercially intriguing peripheral lowlands. Borderlands such
as Chapare, and the perceptions that envelop them, still provide an imprint
for modern Bolivia. However, it is easy to overstate their place in the con-
temporary state, as they are better integrated in the Plurinational State of
Bolivia in the early twenty-first century than at any time in their history.
In the context of this essay, their importance is that they always have been
pivot points in lowland–highland circuits of exchange, the latest manifesta-
tion of this as part of international cocaine supply chains.
A stark, modern-day definition of Chapare is that it is an approximately 200
km long, northwest–southeast trending colonization zone in Cochabamba
Department (Bradley and Millington 2008). Somewhat confusingly, the name
is also given to a major river and one of 16 provinces in Cochabamba Depart-
ment (only part of which comprises part of the colonization zone itself!). It is
contiguous with the Yapacani colonization zone in Santa Cruz Department
to the southeast; and approximately 170 km of roadless lowland tropical for-
est separates it from the southernmost part of the Alto Beni colonization zone
in La Paz Department. All three are in Bolivia’s borderlands. Fifer (1967),
writing about Bolivia’s pioneer fringe, identified another borderland and col-
onization zone that extended from Santa Cruz to the Argentine border. Most
settlers moved to that borderland in the 1920s and 1930s, compared to the
1950s onwards in Alto Beni, Chapare, and Yapacani.
What makes Chapare different from Alto Beni and Yapacani is that it
rapidly evolved into a major coca growing and basic cocaine paste produc-
tion area, i.e., a coca and a cocaine frontier. It remains so today. The estab-
lishment and early development of this frontier provides the impetus for
the research and scholarship that underpins this essay, which is structured
around:

• an analysis of coincident commercial and political interests;

• the development of administrative and political structures;
• the establishment of coca as an export crop;
• the evolution of the frontiers of colonization in the 1950s and 1960s; and
• an analysis of a time series of coca production estimates.

Coinciding commercial and political interests

As far as Chapare is concerned, the commercial interests of the Cocha-
bamba élite have more or less evolved harmoniously with departmental
and national political agendas. The region has been pivotal to almost two
86  Andrew C. Millington
centuries of attempts to grow highland–lowland circuits of exchange from
Cochabamba to capture commercial trade with the northern Bolivian or
Mojos lowlands, which became the Department of Beni in 1842. In the
nineteenth century, Cochabamba’s mercantile class wanted access to rub-
ber exports from Beni, while simultaneously investing in a two-way com-
mercial trade with the “rubber boom economy” in Beni. As late as the
1960s, Fifer (1967) recorded a trade imbalance, with Cochabamba export-
ing 2,760 tons (2,503 metric tons [MT]) of goods to the lowlands, while
importing only 840 tons (762 MT). Maybe this suggests that the long-term
ambition of Cochabamba merchants was to expand their markets, rather
than being a link on a rubber supply chain. But political interests were also
in play.
Commercial and political interests backed a succession of expeditions
aimed at finding feasible routes from the Cochabamba highlands to Beni
in the nineteenth and early twentieth centuries. These had to contend
with two formidable obstacles: an approximately 100 km wide belt
of highly dissected, mountainous terrain, and seasonally flooded low-
lands forests and swamps. Military-led, government-sponsored exped-
itions, e.g., those led by General Federico Ramón between 1919 and
1922 (Pinto Parada 2001) echo to the politics of nation building and
securing Bolivia’s peripheral regions and national borders. Later exped-
itions searched for more efficient routes and reflected commercial rival-
ries more than overarching geopolitical motives (van der Berg 2008).
Regardless of their motivations, the routes and the settlements they
spawned firmly signaled Chapare as a frontier of settlement that enjoyed
a wider sense of purpose than either Alto Beni or Yapacani.2 The devel­
opment of highland–­lowland communications through Chapare opened
it up to commerce while simultaneously throwing a cloak of strategic
geopolitical significance over it.
Another theme that had a significant bearing on the settlement frontier in
Chapare was the “competition to connect” the cities of Cochabamba and
Santa Cruz. Rival commercial and political interests from Cochabamba, La
Paz, Oruro, and Santa Cruz vied with each other to make this connection
by road or rail. The impetus to build a railway from Cochabamba to Santa
Cruz increased after the line from Oruro to Cochabamba was opened in
1917. All three routes proposed were technically complex and, therefore,
expensive to construct. However, the main stumbling blocks were the com-
peting political interests backing each route (Rodríguez Ostria 2012). Work
started a decade after the Oruro–Cochabamba line opened on a proposed
147 km stretch of land (Zubieta 2012); but it only reached Aiquile—60 km
from Cochabamba—before it was abandoned at the start of the Chaco War.
In the first half of the twentieth century, road connections between the two
cities were limited to a series of barely passable tracks, but an improved
route through the Cochabamba highlands and Samaipata reached Santa
Cruz in the 1950s.
 Coca frontiers and cocaleros in Chapare  87
Competition between road and rail interests intensified after World War II
(see for example, articles in the Cochabamba newspaper El Imparcial
during the 1940s). However, the backers of a lowland road route through
Chapare were in the ascendency by this time, and funding to extend the rail-
way from Aiquile to Santa Cruz never materialized. A road was constructed
in eastern Chapare between existing road heads at Villa Tunari and Puerto
Grether during the late 1950s and 1960s. At 160 km from Cochabamba,
Villa Tunari (Figure 4.1) is a key settlement in the history of Chapare and
remains important to this day. It developed around Mission San Antonio
at the confluence of the Espíritu Santo and San Mateo rivers. It was also
located on the old road from Cochabamba to Todos Santos that was part
of the Cochabamba-Beni network of trade routes at a point where traffic
had to cross the Río Espíritu Santo. Puerto Grether is a moribund port on
the Río Ichilo, about 140 km from Santa Cruz at the western end of the
Yapacani colonization zone, now bypassed by the main road. Evidence of
the construction of this highway can be seen clearly on the 1966 Corona
photography (Figure 4.2). It is referred to as the Santa Cruz Highway in this
chapter. Its official designation is Ruta Nacional 4, and could be thought of
as southernmost manifestation of the Carretera Marginal de la Selva dis-
cussed in earlier chapters.

Areas of settlement and forest clearance in the Chapare lowlands,

Figure 4.1 
­September 1966
Note: All roads and major tracks are mapped. Only selected settlements and the major rivers
are labelled. The map is based on interpretation of Corona KH4 imagery, and overlain on a
shaded relief map to differentiate the lowlands from the most northerly ranges of the Andes
in this area.
88  Andrew C. Millington

Figure 4.2  Sample Corona photographs of Bolivian Amazon, 1966

Colonization and the development of administrative

and political structures
Chapare witnessed many attempts at colonization during the twentieth
century, and any analysis of the establishment and development of a coca
frontier (and its cocaleros) needs to be grounded in an analysis of policies
that influenced colonization and its failures and successes. The Chapare set-
tlement model is based on poles of colonization that have grown to become
2,000–5,000 person agricultural, commercial, and administrative centers.
An ever-expanding network of roads and tracks linked surrounding rural
communities to the colonization poles. Settlements have become established
 Coca frontiers and cocaleros in Chapare  89
and agricultural land cleared as this feeder road network has grown over
five decades, (see Figure 4.2) (Millington et al. 2004; van Gils and Loza
Armand Ugon 2006).
Very few of these settlements existed before roads were constructed in the
twentieth century. They are the material culture of a recently colonized land-
scape and, as this chapter argues, the product of interlinked agents. Perhaps
the most interesting of these, in the context of an essay on coca frontiers,
is Shinaota. This town is reputed to have been Chapare’s most important
coca and cocaine trading center in the 1970s. It evolved from Colonia Pres-
idente Busch, an agrarian settlement established in 1938 on the banks of the
Río Coni, which, in 1940, had a population of 130 (El Imparcial 1940). A
company had been formed to settle Polish immigrants there by this time,
which explains its alternative name of Puerto Polonia. The outbreak of
World War II put brakes on those plans, nonetheless it grew through inter-
nal migration.
Shinaota exemplified the kinds of obstacles posed to settlers in the bor-
derlands by the administrative divisions that independent Bolivia inherited
from the colonial state. Shinaota is in Tiraque Province: a province named
after its administrative capital that was founded in 1566 in the Cochabamba
highlands. When the administrative division was created, the center of grav-
ity of population was in the highlands, as were the economic, political, and
administrative foci. The Chapare lowlands, which were also part of the
province, were “unpopulated” and it was inconceivable to people at that
time that they ever would become populated. However, as Chapare was
settled during the twentieth century, towns were founded, but the political
and administrative structures were very slow to adapt to this demographic
shift. By the time Shinaota had grown into a relatively sizeable town in the
1970s, administrative and political structures had still not caught up with
this new demographic reality. People with administrative business in the
province had to make a journey of at least a day to reach Tiraque. Even now,
this journey takes the best part of a day, although Shinaota has evolved into
a separate municipality that administers the lowlands of Tiraque Province
(Table 4.1).
The history of the administrative geography of Shinaota is far from unique.
All towns and villages in the lowlands of the three provinces of the Chapare
lowlands (Table 4.1) had major issues connecting to highlands towns in the
same provinces, as this vivid description of Pojo Municipality illustrates:

Pojo Valles esta separado de Pojo Trópico (ahora constituido en el

Municipo Entre Ríos) … No existe conexión vial directa entre Pojo
Valles y Entre Ríos (Pojo Trópico). Pojo Trópico es accessible desde la
ciudad de Cochabamba a traves de la carretera … pasando por Sacaba,
Colomi, Villa Tunari, Shinaota, Chimoré y Entre Ríos.. [The Pojo valleys
[highlands] are separated from the Pojo Trópico [lowlands] (now consti-
tuted as Entre Rios Municipality) … There is no direct road c­ onnection
90  Andrew C. Millington
Table 4.1  Cochabamba Department: provinces with municipalities in the Chapare
lowlands in the 2012 census

Province1 Municipalities2 Municipalities3 predominantly

predominantly in the in the Chapare lowlands
Cochabamba yungas,
cordillera or high valleys

Chapare Sacaba Villa Tunari

Colomi
Tiraque Tiraque Shinaota
Carrasco Totora Chimoré
Pocona Entre Ríos
Pojo Puerto Villarroel

Notes:
1
 ach of these provinces extends from highland areas in the south to the lowlands in the north.
E
There are 16 provinces in Cochabamba Department, these three comprise 52.5 percent of the
department’s land area but only 25.5 percent of the population (2012 census).
2
Each municipality is named after its main town and administrative center.
3
Other settlements classed as “urban” in the 2012 census (INE 2012) are Eterasama and Villa
14 de Septiembre in Villa Tunari Municipality, Bulo Bulo in Entre Ríos, and Valle Sajta in
Puerto Villarroel.
Sources: INE (2016), Camacho (2005), Mamani (2004)

between the Pojo valleys and Entre Rios (Pojo Trópico). Pojo Trópico
is accessible from the city of Cochabamba by road … passing through
Sacaba, Colomi, Villa Tunari, Shinaota, Chimoré and Entre Ríos..]
(Peredo 2004, 11)

The journey described by Peredo still takes almost a whole day and would
have been longer in the 1960s, even after the Santa Cruz Highway had been
constructed. These administrative anachronisms are only slowly breaking
down, and provincial sub-regions (or municipalities) in the Chapare low-
lands have only been established in the last two decades.
Importantly, in the context of the coca frontier, an alternative hierarchical
administration developed based on colonos’ social organization. This com-
prised sindicatos, centrales and federaciones. Sindicatos are dispersed rural
settlements, which are recognized in the population census (Table 4.2). This
hierarchical structure exercised strong local political and administrative
control over commercial activities and political affairs in Chapare, though
the structure was overtly agrarian.
Arguably, the gulf that opened between weak state administrative control
and strong local political control was a critical factor that enabled coca and
cocaine to gain a foothold in Chapare during the early colonization phase,
when international cocaine trafficking chains were being established in the
1950s and 1960s. Government oversight and influence in Chapare was slight,
while simultaneously the strong political organization around sindicatos,
 Coca frontiers and cocaleros in Chapare  91
Table 4.2  Socio-political organization and settlement geography for Villa Tunari
Municipality in the 2001 census

District Social Number of Number of Number of Total

organisations centres of sindicatos2 centrales3 population
(Federaciónes) population1

Villa Tunari4 1 0 0 2,632

Chipiriri Federación 8 32 4 5,282
Villa 14 de Especial de 3 40 2 8,153
Septiembre Trabajadores
San Francisco Campesinos 7 39 4 3,562
Eterazama del Trópico de 5 23 3 5,415
Samuzabety Cochabamba 5 39 5 4,783
Isinuta 4 27 2 4,966
Nueva Aroma 12 46 7 7,508
Paractito Federación 3 25 3 4,093
Cristal Mayu Especial 2 45 3 3,742
Paracti Yungas del 2 37 5 3,693
Chapare
Notes:
1
Population centers are nucleated settlement, in effect small towns.
2
The average number of people in each sindicato ranges from 83 in Cristal Mayu to 235 in
Eterazama, the average for the municipality is 151.
3
The average number of people in each central ranges from 739 in Paracti to 4077 in Villa 14
de Septiembre, the average for the municipality is 1595.
4
Villa Tunari is the only urban center in this municipality and as such was a district without the
administrative structures of rural areas.
Source: INE (2001)

centrales and federaciones was becoming entrenched. This alternative

administration was influenced by political lessons that many internal
migrants to Chapare had learned in the highland mines (Sanabria 1993)
and by participating in the post-1952 agrarian revolution. Both antecedents
were socialist in outlook and in opposition to the ancien régime and succes-
sor La Paz governments. Given that context, why did the government allow
Chapare to develop relatively unhindered? This is a particularly intriguing
question given contemporary political influences in Chile. Though concerns
over Communist influences in Cochabamba date from earlier times, it was
a matter of grave concern (to the press anyway) during and immediately
after World War II. El Imparcial reported talks by left-wing mine leaders
from Chile and meetings of the Falange Socialista Boliviana and the Unión
Socialista Republicana between 1940 and 1947.
Healy (1988) has argued that the weak state apparatus was more than a
local phenomenon afflicting Chapare, likening it to a national disease caused
by an extended legacy of unstable government. The weak state allowed drug
trafficking to prosper because the state was at a disadvantage when it came to
dealing with cocaleros, who are members of the peasantry, and were part of
92  Andrew C. Millington
a mobilized force that was “…well-organized and combative,” (Healy 1988,
107) with high levels of popular participation and organization through
rural unions since 1952 (Durand Ochoa 2014; Salazar Ortuño 2009).
The fact that sindicatos and federaciones, with their strong political over-
lay, filled the vacuum left by the state’s inability to extend its administration
fully in Chapare may have encouraged the region in developing a strong
identity. As this alternative structure had at least some of its roots in the soil,
it provided the ideal seedbed for activities around coca and then allowed
Chapare to become the geographical source of narco-trafficking. The for-
mation of coca-grower federations to succeed the agricultural federations
after the Ley del Régimen de la Coca y Sustancias Controladas was passed
in 1988 adds weight this argument.
The “administrative independence” of Chapare may have been fiscally
prudent for the government, given its low population, but other reasons
need to be explored to substantiate this. These could include concerns over
secession; something that may have been heightened by improved connec-
tions to Santa Cruz along the new highway and the coca-fueled economic
boom Chapare experienced during the 1970s. Another line of inquiry
might be to what extent the state wanted to retain strong central control of
planned settlement and colonization across the nation. This many have been
important after the global tin crisis in the mid-1980s, when new waves of
miners migrated to Chapare. Though the timing is not synchronous with the
argument above, miners from Catavi and Llallagua obtained departmental
assistance to establish Colonia La Libertad in 1949 (El Imparcial 1949).
Concerns over the growth of these self-administering pockets of left-leaning
activists in the region may be pertinent as well.
A further argument could be that while Chapare was perceived as periph-
eral by central government in the early 1950s, this mindset had not changed
much by the 1960s or 1970s. Analyses of the failed attempts to control the
coca and cocaine trade in the 1970s are also pertinent. Bolivia did relatively
little in terms of sustained efforts to eradicate coca in the 1970s. Policies
existed but were not pursued, and so far no archival evidence had been
found to argue that U.S. government agencies vigorously pursued the Boliv-
ian government on these matters. This may be because the military regime
in Bolivia was a key partner in the pan-continental fight against Commu-
nism. This argument could be extended into the 1980s, when the high Cold
War military regimes of the previous decades were replaced by right-wing,
neoliberal regimes compliant with World Bank policies and the Reagan
administration. Anti-narcotic policies were brought to the table, but tangible
actions on the ground were only spasmodically effective.

Establishing coca as an export crop

Lee and Clawson (1993) note that Alto Beni, Chapare, and Yapacani were
all coca and cocaine producing regions by the late 1980s and early 1990s.
However, their observation needs to be considered carefully. Archival
 Coca frontiers and cocaleros in Chapare  93
records show that Yapacani was a much less important coca and cocaine
region than Chapare, though as they are contiguous there is the possibil-
ity their production may, at times, have been combined. There is also the
strong possibility that the well-established coca plantations of the La Paz
yungas were included as part of Alto Beni. The most pertinent question
is: Why did coca become well established as a major crop in Chapare,
but was much less important in the Alto Beni and Yapacani colonization
zones?
Coca has been grown for centuries in the sub-tropical mountain valleys
or yungas in La Paz and Cochabamba Departments. The plantations in the
Cochabamba yungas were part of haciendas based in Cochabamba high-
lands (Meruvia 2000). Coca is still grown in the communities that evolved
from these haciendas, e.g., Arepucho and Icuna. Their geography—on the
east slope of the Andes adjacent to and immediately south of Chapare—is
important in understanding the commercialization of coca in Chapare. The
geographical location and ecological situation of the long-established La
Paz coca plantations in relation to the Alto Beni are similar of those of the
Cochabamba yungas and Chapare, yet lowland Alto Beni has not become a
coca frontier.
Loza-Balsa (1992) describes Yapacani as “…un caso típico de culitvo
forzado de coca” (Loza-Balsa 1992, 11); he also notes that although the
planting distances between bushes were high due to less than favorable con-
ditions, settlers have converted most of the land to coca. This description is
intriguing, as no statistical records of coca plantation areas or production
exist, and discussions of a Yapacani coca frontier are virtually absent in
the literature. The Yapacani area receives much less rainfall that most of
Chapare and probably not enough to grow the coca varieties that flourish in
Chapare and the yungas. Field surveys conducted by the author since 2001
and the UNODC Bolivia Coca Monitoring Surveys indicate that southeast
Chapare, which is contiguous with Yapacani, has very low coca cultivation
density compared to other parts of Chapare. As Loza-Balsa does not cite
evidence for the statements he made, a question mark remains possible about
coca cultivation in Yapacani.
But it is the geography of the yungas plantations and the Chapare low-
lands in the context of transportation networks that is the critical factor
in explaining the emergence of a fully-fledged, sustainable coca frontier in
Chapare but not in Alto Beni and Yapacani. Two developments are key.
First, the evolution of Cochabamba–Beni circuits of exchange, which
included the search for routes from the city of Cochabamba through the
yungas coca plantations to Chapare (Pinto Parada 2001; van der Berg 2008;
Rodríguez Ostria 1972). These directly linked yungas cocaleros to Chapare’s
new settlers. Second, the establishment of the Santa Cruz Highway, which
became important in moving coca leaf and coca paste to Santa Cruz and
Cochabamba.
Compared to Alto Beni and Yapacani, Chapare is a communications hub.
La Paz is still only linked to Alto Beni by a 310-long km tortuous road
94  Andrew C. Millington
through the yungas to Yucumo at the southern end of Alto Beni. Though
this road passes through the La Paz yungas plantations. Historically, coca
leaf has been transported to La Paz city, and the orientation of hacendados
was toward the capital. This meant that there was less impetus to develop
roads from La Paz to lowland Alto Beni than there was from Cochabamba
to Chapare. The geography of the Yapacani colonization zone is radically
different. It is an extension of settlement that was centered on Buena Vista
and San Carlos, towns that evolved from eighteenth century missions. These
towns were, and still are, in the economic, political, and religious orbits of
Santa Cruz, so much so that they had no other overland connections until
the Santa Cruz Highway was opened. No roads traversed the Yapacani yun-
gas to the highlands in the early and middle twentieth century—a situation
that still prevails.
Historically, the Cochabamba yungas comprised at least three dis-
tinct areas of coca plantations between approximately 1,000 and 1,500
meters above sea level: the former Canton of Vandiola in modern-day
Tiraque Province, and the Icuna and Arepucho Cantons in the Totora
Municipality (Carrasco Province). Arepucho and Icuna have persisted
as coca-growing areas from at least the nineteenth century. They were
recognized as such in Article 9 of Ley 1008, Ley del Régimen de la
Coca y Sustancias Controladas (Government of Bolivia 1998), which
loosely defined traditional coca-growing regions. The fragmentary evi-
dence available suggests the haciendas in the Vandiola yungas had been
abandoned by the late twentieth century. Each of these areas has con-
nections to a highland town with Spanish colonial origins. Vandiola
was connected to Tiraque, and Arepucho and Icuna with Totora. In the
late nineteenth and early twentieth centuries, routes between Arepucho
and Icuna, and Totora, were explored in connection with commercial
trade between Cochabamba and Beni. One expedition reported extensive
coca growing in the yungas. However, no feasible connection was found
between the yungas and the lowlands.
The commercial motivations to develop coca as a commercial crop were
articulated through two important departmental organizations. The Comite
Pro-Cochabamba and the Junta de Propetarios de Yungas (JPY). JPY is the
Cochabamba equivalent of SPY (Sociedad de Propietarios de Yungas), the
better-known organization of coca estate owners in the La Paz yungas (Cusi
and Flores 2007). At the present time, most of the information about JPY
has been obtained from El Imparcial.3 It was founded at the turn of the
twentieth century (Rodríguez Ostria 1972), was active in the 1940s, and
elected members to the national Comité de la Aduana de la Coca as late as
1949. A fundamental difference existed between the JPY and the SPY. The
latter’s attention was mainly oriented toward La Paz, e.g., it was influen-
tial in state formation in the early years of the Republic (Cusi and Flores
2007). However, the Junta de Propetarios de Yungas focused on the Depart-
ment of Cochabamba and its interests overlapped with those of Comité
 Coca frontiers and cocaleros in Chapare  95
­ ro-­Cochabamba in attempting to connect the department’s economy to
P
those of Beni and Santa Cruz Departments.
One of the earliest reports of coca being grown in Chapare comes from
the early twentieth century. Blanes and Flores (1982), in a review of early
sources, recorded that 200 hectare land parcels had been opened along “…
the road that had been constructed from Cochabamba through Mission
San Antonio (now Villa Tunari) to Todos Santos.” Five hundred families
had settled there and were cultivating well-known tropical crops and small
amounts of “secondary crops”—listed as coca, and sugar cane for alcohol
and black sugar. Though this is very important because it is one of the ear-
liest records of coca cultivation in Chapare, it is shrouded in the geographical
imprecision that characterizes almost all descriptions of pre-1950s coca cul-
tivation in Chapare. In this case, the imprecision is related to the reference
that farms had been opened along the old Cochabamba–Todos Santos road.
This coca could have been cultivated 20 km out from the mountain front in
the lowlands adjacent to Todos Santos (the area known as Alto Chipiriri), or
in the lowlands along the Espíritu Santo and San Mateo lowlands south of
Mission San Antonio (Villa Tunari), where settlement had also been encour-
aged. Figure 4.1 shows that approximately 3,000 hectares of the San Mateo
and Espíritu Santo lowlands were cultivated in 1966, while in Alto Chipiriri
the cultivated area was approximately 24,500 hectares. This means that the
10,000 hectares of cultivation implied by Blanes and Flores’ (1982) must
have included parts of Alto Chipiriri and is prima facie evidence of coca
being cultivated in the Chapare lowlands in the early twentieth century. It
is interesting to compare this with Francisco de Viedma’s earlier descrip-
tion of the Chapare lowlands: “…la coca no se desarolla, rinde rapidamente
pero luego se muere, no se tienen sitios donde guardarla…” (Viedma 1836,
no. 361: 32).

Frontiers and colonization in the 1950s and 1960s

The 1950s and 1960s are relatively obscure in many accounts of coloni-
zation, socio-economic development and the development of Chapare as a
coca and cocaine frontier. While the 1970s “coca boom” stimulated a paral-
lel writing boom, much of what was written about the period before 1970 is
descriptive and recycles earlier authorities (e.g., Sanabria 1993; Soux 1993).
A conventional argument is that previously unsuccessful attempts at colo-
nization in Chapare were overcome in the 1960s because new and better
methods enabled farmers to market their produce and local healthcare pro-
vision, including malaria eradication, improved.
Blanes and Flores (1982) argue that spontaneous colonization that had
benefited from a UN assistance program was occurring along the Santa Cruz
Highway by 1953. There is field evidence for this timing, e.g. the Comunidad
El Carmen primary school (Figure 4.1) was founded in 1953, according to
school signs. Spontaneous colonization was replaced by formal approaches
96  Andrew C. Millington
Table 4.3  Settlements in the Chapare colonization zone, 1980

Type of Number Families % Area % Mean area per

settlement (ha) family (ha)
(asentamientos)

Spontaneous 17 44,434 69.2 926,647 58.7 20.8

Directed 10 17,890 27.9 545,087 34.5 30.5
Established with 3 1,592 2.5 106,494 6.7 66.9
convenio
Total 30 63,916 100 157,8228 100 24.7

Source: Blanes and Flores (1982), Table 4.

to settlement according to the Plan Nacional de Desarollo 1962–1971 and

the Plan de Colonización in 1963. The movement of 8,000 families to Alto
Beni, Chapare, and Yapacani in 1963 generated so much concern that the
government established the Instituto Nacional de Colonización (INC). Col-
onization was particularly vigorous between 1965 and 1971 and, after the
INC was formed, a mix of spontaneous and planned settlement became the
standard (Table 4.3).
A relatively under-used data source in studies of colonization fron-
tiers is introduced in this research: space photography. A number of U.S.
space-based reconnaissance missions between 1960 and 1972—known
collectively as the Corona Program—acquired high-spatial resolution pho­
tographs for intelligence and mapping aims. These were declassified by the
Clinton administration in 1995, and have been used to extend time series
of land-use and land resource analyses a decade before 1972, when the first
Landsat 1 imagery became available (Tappen et al. 2000; Lautenbach et al.
2010). They have been used in archaeology (Philip et al. 2002), but not in
geographical analyses of settlement in this historical period to the author’s
knowledge. Details about the Corona Program can be found in McDonald
(1995, 1997) and Ruffner (1995).
Panchromatic photographs shot on September 23, 1966 from the
Corona KH4A mission are used in this study. An advantage of the Corona
photography is its high spatial resolution, approximately 3 meters. Scanned
photographs are available from the USGS Long Term Archive (https://lta.
cr.usgs.gov/declass_1) at US$30 per frame (as of September 2017). The
scanned photographs used in this research were geo-rectified to ground
points with known stable latitude and longitude coordinates obtained
from Google Earth© Pro in Erdas Imagine 2015, e.g., the crossroads in
the upper photography in Figure 4.2. The geo-rectified photographs were
exported to ArcGIS v.10.4 and the cleared areas, roads, tracks, and river
digitized on-screen between Río Eterasama and Río Ivirgarzama (cf. exam­
ples in Figure 4.2).
 Coca frontiers and cocaleros in Chapare  97
Geo-rectified Corona photography provides clear details of the settled
areas to the extent that individual fields can be distinguished from each
other. However, geo-rectification in the lowland forest area that had not
been settled in 1966 is highly problematic because there are no stable
ground control points. It is possible to find stable ground control points,
e.g., river bends and confluences, in the mountainous forested area. Clear
tonal definition between forest, cleared fields, and roads has enabled the
exact boundaries of land clearance around the early settlements to be
mapped.
Figure 4.1 is, therefore, the earliest geometrically accurate map of settle-
ment in Chapare and consequently is an important marker in the chronol-
ogy of establishing the coca frontier. Three distinct areas had been settled by
1966, three years after the establishment of the INC:

1 An area of high cultivation density in the Espíritu Santo and San Mateo
lowlands southwest of Villa Tunari (Figure 4.1), where coca had been
successfully grown as a commercial crop in the early twentieth century.
This zone traces part of the old route used to transport produce, includ-
ing coca, from the Vandiola yungas to the Cochabamba–Todos Santos
road and onwards to the highlands.
2 A number of settlements in the early stages of establishment in the low-
lands between Villa Tunari and Río Chimoré. Some were strung out
along the putative Santa Cruz Highway; others were on secondary
roads to the north and south. Active colonization is evident even to
the extent that roads newly cut into primary forest are clear, though no
settlers had arrived and no fields had been cleared (Figure 4.2).
3 Alto Chipiriri, northwest of Villa Tunari, was served by a network of
feeder roads to the west of the Villa Tunari–Todos Santos road (Figure
4.1). The southern part of Alto Chipiriri was served by the track from
Villa Tunari to the Mojos through Moleto that was explored by General
Ramón in 1919 (Pinto Parada 2001; South 1977). The high cultivation
density in this area, and in the Espíritu Santo-San Mateo lowlands, in
September 1966 indicates that much colonization in these two areas
had taken place before the INC was established.

The approach that has been taken in Chapare is to map forest clearance
for agriculture. In tandem with other researchers, the author has compiled
a time sequence of space photography and satellite imagery from 1966 to
2016, supported by field surveys and interviews with farmers. This has
enabled accurate maps of the development of colonization and agricul­
tural (coca) frontiers to be constructed (Bradley and Millington 2008; Nag-
ombi 2017). Most of the analyses reported by these authors focuses on the
period after 1990. A key finding is the stop–start nature of forest-to-agri-
culture conversion, the rates of which are dependent on the effectiveness of
anti-narcotics policies. A quantitative assessment of deforestation from this
98  Andrew C. Millington
Table 4.4  Chapare: remaining forest cover by sector between 1966 and 1988

Sectors Distance from % forest % forest % forest % forest

Villa Tunari1 remaining, remaining, remaining, remaining
to mid-point of 1966 1975 1988 2014
sector (km)

Along the Ruta Nacional 4 (Santa Cruz Highway): east of Villa Tunari
Shinaota- 18 97.8 92.1 86.1 71.2
Chimoré
Mariposas 49 99.8 94.3 76.1 46.6
Ivirgarzama 62 99.9 86.4 54.9 20.0
Ichilo 96 99.2 98.2 92.3 51.2
Along the Villa Tunari-Moleto road: west of Villa Tunari
Chipiriri  9 77.4 48.5 58.6 66.7
Eterasama 18 99.3 87.4 78.8 36.2
Isinota 34 n.d.2 98.8 74.4 57.7
TIPNIS3 69 99.3 n.d.2 97.0 76.4

Note: Forest cover data for 2014 is included for reference.

1
 illa Tunari is chosen as the reference point for distance measurements and it is the point at
V
which the road from Cochabamba exits the mountains onto the Chapare lowlands.
2
n.d. – no data
3
TIPNIS - Territorio Indígena y Parque Nacional Isiboro-Sécure

sequence of maps between 1966 and 1988 is presented in Table 4.4. This
shows clearly how the agricultural (coca) frontier in Chapare changed after
1966. In particular:

1 the amount of agricultural land in all but one of the geographical regions
(sectors) progressively increased. The exception was Alto Chipiriri—one
of the two very early areas of colonization—where significant amounts
of forest had been cleared by the late 1970s;
2 the amount of forest cleared by 1988 generally decreases with the dis-
tance from Villa Tunari; and
3 clearance was greater in eastern Chapare, along the new Santa Cruz
Highway, than in western Chapare. There is an exception to this gen-
eral pattern. The Ivirgarzama sector was deforested earlier than other
sectors in eastern Chapare because of the construction of the road to
Puerto Villarroel, a key port on the Río Ichilo. This also encouraged
settlement south of the highway in Valle Hermoso.

Coca production estimates

It has already been noted that the volume of research into coca and cocaine
in Bolivia increased markedly during the 1970s. Four themes character-
ize this research. First, much attention was paid to obtaining estimates of
the areas of coca cultivated and the amount of coca leaf produced. Other
estimates, such as the amounts of coca eradicated, were also collected but
 Coca frontiers and cocaleros in Chapare  99
reported less frequently. Second, there was a flurry of microeconomic analy-
ses comparing coca and alternative crop production. A third theme focused
on public health and was manifest in surveys of coca chewing and cocaine
use. Finally, there was a strong security theme associated with eradication,
trafficking, and violence.
The first theme is of particular relevance in this chapter, as coca produc-
tion statistics are another potentially important line of evidence that can be
used in establishing the timing and geography of coca frontiers in Bolivia.
However, these data need to be scrutinized carefully because the statistics
are fraught with issues, e.g., that:

1 they were not collected every year;

2 reporting by department or sub-regions within departments was
inconsistent;
3 they were collected using different methodologies (the details of which
were often not provided); and
4 they were published by different agencies whose positions have to be
considered.

The earliest estimates of production in the time period of interest are

national estimates of coca leaf production between 1937 and 1946. These
ranged from 1,620 (1943) to 6,733.2 (metric) tons (1937) (The Commission
of Enquiry into the Coca Leaf 1949). The first estimates of production by
department and sub-regions were from 1948 when values were reported for
the La Paz yungas (3468 tons), the Cochabamba yungas (831 tons) and Cer-
cado, Sara and Vallegrande Provinces in Santa Cruz Department (50 tons)
(Anon 1950). Pérez (1952) considers this production came from approxi-
mately 6,000 hectares in the La Paz yungas and 600 hectares in the Cocha-
bamba yungas. This would approximate gross yields of 0.6 and 1.4 tons per
hectare in the La Paz and Cochabamba yungas respectively.
Estimates spanning the late 1950s and early 1960s based on tax dec-
larations of dried leaves traded through the legal market are presented in
Table 4.5. Like many estimates, these are underestimates, as they did not
include coca leaf that illegally entered the cocaine trade.
Estimates of the area of coca cultivated in Cochabamba have been calcu-
lated from 1956 to 1963 (Table 4.5) using the average coca yield of 1.4 tons
per hectare (Pérez 1952). The area in 1956 would have been 639 hectares,
which is similar to the 600 estimated by Pérez six years earlier. Coca produc-
tion by 1963 suggests that the area under coca had expanded by approxi-
mately 1,023 hectares between 1950 and 1963. If it is assumed that the coca
cultivation areas in 1950 were in the well-established Cochabamba yungas,
the Espíritu Santo and San Mateo lowlands, and Alto Chipiriri, it is clear
that the increase in the area cultivated from 1956 would mainly have taken
place in the Chapare lowlands, rather than in the yungas. This ­argument is
supported by a sketch map of coca areas drawn by Decio Parreiras (Par-
reirras 1961) showing coca being grown in the lowlands in the vicinity of
100  Andrew C. Millington
Table 4.5  Coca leaf production 1956–1963 in metric tons

Cochabamba La Paz Area of coca cultivated

Department Department in Cochabamba
Department (ha)1

1956 894 2,520 639

1957 1,756 2,677 1,255
1958 1,495 2,570 1,068
1959 1,441 2,064 1,030
1960 1,549 1,881 1,107
1961 1,680 1,657 1,200
1962 1,818 2,377 1,299
1963 2,272 1,730 1,624

Notes: Based on tax declarations.

1
 hese estimates are based the average coca yield for Cochabamba in the late 1940s of 1.4 tons/
T
hectare (Pérez 1952)
Source: Rodriguez (1963)

Agrigento, Alto Chipiriri, Puerto Aurora, and Shinaota, as well as El Palmar

and Villa Tunari (Figure 4.1).
At least one documented trafficking route was in operation by 1957 along
which dried leaf and coca paste from Cochabamba, Santa Cruz, and Sucre
Departments was being taken overland to Corumba in western Brazil and
by air to São Paulo and Rio de Janeiro (Parreirras 1961). There is a slightly
earlier history. Gootenberg (2008) has clearly identified cocaine traffic from
La Paz and Cochabamba, through existing legal commercial networks, to
the U.S. through northern Chile in the early 1950s. U.S. Department of State
archives record U.S. Embassy reports of raids on cocaine factories gleaned
from Bolivian newspapers during the 1950s. Most were in metropolitan La
Paz. It could be assumed they were supplied with leaf from the La Paz yungas,
using the well-established routes that supplied the domestic chewing market,
a putative legal export trade, and the illegal cocaine trade.4 But the same
records also show that cocaine was manufactured in Cochabamba in the
early 1950s (Table 4.6). Using the same calculus of implied transportation
costs, it can be assumed that either Chapare or Cochabamba yungas leaf sup-
plied factories reported in Cochabamba (Table 4.6). Regardless of the geo-
graphical origin of leaves, by 1950 Cochabamba had been incorporated into
drug production and trafficking networks, which supplied Cuba, the U.S.
and beyond, that were supposedly controlled by American organized crime
(according to the sometimes doubtful U.S. Federal Bureau of Narcotics, FBN):

The fountainhead of “the mother lode of narcotics” is in Bolivia, in the

city of Cochabamba. There is an enormous plant ostensibly distilling
alcohol, but in reality refining narcotics to be sold the world over. ... The
“enterprise” is under the auspices of the Lucky Luciano syndicate, and
the “plant” is in the charge of two Brooklyn “executives” the Messers.
 Coca frontiers and cocaleros in Chapare  101
Frank de Piano and Michael Tramontana who travel in the most exclu-
sive Cochabambian [sic] society.
(Confidential source note appended to letter from Acting District Super-
visor Irwin I Greenfield to H.J. Anslinger, Commissioner U.S. FBN,
March 31, 1950)

There are no records of cocaine factory raids in Bolivia in the U.S. Depart-
ment of State for the 1940s, and the information for the 1950s is sparse.
Therefore, the greater volume of reports in the archives from the 1960s
(Table 4.6) may indicate the expansion of coca cultivation in Chapare and
its incorporation in drug trafficking. The coca factories around Santa Cruz

Table 4.6  Reports of Cocaine factories in Cochabamba, Santa Cruz, and Tarija
departments, 1950–1967

Date Location Despatch reference Notes

1950 Cochabamba, Appendices to Appendices from

11 km to W Greenfield letter, confidential source.
Mar 31, 1959 Factory allegedly part of
Lucky Luciano syndicate
1951 Cochabamba Embassy Despatch Subsidiary of a La
424 (La Paz), Paz organization.
Nov 28, 1951 No consulate in
Cochabamba at this time
Late February, Cochabamba, Embassy Despatch Discovered due to a fire
1953 near 576 (La Paz),
Mar 5, 1953
1960 Cochabamba Giordano letter, Allegedly controlled by
May 10, 1960 communist elements
related to national
government
May, 19 1965 Santa Cruz, 3 Airgram A-10, Report on all narcotics
km to N Nov 18, 1965 activity between May
1964 and Nov 1965
May, 1964 Santa Cruz, near Airgram A-10, Report on all narcotics
Nov 18, 1965 activity between May
1964 and Nov 1965
March 1965 Santa Cruz, 60 Airgram A-10, Report on all narcotics
km from Nov 18, 1965 activity between May
1964 and Nov 1965
25 June, 1965 Santa Cruz, eastern A-42 July 20, 1966
part of city
1 July, 1966 Paurito, 27 km SE A-42 July 20, 1966 Two factories
of Santa Cruz
9 July, 1966 Cotoca, 32 km A-42 July 20, 1966
E of Santa
Cruz
20 July, 1965 Santa Cruz Bi-annual summary, Case 536
1/1/66-6/30/66

(Continued)
Table 4.6  (Continued)

Date Location Despatch reference Notes

October, 1965 Portachuelo, 60 Airgram A-10, Report on all narcotics

km NW of Nov 18, 1965 activity between May
Santa Cruz 1964 and Nov 1965
October, 1965 Cochabamba, Airgram A-10, Report on all narcotics
near Nov 18, 1965 activity between May
1964 and Nov 1965
October, 1965 Cochabamba, Airgram A-10, Report on all narcotics
near Nov 18, 1965 activity between May
1964 and Nov 1965
28 January, Oruro, southwest Bi-annual summary, Case 552
1966 part of city 1/1/66-6/30/66
7 February, Trapiche, Cotoca Bi-annual summary, Case 551
1966 Canton, Santa 1/1/66-6/30/66
Cruz Dept.
7 February, Cotoca Airgram A-10, Feb Same as above check
1966 16, 1966 names. El Tinterillo
aka Luis Artega
9 March, 1966 Santa Cruz Airgram A-24,
Mar 18, 1966
15 March, 1966 Santa Cruz Bi-annual summary, Case 554
1/1/66-6/30/66
10 April, 1966 Cotoca Canton Bi-annual summary, Case 597
1/1/66-6/30/66
20 April, 1966 Cotoca Airgram A-31, Allegedly connected
May 5, 1966 to an international
smuggling ring in
Brazil.
21 May, 1966 “El Bajio” near Bi-annual summary, Case 596
Santa Cruz 1/1/66-6/30/66
27 May, 1966 El Cuquito, Bi-annual summary, Case 600
Santa Cruz 1/1/66-6/30/66
29 May, 1966 “Tres Palmas”, Bi-annual summary, Case 598
Santa Cruz 1/1/66-6/30/66
26 June, 1966 Santa Cruz Bi-annual summary, Case 601
1/1/66-6/30/66
30 June, 1966 Viena, 30 km Bi-annual summary, Case 602
from Santa 1/1/66-6/30/66
Cruz
9 July, 1966 Santa Cruz Bi-annual summary, Case 628
1/7/66-12/30/66
25 September, Cochabamba Bi-annual summary, Case 329
1966 1/7/66-12/30/66
2 October, No location Bi-annual summary, Case 653, two separate
1966 given 1/7/66-12/30/66 raids
7 November, Cochabamba Bi-annual summary, Case 655
1966 1/7/66-12/30/66
17 November, Santa Cruz Bi-annual summary, Case 657
1966 1/7/66-12/30/66
Cotoca, 32 km E Cotoca, 32 km Airgram A-31,
of Santa Cruz E of Santa May 5 1966
Cruz
Table 4.6  (Continued)

Date Location Despatch reference Notes

Early “40 Arroyos”, Airgram A-20,

November, Chapare Dec 5, 1966
1966 Province,
Cochabamba
Dept.
“Brecha 13”, “Brecha 13” on Airgram A-20,
on Santa Santa Cruz- Dec 5, 1966
Cruz-Camiri Camiri road
road
“Monetgrina” “Montegrina” Airgram A-20,
13 km on 13 km along Dec 5, 1966
Santa Cruz- Santa Cruz-
Cochabamba Cochabamba
road road
5 January, 1967 “Palmar Airgram, A-27, J
Viriua”, 4 km an 24, 1967
N of Santa
Cruz
January 1967 Bermejo, Tarjia Airgram A-27,
Dept. Jan 24, 1967
January, 1967 Between Airgram, A-27,
Santandita Jan 24, 1967
and Yacuiba
8 February, Yacuiba, Tarjia Airgram, A-30,
1967 Dept. Mar 27, 1967
13 March, Santa Cruz, Airgram, A-30, Supplied with raw
1967 outskirts Mar 27, 1967 cocaine from a plant
near Cochabamba
8 April, 1967 Cotoca, 32 km Airgram A-35,
E of Santa May 12 1967
Cruz
April-May, Cotoca, 32 km Airgram A-35,
1967 E of Santa May 12 1967
Cruz
April-May, San Carlos,
1967
70 km NW of Airgram A-35, Also in Operations
Santa Cruz May 12 1967 Memo. July 11,
1967
April-May, “Urbo” near Airgram A-35,
1967 San Carlos, May 12 1967
70 km NW of
Santa Cruz
April-May, “El Carmen”, S Airgram 35,
1967 of Santa Cruz May 12 1967
May, 1967 Claro Cuta, near Operations Memo.
Santa Cruz July 11 1967
May, 1967 “Tierras Operations Memo.
Nuevas”, 5 km July 11 1967
S of Santa Cruz
104  Andrew C. Millington
that appear in the records in this decade were undoubtedly mainly supplied
by Chapare-grown leaf carried along the new Santa Cruz Highway. This
heralded the link between the peasantry in Chapare with the modernizing
wealthy region of Santa Cruz. The greater number of raids reported in the
1960s than in previous decade could also be due to the establishment of an
American Consulate in Cochabamba, whose region of authority included
the departments of Cochabamba, Santa Cruz, and Tarija. The impetus to
report may have been elevated once systematic reporting of narcotics activ-
ity was implemented.
Table 4.7 provides a time series of coca production estimates the author
has logged from various sources that cover the time frame of this essay. The
most salient points in the context of this chapter are the growth in coca leaf
production in Cochabamba Department, and that Cochabamba production
had overtaken that of La Paz by 1963. An obvious feature of Table 4.7 is that

Table 4.7  Coca production 1940–1990: Bolivia and Cochabamba and La Paz
Departments in metric tons

Cochabamba La Paz Bolivia Difference1 (% of

national estimate)

1940 3,608.4
1941  
1942 3,474.7 3,893.0
1943 3,296.8 3,642.7
1944 3,352.3 3,704.0
1945 6,253.7 4,830.0
1946 7,312.6 4,039.9
1947 6,693.1 3,697.7
1948 6,276.0 3,467.2
1949 6,134.0 3,338.8
1950 5,950.8 2,929.4
1951 6,322.0 7,336.0
1952 3,608.7 3,492.7
1953 3,141.7 3,987.3
1954 2,612.5 3,471.4
1955 2,503.7 2,886.6
1956 894.2 2,332.2 2,766.4 459.9 (16.6%)
1957 1,756.2 2,477.0 2,576.8 1,656.3 (64.3%)
1958 1,495.2 2,377.9 2,736.8 1,136.3 (41.5%)
1959 1,441.7 2,064.9 2,627.4 879.3 (33.5%)
1960 1,550.0 1,881.1 4,947.4 1,516.4 (30.7%)
1961 1,680.0 1,657.9 3,337.9 0
1962 1,818.4 2,377.2 4,195.6 0
1963 2,273.0 1,730.6 4,800.0 796.5 (16.6%)
1964 4,900.0 4,960.0
1965 5,491.0
1966 1,964.0 5,330.0
1967 1,975.0 6,459.5
1968 1,976.0 4,219.5
 Coca frontiers and cocaleros in Chapare  105
Table 4.7  (Continued)

Cochabamba La Paz Bolivia Difference1 (% of

national estimate)

1969 4,506.2
1970 7,094.7
1971 6,427.0
1972 8,406.0
1973 8,816.4
1974 11,458.5
1975 11,800.0 3,928.7 11,172.5 4,566.3 (40.8%)
1976 13,755.0 2,970.6 11,171.7 5,554.0 (49.7%)
1977 11,786.7
1978 26,754.7
1979 27,893.0
1980 36,464.0
1981 55,735.0
1982 65,464.0
1983 63,788.5
1984 70,532.5
1985 10,276.4 99,600.0
1986 44,748.5
1987 96,245.5
1988 35,000.0 13,900.0 98,895.5 49,995.5 (50.6%)
1989 38,500.0 14,400.0 65,406.3 12,506.3 (19.1%)
1990 36,000.0 14,300.0 102,855.5 52,555.5 (51.1%)

Note: Values that are clearly erroneous have not been included.
Sources: Aduana de la Coca (1942, 1948, 1949, 1950, 1951, 1952, 1953, 1954, 1955, 1958,
1959), Blanes and Flores (1982), Instituto Nacional de Estadísticas (INE) (1980), Carter and
Mamani (1986), MACA (1983) Menses (1945), Oficina de Estatísticas Económicas y Estatísticas
(1976), Pérez (1952), South (1977).

while the national data is almost complete, corresponding departmental and

sub-regional data are intermittent. It appears that no data exist between 1959
and 1964; and although there are records for 1965, after 1966 departmental
level are generally not reported again until the late 1980s. The absence of
verifiable estimates from Cochabamba Department makes reconstruction of
the coca frontier in Chapare difficult from statistical sources alone.
There is a further issue of internal inconsistency between data for the
years when departmental and national data are available. While it can be
assumed that almost all coca referenced in Table 4.7 was grown in Cocha-
bamba and La Paz Departments, the departmental estimates rarely add up to
the national total. Moreover, there is no systematic pattern in the differences
between departmental and national estimates with time or the amounts pro-
duced. This is another roadblock to refining knowledge of the development
of the Chapare coca frontier with statistical data. One solution could be to
subtract the La Paz production estimates from those for the entire country.
However, confidence in the estimates derived in this way is low. For example,
106  Andrew C. Millington
Table 4.8  Coca leaf production, 1956–1958 in metric tons

Tax records Aduana de la

Coca records

1956 2332.15 2520.816

1957 2476.95 2677.332
1958 2377.92 2570.296

Note: Based on Aduana de la Coca and tax declaration records for

La Paz Department
Source: Rodríguez (1963)

Cochabamba estimates produced by subtraction between 1942 and 1955

were plausible between 1942 and 1944 (they ranged from 345.9 to 418.3
tons), but implausible between 1945 and 1950 when the national produc-
tion estimates for Bolivia were less than those for La Paz Department alone!
However, glimpses of an important trend can be seen in Table 4.6. That
is the growth in the importance of coca production in Cochabamba rela-
tive to La Paz. As has been argued already, any expansion of production
in Cochabamba Department must have meant an expansion of coca in the
Chapare lowlands. Therefore, these data, despite the caveats that need to be
applied to them, provide further evidence for the timing, development, and
expansion of the Chapare coca frontier. Bascopé (1982) noted that although
other varieties of coca had been grown in other parts of the world compa­
rable to the ecological conditions of the La Paz yungas, including Chapare,
until 1952 none of these areas had acquired the importance of the La Paz
yungas. The fragmentary data from the early 1960s and mid-1970s shows
Cochabamba coca production clearly exceeded that of La Paz. This is firm
evidence that a coca frontier had opened in Chapare by the early 1960s and
had expanded significantly by the mid-1970s.
A further caveat is that there are likely biases in the area estimates due to
the different organizations collecting and reporting these data. This is illus-
trated by two sets of estimates from La Paz Department (Table 4.8), which
show that tax declarations for coca production were systematically lower
than the custom levies on leaf traded. Differences in estimates, however,
became highly contested once it was clear that coca was being grown for
cocaine as well as chewing. This is particularly true from the 1970s onwards
and remains so to the present day. These and other caveats mean that these
data should be regarded as relative rather than absolute until further analy-
ses have been carried out.

The Chapare frontier: emergence, establishment, and persistence

The development of the Chapare coca frontier can summarized around a
temporal framework of emergence, establishment, and persistence.
 Coca frontiers and cocaleros in Chapare  107
Emergence: An embryonic frontier opened in Chapare as it evolved into
one of Bolivia’s key highland–lowland communications hubs during the
nineteenth century. By the early twentieth century, the Chapare lowlands
had been opened to agricultural colonization. Coca was established at the
outset of settlement in the Chapare lowlands, albeit as a secondary crop, and
the available evidence points to Alto Chipiriri and the Espíritu Santo and
San Mateo lowlands being the loci of early settlement and cultivation. While
it is clear that a small region of central Chapare had become a colonization
frontier by the early twentieth century, it is difficult to describe the area as a
coca frontier, as the plant was only a minor part of the crop mix. As the coca
grown was destined for the domestic chewing, neither can Chapare be described
as a cocaine frontier at this time.
Its relationship to the coca plantations in the adjacent Cochabamba yun-
gas is uncertain. It may be that the Junta de Propetarios de Yungas may have
had a role in the establishment of the agricultural settlements in Alto Chip-
iriri and the Espíritu Santo and San Mateo lowlands. However, until more
material can be unearthed on this organization, it is speculation.
Establishment: Poles of agrarian settlement were established in Chapare
from the late 1930s as the government tried to attract overseas migrants
and highlands communities began to migrate to the lowlands en masse in
family units. After the often chaotic break up of old estates during the post-
1952 agricultural revolution, a gap between weak state bureaucratic control
in Chapare and strong local-level organization grounded in agrarian politics
opened up. This was probably a critical factor in enabling coca and cocaine
to become established. But mirroring arguments made elsewhere, myriad
small failures in modernization and development in these new settlements—
particularly in transportation and health provision—created further insta-
bility amongst colonists. An uneasy vacuum was developing. By the early
1950s these settlements were being established simultaneously with cocaine
manufacturing taking place in Cochabamba and the establishment of cocaine
trafficking routes to North and South American cities. By 1950 there is little
doubt that central Chapare was simultaneously a colonization and coca fron-
tier and, on the basis that coca leaves from Chapare were being processed for
cocaine, it was also an embryonic cocaine frontier.
Further waves of colonization led to more settlements being established as
a network of roads and tracks penetrated the forests. This was partly related
to the construction of the Santa Cruz Highway, which later facilitated the
transport of coca leaf to cocaine factories in Santa Cruz Department, but
which at this time extended the coca frontier eastward to the banks of the
Río Ivirgarzama, where Ivirgarzama is now located, by the end of the 1960s.
The frontier also encroached in the northwest along the road to Moleto. Sig-
nificant amounts of clearance took place between Puerto Villarroel and Valle
Hermoso in the 1970s, and between Villa Tunari and Chimoré. In summary,
a well-established colonization, coca, and cocaine frontier extended from the
west of Alto Chipiriri to Río Sajta by the end of the 1960s.
108  Andrew C. Millington
Persistence: In the two decades after the 1970s there was further settle-
ment, forest clearance, and illicit coca cultivation as the frontier encroached
further into forests in response to the expanding feeder road network, and
as oil and gas exploration lines were cut. Expansion occurred in three direc­
tions: to the east beyond the Río Ivirgarzama to the border with Santa Cruz
Department with easy transport to the economically burgeoning and buoy-
ant Santa Cruz Department; to the west into what is now the Territorio
Indígena y Parque Nacional Isiboro-Sécure (TIPNIS); and north into season-
ally flooded lowlands. The unstable vacuum from failed modernization and
development that had been building up from the end of World War II was
being filled by a coca-cocaine agroindustry. All of Chapare was a coloniz-
ation, coca, and cocaine frontier in the 1980s and 1990s.
To summarize, during the twentieth century, a series of asynchronous,
geographically overlapping frontiers emerged and shifted over time within
the forested tropical lowlands that covered Chapare. Different economic,
demographic, and political institutions and agents had forged these fron-
tiers. Whilst a modest colonization, coca, and cocaine frontier had been
established by the end of the 1960s, there was no guarantee that it would
grow or even survive. But then the young, the hip, and the rich discovered
the pleasures and status of cocaine. Global demand skyrocketed. Chapare,
with a legacy of coca linked to pre-1970s narco-trafficking networks, and
weak government oversight, was ideally suited with space to expand and
ultimately sate the appetites of North American and European consumers.
A perfect storm had brewed, as the memoirs of Anselmo Andreotti, a
priest posted to La Victoria from 1969 to 1980, describe. La Victoria was
like other colonies in his parish—largely occupied with drug trafficking
(Anderotti 2003). The colonists, the refugees of failed modernization in
Chapare, had become cocaleros, and the colonization frontier had become a
persistent coca and cocaine frontier.

Notes
1 The authors would like to thank the staff at Stanford University Library; the
National Administration Records and Archives, College Park, Maryland; the
Inter-American Institute, Berlin; the National Archives, Sucre; and the Municipal
Archives, Cochabamba are thanked for their help; in particular Wenceslaw Fer-
ruffino in Cochabamba. Flinders University provided travel funds and an Out-
side Studies Program Grant to visit libraries and spend a sabbatical in Bolivia.
Early field work was funded by the European Union. Many ideas were discussed
with Andrew Bradley, Félix Huanca, and Danny Redo. Eric Lambin hosted my
visit to Stanford. Emmanuel Bambe, Robert Keane, and Blythe Schembri worked
with me on Corona data. Paul Gootenberg suggested I turn my attention to the
historical development of Chapare and invited me to present this research at the
34th LASA Congress in New York (2016). Andrew Ehrinpreis commented on the
conference paper.
2 The early routes comprised overland tracks from Cochabamba to river ports,
then by boat to Beni. Not all settlements were new, e.g., Chimoré, Todos Santos,
 Coca frontiers and cocaleros in Chapare  109
and Villa Tunari grew out of old missions, and Moleto was a “núcleo central” of
the Yuracare (Pinto Parada 2001).
3 References to JPY are fleeting compared to the catalogued SPY archive. There
was no accessible material on JPY in the Municipal Archives in Cochabamba in
2015. A lot of material in this archive still has to be catalogued.
4 Four letters concerning samples and imports of Bolivian coca leaves were received
by U.S. companies and import agents dated between December 17, 1955 and
October 23, 1956 in the U.S. Bureau of Narcotics files.

Bibliography
Aduana de la Coca. 1942. Estadística general de exportación de las provincias de
Nor y Sud Yungas e Inquisivi. La Paz: Editorial Sport.
Aduana de la Coca. 1948. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1948. La Paz: Aduana de la Coca.
Aduana de la Coca. 1949. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1949. La Paz: Aduana de la Coca.
Aduana de la Coca. 1950. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1950. La Paz: Aduana de la Coca.
Aduana de la Coca. 1951. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1951. La Paz: Aduana de la Coca.
Aduana de la Coca. 1952. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1952. La Paz: Aduana de la Coca.
Aduana de la Coca. 1953. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1953. La Paz: Aduana de la Coca.
Aduana de la Coca. 1954. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1954. La Paz: Aduana de la Coca.
Aduana de la Coca. 1955. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1955. La Paz: Aduana de la Coca.
Aduana de la Coca. 1958. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1958. La Paz: Aduana de la Coca.
110  Andrew C. Millington
Aduana de la Coca. 1959. Estadística general de exportación de productos agricolas
de las provincias Nor y Sud Yungas, Inquisivi, Larecaja, Caupolicán, Muñecas,
Murillo y Loayza, cuya recaudación de Impuestos corre a cargo de la Aduana de
la Coca, correspondiente a la gestión de 1959. La Paz: Aduana de la Coca.
Alcaraz del Castillo, Franklin E., Rosse M. Soliz, and Julia Zuaro. 2001. La Coca del
Siglo XX en Bolivia. La Paz: Centro Latinoamericano de Investigación Científica.
Anderotti, Anselmo. 2003. Once años en el Chapare 1969–1980. Cochabamba: Edi-
torial Verbo.
Anon. 1950. “Coca Chewing, Geography and Nutrition.” Bulletin of Narcotics 4: 2–13.
Bascopé, Rene. 1982. La veta blanca. La Paz: Ediciones Aquí.
Blanes, José, and Gonzalo Flores. 1982. ¿Donde vá el Chapare? Cochabamba: Cen-
tro de Estudios de la Realidad Económica y Social.
Bradley, Andrew V. and Andrew C. Millington. 2008. “Coca and Colonists: Quan-
tifying and Explaining Forest Clearance Under Coca and Anti-narcotics Policy
Regimes.” Ecology and Society 13(1): 31. www.ecologyandsociety.org/vol13/
iss1/art31/. Last accessed 17 March 2018.
Camacho, Miriam. 2005. Características sociodemográficas y económicas del
Municipo de Chimoré. Cochabamba: Centro de Estudios de Población, Univer-
sidad Mayor de San Simón.
Carter, William F., and Mauricio Mamani. 1986. Coca en Bolivia. La Paz: Editorial
Juventud.
Centro Cuarto Intermedio (CCI). 2011. “Los trasfondos del TIPINIS: la complejidad
del conflict.” Estudios & Ensayos XI 36: 1–105.
Cusi, Raúl, and Miguel Flores. 2007. ADEPCOCA Hechos que hicieron Historia.
La Paz: ADEPCOCA.
de Viedma, Francisco. 1836. Descripción geográfica y estadística de la provincia
de Santa Cruz de la Sierra. Buenos Aires: Imprenta del Estado. Online version
published by Biblioteca Virtual Miguel de Cervantes. www.cervantesvirtual.com/
obra/descripcion-geografica-y-estadistica-de-la-provincia-de-santa-cruz-de-la-si-
erra--0/. Last accessed 23 August 2017.
Durand Ochoa, Ursula. 2014. The Political Empowerment of the Cocaleros of
Bolivia and Peru. London: Palgrave MacMillan.
El Imparcial (Cochabamba). 1940. “Colonia Presidente Busch.” July 12, 1940, 3.
El Imparcial. 1949. “Colonizadores para el Chapare.” May 11, 1949, 1.
Fifer, Valerie J. 1967. “Bolivia’s Pioneer Fringe.” Geographical Review 57(1): 1–23.
Gootenberg, Paul. 2008. Andean Cocaine: The Making of a Global Drug. Chapel
Hill NC: University of North Carolina Press.
Government of Bolivia. 1998. Ley del Régimen de la Coca y Sustancias Controladas.
Ley N° 1008 de 19 de Julio de 1988. www.dgsc.gob.bo/normativa/leyes/1008.
html. Last accessed 22 July 2017.
Healy, Kevin. 1988. “Coca, the State, and the Peasantry in Bolivia, 1982–1988.”
Journal of Interamerican Studies and World Affairs 30(2/3): 105–26.
Instituto Nacional de Estadísticas (INE). 1980. Bolivia en cifras. La Paz: INE.
Instituto Nacional de Estadísticas (INE). 1982. 1982 Atlas Censal. La Paz: INE.
Instituto Nacional de Estadísticas (INE). 2001. Censos Anteriores. http://censosbo-
livia.ine.gob.bo/webine/censos-anteriores. Last accessed 1 July 2017.
Instituto Nacional de Estadísticas (INE). 2012. Censo de Población y Vivenda 2012.
La Paz: INE. www.censosbolivia.gob.bo Last accessed 18 March 2018.
 Coca frontiers and cocaleros in Chapare  111
Instituto Nacional de Estadísticas (INE). 2016. Censos 2012. http://censosbolivia.ine.
gob.bo/webine/article/ine-concluyó-procesamiento-de-datos-del-censo-2012-pa-
ra-medir-no-sólo-cuántos-somos-sino. Last accessed 1 July 2017.
Lautenbach, Sven, Caroline Kugel, Angela Lausch, and Seppelt Ralf. 2010. “Analysis
of Historic Changes in Regional Ecosystem Service Provisioning Using Land Use
Data.” Ecological Indicators. doi:10.1016/jecolind.2010.09.007
Lee, Renssalaer, and Patrick Clawson. 1993. “Crop Substitution in the Andes.”
ONDCP Paper. 1–72. Washington, DC: Office of National Drug Control Policy.
Loza-Balsa, Gregorio. 1992. Monografía de la coca. La Paz: Sociedad Geográfica
de La Paz.
MACA (Rural Development Services). 1983. Nuestros conocimientos. Prácticas
agropecuarias tradicionales en Bolivia. La Paz: MACA.
Mamani, María L. 2004. Características sociodemográficas y económicas del Muni-
cipio de Tiraque. Cochabamba: Centro de Estudios de Población, Universidad
Mayor de San Simón.
McDonald, Robert A. 1995. “CORONA: Success for Space Reconnaissance. A Look
into the Cold War, and a Revolution for Intelligence.” Photogrammetric Engi-
neering and Remote Sensing 61: 689–720.
McDonald, Robert A. 1997. Corona between the Sun and the Earth. The First NRO
Reconnaissance Eye in Space. Bethesda MD: American Society for Photogram-
metry and Remote Sensing.
Menses, Raúl R. 1945. En Bolivia está Yungas: Un studio sobre turismo, geografía,
economía. La Paz: Editorial “Económica” Chulumani.
Meruvia, Fanor. 2000. Historia de la coca. Los Yungas de Pocona y Totora (1500–
1900). La Paz: Plural/CERES/Alcaldia de Totora.
Millington, Andrew C., Ximena Vélez Liendo, and Andrew V. Bradley. 2004. “Scale
Dependence in Multi-temporal Mapping of Forest Fragmentation in Bolivia:
Implications for Explaining Temporal Trends in Landscape Ecology and Appli-
cations to Biodiversity Conservation.” ISPRS Journal of Photogrammetry and
Remote Sensing 57(4): 289–99.
Nagombi, Edmond. 2017. “Quantifying Deforestation Rates in a Cocaine Source
Region in Bolivia: A Study to Verify Rates of Land-Use Change Under Pro-coca
Policies.” Masters of GIS thesis, Flinders University, Adelaide.
Oficina Nacional de Estudios Económicas y Estadística. 1976. Boletin de la Oficina
Nacional de Estudios Económicas y Estadística 1976. La Paz: Bolivia.
Parreirras, Decio. 1961. “Data on the Illicit Traffic in Cocaine and Coca Leaves in
South America, with an Annex on Narcotics Control in Brazil.” Bulletin of Nar-
cotics 2: 10–15.
Peredo, M. 2004. Características Sociodemográficas y Económicas: Municipio de
Pojo. Cochabamba: Centro de Estudios de Población.
Peréz, R. 1952. “The Coca Question in Bolivia.” Bulletin of Narcotics 2: 10–15.
Philip, Graham, Daniel Donoghue, Anthony Beck, and Nikolaos Galiatstatos. 2002.
“CORONA Satellite Photography: An Archaeological Application from the Mid-
dle East.” Antiquity 96: 109–18.
Pinto Parada, Rodolfo. 2001. Rumbo al Beni. 2nd edition. La Paz: Proyecto de Pavi-
mentación Carretera Santa Cruz-Trinidad, Tramo San Pablo-Casarabe.
Rodríguez, Argos A. 1963. “Possibilities of Crop Substitution for the Coca Bush in
Bolivia.” Bulletin of Narcotics 3: 13–23.
112  Andrew C. Millington
Rodríguez Ostria, Gustavo. 1972. Historia del Trópico Cochabambino. Cocha-
bamba: Prefectura del Departamento de Cochabamba.
Rodríguez Ostria, Gustavo. 2012. “¡Ferrocarril o nada! La lucha por la locomotora
en Cochabamba 1892–1927.” Revista Estudios del ISHiR (Investigaciones Socio
Históricas Regionales) 2(3): 1–29.
Ruffner, Kevin C. (ed.). 1995. Corona: America’s First Satellite Program. Washing-
ton, DC: Central Intelligence Agency (CIA).
Salazar Ortuño, Fernando B. 2009. Movimentos sociales en Torno a la producción
de coca en Bolivia. La Paz: IESE Universidad de San Simon.
Sanabria, Harry. 1993. The Coca Boom and Rural Social Change in Bolivia. Ann
Arbor, MI: University of Michigan Press.
South, Robert B. 1977. “Coca in Bolivia.” Geographical Review 67(1): 22–33.
Soux, María L. 1993. La Coca liberal. Producción y circulación a principios del siglo
XX. La Paz: Cocayapu/CID.
Tappen, G. Gary, Amadou Hadj, Eric. C. Wood, and Ronald W. Lietzov. 2000. “Use
of Argon, Corona, and Landsat Imagery to Assess 30 Years of Land Resource
Changes in West-Central Senegal.” Photogrammetric Engineering and Remote
Sensing 66(6): 727–35.
UNODC (United Nations Office on Drugs and Crime). 2005. Bolivia Coca Cultiva-
tion Survey 2004. La Paz: UNODC/GoB.
UNODC (United Nations Office on Drugs and Crime/Estado Plurinacional de
Bolivia). 2016. Monitereo de Cultivo de Coca 2015. La Paz: UNODC/GoB.
van Gils, Hein A.M.J., and Analia V. Loza Armand Ugon. 2006. “What Drives
Conversion of Tropical Forest in Carrasco Province, Bolivia?” Ambio 35(2):
81–85.
van der Berg, Hans. 2008. En busca de una senda segura. La Paz: Plural/Universidad
Católica Boliviana.
Zubieta, Luis R.G. 2012. Ferrocarriles Bolivia. Del anhelo a la frustración. Desa-
rollo, producción y dependencia. www.ferrocarrilesbolivia.com. Last accessed 23
August 2017.

Archives Consulted
US National Archives and Records Administration, College Park MD
American Consulate, Cochabamba 1965
Dept of State Airgram A-10, 000254 Control of Narcotics Drugs – Narcotics
Activ­ities in Cochabamba Consular District Ref: 7 FAM 482.3-2 November 18,
1965.
May 1964 to November 1965.
American Consulate, Cochabamba 1966
Department of State Airgram A-31, 000254 – Control of Narcotic Drugs Ref: 7 FSM
482.3-2 May 5 1966.
American Consulate, Cochabamba 1967
Dept of State Airgram A-29, 000254 Control of Narcotics Drugs Ref: 7 FAM
482.3-2 January 24, 1967.
Bi-annual summary of investigations of manufacture, traffic and consumption of
narcotics in Bolivia, January 1 to June 30, 1966, July 1 to December 30, 1966.
 Coca frontiers and cocaleros in Chapare  113
0660 Foreign Countries: Bolivia (file 1 1933–1955)
Letter, 31 March 1950 from Irwin I Greenfield Acting Dist. Supervisor to H. J.
Anslinger. Notes appended: (i) Information from a Confidential Source March
27, 1950, (ii) Information from a Confidential Source June 19 1950.
American Embassy, La Paz, Despatch 424 to Dept. of State, Washington, DC.
November 28, 1951.
0660 Foreign Countries: Bolivia (file 2 1956–1963)
American Embassy, La Paz, Airgram A-10 to Dept. of State, Washington, DC.
November 18, 1965: 000254 Control of Narcotic Drugs – Narcotic Activities in
Cochabamba Consular District.
American Consulate, Cochabamba, Airgram A-10 to Dept. of State, Washington,
DC. February 16, 1966: 000254 – Control of Narcotic Drugs.
American Consulate, Cochabamba, Airgram A-24 to Dept. of State, Washington,
DC. March 18, 1966: 000254 – Control of Narcotic Drugs.
American Consulate, Cochabamba, Airgram A-42 to Dept. of State, Washington,
DC. July 20 1966: 000254 – Control of Narcotic Drugs.
American Consulate, Cochabamba, Airgram A-20 to Dept. of State, Washington,
DC. December 5, 1966: 000254 – Control of Narcotic Drugs.
American Consulate, Cochabamba, Operations Memo to Dept. of State, Washing-
ton, DC, July 11, 1967: Control of Narcotic Drugs 000254.
Box 3060, File SOC/1/1/67
American Consulate, Cochabamba, Airgram A-27 to Dept. of State, Washington,
DC. January 24, 1967: 000254 – Control of Narcotic Drugs.
American Consulate, Cochabamba, Airgram A-30 to Dept. of State, Washington,
DC. March 27, 1967.
American Consulate, Cochabamba, Airgram A-35 to Dept. of State, Washington,
DC. May 12, 1967: 000254 – Control of Narcotic Drugs.
Box 3196 1964-1966
American Consulate, Cochabamba, Airgram A-31 to Dept. of State, Washington,
DC. May 5, 1966: 000254 – Control of Narcotic Drugs.

Newspapers consulted:
El Imparcial (Cochabamba) 1940–1949.
5 Economic development
policies in Colombia
(1960s–1990s)and the turn
to coca in the Andes Amazon
Jennifer S. Holmes, Viveca Pavón, and
Sheila Amin Gutiérrez de Piñeres1

Introduction
This chapter provides an overview of the economic policies and economic
cycles that helped push poor migrant Colombian farmers into coca produc-
tion, an almost entirely new crop, over the three decades between the 1960s
and 1990s. First, it reveals that apart from classic questions of Colombia’s
political vulnerability to both illegal drugs and violence stressed by previous
studies (Thoumi 2003), a political economy was in play. Colombia became
ripe for illicit coca because of the specific dynamics of economic inequalities
and insecurities created by this cyclical political economy of coffee, and by
its associated practices of business illegality and speculative boom investing.
The second part of the chapter examines the effects of this political economy
in the three main areas where migrants flocked to after the mid-1970s for
emerging coca livelihoods: the new Amazonian departments of Guaviare,
Putumayo, and Caquetá. These case studies reveal a lack of sustained gov-
ernmental support in such basic local services as education and security,
and dynamics of social inequality in such refuge regions as well. The next
chapter, by María Clara Torres, follows up on this one with a specific ethno-
graphic history of campesino migration to Meta Department, where official
Colombian colonization projects of the 1960s also failed, leading to the
widespread adoption of coca.

A brief history of coca production in Colombia

In the nineteenth and twentieth centuries, coca was not as widespread in
Colombia as in other Andean countries, and it was only present in the south-
ern departments of Cauca and Huila, mostly oriented to indigenous users.
In Colombia, use of coca was quite limited, compared to tobacco, chicha (a
fermented corn beverage), and several other drugs, including some halluci-
nogens (Bula 1988). The traditional use of these drugs was limited to rural
regions. By 1938 the sale of coca leaves was limited to medically prescribed
doses filled only in pharmacies. By 1941 all new coca plantations were pro-
hibited (Rubio 1988), a sign of how exotic and expendable the plant was to
 Economic development policies in Colombia  115
Colombia’s urban elites, unlike Bolivia or Peru. In March 1947, President
Ospina, responding to new international initiatives against coca, signed a
decree that prohibited the possession, growth, distribution, and sale of coca
(and marijuana), ordered the destruction of existing plants, and established
jail and fines for those who violated the decree. The decree, however, was
not universally enforced and was ultimately delayed by one year. The Cauca
hacendados were the main opponents of the decree. By the time the one-
year stay expired, the Public Health Minister Dr. Bejarano resigned and the
decree was never reissued or enforced (Thoumi 1995). Prior to the 1970s,
then, coca was a quite scarce peasant crop in Colombia.
Psychoactive drug manufacturing before the late 1960s in Colombia
was not well known. In one colorful instance of the late 1950s, a group of
Colombian smugglers established a laboratory in Medellín to refine cocaine,
heroin, and morphine for export to Havana, where the American mafia
managed the distribution. This laboratory was discovered and destroyed by
the Colombian secret service in conjunction with the FBI (Thoumi 1995).
But in the early 1970s, Colombia suddenly transformed into the world cen-
ter of illicit cocaine processing and marketing chains displaced from their
longer but modest roots in Peru, Bolivia, and Chile (Gootenberg 2008, 303–
306). Coca paste or base was still imported from either Peru or Bolivia.
Colombia’s rapid success in manufacturing and marketing the drug led to
the spread of domestic coca-growing, which was embraced as a cash crop
in depressed and isolated regions of the country. During the “lost decade”
of the 1980s, the industry, to capture greater profits, became even more
vertically integrated.
The Caguán region in Caquetá and the Guaviare precinct began to grow
coca in the mid-1970s. Coca cultivation attracted a variety of migrants in
search of quick profits. Some came with urban backgrounds and criminal
experience. Some of the urban migrants already linked to the drug trade had
higher education, like chemists who came to work for laboratories. Other
migrants came from the emerald mining community, which had violent and
illegal tendencies.  Others who had migrated previously took advantage
of the new cash crop with yields that outpaced those of other agricultural
opportunities. This mix of migrants with varying backgrounds brought
experience, violence, and desire for profits together in search of a livelihood.
This combination of migrants and systemic economic failure provided a fer-
tile ground for coca cultivation, especially in the absence of other alterna-
tives to earn a living either in the region or nationally.

Historical overview of the Colombian economy

and attempts at agricultural reform
Exploring the state of the agricultural sector throughout Colombia is a key
to understanding the ultimate failure of mid-twentieth century development
in the Colombian Amazon. While prior to 1966, the Colombian economy
116  Jennifer S. Holmes et al.
was characterized by high inflation, from the mid-1960s through the late
1990s the Colombian economy was defined by coffee prices and the govern-
ment’s response to them (Amin Gutiérrez de Piñeres and Ferrantino 2000).
Beginning in 1966, favorable world economic conditions and moderate
fiscal and monetary policies stimulated overall growth. During the coffee
bonanza (1975–1982), government spending increased dramatically as eco-
nomic growth decelerated. From 1966 until 1974, annual growth rates of
real GDP averaged 6.57 percent, while during the coffee bonanza real GDP
growth decelerated to 4.05 percent annually. After 1983 through the 1990s,
and as coffee prices became highly unstable, annual growth rates of GDP
fell to an average of 3.3 percent. While still respectable by Latin Ameri-
can standards, these lower growth rates justified constraints on government
spending on social programs, including commitments to provide services to
the far-flung Amazonian territories.
The overall numbers, however, belied important challenges in the agricul-
tural sector. In the late 1960s, President Lleras Restrepo began a program
of austerity that included trade and exchange rate controls, tight credit pol-
icies, tax reforms, a balanced budget, and an emphasis on export expansion
and import substitution (Worldmark 1984). By that point, the stagnation
of real wages in the agriculture sector had lasted more than two decades
(Schultz 1969) making the pain of an austerity program even more acute.
The stagnant conditions for agriculture were compounded by widely
flouted currency controls that invited illegality. Even smuggling of coffee
exports became common. From 1931 to 1991, Colombia had exchange
controls that made it illegal for residents to hold foreign currencies or
other financial assets in or outside the country. However, there was wide
circulation of foreign exchange across the country, and the Central Bank
allowed for black market exchange purchases through its “sinister window”
(Thoumi 1995). This practice simplified money laundering, and continued
to contribute to the disrespect of the law. Over several decades, the wide-
spread use of such “illegal” exchanges removed the social stigma from most
illegal economic activity in Colombia.
In the mid-1970s, the first “coffee bonanza” led to increased export reve-
nues and the government embarked on a spending spree that proved unsus-
tainable (Garcia Garcia and Montes Llamas 1988). While positive world
economic conditions provided the basis for stable and consistent growth
in Colombia, the debt crisis of the early 1980s forced the Colombian gov-
ernment to implement a stabilization program in 1984. This included sig-
nificant reductions in spending and the devaluation of the Colombian peso.
However, urban import substitution was still the major trade policy focus.
A reprieve came in 1986–1987 with a small coffee bonanza. Then in 1988
coffee prices started falling again, forcing the Colombian government to
enact a comprehensive stabilization program.
In response to falling coffee prices, President Barco began a trade liber-
alization strategy in the late 1980s. But it was not until President Gaviria
 Economic development policies in Colombia  117
took office that the Colombian economy embraced trade liberalization as
its major economic policy. Gaviria eliminated most nontariff barriers and
dramatically reduced tariffs; implemented a program of financial liberaliza-
tion; eliminated the previous system of exchange rate controls; and permit-
ted free flows of foreign exchange and international capital. Furthermore,
incentives were implemented to attract foreign direct investment. Barco and
Gaviria’s plans to advance economic change, increase foreign capital and
privatization, were implemented in a context of fiscal austerity. Unfortu-
nately, “this opening process did not increase exports, create more employ-
ment or encourage substantial growth” (Tokatlián 1996, 132). In fact, U.S.
investment dropped from the late 1980s to the early 1990s and exports did
not materialize as expected.
It is worth noting that, in parallel, the Colombian government embarked
on a number of different policy experiments as a response to rural vio-
lence and low global coffee prices. During “La Violencia,” Colombia, like
many Latin American countries, enacted a fixed exchange rate and import-­
substitution policies to stimulate growth. The 1961 Ley Agraria (Law 135),
aimed to increase economic activity and reduce unequal distribution of the
“propiedad rústica.” However, it did not provide benefits for sharecroppers
and, as outlined in other chapters, the lands that were distributed were of
poor soil quality, limiting harvest to only a few years. This was further com-
plicated by poor market access for the crops. Finally, large landowners man-
aged to gain access to the fertile lands near the navigable rivers (LeGrand
1988).
Despite general efforts to increase agricultural productivity and bring new
lands into production, there was a general failure to incorporate new lands
into the economy (American Journal of Comparative Law 1968, 48). There
was also a push for more agricultural development in the 1960s, especially
as food prices rose while exports fell. Some estimates from the 1960s sug-
gested that Colombia needed to double its food supply to keep up with
population growth (Adams 1969, 535). The 1971 development plan called
for encouraging exports, especially non-traditional goods, and increasing
agricultural productivity and the real income of farmers (ACEP 1974, 143),
in addition to a major push for urban development.
Other reforms were attempted in the 1990s to restructure the Colombian
economy to restart growth. This restructuring generated optimism in rural
areas. The anticipation that deregulation and export promotion would pro-
duce growth that import substitution had stopped delivering raised hopes in
rural areas. While the 1990s export promotion polices and global focus did
increase primary product exports (Cerrutti and Bertoncello 2003, 10), they
also unwittingly created an environment favorable to the cultivation and
export of illicit goods.
The 1993 Ley Agraria (Law 101) provided credit subsidies to small farm-
ers, but it also created the Incentive to Rural Capitalization (ICR) that pro-
vided as much as 40 percent of the cost of investments. Larger farmers,
118  Jennifer S. Holmes et al.
who were ineligible for other programs, could take advantage of the ICR
(Guterman 2007, 21). Jaramillo (2002) concludes that the standard of living
for rural populations increased from 1990 to 1997. However, this improve-
ment in the living standards of rural farmers was hardly uniform. Devel-
opment plans favored commercial agriculture at the expense of the small
shareholder agriculture (Martín Taborda 2002, 119). Commercial agricul-
ture, such as flower growing, was in general found in developed regions near
larger cities such as Medellín and Bogotá. Hence, the overall increase was
likely a result of improved labor opportunities in the commercial agriculture
sector. The overall effect was to improve conditions for large farms and the
agro-industry, while resulting in very few benefits to small farmers (Guter-
man 2007). This unequal distribution of assets made other crops, such as
coca, a more reliable alternative to traditional, legal crops especially in dis-
tant rural regions of the country.
Colombian businesses have traditionally operated under the expecta-
tions of high short-term profits. Ocampo and Villar (1994, 61) have argued
that successful booms in the nineteenth century, as well as segmented mar-
kets created by limitations on internal transportation, created incentives
for entrepreneurial activity with little long-term investment. These entre-
preneurs invested in commerce that yielded high short-term profits and a
quick turnover. This mentality was labeled “production-speculation.” This
business style was reinforced in the 1970s by the capital market liberal-
ization of 1970–1974 and the coffee boom of 1975–1978 (Thoumi 1995).
The quick profit mentality was critical to the spread of coca in Colombia,
given that other agricultural goods were experiencing price decreases by
the late 1970s.

Before the coca explosion: trends

Demographic trends and colonization efforts
Typical Latin American migration patterns follow a rural-to-urban pat-
tern over the last few decades (Rodríguez Vignoli 2004; ECLAC 2007).
In Colombia, about one-third of the rural under-40-year-old population
in 1950 had migrated by 1964. However, Colombia has also experienced
flows to areas of new colonization or otherwise relatively unpopulated land.
New insecticides and major road construction opened up new areas for col-
onization after World War II (Adams 1969, 528). In fact, in the 1960s and
1970s, more than a third of the internal migration was to rural areas, not
urban areas. Most of these migrants were men with the lowest educational
attainment (Martine 1975). It was not until the mid-1980s that Colombia
shifted from a predominantly rural to an urban economy (Thoumi 1995).
In Caquetá, the migration was based on the cocaine bonanza of the 1970s
(Centro Nacional de Memoria Histórica 2013). This migration was more
than just seasonal migration, as needed for legal agriculture. In contrast,
 Economic development policies in Colombia  119
Table 5.1  Population trends and fraction of previous estimates (percentage)

Dept. 1960 % 1970 % 1980 % 1990 % 2000

Caquetá 81,488 182 148,318 165 244,072 10 342,285 122 418,998

Guaviare — — 28,681 288 82,581 142 117,189
Meta 126,249 176 222,804 175 389,781 149 581,135 121 700,506
Putumayo 42,650 148 63,313 117 137,650 185 254,699 131 332,434
National 16,480,383 133 22,061,215 116 27,737,900 123 34,261,565 118 40,403,958

Source: Compiled from multiple DANE, Boletín de Estadística, Bogotá http://biblioteca.dane.

gov.co/biblioteca/categories/90/

migration into Guaviare and Putumayo was originally based on long-term

industries such as fishing and agriculture, and petroleum extraction.
All four departments grew faster than the national average from 1950
through 2000, due to colonization efforts, especially in Caquetá (Table 5.1).
The map (Map 5.1) of population change in Colombia illustrates how dra-
matic the population change was in these departments.
These four departments demonstrate the “sizable” rural–rural migration
and smaller urban–rural flows toward the “empty quarters,” where peasants
and other migrants have squatted on unoccupied land” (Thoumi 2003, 86).
These are isolated regions far away from the markets that would be neces-
sary to support legal economic activity. There are also significant hindrances
to creating a reliable transportation network and a weak, if not absent, state.
These are longstanding challenges that have not been successfully addressed.
By the 1980s and 1990s, it is estimated that 90 percent of the coca in Colom-
bia was cultivated in colonization areas (Osorio Perez 2000, 2). Coca cultiva-
tion for the global cocaine market is thought to have attracted smallholders
to lowland Amazonian forests (Hecht 1993; Kaimowitz 1997; Young 1996).
For example, between 1985 and 1993 the population in Caquetá increased
by 39 percent and it was the growing coca trade that was the attraction
(UNODC 2015, 82). However, there was already a stock of colonists in situ
who were looking for a successful cash crop as well.
Government-assisted migration in the 1960s renewed the colonization
front, but did not improve local infrastructure (Arcila et al. 1999). The pop-
ulation in these regions is dispersed and has seasonal movements. The dis-
parity between the general population trends and preschool enrollment was
extreme. In fact, the trends are opposite. The following four tables describe
preschool and primary school enrollment and tally preschool and primary
school teachers, as a rough approximation of services and infrastructure avail-
able to the new residents. Tables 5.2 and 5.3 focus on rural preschool enroll-
ment and the number of preschool teachers. As the population increased,
rural preschool enrollment actually dropped in all departments other than
Guaviare (Tables 5.4 and 5.5). With the same exception, the number of rural
preschool teachers dropped as well. While these departments saw an influx
of residents, government services did not increase to match local needs.
120  Jennifer S. Holmes et al.

Map 5.1  Colombian population growth from 1951 to 2005 (percentage)

Table 5.2  Rural preschool enrollment

Department/Year 1977 1980 1984 1987

Caquetá 424 208 144 178

Guaviare 5 11 89
Meta 421 59 35 148
Putumayo 127 30 78 29
National 12,934 8,915 12,958 21,463

Source: DANE Boletín de Estadística 434 May 1989 Table 1A, p. 179
 Economic development policies in Colombia  121
Table 5.3  Rural preschool teachers

Department/Year 1977 1981 1984 1987

Caquetá 17 7 8 12
Guaviare 0 0 3 5
Meta 20 2 4 8
Putumayo 3 0 4 2
National 547 508 622 1,013

Source: DANE Boletín de Estadística 434 May 1989 Table 2A, p. 186

Table 5.4  Rural primary school enrollment

Department/Year 1978 1981 1984 1987

Caquetá 19,564 19,408 11,933 15,166

Guaviare 0 2,933 2,934 3,611
Meta 14,483 15,271 22,761 24,788
Putumayo 11,574 5,581 11,498 2,700
National 1,234,218 1,266,966 1,210,655 1,212,493

Source: DANE Boletín de Estadística 434 May 1989,Table 15A, p. 241

Table 5.5  Rural primary school teachers

Department/Year 1978 1981 1984 1987

Caquetá 683 757 465 614

Guaviare 0 143 161 168
Meta 460 474 780 874
Putumayo 469 236 492 134
National 39,496 42,838 42,818 47,717

Source: DANE Boletín de Estadística 434 May 1989 Table 16A, p. 249

Despite evidence that mostly men participated in rural-to-rural area

migration, these trends are still revealing. They highlight the extent to which
the typical government services had not kept pace with the population
growth, nor were designed to support families. There are slightly different
trends with primary school enrollment and the number of teachers. Primary
school enrollment decreased in Caquetá and Putumayo in this time period,
but increased in Guaviare and Meta.
The number of teachers was slightly lower in Caquetá in this period,
almost doubled in Meta, and declined by almost three-quarters in Putu-
mayo. The trends in education could be a reflection of the infiltration of
the region by the FARC and increased violence that led to displacement of
families. These school enrollment and staffing data approximate the lack of
122  Jennifer S. Holmes et al.
infrastructure and social services that are needed to support a stable popu-
lation that intends to invest in the local economy.

Regional economic development in four departments

Difficulties in measurement arise from the agglomeration of the eventual
nuevos departamentos (Guainía, Vichada, Vaupés, Guaviare, and Putu-
mayo) into the general category of precincts and intendancies (intendencias
y comisarias) in the early period. Despite efforts to encourage the licit econ-
omy in these departments since the 1960s, the licit sector remained very
small compared to the rest of the country (Table 5.6). We examine sectoral
GDP for these departments, when possible.
Examining national trends in illegal production illustrates how these
departments are also characterized by greater volatility in growth rates
due to the impact of the illicit economy. Figure 5.1 illustrates deviation in
growth rate of GDP in constant Colombian pesos (COP) with varying base
years that correspond to including or excluding the illicit economy. Base
year 1975 and 2005 figures do not include estimates of coca production or
other illicit activities, whereas 1994 base year figures do. Examining the data
for overlapping base years reveals deviations between data that do (base
year 1994) and do not (base years 1975 and 2005) include illegal markets.
For the 2005 base year data, DANE decided to estimate separately illicit
activity and not report it in GDP, as in Figure 5.2 (DANE: Dirección de
Síntesis y Cuentas Nacionales (DSCN) Grupo de retropolación, May 2013).
Figure 5.1 illustrates the deviation in growth rates between years with data
in 1975 and 2005 base years compared to 1994 base years to approximate
the impact of the illicit economy. The data in Figure 5.2 are divided into
agriculture and industrial production related to the illicit economy. The
data are DANE estimates for the country as a whole regarding the illicit
economy. “Agricultural” corresponds to the cultivation of the crops, while
“industrial” represents the processing of the crops into cocaine and heroin,
for example. The volatility of the economy for the country as a whole is
­significantly less than for the departments studied here (Figure 5.1). Vola-

Table 5.6  GDP in constant 1975 pesos (1980–1989)

Department/Year 1980 1983 1986 1989

Caquetá 3,032 3,194 3,358 3,999

Meta 8,106 8,144 9,578 13,278
Precincts and 8,282 10,266 14,965 19,908
intendancies
National 525,765 551,380 621,781 705,068
Source: DANE Cuentas Regionales downloaded on 2017/12/09 www.dane.gov.co/index.php/
estadisticas-por-tema/cuentas-nacionales/cuentas-nacionales-departamentales
 Economic development policies in Colombia  123

Figure 5.1 GDP growth rate in constant dollars, including or excluding the illicit
economy of Columbia
Source: downloaded on 2017/12/09 www.dane.gov.co/index.php/estadisticas-por-tema/cuen-
tas-nacionales/cuentas-nacionales-departamentales

tility creates uncertainty that makes long-term investment in the land more
costly and lowers return on investment in alternative crops. Coca grows nat-
urally with minimal investment in the land and provides multiple harvests in
a year, making it an ideal crop in the face of certainty and vague land tenure
laws. Hence, economic volatility increases the likelihood that campesinos
would choose coca over crops that require long-term investments in land
and infrastructure.

Department case studies

Subsequent discussion of Meta is limited, as previous authors (e.g., UNODC
2010) and Chapter 6 (this volume) will examine it in detail. Guaviare and
Putumayo are nuevos departamentos, which were created in 1991. Guaviare
had a “comisaría” status, split off from Vaupés in 1977 before becoming
a department in 1991. Putumayo, Caquetá, and Guaviare are also among
the most hard-hit in terms of violence since 1990. Meta had less intense,
but still significant violence in this later time period. Both Putumayo and
124  Jennifer S. Holmes et al.

Figure 5.2  GDP estimates of the Colombian national illicit economy in constant COP
Source: SIMCI - Cálculos DANE

­ uaviare experienced a failed development plan in the mid-1990s—Pro-

G
grama de Desarrollo Alternativo (PDA)—due to low local administrative
capacity, few contacts with the local community, and the inability to differ-
entiate the stable local population from the cocalero and drug trafficking
population (Posada 1995, 7). The agricultural practices used by colonists in
Guaviare and Caquetá were not sustainable (Andrade 1988, 56). Both Gua-
viare and Putumayo have been colonization zones affected by the increased
power of large landholders simultaneous with the decrease of small agrarian
landholdings. The uneven balance of power started the mass expulsion of
campesinos and indigenous to wastelands in search of better living condi-
tions that were not available to them within the agricultural frontier (Pérez
Martínez 2004, 72). These departments have in common provincial farming
technology, migrant populations, land tenure systems that do support long
term investment on the part of small farmers, lack of government supported
infrastructure, and a strong guerrilla and paramilitary presence.

Guaviare
The early twentieth-century Guaviare economy was based on rubber, exotic
pelts, and fishing, although the economic activity was not stable. Guaviare is
 Economic development policies in Colombia  125
heavily protected (on paper if not in practice) for conservation (Nukak, Chirib-
iquete) and as an indigenous reserve (La Fuga). A wave of new residents arrived
in the 1950s as people fled the violence in neighboring departments. There was
a campaign to encourage people to move to El Retorno, with the intention of
reducing the population in the major urban centers. They used Radio Sutatenza
to publicize the program that included messages of “land without men for men
without land.” The campaign even included free airplane rides, courtesy of the
Colombian Air Force. It was part of a grand development project to modernize
the campesino economy. The goal of colonization in the 1960s was to pro-
mote cattle ranching and cacao plantations, in addition to an expanded fishing
industry. Due to campesino complaints, in 1969, 180,000 hectares of land were
removed from the forest reserve and opened up for development. In 1970, efforts
intensified with the integrated rural development plan (DRI), but the attempts
to increase campesino productivity through technology transfer, credit, and
technical assistance failed because of insufficient support and inclement condi-
tions (Del Cairo and Montenegro-Perini 2015). During 1973–1977, corn was
planted and more ranches established. New colonists arrived (Etter 1987, 2).
The municipality of San José was established in 1976, and in 1977 Guaviare
became a separate comisaría. Major legal agricultural activities were ranch-
ing and fishing, with later emphasis on developing plantain, cassava, cacao,
corn, rice, peach-palm oil, palm oil, moon fruit, copoazú, caimarona grape,
and other various tropical fruits and nuts (Cárdenas and Junguito 2009, 9).
Nevertheless, the coca boom dwarfed the growth of this licit economy (Gober-
nación de Guaviare 2016, 8).
Guaviare’s demographics can be described by five epochs. First, slow
growth until the 1950s. Second, growth from people fleeing the violence
from the civil war in other parts of the country. Third, directed colonization
through 1967. Fourth, the introduction of migrants attracted to the drug
trade and coca cultivation, the flow of migrants that is still present today,
and fifth, the advance of colonization to the south and the west (Gober-
nación de Guaviare 2011, 13). Most of the growth in the 1970s and early
1980s was driven by the attraction to the emergence of coca as a cash crop
(Gobernación de Guaviare 2011, 16).
The department is very isolated, with only one paved road connecting
the capital San José del Guaviare to Villavicencio, the department capital
of Meta, in the twenty-first century. However, even as late as the 1980s, the
department was accessible by road only during the three months of the dry
season. Main cities within Guaviare have to be accessed either through dirt
roads, or by air or river. The low employment in the area and the physi-
cal isolation from markets prevented the take-off of legal agriculture in the
department, and most residents were limited to subsistence agriculture until
the cultivation of marijuana and, later, coca (Molano 1987).
In the 1970s, during the boom of marijuana cultivation, known as the
“bonanza marimbera,” some drug traffickers settled in Guaviare. Around
1977, coca cultivation began (Etter 1987, 2). By the end of the decade, there
was such an oversupply of coca that prices fell, leading some people to leave
126  Jennifer S. Holmes et al.
Table 5.7  Agriculture, ranching, and fishing as sectoral GDP (in constant 1975
million pesos)

Department 1980 1983 1986 1989

Caquetá 1,583 1,642 1,800 2,233

Meta 4,150 3,427 3,814 5,007
Precincts and 4,082 4,704 4,705 5,963
intendancies
National 119,314 124,196 132,792 151,423
Source: DANE Boletín de Estadística 457, April 1991, Table 10, p. 216–222, DANE Boletín de
Estadística 463 Oct. 1991 Table 10, p. 212

Guaviare or turn to other crops or sources of income. In the 1980s, the

FARC first front moved in. They encouraged and supported coca cultiva-
tion. Local campesino investment in the licit activities decreased. When coca
prices fell in 1986 and 1987, more attention turned to ranching and logging
(Table 5.7). Attempts at eradication started in 1994, followed by a short-
lived agreement to suspend eradication for crops smaller than 3 hectares.
After this agreement was not honored, the campesinos did not receive com-
pensation or assistance to shift away from coca cultivation, and the FARC
increased recruitment and support for returning to coca cultivation (Posada
1995, 5). In 1997, marches and protests against fumigation—involving tens
of thousands of cocaleros—took place in Guaviare (Osorio Perez 2000, 12).

Putumayo
Putumayo was an intendencia prior to 1991. The department was one of the
early adopters of coca cultivation. It is estimated that the first leaves were
harvested in the region around 1978 (Ramírez 1998; Ramírez 2001). How-
ever, Maria Victoria Rivera (1994) dates the first commercial coca agron-
omies to 1983, setting a five-year gap between the start of coca’s growth
and its commercialization. The coca grown originally came from Peru, and
was eventually harvested across the entire department of Putumayo (Rivera
Flórez 2005). The patterns of cultivation were associated with availabil-
ity of goods and labor, mirroring routes of colonization along the rivers
(Fundación Alisos 2011, 38). Coca cultivation in the department increased
the amount of land used for agriculture, altered population patterns, and
increased the size of urban centers. Coca was not evenly dispersed, however,
and instead was concentrated in Bajo Putumayo, an area with a higher inci-
dence of violence.
Petroleum played a significant role in both the development of Putumayo
and the introduction of coca into the region. The petroleum expansion started
in this department in 1963 with the concession of one million ­hectares to
the Texas Petroleum Company. The expansion brought with it job openings
in oil, which increased migration towards Putumayo (González 1989, 193–
 Economic development policies in Colombia  127
194). Puerto Asís hosted Texaco and quickly became the migration hub of
the department. By 1964 Texaco had around 1,000 employees. These work-
ers received four times the salary of an agricultural worker in Putumayo,
making it a desirable job for workers in neighboring areas. Yet the petroleum
company was not the largest source of employment in the area; instead, col-
lateral activities such as bars and brothels filled this role (Domínguez 2005,
270–271). Because petroleum companies were legally American territory
and local workers could not live within this land, the employees established
communities surrounding Texaco. These settlements emerged quickly to
meet employee demands, and were not structured or organized in any way.
To meet population demands, other migrants initiated infrastructure like the
building of roads, warehouses, department stores, and the introduction of
larger transportation mechanisms. This growth also included illicit activities
such as brothels and cantinas. Even though there was no established plan-
ning in the city by the late 1960s, Puerto Asís started the construction of
roads to facilitate the entrance of heavy machinery to construct the Transan-
dino Pipeline. Eventually the settlements in Putumayo that boomed during
the petroleum “fever,” such as La Hormiga, Puerto Caicedo, and La Dorada,
became municipalities during the coca bonanza.
The increased flow of migration to Putumayo meant that it stopped
being an empty space west of the Amazon. Instead, what it experienced
was unprecedented growth. Between 1973 and 1985 there was a population
increase of 145 percent for Putumayo, while the rest of the country only
experienced a population growth of 31 percent. This growth was concen-
trated in the lower (Bajo) Putumayo, which was also the area where coca
cultivation started. During the same period, agricultural workers lacked
agricultural reform and were facing protectionist measures that lowered
the price of their potential products by 54 percent (de Rementeria 2001).
As a result, cultivation of legal crops such as plantain, corn, and cassava
decreased (Fundación Alisos 2011).
Bajo Putumayo experienced a population growth of 300 percent versus
the 145 percent of the entire department. Bajo became a region with a high
incidence of violence and the establishment of coca plantations. Once the
petroleum companies left the department, the only other experience the
workers had was with agriculture and coca became the crop of choice.
Both coca and petroleum colonization originated with a get-rich-quick or
boom mentality, and a transitory population that did not require long-term
investment in the region. This exploitive mentality limited the amount of
community organization and opened the possibility of producing illegal
products for higher incomes. Some researchers believe that people who pop-
ulated Putumayo were interested in colonizing and establishing agricultural
communities and were not simply looking for easy money (González 1998,
90; Ramírez 2001). Regardless of their original motivations, Putumayo
­experienced rapid and uncontrolled colonization and the search for a quick
income by at least some colonizers is unsurprising (González 1998, 90).
128  Jennifer S. Holmes et al.
Caquetá
Prior to 1991, and despite its size and history of colonization, Caquetá was
an intendencia. Caquetá is in the northeast region of the Amazon, and the
transition between the Andes and the Amazon. Caquetá’s extensive river
system connects not just parts of the department, but also the entire country,
making it an ideal hub for narco-trafficking (UNODC 2015, 33). Prior to the
coca boom, the main economic activities of Caquetá were small-scale min-
eral extraction, primary agricultural products, and fish and forest c­ ultivation
(UNODC 2015, 72). Large-scale investments were unsustainable, given the
poor transportation infrastructure and lack of access to credit and technol-
ogy to aid increased yields.
A fragile licit economy and political instability created conditions favor-
able for illicit production. While coca cultivation has been present in Caquetá
since the 1970s, when the FARC arrived it also became the center of cocaine
production and trafficking (UNODC 2015, 35). All 16 municipalities of
Caquetá were affected by coca cultivation, and changes in production were
directly related to the degree of FARC control over the region. In the 1990s,
the FARC charged campesinos a 10 percent tax on coca crops (UNODC
2015, 63). The instability and violence in the area made coca cultivation the
crop of choice. Coca could be harvested twice a year with minimal invest-
ment. In 1997, the paramilitaries infiltrated the department in an attempt
to take over the illegal drug trade. This was a period marked by heightened
violence and instability. During the 2000s, as state control increased, coca
planting declined (UNODC 2015, 36).
There is a history of crop substitution related to the price of coca and
regional stability. In 1989, as coca prices were falling, coca farmers
approached Jacinto Franzoi—a local priest known for publicly speaking
against coca and the FARC—for help in finding an alternative crop. The
farmers planted cacao, which takes three years to yield a harvest (Redacción
El Tiempo 2004). By 1992, when the cacao crop was to be harvested, the price
of coca began to rise and some farmers abandoned their cacao production,
while others sold their product to Franzoi, who paid above-market prices.
Eventually, all farmers abandoned cacao in favor of coca, importing virtu-
ally all food from Florencia, the capital (Redacción El Tiempo 2004). Along
with the military control of the region, interest in the cacao had returned by
2002. Improving on such past experiences, in 2007 the Colombian govern-
ment supported an alternative to coca cultivation: production of arowana,
a native aquarium fish in the region and in high demand in China (UNODC
2015, 51). Given robust demand and a variety of local investors, arowana
exports began in 2013 and have remained in place through 2017 (Quin-
tero 2017). Post-2011, plantain, rubber, sugarcane, cassava, and maize each
had more hectares cultivated than coca. Therefore, the success of alternative
development programs depends on both long-term government support for
infrastructure and security, as well as the demand for licit alternatives.
 Economic development policies in Colombia  129
Implications
An overview of three departments, Guaviare, Putumayo, and Caquetá
reveals that coca became the crop of choice because of the circumstances
and context in which farmers had to make decisions. These departments
have in common provincial farming technology, migrant populations, land
tenure systems that do not support long-term investment on the part of
small farmers, lack of government supported infrastructure, and strong
guerrilla and paramilitary presences. Migrant inflows matched the coca
bonanzas, and outflows of families followed the increase in violence. Coca
cultivation was more a response to the lack of viable economic alternatives
and inequalities rather than a decision to participate in the illicit economy.
However, by the 1990s, the establishment of the FARC and the infiltration
of the paramilitaries created an environment where the campesinos really
only had one choice and that was to cultivate coca. Given that coca requires
minimal investment in the land and the land tenure system for small farm-
ers was insecure in these departments, coca cultivation is not a surprising
choice. The lack of physical infrastructure, which would not support high
valued-added agriculture, coupled with an extensive river system that con-
nected major drug-trafficking routes, made these regions prime locations for
cocaine processing. In order for future alternative development plans to be
successful and sustainable in this region, the Colombian government will
have to invest not only in physical, but also social infrastructure creating an
environment that encourages local long-term investment. Without providing
support beyond security alone, the transition from illicit to licit economy
will remain difficult, tumultuous, and unlikely.

Note
1 We would like to thank all those in Colombia who have supported our fieldwork
over the last two decades.

References
ACEP (Asociación Colombiana para el Estudio de la Población). 1974. Población
de Colombia. Bogotá: Asociación Colombiana para el Estudio de la Población.
Adams, Dale W. 1969. Rural migration and agricultural development in Colom-
bia. Economic Development and Cultural Change 17, 4: 527–539.
American Journal of Comparative Law. 1968. Colombia: Agrarian Reform 438, 48.
Amin Gutiérrez de Piñeres, S. and M. Ferrantino. 2000. Export Dynamics and Eco-
nomic Growth in Latin America: A Comparative Perspective. London: Ashgate
Publishing.
Andrade, Angela. 1988. Desarrollo de los sistemas agrícolas tradicionales en la Ama-
zonia. Boletín Museo del Oro 21: 39–59.
Arcila, Oscar H., Gloria I. Gonzalez, and Carlos A. Salazar. 1999. Guaviare:
población y territorio. Bogotá: Tercer Mundo Editores.
130  Jennifer S. Holmes et al.
Bula, Mayra. 1988. Antecedentes en las Culturas Nativas. In Historia de la Droga-
diccion en Colombia, compiled by A. Perez, 9–28. Bogotá: Tercer Mundo-Edi-
ciones Uniandes.
Cárdenas, Mauricio and Roberto Junguito. 2009. Nueva introducción a la economía
colombiana. Bogotá: Fundación para la Educación Superior y el Desarrollo
(Fedesarrollo).
Centro Nacional de Memoria Histórica. 2013. Caquetá: conflicto y Memoria.
Bogotá: Imprenta Nacional de Colombia.
Cerrutti, Marcela and Rodolfo Bertoncello. 2003. Urbanization and Internal Migra-
tion Patterns in Latin America, 1–24. Argentina: Centro de Estudios de Población.
DANE: Dirección de Síntesis y Cuentas Nacionales (DSCN). 2013. Grupo de ret-
ropolación. Gobierno de Colombia. Bogotá, Colombia. www.dane.gov.co/
index.php/acerca-del-dane/informacion-institucional/organigrama/direccion-
de-sintesis-y-cuentas-nacionales-dscn
de Rementeria, Iban. 2001. La Guerra de las Drogas. Bogotá: Editorial Planeta
Colombiano.
Del Cairo, Carlos and Iván Montenegro-Perini. 2015. Spaces, peasants and environ-
mental subjectivities in guaviare. Memoria y Sociedad 19, no. 39: 49–71.
Domínguez, Camilo. 2005. Amazonia colombiana, economía y poblamiento. Bogotá:
Universidad Externado de Colombia.
ECLAC (Economic Commission for Latin America and the Caribbean). 2007. Social
Panorama of Latin America 2007. Santiago, Chile: United Nations.
Etter, Andrés. 1987. Seguimiento de la colonización en la Amazonia Colombiana el
Caso de San Jose del Guaviare (1973–1986). Bogotá: Unidad de Levantamientos
Rurales, Subdirección de Docencia e Investigación (CIAF).
Fundación Alisos. 2011. Retos para un desarrollo sostenible: transformaciones en la
Amazonia colombiana. Bogotá: Fundación Alisos. http://pdf.usaid.gov/pdf_docs/
PA00KK7H.pdf
Garcia Garcia, Jorge, and Gabriel Montes Llamas. 1988. Coffee Boom, Government
Expenditure, and Agricultural Prices: The Colombian Experience. Research
Report no.68. Washington, DC: International Food Policy Research Institute.
Gobernación del Guaviare. 2011. Plan de Desarrollo Departamental Por Un Gua-
viare Incluyente 2010–11. San José del Guaviare, Colombia: Departamento del
Guaviare.
———. 2016. Plan de Desarrollo Departamental 2016–2019. Guaviare Paz y Desar-
rollo Social. San José del Guaviare, Colombia: Departamento del Guaviare.
González, José Jairo. 1989. Caminos de Oriente. Aspectos de la colonización con-
temporánea del Oriente colombiano. In Revista Controversia. Un país en con-
strucción. Poblamiento, problema agrario y conflicto social. Número 151–152.
Bogotá: CINEP.
———. 1998. Amazonía Colombiana. Espacio y Sociedad. Bogotá: CINEP.
Gootenberg, Paul. 2008. Andean Cocaine: The Making of a Global Drug. Chapel
Hill, NC: The University of North Carolina Press.
Guterman, Lisa. 2007. Distortions to Agricultural Incentives in Colombia. Agricultural
Distortions Working Paper 14, December 2007. Washington, DC: World Bank.
Jaramillo, Carlos Felipe. 2002. Crisis y transformación de la agricultura colombiana,
1990-2000. Bogotá: Banco de la República, Fondo de Cultura Económica.
Hecht, Susanna B. 1993. The logic of livestock and deforestation in Amazonia. Bio-
science 43: 687–695.
 Economic development policies in Colombia  131
Kaimowitz, David. 1997. Factors determining low deforestation: the Bolivian Ama-
zon. Ambio 26: 536–540.
LeGrand, Catherine. 1988. Colonización y Protesta Campesina en Colombia (1850–
1950). Bogotá: Universidad Nacional.
Martín Taborda, Jorge Iván. 2002. Colonización y Recomposición Campesina en
Guaviare 1960–1998. Memoria y Sociedad 7, 13: 117–158.
Martine, George. 1975. Volume, characteristics and consequences of internal migra-
tion in Colombia. Demography 12, 2: 193–208.
Molano, Alfredo. 1987. Selva adentro. Una historia oral de la colonización del Gua-
viare. Bogotá, Colombia: El Áncora Editores.
Ocampo, José A., and Leonardo Villar. 1994. Colombian Manufactured Exports,
1967–1991. In Manufacturing for Export in the Developing World: Problems
and Possibilities, G.K. Helleiner, ed., 54–98. New York: Routledge.
Osorio Perez, Edilma. 2000. Viejas y nuevas ruralidades a partir de la migraciones
internas; algunas reflexiones desde la realidad colombiana. Bogotá: Pontificia
Universidad Javeriana. Seminario Internacional. http://bibliotecavirtual.clacso.
org.ar/ar/libros/rjave/mesa1/osorio.pdf
Pérez Martínez, Manuel Enrique. 2004. La conformación territorial en Colombia:
entre el conflicto, el desarrollo y el destierro. Cuadernos de Desarrollo Rural 5:
61–90.
Posada, Alberto Reyes. 1995. La erradicación de cultivos: un laberinto.  Análisis
politico 24: 66–72.
Quintero, Andres. 2017. Peces de Caquetá nadan en varias metrópolis de Asia. Portafo-
lio, August 8, 2017. www.portafolio.co/negocios/caqueta-conquista-exportaciones-
de-peces-en-asia-508513
Ramírez, María Clemencia. 2001. Entre el estado y la guerrilla: identidad y ciu-
dadanía en el movimiento de los campesinos cocaleros del Putumayo. Bogotá:
Instituto Colombiano de Antropologia e Historia (ICAHN), Colciencias.
Ramírez, Roberto. 1998. Conflictos sociales en el Putumayo. In Conflictos regio-
nales: Amazonia y Orinoquia, J.J. González, ed., 71–130. Bogotá: Iepri, Fescol.
Redacción El Tiempo. 2004. En El Remolino de La Coca y El Cacao. El Tiempo (Bogotá)
December 5, 2004. www.eltiempo.com/archivo/documento/MAM-1523460
Rivera, María Victoria. 1994. Los Proyectos de Sustitución de Cultivos en los Depar-
tamentos de Caquetá, Guaviare, y Putumayo. In Drogas, Poder y Región en
Colombia, R. Vargas, ed., 233–282. Bogotá: Cinep.
Rivera Flórez, Guillermo. 2005. Cultivos de Coca, Conflicto y Deslegitimacion del
Estado en el Putumayo. Bogotá: Universidad Externado de Colombia.
Rodríguez Vignoli, Jorge. 2004. Migración interna en América Latina y el Caribe:
estudio regional del período 1980–2000. Serie población y desarrollo No. 50.
Santiago, Chile: Naciones Unidas Comisión Económica para América Latina y
el Caribe (CEPAL).
Rubio, María I. 1988. Siglo XX: 1900–1960. In Historia de la Drogadicción en
Colombia, A. Pérez G., ed. Bogotá: Tercer Mundo, Ediciones Unidades.
Schultz, T. Paul. 1969. Population Growth and Internal Migration in Colombia.
Santa Monica, CA: Rand Memorandum 5765.
Thoumi, Francisco E. 1995. Political Economy & Illegal Drugs in Colombia. Studies
on the Impact of the Illegal Drug Trade. Boulder, CO: Lynne Rienner Publishers.
———. 2003. Illegal Drugs, Economy, and Society in the Andes. Washington, DC:
Woodrow Wilson Center Press.
132  Jennifer S. Holmes et al.
Tokatlián, Juan Gabriel. 1996. Drug Summitry: A Colombian Perspective. In Drug
Trafficking in the Americas, B.M. Bagley and W.O. Walker, eds., 131–157. Miami:
University of Miami, North–South Center Press.
UNODC (United Nations Office on Drug and Crime). 2010. Transformación socio-
económica y biofísica asociadas con los cultivos ilícitos en la region Sur del
Meta-Guaviare 1990–2009. Bogotá: UNODC, Embajada de la Republica Fed-
eral Alemana, and Ministerio de Defensa Republica de Colombia.
———. 2015. Caracterización Regional de la problemática asociada a las drogas
ilícitas en el departamento Caquetá. Bogotá: UNODC, Ministerio de Justicia y
del Derecho de Colombia. https://www.odc.gov.co/Portals/1/publicaciones/pdf/
oferta/caracterizacion/RE0629JUN2015-caqueta.pdf
Worldmark Encyclopedia of the Nations. 1984. Detroit, MI: Gale Research, Allen.
Young, K.R. 1996. Threats to biological diversity caused by coca/cocaine deforesta-
tion. Environmental Conservation 23: 7–15.
6 The making of a coca frontier
The case of Ariari, Colombia
María-Clara Torres

On December 21st, 1961, representative Anibal “El Tuerto” Prado from the
left-wing Revolutionary Liberal Movement (MRL) delivered a speech as the
Colombian land reform bill was being voted on in Congress. He admon-
ished that an agrarian reform could have unforeseen consequences by rais-
ing expectations that could not be fulfilled:

This bill is a statute to encourage colonization and in no way will it

solve the vexing problem of the distribution of property. Honorable
Congressmen, the day when you ask yourselves why you failed, that
day, we will be back and our numbers will be in the thousands.1

This chapter, following on the last contextualizing chapter, continues to

grasp how world market demand for illicit cocaine produced such a rapid
and dynamic response in Colombia, a country with few historical roots
in indigenous coca culture. Although the plant has been grown since pre-­
Hispanic times by some indigenous groups, the massive spread of coca fields
and its linkage to cocaine production is a dramatically recent phenomenon.
Compared to other Andean nations, Colombia’s indigenous population is
small, 3 percent according to the latest census (Departamento Administra-
tivo Nacional de Estadística [DANE] 2007). This meant that the traditional
use of coca had a limited impact on highland, and now urban, Colombian
society as a whole. The shift of the illicit capitalist economy to Colombia
meant a key historical change in the global cocaine commodity chain in the
late 1980s and 1990s. For the first time, Andean coca capitalism was sepa-
rated from indigenous coca traditions (Gootenberg 2008).
This chapter explores how the nation’s long unresolved agrarian prob-
lems, the legacies of Cold War agrarian reforms, and counterinsurgency cam-
paigns interacted with small-scale frontier colonist economies and cultures
to produce a specific regional path to coca-cocaine in Colombia. It delves
into the historical trajectory of coca-cocaine in the Ariari region, part of the
Department of Meta, from the late 1950s to the 1980s. Meta is the fourth
of Colombia’s early coca poles contextualized in Chapter 5. The chapter
analyzes how grassroots communities living in the frontier enclosed by the
134  María-Clara Torres
rivers Ariari, Guayabero, Duda, and Güejar began their involvement with
coca and how this regional site became a major drug hotspot by the 1980s.
First, the complex interplay between U.S. Cold War development policies
and Colombian politics from the late 1950s to the 1980s is crucial to under-
stand the making of a coca frontier in Ariari. Launched in 1961 by the Ken-
nedy administration as a hemispheric policy, the Alliance for Progress (AFP)
promoted political and economic modernization in order to foster develop-
ment and stable middle classes “inoculated” against the Cuban Communist
revolution (Grandin 2004). The communist threat persuaded the United
States government to support Latin American social reforms it had previ-
ously opposed (Grandin 2004; Winn 1992). A series of agrarian reforms
took place in Bolivia, Peru, and Colombia at different times throughout
the Cold War. Yet, agrarian reforms were implemented unequally across
the Andean region as domestic actors embraced, adjusted, and reformu-
lated Cold War political objectives and ideologies to fit their own purposes
(Mayer 2009; Morales 1989; Sanabria 1993; Zamosc 1986). The agrarian
reforms therefore unfolded in their own ways in each country.
In Colombia, only a portion of the national elites was ready to adopt the
AFP’s premises and goals. For the National Front Coalition, an agrarian
reform seemed necessary to pacify and restore political control over the
rural population after the civil war between Liberals and Conservatives,
known as La Violencia (1948–1958). During that period, rural folk faced
off against each other under the elite’s complicit and even encouraging gaze.
The civil war’s death toll, never confirmed, may have reached between two
hundred thousand and four hundred thousand casualties and left eight hun-
dred thousand homeless and landless (Henderson 1985; Oquist 1980; Ortiz
1985; Roldán 2002; Sánchez and Meertens 2001; Uribe 1990). Between
1958 and 1974, the Liberal and Conservative ruling class enacted the
National Front, a power-sharing coalition between mainstream Liberals
and Conservatives, meant to eliminate partisan violence. Such a coalition
guaranteed presidential alternation and bureaucratic parity in the legisla-
tive, judiciary, and executive branches. The National Front elites embraced
the AFP’s agrarian reform in their own terms, as a means to rebuild the
clientele-based power of these two traditional parties in the countryside
(Zamosc 1986).
Yet, the Colombian bipartisan ruling class conceived colonization of the
frontier lands as a remedy to the heavy toll of rural dispossession and dis-
placement resulting from the civil war, while avoiding the political prob-
lem of expropriation, particularly in the nation’s political power base in the
Andean and inter-Andean core. State colonization of the tropical lowlands
was a domestic experiment that preceded the hemispheric reformist move-
ment. Between 1958 and 1961, the Rehabilitation Program fostered in Ari-
ari a project of “direct” migration, including land allocation, credits, and
tertiary road building for La Violencia’s victims. This domestic experiment
set the limits and scope for the upcoming AFP’s agrarian reform.
 The making of a coca frontier  135
Second, this chapter delves into the impact of Cold War counterinsur-
gency warfare on the emergence of coca-cocaine production in Colombia.
From the 1960s to the 1980s, modernizing reforms alternated with epi-
sodes of brutal state repression in the hemisphere. According to historian
Michael Latham (2011), U.S. policymakers believed that counterinsurgency
would create the stability necessary for economic modernization. Yet, as
he himself notes, counterinsurgency was not simply imposed from outside,
as “[domestic] leaders found a ready vocabulary with which to legitimate
their authority, identify threats, and demand increased U.S. economic and
security assistance” (Latham 2011, 124). Colombia’s long unresolved
agrarian problems, exacerbated by La Violencia, were caught in global
Cold War tensions. In the aftermath of the civil war, in 1964–1965, the
last rural enclaves of peasant self-defense were bombarded with napalm
by the Colombian army in an operation sponsored by the U.S. govern-
ment in the southern provinces of Colombia, including specific sites of
the area enclosed by the rivers Ariari, Duda, and Guayabero in the Meta
Department. Some Conservative and Liberal party elites justified this mili-
tary operation by decrying these nuclei of armed peasants as “independent
republics.” As a result, peasant self-defense groups transformed themselves
into the communist FARC guerrilla by the mid-1960s (González 1992;
Karl 2017). This well-organized armed group, joined by thousands of fron-
tier colonists, became an intrinsic part of the regional identity. Since the
1980s, this insurgency has played a central role in the spread of commercial
coca, as illicit crops became an increasingly strategic funding source in a
war against the Colombian establishment. With the spread of coca crops
in the mid-1980s, the FARC reinforced its essential role as a guardian of
local authority by setting prices and sometimes shielding the small coca
producer from exploitation by drug traffickers (Gutiérrez 2005; Jaramillo
et al. 1989). Thus, state coercion helped shape the very subversive dangers
it hoped to avoid.
Third, this chapter examines the contentious historical processes by
which local rural actors situated at the base of the global cocaine com-
modity chain, created and reflected evolving meanings of their livelihoods.
The (illicit) agro export boom not only produced marketable goods, but
also collective identities. Destitute and displaced frontier settlers created
a mestizo and colono coca culture. Furthermore, the logic that a substan-
tial portion of the interior rural populations could be transferred to public
lands on the frontier with minimal social conflict and without recourse
to an effective agrarian reform proved illusory in Ariari. Instead, a mov-
ing frontier opened up, pushed, and eventually transformed into a leading
drug hotspot. This case study shows how Colombia’s transformation into
a major coca-cocaine country rose from the ground up. Today’s small coca
producers, numbering in the ten thousands, are the historical result of the
unresolved legacy of the rural tensions that unfolded between the 1950s
and the late 1980s.
136  María-Clara Torres
The rehabilitation program in Ariari (1958–1961)
State colonization projects in Ariari preceded the U.S. hemispheric reform
movement. In 1958, prior to the Alliance for Progress-inspired Agrarian
Reform, the former General Secretary of the Organization of American
States, Colombian President Alberto Lleras Camargo, created the Special
Rehabilitation Commission. This program was an outcome of what historian
Robert Karl (2017) denominated the “Creole Peace” between Liberals and
Conservatives. After the bloody years of the civil war known as La Violen-
cia, the project “presented people an opportunity to rehabilitate themselves,
their regions, and their nation” (Karl 2017, 83). The Rehabilitation Program
included two essential components: the advancement of development and
the furtherance of pacification. On the one hand, its designers envisioned
expanding the state’s territorial reach to the areas most severely affected
by the warfare, through the construction of physical infrastructure and the
opening of roads (Karl 2017). On the other hand, its architects conceived
achieving peace by offering an amnesty to those who had committed crimes
during the armed struggle. They also anticipated the return of the civil war’s
victims uprooted and dispossessed of their rural property, but they promptly
encouraged the alternative of resettlement in Colombia’s internal frontiers,
if the option of repatriation failed (Karl 2017; Sánchez 1988).2
The Ariari region of the Department of Meta, was considered one of
those internal frontiers, where victims could be resettled if repatriation
to the Andean heartland failed. The Rehabilitation Program thus became
the first National Front colonization experiment in the region of Ariari.
Yet migration here was not limited to the state’s enterprise, as rural folk
had flocked spontaneously to Ariari during and after the civil war of the
1950s. They were attracted to this region by the prospect of acquiring land,
refuge, and protection. When the Special Rehabilitation Commission was
put in place, blue-Conservative, red-Liberal and Communist territories had
already divided the region, as victims and fighters had settled according to
political party affiliation throughout the 1950s. The upper course of the
Ariari River was mostly a Conservative stronghold. In my interviews, I
was told that one of the rural hamlets, Guamal, became infamously known
as “Puerto Chamizo” (Wooden Slab Port), for being a refuge destination
for “pájaros” (birds), as the paid thugs of the Conservative landowners
were dubbed. Liberal guerrillas dominated the middle course of the Ariari
River, including Nueva Granada, San Martín, San Juan de Arama, and
San Antonio del Ariari. The Communist peasants settled some hundred
kilometers to the west, enclosed by the Amazonian rainforest, the savannas
and the steep topography of the Macarena National Park.3 Incursions into
areas deemed within the enemy’s sphere of influence could be fatal. The
aforementioned different political colors had already tinted, organized,
and divided Ariari by the time the program designed in Bogotá was imple-
mented in the region.
 The making of a coca frontier  137

Map 6.1  Colombian cordilleras and political division

Although the incidents related to La Violencia did not end in the coun-
tryside until the mid-1960s, the enactment of the National Front in 1958
allowed for moments of respite in the region. In the late 1950s, Ariari’s Con-
servative and Liberal commanders reached an agreement, inconceivable until
then, to handle the mass arrivals. They established a safe corridor that could
be used at designated times, providing the migrants safe passage through
the Conservative territories, taking them to Liberal-controlled zones. I inter-
viewed Alfonso Castaño Arcila, who recalled that his family was stranded
for a couple of days at the outer fringes of the blue-Conservative zone, until
138  María-Clara Torres

Map 6.2  Meta Department and main western rivers

they were finally authorized to use the corridor to travel to the red-Liberal
hamlet, known as San Antonio del Ariari (today Fuentedeoro).
In 1959, the Special Rehabilitation Commission selected two Ariari
River Basin hamlets, which were part of the recently created municipality
of Nueva Granada. Liberal President Lleras Camargo prioritized these two
mushrooming villages, as they were positioned at the spearhead of the Lib-
eral frontier, hundred of miles away from the influence of the Communist
settlements (Londoño 1989). In the early stages of the National Front, the
Liberal intelligentsia of Bogotá viewed the far-flung Communist colonies as
 The making of a coca frontier  139
more menacing than the Conservative bands. The Liberals increasingly per-
ceived these remaining pockets of armed fighters as a threat to the bipartisan
pacification project.4
Furthermore, the Rehabilitation Commission gave authority to the Agrar-
ian, Industrial and Mining Bank (Caja de Crédito Agrario, Industrial y Min-
ero) to implement the colonization project. This process involved selecting
the families who were to be sent to Canaguaro and Avichure, choosing,
and distributing plots of public lands, providing loans, offering technical
assistance, and building tertiary roads. Some of the early 1960s settlers that
I interviewed in the small hamlet of Canaguaro were originally from the cof-
fee region, specifically the Viejo Caldas Department,5 one of the areas most
severely disrupted by the civil war. Although they themselves had not been
state-­directed settlers, some of their family members were. They recalled
that, in the late 1950s, state-sponsored radio announcements frequently
interrupted the regularly scheduled broadcast, inviting people from the cof-
fee hillsides to migrate and “colonize” Colombia’s tropical lowlands.
The Rehabilitation Program established the Agrarian Bank and the
Comisariato. These were often the sole official agencies that provided agri-
cultural services to Ariari’s remote areas in the late 1950s and early 1960s.
Yet the local population perceived the role of these two agencies in the com-
munity very differently. On the one hand, the Agrarian Bank established
an office in the town of Nueva Granada and rural dwellers had to travel
hours to get there. Its loan program stirred doubts as to whether the agency
was there to protect the settlers’ interests or whether it was a threatening
bureaucracy that could deprive them of their lands if they failed to meet
their debt obligations. On the other hand, the Comisariato is perhaps one of
the official institutions of the Rehabilitation Program that settlers remember
with nostalgia. The Comisariato set up a warehouse and a store in the bur-
geoning hamlets of Canaguaro and Avichure. It played a key role in support-
ing the overall process of colonization, as its services were offered not only
to state-directed colonists but also to the hoards of spontaneous migrants,
who were also civil war’s victims. The Comisariato provided colono fami-
lies with low-cost food, tools, medications, fungicides, seeds, and fertilizers
(Hormaza 2016). But most importantly, it purchased the harvests at a stable
and predetermined price, usually higher than what the middlemen would
pay. The Comisariato offered protection from price fluctuation; in doing
so, it promoted the peasant production of rice and corn agriculture. This
grassroots official institution stood as a bulwark against the exploitation of
abusive middlemen, and protecting the incipient colono economy.
Furthermore, the Rehabilitation Program was committed to expand the
state’s territorial reach and to foster development through the construction
of physical infrastructure. Karl suggests that the program had a special
emphasis on road construction; this “would foster market and political inte-
gration, meeting the unending local clamor for roads while also connecting
the capital with centers of violence” (2017, 82). In Ariari, these aspirations
140  María-Clara Torres
materialized in the construction of the Guillermo León Valencia6 bridge and
in the construction of a network of tertiary roads within the Liberal-held
territory.7 New local roads connected the village of Nueva Granada—which
was already becoming the center of Ariari—to the new hamlets of the river
basin. In addition, public work construction created jobs for newcomers.
A spontaneous settler still living in Canaguaro, Gustavo Jaramillo, recalled
that he had his first job experience opening roads for the Rehabilitation
Program, in which he learned skills that he later used to work his way up
through the blue-collar ladder and was finally appointed topographer for
Caminos Vecinales, the official agency in charge of building tertiary roads
in rural Colombia.8
Yet, the Rehabilitation Program also intensified the fight over the posses-
sion of frontier land. The policy designed in Bogotá neglected the fact that
colono families had occupied considerable tracts of land during and after La
Violencia. The settlers had defined boundary lines between adjoining neigh-
bors; some had even purchased the plots from a previous squatter including
co-signed notes. They considered those notes as precious, regardless of the
fact that they were not legally binding. The national state was adjudicating
lands that were fallow only in the official records; in doing so, it was dis-
rupting informal local agreements.
Likewise, tensions between directed and spontaneous colonists arose, as
the latter were not eligible for some of the benefits offered by the state col-
onization project, particularly the allocation of 50 hectares per family and
bank loans to bring the land into production. An early 1960s female colonist
I interviewed in Canaguaro narrated the painstaking efforts made by those
families who did not get any of these government benefits and still managed
to buy a small property and remain in the area. She stressed:

When I arrived here, there was mud everywhere, mud and hollows left
by the animals. Many mules carried rice and corn, which were the food
staples most people produced here. My husband bought a piece of land
with the money that he earned from working. And we still have that
land. My husband passed away many years ago and neither my chil-
dren nor I’ve wanted to sell it, even though we have only twenty-two
hectares.

This woman could be counted among the successful group of spontaneous

pioneers, who became proprietors and remained in the area despite living in
proximity to waves of violence since the 1960s. She blamed directed colo-
nists for their irresponsibility and for not taking advantage of the opportuni-
ties offered by the government: “they accepted the loans [from the Agrarian
Bank] and drank it in beer. When they had spent all their money, they left
and returned to the cities,” she explained.
Historian Robert Karl suggests that the Rehabilitation Program had a
similar outcome in the neighboring Department of Tolima. By 1961, the
 The making of a coca frontier  141
Agrarian Bank went bankrupt from high default rates. Rural folk who had
received loans from the program refused to make payments as they believed
the money was some kind of reparation for the moral and material harm
suffered during La Violencia: “Though Colombian law made no formal
provision for the payment of reparation to victims—posits Karl—a size-
able number of Rehabilitación loan recipients proceeded as they believed
was just, treating government assistance as a de facto form of reparation”
(2017, 112).
In my fieldwork in Canaguaro in 2016, I could not find any land recipient
of the Rehabilitation Program still living there, as they all had abandoned
their parcels many years ago. Most of them returned to the urban interior,
I was told. Early 1960s Canaguaro spontaneous settlers recalled that this
brief state experiment had failed, as many directed colonists were impov-
erished urban dwellers from the Andean heartland of the country, such as
shoemakers, barbers, and shopkeepers, who had neither the endurance nor
the skills to cope and survive in the tropics and to carry on a settler life.
According to Londoño (1989), some “directed” colonists who failed to meet
their debt obligations with the Agrarian Bank due to a bad harvest quickly
abandoned the land and got themselves into the Serranía of the Macarena,
hoping that the government would “forget” the bill.
The Rehabilitation Project was a state-led colonization and development
project that predated the Alliance for Progress in Ariari. The bipartisan
elite conceived it as a remedy to the heavy toll of peasant displacement and
dispossession resulting from the civil war. Yet, directed colonization repre-
sented only one form of exodus among others in the region; the influx of
spontaneous settlers overflowed and rapidly buried the official initiative.
Despite this, the Rehabilitation Program left three important legacies: the
Comisariato offered stable prices for the colonists’ harvest, the Guillermo
Leon Valencia Bridge connected the Ariari River Basin to the flourishing
town of Granada, and a network of tertiary roads surrounded the munici-
pality of Granada still existing to this very day. This state-led experiment set
the limits and scope for the upcoming Agrarian Reform.

The agrarian reform in the Colombian Congress

The Cuban Revolution was the turning point. It persuaded the U.S. to sup-
port the Latin American social reforms it had previously opposed. In 1961,
President John F. Kennedy created the Alliance for Progress, a hemispheric
reform program aimed at decreasing urban and rural inequality. According
to Peter Winn (1992), the AFP represented the Marshall Plan that Latin
American elites had yearned for since World War II. The United States prom-
ised twenty billion dollars in investment and loans, but Latin America’s gov-
ernments had to invest billions of their own as well (Winn 1992). Thereafter,
the unresolved Colombian agrarian problem, exacerbated during La Violen-
cia, was trapped in Cold War anxieties and internationalized.
142  María-Clara Torres
Moreover, land reform became a sharp point of contention in domestic
politics, especially within the Colombian Congress where the project had to
gain the approval of its members. To carry it out, the bipartisan government
sought the support of the Catholic Church, the military, and various factions
of the Liberal and Conservative parties. Notwithstanding, two extremes of
the political spectrum represented in the legislature were strongly opposed
to enacting a national land reform. For distinct reasons, a faction within the
Conservative Party, known as Laureanistas, and the more radical left wing
of the Liberal Party, the Revolutionary Liberal Movement (MRL), refused
to endorse the project. Thus, the law reflected the thinking of moderate and
practical reformers who wanted to regain control of the countryside and to
hopefully modernize agriculture.
On the far right, Laureanistas Conservatives fiercely opposed the expro-
priation of private lands without sufficient cash compensation. According to
their main spokesperson in Congress, Senator Álvaro Gómez Hurtado, the
state was the largest owner of fallow or underdeveloped lands. Therefore,
it lacked the moral authority to expropriate underused private lands. The
state itself had proven incapable of converting uncultivated public domains
into productive plots with adequate infrastructure. Instead of carrying
out a program of expropriation, the extreme right-wing’s solution to the
long-term land tenure problem was to relocate thousands of rural fami-
lies living on overcrowded smallholdings to government held lands (Gómez
Hurtado 1961).
Radical opposition also came from the left. Inspired by the bearded youths
who had descended from the Sierra Maestra in Cuba, a dissident wing of the
Liberal Party known as the Revolutionary Liberal Movement (MRL), cham-
pioned a Cuban model of Agrarian Reform that would forcefully expro-
priate private estates without compensating landowners (Costello 2015;
López Michelsen 1961). The MRL leader, Senator Alfonso López Michelsen
(1961), argued in favor of transferring land titles from absentee landowners
to sharecroppers and tenants. From his perspective, the strategy of trans-
ferring land title deeds from absentee landowners to rural laborers located
on the fertile lands of the nation’s core was more effective than large-scale
colonization projects in Colombia’s periphery. According to him, the fam-
ilies already occupying and working on the lands knew how to survive on
them—what crops were profitable and how the markets functioned. Con-
gressman López Michelsen pointed out that, “if those lands [on the back-
country] remain uncultivated, it is precisely because they have difficulties,
and now the elites want to give them to landless peasants” (1961, 100). In
addition, MRL Senator López Michelsen contended that the ruling coali-
tion would only make superficial reforms in order to be granted loans from
Kennedy’s Alliance for Progress. In his view, mainstream Liberals and Con-
servatives who were rallying in the National Front Coalition were trying
to “make tortillas without breaking the eggs.”9 Thereby, the agrarian bill
was “a project of illusions” and was doomed to fail. The MRL would then
 The making of a coca frontier  143

Figure 6.1 Mid-1960s liberal press presented the AFP agrarian reform as a showcase
of success. The photo caption reads: This campesino was a few steps
away from finding himself in a conflict zone. Now he is working the land.
2nd March 1964 edition of El Tiempo (Courtesy of El Tiempo).

undertake the revolution with the disaffected masses of peasants that were
aggrieved by an unsuccessful reform (López Michelsen 1961).
However, those who actively favored land reform in the National Front
Coalition, such as moderate Liberal Senator Alberto Galindo, had a more
pragmatic view:

worse than undertaking an agrarian reform with flaws that can be cor-
rected as we go, is to not make any reform. The peasants of this country
are not going to wait indefinitely until an agreement is reached between
the extreme right and left wings.10
144  María-Clara Torres
Indeed, at the end of La Violencia, the last enclaves of communist peas-
ant self-defense had refused to demobilize their units, posing a threat to
the National Front Coalition’s broader pacification project.11 An agrarian
reform was therefore deemed necessary to neutralize rural discontent. Con-
gressional voices from the ruling coalition, such as moderate Conservative
Antonio Álvarez Restrepo, urged government to enact the land reform bill
and warned that, “in the desperate countryside, the nuclei of criollo commu-
nism are brewing.”12
The Colombian Social Agrarian Reform Law was passed a few days before
Christmas, on December 21, 1961, after a long year of debates and negotia-
tions in Congress. Yet, the resulting document clarified that the expropriation
of adequately exploited private lands was the last resort, implemented only
after other options had been exhausted. Colonization of public domains in
the nation’s lower land periphery was ultimately chosen as the first and fore-
most priority to solve “the agrarian problem.” This option intended to help
rural families attain landownership while avoiding the political problem of
expropriation in the highlands, the seat of elite landed power.

The AFP’s Meta N°1 Project

The U.S.-sponsored agrarian reform focused on several internal frontiers
to conduct pilot projects; most of them coincided with the regions previ-
ously singled out by the Rehabilitation Program.13 In 1964, Project Meta
N°1 replaced the Ariari Rehabilitation Program. The first National Front
colonization project in Ariari was discredited, as most directed colonists
abandoned the area very quickly. The government suspended the promotion
of colonization in Ariari and focused its resources and efforts on allocating
land titles to spontaneous colonists already living in the area. The Agrarian
Bank’s responsibilities were ceded to the newly created Institute for Agrarian
Reform (Incora). The latter became the official entity in charge of allocation
of land titles and the division of large estates into plots.
According to a survey conducted by Incora, a large number of migrants
from all parts of Colombia had entered Ariari by the early 1970s. People
from Tolima and Cundinamarca, the largest segment, made up about 60
percent of the total number of migrants (Table 6.1).
In 1972, more than 50 percent of the families had acquired their lands by
purchasing farm assets from a previous occupant and only 35 percent by
clearing the land (Table 6.2). This empirical data confirms that, by the early
1970s, a large section of Ariari had already been exploited and pressures
over public lands had increased.
The size distribution of landholdings at the fronter sharply contrasted
with that of the country at large. In 1972, 6.9 percent of farms in the three
most dynamic government colonization regions (Meta, Caquetá and Arauca)
had less than 10 hectares of land, representing 0.6 percent of the total area
surveyed. The majority of the farms, 74.7 percent, had lots ranging from 10
 The making of a coca frontier  145
Table 6.1  Origins of Meta migrants in 1972

Place of origin1 Percent of total

Tolima 42.0
Cundinamarca 19.0
Valle 14.4
Caldas  9.0

Note:
1
Only the first four locations were cited in INCORA (1974).
Source: INCORA (1974)

Table 6.2  Forms of land acquisition, Ariari, 1972

Number Percent

Burning, clearing and weeding 19 35.2

Purchasing farm assets (crops, 28 51.8
animals and installations)
Combination 6 11.1
Others 1 1.9
Total 54 100

Source: Adapted from INCORA (1974)

to 100 hectares, encompassing slightly more than one half of the area (51.6
percent). In addition, 18.4 percent of farms had more than 100 hectares,
occupying almost the other half of the territory under survey (47.8 percent).
The land distribution in those colonization regions contrasts with the acute
concentration of land in the rest of Colombia (Table 6.3). According to the
Agricultural Census, in 1971, the vast majority of farms in Colombia (73
percent) had less than 10 hectares, occupying seven percent of the cultivable
land (Departamento Administrativo Nacional de Estadística 1971). Only 22
percent of the farms ranged in size from 10 to 100 hectares, representing 25
percent of the total area. Finally, 4.3 percent of the farms had more than 100
hectares, concentrating 67.5 percent of the cultivable land (Departamento
Administrativo Nacional de Estadística 1971). While nationally 4% of large
farms held 67% of the land, in the three colonization regions 18% of the
large farms concentrated only 48% of the land.
In the early 1970s, the agricultural frontier held a more equitable system
of land tenure than the rest of the rural country. The new colonized areas
produced a “land distribution diamond,” that is, the majority of the farms
represented farms between 10 and 100 hectares, but only few were larger
than 100 hectares, or less than 10 hectares. By contrast, the country had an
“inverted land distribution pyramid.” At the top many farms were small-
holdings, but few farms were very large, and thus at the bottom. Between
146  María-Clara Torres
Table 6.3  Land distribution in Incora-sponsored projects, 19721

Size of farms Farms Hectares

(in hectares)
Number Percent Number Percent

Less than 10 12 6.9 81 0.6

11–30 23 13.2 526 3.9
31–50 60 34.5 2,655 19.7
51–100 47 27.0 3,776 28.0
101–200 24 13.8 3,752 27.8
201 or more 8 4.6 2,681 20.0
Total 174 100.0 13,471 100

Note:
1
Includes Projects Meta n°1, Caquetá n°1, and Arauca n°1.
Source: INCORA (1974)

1964 and 1972, Incora granted 3,763 land titles that covered 165,496 hect-
ares to Ariari colonos (INCORA 1974). I argue, however, that land alloca-
tion on the frontier largely exceeded the government’s official titling efforts.
The next section will address the way settlers created their own informal
arrangements to distribute land and labor.

Frontier colonist life

Upon their arrival, migrants either found jobs as occasional wageworkers
or became partijeros (a form of sharecropping). The partija was a system of
agricultural production widespread in Ariari. It was an informal agreement
in which a landholder provided seeds, distributed fertilizers, and allowed
a newcomer to use a portion of the land. In return, the newcomer known
as partijero, was responsible for burning, clearing, and weeding the forest;
planting cash crops; and for providing his own tools and mules. At harvest
time, the landholder recovered the amount he invested and then the remain-
ing was split in half by the two parties.
The partija system could be considered highly exploitative, because if the
harvest was smaller than expected or if the price plummeted the landholder
would be prioritized. He would regain the money first and the sharecropper
would gain 50 percent of whatever was left (if anything was left). It was
likely that in case of flood or price decline the landholder would recover the
money invested but the cropper would not get his share back. In spite of
this, the former partijeros I interviewed in Ariari, found that it also worked
as a mutual beneficial agreement. Geographer Robin Marsh (1983) noted
that the landholder had abundant land, but little money for wages. He was
therefore unable to employ farm laborers, who were paid daily, to work the
entire land. At the same time, the cropper had labor, but no land and no cap-
ital (Marsh 1983). The partija thus provided newcomers with land access
 The making of a coca frontier  147
and cash. It was a way for dispossessed peasants to earn a living from land
occupied by someone else and to carry out subsistence agriculture to sup-
port their families upon their arrival to the colonization area (Marsh 1983).
People on both sides of this arrangement shared a similar history; they were
both semi-literate and impoverished peasants from the interior; they were both
uprooted and dispossessed during La Violencia; they were both seeking refuge
and attracted to Ariari by the hope of acquiring cheap and abundant land; and
neither could really claim legal property titles on the new land. The difference
between the two parties was a decade or less of arrival to the region, which cre-
ated two groups of people: “semi-established” public land settlers and “new-
comers.” As long as the state or an overriding private interest did not step in
and question the agreement between neighbors, the de facto “owner” could
determine who could use the land, under what conditions, and for how long.
The partija required continual access to new land, as well as an unstable
and poorly defined land tenure regime. It came into wide use in agricultural
frontier zones such as Ariari, where there was an abundance of public land
occupied by untitled settlers. It was also possible there, inasmuch as the pre-
dominant cash crops in those years—hand-sown corn and rice production—
did not require long-term investment. Typically, the land would no longer
produce after a couple of harvests, as cultivation without fallowing or fertiliz-
ing depleted the soil of its nutrients. With the money earned as partijeros, they
would then continue into the forest frontier, open up the land and establish a
separate farm farther away, repeating this cycle time and time again. Only the
most successful partijeros would manage to acquire land titles, build capital
and become prosperous farmers at the end of their lives. Only a few of them
became small proprietors who could establish successful agricultural farms.
The image of the colono burning and clearing the forest with a machete
and an axe, venturing time and time again into the hinterland, tells a story
of struggle for survival against market forces. Yet, it also tells us about the
existence of an ethos of the partijero—a culture that the cocalero or small
coca producer would inherit from the 1980s onward.
The following testimonies are offered in an attempt to capture the voice
of early colonists and to give proper evidence of their lived experience. They
come from Maria Ligia Barbosa, one of those exceptional small farmers still
living in Canaguaro, who had survived La Violencia and had witnessed the
Agrarian Reform. She was born in 1931 and arrived to Canaguaro (Meta)
in 1961 from the municipality of Chaguaní (Cundinamarca). At the time of
my interview, she was 85 years old and was living alone on a small-sized
farm that her husband had managed to buy some time after their arrival.
He had already passed away, and her children had left the house and had
started their families. Some of them were cultivating illicit coca in the out-
skirts of the Serranía of the Macarena National Park. She shared her story:

When the Violencia broke out, most people were forced to hastily flee
our homes, leaving everything behind. We abandoned our farmland
and only took with us our six children and three beasts. My husband
148  María-Clara Torres
traveled on the back of an ox, while my six children and I traveled
on a friend’s old truck. Our friend dropped us off on the bank of the
Ariari River, as he believed that the truck could not ford it. I clung
desperately onto my friend’s neck and reproached him: “you claim to
be our good friend, how could you do this to us? Aren’t you ashamed
to leave us stranded here in the middle of nowhere?” She emphatically
said: “Return us where we can die.” He replied: “but what can I do? I
have already brought you this far and I cannot take you back.” And so
he left us there. I held my children’s hands and started to cross the river
her gaze still fixated at the truck. My husband pointed at the other bank
of the river and said: “go on walking, mija, go on walking.”

Already exhausted from the endless hardship of colonization, Maria Ligia

refused to venture any further away. A man once proposed to give her hus-
band 10 hectares of land farther away if her husband gave him 1 hectare
of their land. She declared outrageously, “well, I do not accept that. If you
ask me as the woman of this family, I do not agree.” Her husband received
offers from numerous buyers. One day she told him decisively, “look, mijo,
if you’re bored with me, get yourself another woman who follows you wher-
ever you want. Do you know where I am going?” She urgently answered
her own question, “I’m going back to my native land where my family is.
I will not walk any further to waste the little youth that’s left in me.” “I told
my husband and I kept saying to him: ‘I am not taking another step into the
wilderness. I’m not going to squander my life back there, now that I have
a new life here.’” Indeed, the decision of certain household women such as
Maria Ligia, might have contained the havoc that colonization brought to
the frontier by slowing down the pace of deforestation
Yet, when coca crops started to spread in the 1980s and 1990s, it caused
certain moral conflicts within peasant communities among different gen-
erations. Mrs. Ligia’s stance, representative of the first generation of mid-­
century settlers, was to stay away from illegal coca. She instilled this form of
moral thinking to her descendants, by stressing the importance of self-sac-
rifice, rectitude, and integrity. She warned her sons about the dangers and
risks of growing coca, bitterly arguing with them and trying to persuade
them to stay away from that “wicked plant,” as she referred to it. “You were
born poor, you can die poor too,” she kept on saying. Notwithstanding,
like many other of their generation, one of her children was willing to take
the risk. Turning into a coca grower did not conform to her expectations.
She resented that her son had somehow betrayed the norms of traditional
peasant morality.
Maria Ligia did not explicitly connect the lack of economic opportuni-
ties for young rural people and their resort to illicit coca. Nonetheless, she
blamed state authorities for the community´s long-lasting poverty and lack
of basic services. She recalled a meeting she had a couple of years ago with
then Governor Alan Jara to demand electrical service. She shared her story
 The making of a coca frontier  149
as follows: “The Governor covered his face with his hands and exclaimed:
‘mi amor, it’s shameful that at this point in your life you still have no elec-
tricity!’” Her heart was filled with joy when she heard those words. A state
authority had finally recognized the hardship of their living conditions and
something could be done now. He immediately scribbled something down,
giving her a note that he said she could use to demand electrical service
at the Departmental Governor’s Office. She knocked on everyone’s office
doors, but no one gave her a response. That happened three years ago. “The
Governor is not in office anymore; I have his note, but still no electricity”
Maria Ligia declared with outrage. At the end of the interview, the 85-year-
old colonist woman told me that one of her children had offered her to leave
the farm and join his new family in the village of Canaguaro, where they
could take better care of her during the last years of her life.

Coca resiliency and spread

Beginning in the 1980s and continuing through the 1990s, the coca-cocaine
boom had dramatic consequences on the Ariari frontier, notably on the dis-
tribution and the uses of the land. As Granada developed into the largest
municipal seat in Ariari, the surrounding rural land increased in value and
became a means of money laundering. An emergent wealthy class of rural
planters increasingly associated with drug trafficking, appropriated and then
concentrated the land. Former recipients of titles from the agrarian reform
sold their properties with the hope of using the seed capital for starting a
small business in a town or to switch to cattle grazing in another plot fur-
ther away. Meanwhile, untitled colonists sold or bargained their farm assets
(fences, crops, animals, water spring sources and buildings) in exchange for
larger tracts of land in the immediate hinterland located in the National
Park Serranía of the Macarena. In the area surrounding Liberal Granada, a
frontier peasant economy based on subsistence crops, hand-sown rice and
corn production, was replaced by (narco) entrepreneurial agriculture dom-
inated by large scale and capital-intensive plantations of rice, palm oil, and
extensive cattle ranching. Although pressures over public lands were appar-
ent in the early 1970s, severe land concentration occurred in the span of a
decade. By 1984, the land Gini coefficient in Meta rose to 0.88, the high-
est among Colombia’s departments, followed by Casanare at 0.84 and by
Cauca at 0.82 (Instituto Geográfico Agustín Codazzi 2012). From 1960 to
1984, the region went from being an open agricultural frontier with a sys-
tem of land tenure more equitable than the rest of rural Colombia, to having
the most unequal land distribution in the country. 
Why did frontier settlers yield to pressure from this emergent class over
the land in Ariari? For frontier colonists, the struggle for land was not as
crucial as it was for wageworkers and for sharecroppers (arrendatarios)
living on the savannas of the Caribbean Coast. Sociologist Leon Zamosc
(1986) showed that such Colombian rural workers organized themselves
150  María-Clara Torres
and fought against the hacienda owners in the late 1960s and 1970s. Yet, in
Ariari, colonists had recourse to pockets of abundant land in the immediate
hinterland. Under these circumstances, cash was therefore a more valuable
and more privileged resource than land. The sale of farm assets and the sub-
sequent opening of new colonization fronts represented a modest source of
cash. Later on, revenues from coca crops became another incentive to repeat
the cycle; an incentive that fit the colonists’ urge to move further inland. 
The frontier was thus pushed westwards to the immediate hinterland
located at the edges of the Serranía of the Macarena and through to the
heart of the National Park. The rivers Ariari, Guayabero, Duda, and Güejar,
delimited the area covered by Amazon rainforest, savannas, and the moun-
tain range Sierra de la Macarena. In the mid-1950s, peasant Communist
nuclei had sprouted on the banks of the Guayabero River. These enclaves
of peasant self-defense became the Revolutionary Armed Forces of Colom-
bia (FARC) in 1964, after the Colombian military bombarded them with
napalm in a U.S.-sponsored counterinsurgency operation (Arenas 1972;
González 1992; González and Marulanda 1990). 
The first local marijuana plantations appeared in the mid-1970s (a for-
gotten regional relative to the famous marimbero boom of the Guajira
Peninsula), but quickly receded by the end of the decade, when its prices
collapsed abruptly. According to economist Oscar Arcila, the U.S. creation
of a seedless “weed” resulted in a contraction of demand for Colombian
marijuana and caused an overproduction crisis. In 1977, prices dipped so
low that marijuana had to be burnt or thrown into the rivers (Arcila 1989).
That was the end of Macarena colonists’ brief involvement with cannabis.
Nevertheless, sociologist Alfredo Molano contended that “[marijuana] was
a general rehearsal, a frustrated one, but it prepared people for coca culti-
vation, which came almost immediately” (1989, 300–301). By 1978, coca
replaced marijuana plantations in the Serranía of the Macarena National
Park. Emerald dealers tapped into their experience with smuggling and vio-
lence and attempted to control every stage of the lucrative cocaine industry
(Molano 1989). At a location known as La Libertad, between El Retorno
and Calamar, in the Amazonian department of Guaviare, emerald traders
established extensive coca fields protected by heavily armed men; they hired
the local workforce to grow the shrubs, trained specialized groups to refine
cocaine, and ultimately shipped the drug by plane to Villavicencio or Bogotá
(Molano 1989). Yet, colonists soon discovered the secret of coca cultivation
and seeds spread northwards along the rivers, from La Libertad to San José
del Guaviare, and from there to Puerto Rico (Meta) and then to the Serranía
of the Macarena (Molano 1989). Small coca plots were then opened in iso-
lated areas inside the protected park.
The FARC leadership was initially reticent to accept capitalist drug pro-
duction in their regional stronghold, as it would instill an individual enrich-
ment mentality among their grassroots supporters and eventually hurt the
organization’s long-term objectives of overthrowing the Colombian state
 The making of a coca frontier  151
(Cook 2011; Jaramillo et al. 1989). However, at the Seventh Guerrilla Con-
ference held in 1982, the rebel organization embraced commercial coca as a
strategic funding source in the battle against the Colombian establishment
(Cook 2011; Marcy 2010). This decision played a central role in the spread
of illicit coca. 
A census conducted in the Serranía of the Macarena registered a signifi-
cant migratory influx of colonos beginning in the late 1970s and accelerat-
ing from 1982 on (Universidad Nacional de Colombia 1989). From 1982 to
1988, an estimate of 2,053 new families arrived to the protected area; and 7
percent of the 1,019,036 hectares located in the forest reserve were already
carved out into coca and subsistence plots. In 1988, there were 24,878
people living in the national park; the largest fraction, 46 percent, born
in the Department of Meta, followed by 14 percent in Cundinamarca, 12
percent in Tolima, 5 percent in Boyacá, 5 percent in Valle del Cauca, and the
remaining 18 percent in other departments. The 1988 census documented
active colonization fronts and the mushrooming of several villages as cen-
ters for trade of coca paste. These hamlets flourished on the river banks, as
these were oftentimes the only means of transportation for both people and
drugs. Throughout the 1980s, hamlets such as Barranco Colorado emerged
along the Ariari River in the municipality of Puerto Rico; Puerto Toledo
along the Güejar River; Nueva Colombia on the Guayabero River in the
municipality of Vistahermosa; and Jardín de Peñas along the Duda River
in the municipality of Mesetas. These villages owed their rapid growth to
land concentration in other areas of the country, including the Department
of Meta, and to the commercial benefits derived from coca crops (Cubides
et al. 1989).
Coca’s spread was in a sense a form of colono resiliency. For Molano,
“it seemed as if all the sufferings, deprivation, and exploitation to which
the settlers had been subjected, had suddenly claimed revenge. And what a
revenge!” (Molano 1989, 301). News of the illegal crop soon attracted people
from all over the country, who flooded the periphery to grow and process the
leaves. A wide array of people offering goods and services arrived; sellers of
precursor chemicals, shopkeepers, prostitutes, street vendors, as well as tran-
sient workers, such as seasonal coca leaf pickers known as raspachines, were
attracted to this natural reserve transformed into a leading drug hotspot.
However, in Macarena, the FARC’s historic stronghold, the insurgency
enforced limits to narcotics production and trade. It opposed the ­traffickers’
ambitions to create a cocaine complex with large-scale plantations of coca,
sophisticated laboratories, and clandestine airstrips, such as those established
in the early 1980s in other areas of the country.14 In Macarena the guerrilla
fostered small-scale production of coca along with the cultivation of subsis-
tence crops (Arcila 1989; Gutiérrez 2005). According to Oscar Arcila,

colonists were notified that for each hectare cultivated in coca, they had
to plant another in yucca, plantain, and maize. In addition, a so-called
152  María-Clara Torres
Agrarian Thursday instilled that everyone had to dedicate that day of
the week to agricultural activities different from coca cultivation.
(1989, 175)

The FARC guerrillas struggled to preserve the peasant background of the

cocalero. Yet, this was a very difficult task considering that their own rev-
enues as a rebel organization relied increasingly on taxes collected from
narcotics production in the rural areas that they dominated. By the end of
the 1980s, Jacobo Arenas, the ideological leader for the guerrillas, expressed
this ambiguity in an interview to the press:

I am a revolutionary, I fight against capitalism. But I live in a capitalist

system and I understand that capital is needed to undertake revolution.
When the revolution triumphs, we’ll stop right there and we’ll create a
new mode of production, a new conception of life in which exploitation
of men will be abolished.
(Semana 1987)

Until the day of this longed for victory, the guerrilla movement had to adapt
and nuance its ideological opposition to capitalist drug commodities, pro-
viding specific regulations for everyday life in grassroots communities, also
eager to participate in the boom. Ideology was thus materialized in a con-
crete way, as narrated by Jacobo Arenas a couple of years before his death:

We tell coca growers: compañeros (comrades), we are not going to pro-

hibit you from growing coca, because that is something that has become
widespread, and besides, we are not the government to prohibit it. You
may cultivate it, but make sure you buy cattle, set your farms in order,
clean up your houses, and educate your children, because coca may end
anytime. And people are sympathetic to that orientation, and not just
sympathetic, they have to accept it.
(Semana 1987)

The guerrilla organization was driven to drug production because of sus-

tained pressures from below and their own military priorities. The FARC’s
decision to support narcotics production in 1982 at their Seventh Confer-
ence, allowed them to expand geographically and to strengthen their m
­ ilitary
capacities (Cook 2011). By the mid-1990s, Colombia, a country with a neg-
ligible history of coca cultivation, had become the main Andean producer
of coca for the international drug trade. The illicit agricultural economy
fostered a mestizo and colono coca culture. For the past four decades, coca
smallholders have been actively engaged in building municipalities, in party
politics, and have routinely engaged with the national state to demand bet-
ter services and integration as working citizens. They have done all this,
despite their “illicit” status in many sociological senses. 
 The making of a coca frontier  153
Epilogue
In April 2006, the Agrogüejar Peasant Association summoned the Commu-
nity Action Councils based along the Guëjar River to join a protest, which
they called a “Civil Resistance Round.” About 6,000 men, women, and
children, participated in this mobilization. These families were involved in
the small-scale production of coca leaf and paste. They marched from their
farms to Caño Danta, a hamlet located in the National Park Serranía of the
Macarena. They set up improvised plastic tents and stayed for about a month
on the sides of the dirt road that connected Caño Danta to the municipalities
of Vista Hermosa, Puerto Lleras, and Puerto Rico in the Department of Meta
(International Peace Observatory 2006; El Tiempo 2006b). The marchers
demanded a meeting with the then President Álvaro Uribe Vélez to put an
end to human rights violations perpetrated during another version of Plan
Colombia (Tate 2015), this time known as “The Patriot Plan.” Supported
by both Colombian and U.S. funds, the “Patriot Plan” had nearly 18,000
soldiers positioned in the forests of Meta, Caquetá, and Guaviare, to reclaim
the territory considered as the FARC’s historical rearguard, as well as to cap-
ture the guerrilla commanders (Ruiz 2004). Picketers denounced the abuses
committed by the military, and demanded the immediate withdrawal of para-
militaries and the suspension of indiscriminate aerial spraying of glyphosate
on both their legal and illegal crops. In addition, they requested funding for
farming projects and land rights to the plots located within the National Park
Serranía of the Macarena and extending outside its borders. On May 16,
2006, the breaking news appeared on the daily El Tiempo with the title: “The
FARC repeat the coca marches” (El Tiempo 2006a). Indeed, the 2006 pro-
test of Caño Danta replayed the images of the cocalero mobilization under-
taken a decade before throughout the Colombian Amazon region (Ramírez
2011). The then Treasurer of Agrogüejar, Víctor Hugo Moreno, was reported
to the press as stating: “We know that coca is illegal. We will not oppose its
eradication, but the government must stop the fumigation and support our
income-producing projects as alternatives to survive” (El Tiempo 2006b).15 
The demonstrators of Caño Danta did not succeed in bringing President
Uribe to the table as they had hoped, but they did achieve a dialogue with the
then High Commissioner for Peace, Luis Carlos Restrepo. As a result, 480
families agreed to uproot 2,000 hectares of coca fields. In return, the govern-
ment promised immediate food aid, funding for alternative farming projects,
and land titling only on the outer fringes of the protected area (Agrogüejar
2012). The military pressure was very intense before and during the protest.
There were also rumors that if coca growers did not comply with the govern-
ment’s eradication requirement, cocaleros could be prosecuted and separated
from their children, who would be sent to homes run by the Colombian Insti-
tute of Family Welfare. Small coca producers were then forced to commit to
“voluntarily” eradicating 2,000 hectares. From 2006 to 2008, they them-
selves opened their farm gates allowing the security forces access to destroy
the coca fields. This effectively cut off the main source of cocalero income. 
154  María-Clara Torres
The 2010 United Nations Drug Control Program [UNODCP] report
presented the reduction in cultivated areas of coca as a success of another
anti-narcotic campaign known as the Consolidation Plan of the Macarena
(Oficina de las Naciones Unidas contra las Drogas y el Delito 2010). Yet, the
gains in counterinsurgency and counterdrug war were not gains in human
rights. What was considered an achievement in official national and inter-
national circles was experienced on the Ariari–Güejar river basin as anxiety,
uncertainty, food crisis and internal displacement in search of employment.
The village of Puerto Toledo that burgeoned in the 1980s as a center of
trade for coca paste became deserted. The youth migrated in the hope of
finding jobs as agricultural workers on the extensive plantations of oil palm
in Granada, or as temporary employees in the oil industry in Puerto Gaitán.
But the situation was especially hard for older people, who had no option
but to remain in the area. When one of them was asked what he lived on, he
replied: “¡de milagro!” (miraculously!)16 
Eventually, in 2010, the new government of Juan Manuel Santos and the
European Union co-funded an alternative development project in the area,
which included honey, rice, and rubber production. The initiative was intended
to benefit 80 peasant families living on the edges of the National Park. The
sponsors justified the need for alternative development as follows: “with the
active participation of local grassroots organizations in Puerto Toledo, illicit
crops have been almost completely eradicated, therefore creating positive
conditions for the preparation and execution of a pilot ­project of alternative
development” (Colombia, Acción Social 2008, 5). Indeed, state repression had
laid the foundations for alternative development. The project was carried out
simultaneously to another counter-narcotic and counterinsurgency campaign,
this time known as the National Plan of Territorial Consolidation. The initia-
tive was conducted in the midst of combat, bombings, forced disappearances,
detentions, and daily searches conducted by armed forces and paramilitaries.
Yet, its designers expected that at the end of the project, “[it] would have
improved the confidence between the citizens and the state, which would pre-
vent the resurgence of illegal economies” (Colombia, Acción Social 2008, 8). 
In October 2015, four years after the project’s conclusion, I conducted
fieldwork in Puerto Toledo. I could confirm that none of the three production
lines of honey, rice, and rubber existed. The planks to build hives, stacked
up, one on top of the other, had deteriorated from weevil damage and were
rotting in the backyard of a house. Some of the rice seeds were already rot-
ten when they arrived at Puerto Toledo, the remainder did not grow because
it did not fit the agro-ecological conditions of the area, and consequently
the rice thresher provided by the project was never used. Rubber production
was quickly abandoned, as technical assistance stopped at the project’s con-
clusion, and so did the community’s interest in maintaining the plantations.
Moreover, the project’s designers never planned any marketing process for
rubber. After nearly four decades of eking a living out of cultivating the leaf,
coca farmers have very limited experience with any other commercial crop
and can hardly imagine a future without coca.
 The making of a coca frontier  155
Today, new democratizing opportunities are opening up for grassroots
coca communities in light of two special junctions: the crisis of the global
war on drugs and the ongoing peace process in Colombia (Gootenberg
2017). In a speech delivered a few days before the fifteenth anniversary
of Plan Colombia, the Colombian President Juan Manuel Santos, empha-
sized  “the need that the whole world recognizes that the war on drugs,
declared forty years ago, has not been won. If it has not been won after
forty years, it’s because we are doing something wrong” (Agencia Efe 2016).
Hours later, the President explained: “We have failed to offer to the thou-
sands of farmers an alternative to produce something different from illicit
coca. And until that happens, coca production will persist” (Presidencia de
la República 2016). For the first time, the Colombian president publicly
questioned the effectiveness of alternative development programs fostered
by the government and by international cooperation agencies.
Although beyond the scope of this chapter, the 2016 peace accords
between the government of Colombia and the FARC guerrillas have sig-
nificantly advanced understanding of the importance of rural development
policies for coca-growing regions. For the first time in the history of drug
control in the country, long-term development initiatives may involve state
authorities, the demobilized insurgency, and grassroots coca communities.
As we write in 2017, the peace accords represent an opportunity to under-
take long-deferred rural reforms, as well as to guarantee political partici-
pation and representation for communities living in Colombia’s periphery.
Yet, as long as cocaine remains an illegally prohibited substance, there will
be incentives to its production for a criminal black market. As Colombian
journalist Alvaro Sierra (2002) stated, “[illicit] coca can neither be created
nor destroyed, but only displaced.” 

Notes
1 “Relación de debates,” December 21, 1961, Trimester 4, Anales del Congreso de
Colombia, Box 4, Biblioteca Luis Ángel Arango.
2 According to historian Gonzalo Sánchez (1988), the program initially prioritized
the departments located in the nation’s political center most devastated by the
civil war: Caldas, Cauca, Huila, Tolima, and Valle del Cauca. It was only a year
later, in 1959, that internal frontiers such as Ariari, Sumapaz, and Magdalena
Medio, were incorporated.
3 From the 1990s onwards, historians and social scientists referred to this specific
and localized form of settlement led by the Colombian Communist Party, as
“armed colonization” (González 1992; González and Marulanda 1990; Molano
1992; Molano and Fajardo 1989; Ramírez 2001).
4 These self-defense groups became the Marxist-Leninist FARC guerrilla organiza-
tion in 1964.
5 Today’s departments of Caldas, Risaralda, and Quindío, positioned on the Central
Cordillera of the Andes, formed the Viejo Caldas Department from 1905 to 1966.
6 The construction of the bridge over the Ariari River began in 1960, under the Reha-
bilitation Program, but it was inaugurated only years later during the Presidency of
Conservative Guillermo León Valencia, 1962–1966 (Universidad Nacional 1989).
The bridge was considered a significant infrastructure project at that time.
156  María-Clara Torres
7 The construction of the bridge over the Ariari River began in 1960, under the
Rehabilitation Program, but it was inaugurated only years later during the Presi-
dency of Conservative Guillermo León Valencia, 1962–1966 (Universidad Nacio-
nal 1989). It was considered a significant infrastructural project at that time.
8 Karl suggests that the Rehabilitation’s road projects “intentionally eschewed
mechanized labor in favor of sponsoring as many paid positions as possible”
(2017, 81–82).
9 In Colombia, tortillas refer to omelets or Spanish tortillas.
10 “Relación de debates,” December 21, 1961, Trimester 4, Box 4, Anales del Con-
greso de Colombia, Biblioteca Luis Ángel Arango.
11 In 1964, they became the FARC guerrilla organization.
12 “Acta de la sesión del jueves 16 de noviembre de 1961,” November 21, 1961,
Trimester 4, Box 47, Anales del Congreso, Biblioteca Luis Ángel Arango.
13 Incora’s agrarian reform included the following regions: Ariari (Meta #1),
Sarare (Arauca #1), El Retorno (Caquetá #1), Pacífico (Cauca #1) Bajo Putu-
mayo (Putumayo), Urabá (Chocó #1), Magdalena Medio, Casanare (Sudirección
Yopal), Sumapaz (Cundinamarca #4), Huisitó-El Tambo (Cauca #2). (Instituto
Colombiano de la Reforma Agraria 1974).
14 These cocaine complexes operated in Tranquilandia, on the Yarí plains of
Caquetá, and in El Azul, Putumayo. Gonzalo Rodríguez Gacha, an emerald dealer
member of the Medellín Cartel, established these cocaine complexes. The cartel
paid a fee to the FARC guerrillas for each ton of cocaine processed, in exchange
for providing security services. These narcotics-processing centers operated in
Tranquilandia, on the Yarí plains of Caquetá, and in El Azul, Putumayo. DEA
special agent, Johnny Phelps, called them “cocaine industrial park” or the “Sili-
con Valley of cocaine” (Gugliotta and Leen 1989). They were dismantled by the
Colombian army in 1984.
15 In the first trimester of 2006, six murders and 19 forced disappearances had been
reported in the hamlets of Puerto Toledo, then occupied by the military (Comité
Permanente 2006; International Peace Observatory 2006).
16 Interview conducted in corregimiento of Puerto Toledo, municipality of Puerto
Rico (Meta), in October 2015.

References
Agencia Efe. 2016. Santos propone revisar la política antidrogas global para su mayor
efectividad: January 28. www.efe.com/efe/america/politica/santos-propone-re-
visar-la-politica-antidrogas-global-para-mejorar-su-efectividad/20000035-
2823906, accessed February 2, 2016.
Agrogüejar. 2012. Plan de Desarrollo Sostenible de la Zona de Reserva Campesina,
Sector Güejar-Cafre, Puerto Rico, Meta: May. http://issuu.com/centrodedocu-
mentacionanzorc/docs/plan_de_desarrollo_sostenible_de_la_c8173e8e0ae11c,
accessed February 2, 2016.
Arcila, Oscar. 1989. Sectores de actividad económica regional. In La Macarena,
reserva biológica de la humanidad: territorio de conflictos. Cubides, Fernando
Mockus Sivickas, Antanas Avellaneda, Mario González, Henry Arcila Niño,
Óscar Molano, Alfredo Acero, Hugo Pacheco, and Juan Carlos Mosquera Mesa,
Ricardo, eds. Bogotá: Universidad Nacional de Colombia.
Arenas, Jacobo. 1972. Diario de la Resistencia de Marquetalia. Bogotá: Ediciones
Abejón Mono.
Colombia, Agencia Presidencial para la Acción Social y la Cooperación Internacio-
nal [Acción Social]. 2008. Anexo A. Laboratorio de Paz III. Formulario de Solic-
itud de Subvención. Bogotá.
 The making of a coca frontier  157
Comité Permanente por la defensa de los derechos humanos. 2006. 6.000 cocale-
ros del Meta protestan erradicación forzada de cultivos ilícitos: May 19. https://
www.google.es/#q=6.000 cocaleros del Meta protestan contra la erradicación
forzada de cultivos ilícitos, accessed February 2, 2016.
Cook, Thomas. 2011. “The Financial Arm of the Farc: A Threat Finance Perspec-
tive.” Journal of Strategic Security. 4(1): 19–36.
Costello, Thomas. 2015. The Guarantee of Freedom and Dignity? Colombian Land
Reform, the Alliance for Progress, and the United States. Ph.D. dissertation,
Department of History, University of Connecticut.
Cubides, Fernando, Antanas Mockus Sivickas, Mario Avellaneda, Henry González,
Óscar Arcila Niño, Alfredo Molano, Hugo Acero, Juan Carlos Pacheco, and
Ricardo Mosquera Mesa. 1989. La Macarena Reserva biológica de la humani-
dad Territorio de conflictos. Bogotá: Universidad Nacional de Colombia.
Departamento Administrativo Nacional de Estadística [DANE]. 1971. Censo Nacio-
nal Agropecuario 1970–1971. Bogotá: DANE.
Departamento Administrativo Nacional de Estadística [DANE]. 2007. Colombia,
una nación multicultural: su diversidad étnica. Bogotá: DANE.
El Tiempo. 1964. Excelentes resultados ha logrado la reforma agraria en Colombia:
March 3.
El Tiempo. 2006a. Las Farc repiten las marchas cocaleras: May 16. www.eltiempo.
com/archivo/documento/MAM-2027204, accessed February 1, 2016.
El Tiempo. 2006b. Tregua en Panamericana se suspendió por ausencia oficial: May 18.
www.eltiempo.com/archivo/documento/MAM-2028072, accessed February 2, 2016.
Gómez Hurtado, Álvaro. 1961. El Auténtico contenido de una reforma agraria.
In Tierra: diez ensayos sobre la reforma agraria en Colombia. Carlos Lleras
Restrepo, ed. Bogotá: Ediciones Tercer Mundo.
González, José Jairo. 1992. Espacios de exclusión: El estigma de las repúblicas inde-
pendientes 1955–1965. Bogotá: Cinep.
González, José Jairo, and Elsy Marulanda. 1990. Historias de frontera: Colonización
y guerras en el Sumapaz. Bogotá: Cinep.
Gootenberg, Paul. 2008. Andean Cocaine: The Making of a Global Drug. Chapel
Hill, NC: University of North Carolina Press.
Gootenberg, Paul. 2017. “Cocaine Histories and the Diverging Andean ‘Drug Wars’
in Bolivia, Colombia, and Peru.” ContraCorrientes. 15(1): 1–35.
Grandin, Greg. 2004. The Last Colonial Massacre: Latin American in the Cold War.
Chicago, IL: University of Chicago Press.
Gugliotta, Guy and Jeff Leen. 1989. Kings of Cocaine: Inside the Medellín Cartel, an
Astonishing True Story of Murder, Money, and International Corruption. New
York: Simon and Schuster.
Gutiérrez, Omar. 2005. Desarrollo rural alternative y economía política de la coca en
el Meta, 1982–2004. Bogotá: PNUD, ASDI.
Henderson, James. 1985. When Colombia Bled: A History of the Violencia in Tol-
ima. Birmingham, AL: University of Alabama Press.
Hormaza, Ingrid Carolina. 2016. La reforma agraria como ejercicio de planificación:
Experiencias de los proyectos de colonización del INCORA en el Caquetá entre
1964–1974. Masters thesis, Programa de ordenamiento urbano-regional, Univer-
sidad Nacional de Colombia.
Instituto Colombiano de la Reforma Agraria, INCORA. 1974. La colonización
en Colombia: una evaluación del proceso. Bogotá: Instituto Interamericano de
Ciencias Agrícolas.
158  María-Clara Torres
Instituto Geográfico Agustín Codazzi, IGAC. 2012. Atlas de la distribución de la
propiedad rural en Colombia. Bogotá: Imprenta Nacional.
International Peace Observatory. 2006, May 18. Movilización campesina contra el
Plan Patriota. www.peaceobservatory.org/es/423/movilizacion-campesina-contra-
plan-patriota, accessed February 1, 2016.
Jaramillo, Eduardo, with Leonidas Mora and Fernando Cubides. 1989. Coloni-
zación, coca y guerrilla. Bogotá: Alianza Editorial.
Karl, Robert. 2017. Forgotten Peace: Reform, Violence and the Making of Contem-
porary Colombia. Oakland, CA: University of California Press.
Latham, Michael. 2011. The Right Kind of Revolution: Modernization, Develop-
ment, and U.S. Foreign Policy from the Cold War to the Present. Ithaca, NY:
Cornell University Press.
Londoño, Oscar. 1989. Colonización del Ariari: Aproximación a una historia
regional. Villavicencio: Cenessoll.
López Michelsen, Alfonso. 1961. Hacia una verdadera reforma que complete la Rev-
olución en Marcha. In Tierra: Diez Ensayos sobre la Reforma Agraria en Colom-
bia. Carlos Lleras Restrepo, ed. Bogotá: Ediciones Tercer Mundo.
Marcy, William. 2010. The Politics of Cocaine: How U.S. Foreign Policy Has Cre-
ated a Thriving Drug Industry in Central and South America. Chicago, IL: Law-
rence Hills Books.
Marsh, Robin Ruth. 1983. Development Strategies in Rural Colombia: The Case of
Caquetá. Los Angeles, CA: University of California Press.
Mayer, Enrique. 2009. Ugly Stories of the Peruvian Agrarian Reform. Durham, NC:
Duke University Press.
Molano, Alfredo. 1989. El proceso de colonización de la region del Ariari-Güejar. In
La Macarena, reserva biológica de la humanidad: territorio de conflictos. Cubides,
Fernando, Antanas Mockus Sivickas, Mario Avellaneda, Henry González, Óscar
Arcila Niño, Alfredo Molano, Hugo Acero, Juan Carlos Pacheco, and Ricardo
Mosquera Mesa eds. Bogotá: Universidad Nacional de Colombia.
———.1992. Trochas y fusiles. Bogotá: El Áncora.
———. and Darío Fajardo. 1989. Colonización de la Macarena, Yo le digo una de
las cosas. Bogotá: Corporación Araracuara.
Morales, Edmundo. 1989. Cocaine: White Gold Rush in Peru. Tucson, AZ: Univer-
sity of Arizona Press.
Oficina de las Naciones Unidas contra las Drogas y el Delito & Acción Social. 2010.
Des-cifrando las familias guardabosques. Evaluación final. Consolidado nacio-
nal, línea base, seguimiento y línea final. Bogotá.
Oquist, Paul. 1980. Violence, Conflict, and Politics in Colombia. New York: Aca-
demic Press.
Presidencia de la República. 2016. Palabras del Presidente Juan Manuel Santos en la
visita a la Universidad Nacional de Colombia: January 18. http://es.presidencia.
gov.co/discursos/Palabras-del-Presidente-Juan-Manuel-Santos-en-la-­visita-a-la-
Universidad-Nacional-de-Colombia, accessed February 2, 2016.
Ortiz, Carlos Miguel. 1985. Estado y subversión en Colombia: La Violencia en el
Quindío, años 50. Bogotá: Uniandes.
Ramírez, María Clemencia. 2011. Between the Guerrillas and the State: The Cocalero
Movement, Citizenship, and Identity in the Colombian Amazon. Durham, NC:
Duke University Press.
 The making of a coca frontier  159
Ramírez Tobón, William. 2001. “Colonización armada, poder local y territorial-
ización privada.” Journal of Iberian and Latin American Research. 7/2: 63–81.
Roldán, Mary. 2002. Blood and Fire: La Violencia in Antioquia, Colombia, 1946–
1953. Durham, NC: Duke University Press.
Ruiz, Marta. 2004. El río de la guerra. Revista Semana. www.semana.com/nacion/
articulo/el-rio-guerra/66978-3, accessed February 2, 2016.
Sanabria, Harry. 1993. The Coca Boom and Rural Social Change in Bolivia. Ann
Arbor, MI: University of Michigan Press.
Sánchez, Gonzalo. 1988. Rehabilitación y violencia bajo el Frente Nacional. Revista
Análisis Político. 4: 21–42.
Sánchez, Gonzalo, and Donny Meertens. 2001. Bandits, Peasants, and Politics: The
Case of La Violencia in Colombia. Austin, TX: University of Texas Press.
Semana. 1987. Jacobo Arenas habla sobre Ledher. April 6.
Sierra, Álvaro. 2002. “La coca viajera.” El Tiempo: October 13. www.eltiempo.com/
archivo/documento/MAM-1372034, accessed February 1, 2016.
Tate, Winifred. 2015. Drugs, Thugs, and Diplomats: U.S. Policymaking in Colombia.
Stanford: Stanford University Press.
Universidad Nacional de Colombia. 1989. La Macarena, reserva biológica de la
humanidad. Territorio de conflictos. Bogotá: Universidad Nacional.
Uribe, Maria Victoria. 1990. Matar, rematar y contramatar: las masacres de La Vio-
lencia en el Tolima, 1948–1964. Bogotá: CINEP.
Winn, Peter. 1992. Americas: The Changing Face of Latin America and the Carib-
bean. Berkeley, CA: University of California Press.
Zamosc, Leon. 1986. The Agrarian Question and the Peasant Movement in Colom-
bia: Struggles of the National Peasant Association, 1967–1981. Cambridge:
Cambridge University Press.
7 Epilogue
Will governments confront coca
cultivation, or its causes?
Adam Isacson

In March 2017, the new Trump administration released the U.S. govern-
ment’s annual estimate of coca in Colombia. It found a record amount:
188,000 hectares in 2016. (Office of National Drug Control Policy 2017)
The U.S. coca estimate had increased 133 percent from 2013 to 2016. Its
cocaine production estimate for Colombia grew by 202 percent, to 710 tons.
This was a return to levels last seen at the outset of “Plan Colombia” in
2000–2001.
The new Secretary of State, Rex Tillerson, voiced frustration at a June
2017 Senate hearing: “We’ve had discussions with Colombian leadership,
with President Santos, and questioned why—you know—how could this
happen? How could this be?” (U.S. House Committee on Appropriations
2017)
The answer is not complicated. The increase in coca production hap-
pened the same way it usually happens. All along the production chain,
participants persist in searching for shortcuts, with poor results. Small
farmers in agricultural frontier zones opt for the one crop that guarantees
steady, if modest, profits. Elites avoid the hard work of extending the state
to abandoned territories, or rooting organized, crime-tied corruption out
of fragile institutions. Users in consuming countries seek a few moments
of artificial well-being. The U.S. government, partnering with elements in
local elites and security forces, opts for quick-payoff strategies like forced
eradication campaigns. These kill some plants, but leave the problem’s
roots intact.
With few exceptions, we have yet to see a determined, patient approach
that might bring a permanent change to the dynamics driving the coca
economy. One based on dialogue with affected populations, embracing the
complexities of production and consumption. The result of this absence is
evident in the record coca crops of the mid-2010s.
The years of the Barack Obama administration (2009–2016) saw modest
space open for new approaches to illicit drugs. In many U.S. states, attitudes
toward cannabis changed. Attention shifted to the domestic opioid crisis,
with few calls for a hardline response. (Serrano 2016) Amid declining use,
Andean cocaine became an afterthought.
 Epilogue  161
During this period, the State Department “decertified” Bolivia every year,
waiving penalties (U.S. Department of State 2000–2017). Obama admin-
istration diplomats disliked Bolivia’s defiance of the U.S. anti-coca model.
Still, it was hard to argue with the results of Bolivia’s “social control” experi-
ment. Coca cultivation now rarely exceeds 35,000 hectares (or 25,000 in the
UNODC estimate), with about half of it for legal use (U.S. Department of
State 2000–2017; UN Office on Drugs and Crime 2017, 25; Andean Infor-
mation Network 2017). In the Chapare region, constant dialogue with farm-
ers manages illicit coca. In the Yungas coca-growing region, the Morales
government’s plans have encountered more resistance (Andean Information
Network 2010). More than coca, Bolivia’s greater challenge today is cocaine
from elsewhere. “Peruvian officials estimate that 50 percent of all Peruvian
cocaine departs to or through Bolivia via aerial transshipment,” the State
Department reports (U.S. Department of State 2016, 111).
In Peru, Ollanta Humala’s government made initial moves toward an
independent coca model. It stood down in a matter of weeks, though, decid-
ing not to confront the United States (Stone 2012). Peru’s targets for forced
eradication remained high. The U.S. ambassador congratulated Peru’s gov-
ernment and held it up as an example for the region (Government of Peru
2015). But coca cultivation, in the U.S. estimate, has stayed within a range
of 46,500 to 59,500 hectares after a big increase in 2010 (U.S. Department
of State 2017a, 29).
The most interesting recent experience is Colombia’s. Cultivation there has
increased the most, and experimentation with a new approach is beginning.
In 2013–2015 Colombia slowed, then stopped, a program of aerial herbi-
cide fumigation. Colombia now uses the U.S.-granted spray planes to fight
forest fires (Ministry of Defense of Colombia 2015). Over 23 years and 1.8
million hectares sprayed, this program was controversial (U.S. Department
of State 2000–2017). Small-scale coca-growers complained of health effects
and hunger, and raged at the state. They also adjusted, forcing Colombia to
spray more to achieve the same result (Isacson 2015). “I’ve explained to Vice
President Pence that aerial spraying did not bring about the results everyone
hoped for,” Colombian President Juan Manuel Santos said after a meeting
with U.S. Vice-President Mike Pence. “In the year 2007—and I showed him
the numbers—is when we did the highest volume of air spraying of illegal
crops, and that is one of the years when coca production increased the most”
(The White House 2017b).
Negotiations with the FARC guerrilla group drew up the outlines of a
new approach. (The negotiations ended a conflict that coca rents prolonged.
At the same time, coca was a symptom of the same neglect, corruption, and
inequality that fed the conflict.) Colombia committed to a sharp nationwide
increase in investment in rural smallholding production. It agreed to help
communities of coca-growing families transition to licit crops (High Com-
missioner for Peace 2016). Coca-growers and analysts believe the second
commitment will not succeed without the first (Garzón and Bernal 2017, 13;
162  Adam Isacson
Asamblea Constitutiva de la Coordinadora de Cultivadores de Coca, Amap-
ola y Marihuana 2017; González Posso 2017).
While this is hopeful, the political reality is murky. There was a four-
year gap between fumigation’s initial slowdown (2013, after the FARC shot
down two spray planes) and the new strategy’s first steps (2017). During
this interregnum, the prices of basic commodities—especially illicitly mined
gold—plummeted. This made coca appear to be a more attractive income
source. (Semana 2017a) It also coincided with the FARC peace talks.
The result of the four-year gap is evident in the “green” areas of the UN
Office on Drugs and Crime (UNODC) maps of Colombia (UN Office on
Drugs and Crime 2002–2016). The crop is now very easy to find, with recent
plantings climbing up to roadsides and riversides. Also easy to find, though,
is evidence of how little has changed since the launch of “Plan Colombia.”
The 1999 document’s subtitle was “a plan for peace, prosperity, and the
strengthening of the state” (Government of Colombia 1999). But the state
remains weak, if it is present at all.
Almost 300 of the country’s 1,100 municipalities exhibit almost no
government beyond the county seat (Valencia and Obando 2015). An
area comprising about a third of national territory lacks police, roads,
judges, prosecutors, and land titles. Residents—about 15 percent of the
­population—are disconnected from the national economy. As it gets turned
into a portable product with many buyers, coca is an ideal crop under these
conditions.
As this volume notes, Colombia, like Peru and Bolivia, experienced a
series of efforts to build a state presence in its agricultural frontier zones.
In Colombia, these left little behind. Colonization programs in the 1960s
built up departmental capitals and led to the founding of town centers
in what were wilderness areas, but government presence later stalled, or
receded, amid armed conflict and underfunding. As the work of Darío
Fajardo, Absalón Machado, Alfredo Molano, and others had found, vio-
lence and displacement caused much more “colonization” of frontier zones
than government programs (Fajardo 2014; Machado 2009; Molano 2013).
A “National Rehabilitation Plan” of rural infrastructure, agriculture, and
education investments began ambitiously in 1982, but collapsed from
neglect during the César Gaviria administration (1990–1994) (Mercado
1993; Gómez Muñoz 2016). The most ambitious program, a U.S.-backed
National Territorial Consolidation Plan (2005–2015), achieved initial suc-
cess but faded away during the Santos government amid failures of insti-
tutional coordination and the departure of key officials (Isacson 2013). It
became a small, assistentialist government agency that ceased to exist in
2016. “Colombia’s history is riddled with consolidation failures” is the flat
verdict of a 2015 paper from the U.S. National Defense University’s Center
for Complex Operations (Domínguez 2015, 91).
In Colombia’s frontier coca-growing areas, the result is stark. Coca-­
growing communities are poorer than the national average (87.3 percent
 Epilogue  163
of Colombia’s Multidimensional Poverty Index);. 58 percent lack electrical
connections (Garzón and Bernal 2017, 9). Few have even one prosecutor
or judge assigned to them (García Villegas et al. 2013, 12). Eighty percent
of coca is grown within 2 kilometers of a river, a symptom of the absence
or impassability of tertiary or farm-to-market roads (Rico et al. 2017).
A periodic trope in journalists’ coverage of these zones is the coca boom-
town, where all shops and restaurants, lacking access to national currency,
accept payment in coca paste weighed on countertop scales (Garibello 2004;
Vizcaíno and Fares 2016). The Catatumbo region of Norte de Santander
department, two hours’ drive from Cúcuta (metro population 800,000)
and Bucaramanga (1.1 million), is crossed by a mid-twentieth-­century road
along which coca is frequently visible. Built by oil companies and not main-
tained since, it is the region’s only paved surface, and it takes seven hours
to drive its roughly 100 miles. It would be even worse if communities didn’t
maintain it by themselves, charging a toll. State presence is so scarce in rural
Catatumbo that guerrilla groups put up billboards and banners, tagging
most structures with graffiti. In Playa Bonita, Bagadó, about two hours
upriver from the capital of Chocó department, townspeople told the author
in 2014 that they had not seen an official from the municipal mayor’s office
in three years. The author traveled for nine hours up the Caquetá river in
late 2016, seeing much coca but no evidence of state presence outside of
three port towns and a teacher who boarded the boat between towns1.
The peace accord proposes to bring the state, and basic services, to zones
like these. It would undo abandonment through “Development Programs
with a Territorial Focus” (PDET), a program getting underway in 16 regions
within the framework of the accord’s first chapter (rural development).
These call on the government to make 15 years of targeted investments in a
long list of priorities, from road construction to irrigation to food security
to land titling.
But the government’s commitment to implement the accord is uncer-
tain. President Juan Manuel Santos, who made peace his signal policy, is
in the last year of his term. His successor may not share this prioritiza-
tion of rural smallholders. Worryingly, Colombia’s finance ministry has set
aside only about US$53 million (160 billion pesos) for the PDETs in 2018
(Reyes Posada 2017). Either way, the recent record shows a state with a
record of falling far short on its commitments. (Recent examples include the
2007 National Territorial Consolidation Plan, the 2011 Free Trade Labor
Action Plan, and numerous local agreements signed with rural populations
demanding services in 2013 (Isacson 2013; Trumka et al. 2014)). Nobody
knows whether the FARC peace accord will be an exception. If it is, Colom-
bia could in fact achieve permanent reductions in coca cultivation.
Failure to follow through would represent a terrible setback. One of the
clearest historical lessons derived from the case studies in this book is that
when government “territorial consolidation” initiatives fail or are retracted,
campesinos (especially in Colombia) experience a deep sense of “betrayal”
164  Adam Isacson
or “abandonment” by the state—something their distrustful leaders actually
note. And as happened earlier, this might lead into intensified illicit activities
like drugs. Of course, these observations could be tempered by the fact that
throughout Latin America, it is a common political discourse to assert and
feel that “the state” is absent and has let people down, including in much
campesino discourse. Still, experience in Colombia indicates that failures
or half-hearted efforts to improve governance carry a large cost in state
legitimacy, making future efforts to address the problem even more difficult.
That, returning to Secretary Tillerson’s phrase, is “how it happened.” As
this epilogue is written in mid-2017, there is evidence of a glut of cocaine. In
some regions, growers are reportedly even burying coca paste as they wait
for prices to rise again, which could also be a result of market disruption as
FARC buyers demobilize (Semana 2017a). There is little evidence yet that
cocaine abuse is increasing to match supply. U.S. indicators show a modest
increase in reported use, but nothing close to Colombia’s doubling of pro-
duction (CNN Español 2017). The present moment may tell us much about
the extent to which supply can drive demand.
Mid-2017 is also the initial period for a new administration in ­Washington.
Officials and supporters in Congress are urging a return to the past. Secretary
of Homeland Security John Kelly praised “the amazing efforts that countries
like Colombia [before the recent policy change] put behind reducing the
production of cocaine in their case. And Peru is right along with them” (U.S.
Senate Appropriations Subcommittee on Homeland Security 2017). Forced
eradication remains the model.
Patience is thin for the modest experimentation tolerated during the
Obama years. Pressure on Bolivia will continue, likely intensifying over
the Obama administration’s periodic critiques. Little Bolivian or Peruvian
cocaine goes to the U.S. market, though, so pressure will remain diplomatic.
It is in Colombia where the coca issue could rise to a level high enough
to harm the bilateral relationship. Appearing with President Santos in May
2017, President Donald Trump cited “an alarmed—and I mean really a very
highly alarmed and alarming trend,” the increase in Colombia’s coca crop
(The White House 2017a). Assistant Secretary of State for International Nar-
cotics and Law Enforcement Affairs William Brownfield warned in August
2017, “Colombian leadership must find a way to implement a robust forced
manual eradication effort to create a disincentive to coca cultivation and
an incentive to participation in the government’s crop substitution effort”
(Brownfield 2017, 3).
Colombia’s fumigation suspension is unpopular within the Trump admin-
istration and the Republican Party. “The one that’s really concerning is this
massive surge in cocaine production in Colombia over the last year and a
half, which perfectly coincides with President Santos’s decision to suspend
aerial eradication,” Senator Marco Rubio (R-Florida) complained at a June
2017 Senate hearing (U.S. Senate Committee on Foreign Relations 2017).
Secretary Tillerson concurred: “So we have told them, though, we’ve got
 Epilogue  165
to get back to the spraying, we’ve got to get back to destroying these fields.
That they’re in a very bad place now in cocaine supply to the United States.”
Colombia confronts a widespread belief in Washington that it suspended
fumigation as a concession to the FARC. “They stopped aerial eradication
because they didn’t want to upset the peace deal and the FARC,” Rubio said.
“[T]he fact is that the peace process, peace agreement, included provisions
that FARC demanded. And that is to no longer eradicate and spray,” said
Republican Hal Rogers (R-Kentucky) (U.S. House Committee on Appropri-
ations 2017). Rogers chairs the subcommittee that drafts the annual foreign
aid bill.
This is inaccurate. The November 2016 peace accord maintains forced
eradication, even fumigation, for uncooperative coca-growers (High
Commissioner for Peace 2016, 114). Instead, Colombia stopped spraying
because the World Health Organization reported that the herbicide, glypho-
sate, might cause cancer (Guyton et al. 2015, 490–491). The narrative about
FARC appeasement nonetheless persists. So does the belief that Central
America’s MS-13 gang is moving tons of cocaine to the United States (U.S.
Senate Appropriations Subcommittee on State, Foreign Operations, and
Related Programs 2017). This is erroneous—cartels do that—but it feeds
into growing U.S. anti-immigrant sentiment.

“[W]e are going to continue to press them” on eradication, Secretary Til-

lerson has promised. (U.S. House Committee on Appropriations 2017)
In Colombia, this spurs fears that relations with the United States may
again become “narcotized” (Semana 2017b). As in the 1980s and 1990s,
bilateral conversations may begin with alarm over cocaine, rather than
areas of mutual cooperation.

This may not occur, though, for reasons of sheer incongruence. The Trump
administration is not backing up its get-tough rhetoric with money. Its 2018
foreign aid request to Congress would slash the State Department program
that funds coca eradication (U.S. Department of State 2017b). International
Narcotics Control and Law Enforcement (INCLE) would drop by $18
million, to a 17-year low of $125 million (Center for International Policy
2017). The U.S. Southern Command continues to lack resources; it remains
unable to interdict 75 percent of suspected drug shipments it detects in the
Caribbean and eastern Pacific. (U.S. Senate Committee on Armed Services
2017). Instead, the current “drug war” focus is Mexico, the source of almost
all heroin and methamphetamine consumed in the United States.
Whether resourced or not, a single-minded U.S. focus on drugs risks miss-
ing a great opportunity in Colombia. The peace accords briefly open the
door for the larger goal of improved territorial governance, which would
make forced eradication and other shortcuts unnecessary. A retreat in U.S.
assistance, along with a hostile diplomatic posture, could make that goal
more distant.
166  Adam Isacson
Where, then, is the approach to Andean cocaine headed? There is an opti-
mistic and a pessimistic scenario. Even the optimistic scenario, though, is
hardly a utopia.
Both operate within the constraint of cocaine’s continued illegality. As of
mid-2017, consuming countries are not moving toward legalizing or reg-
ulating cocaine. Even where restrictions on cannabis are loosening, popu-
lations tend to view cocaine as “different.” It is more addictive. Memories
of the 1980s–1990s crack plague remain fresh. The belief holds that legal-
ization would increase availability, and thus abuse and social costs. Any
speculation into the immediate future must assume that cocaine prohibition
will continue.
The pessimistic scenario is a return to the past. The Trump administration
and Bogotá’s post-Santos government double down on “shortcuts.” Forced
coca eradication, with renewed aerial spraying, jumps to more than 150,000
hectares per year. The central government, neglecting peace accord com-
mitments, leaves rural frontier areas abandoned. Farmers in coca-growing
zones continue to lack integration into markets, or even titles to their land.
The result in Colombia resembles the mid-2000s heyday of Plan Colom-
bia. Coca cultivation shrinks, but Colombia remains the largest producer in
the Andes. Growers adjust to eradication as a cost of doing business. They
replant, spreading to new parts of the country. Cocaleros organize, with help
from ex-FARC members outraged at the government’s failure to honor the
peace accord. As riot police clash with farmers, ungoverned territories fall
under the violent sway of new organized crime groups. Waves of massacres
and forced displacement follow.
In Bolivia, U.S. pressure on a post-Evo Morales government weakens the
“social control” model. Illicit cultivation increases, spurred by Brazilian
criminal groups. In Peru, eradication outpaces state presence, especially in
the volatile VRAEM region. Coca increases in new areas like Inambari-Tam-
bopata and Marañon. Nationwide, the amount of land planted with illicit
coca stays the same or creeps upward.
Inside the United States the Trump administration discards decades of
lessons learned. Attorney-General Jeff Sessions oversees mass incarcera-
tion of people possessing drugs, including Andean cocaine. People suffer-
ing from addiction are felons, ineligible for many jobs, unable to borrow
for education, and unable to access underfunded treatment. The addicted
population swells. Cocaine remains plentiful. Building more fencing along
the U.S.–Mexico border makes no difference. As today, most of the drug
crosses through under-manned land ports of entry (U.S. House Committee
on Appropriations 2016).
As cocaine supplies remain high, organized crime seeks new markets
around Latin America. Cocaine and crack spread throughout countries with
little history of prior use, like Colombia and Mexico. This worsens crime
and public health in cities. Once again, the pursuit of “shortcuts” leads to
an even worse place.
 Epilogue  167
In the optimistic view, governments resist the temptation to pursue short-
cuts. Political leaders show patience with innovative strategies that do not
yield immediate results.
The most important progress occurs in Colombia, the region’s largest illicit
coca producer. The post-Santos government honors peace accord commit-
ments to invest in the countryside. An incipient state presence establishes
itself in abandoned areas. Unlike the “rehabilitation” and “consolidation”
programs of the past, this time the effort is sustained for 15 years as foreseen
in the peace accords. Plans respond to local circumstances, and are devel­
oped and carried out in consultation with communities through an inclusive
(if fractious and messy) process. Campesino organizations, including those
formed by demobilized guerrillas, bring the effort out of municipal county
seats into rural hamlets, lending the effort the legitimacy that previous top-
down efforts lacked.
Regions still exhibit different dynamics of power relations, economics,
corruption, criminality, ethnicity, and culture. Progress is halting, and there
are many setbacks. Still, Trump administration hardliners, distracted by cri-
ses elsewhere, do not “press” too hard. With land titles, some basic services,
and better access to markets, more farmers find legal crops viable. Colom-
bian coca-growing edges downward. When they come, these gains show
more promise of being permanent.
Bolivia continues cooperating with rural communities to eradicate illicit
coca plots. The “social control” model gains acceptance in the Yungas coca
heartland. Illicit cultivation remains at—or a shade below—the current
level, which is a fraction of Peru’s or Colombia’s.
The optimistic scenario does not foresee major progress in Peru. The
country lacks a watershed event like Colombia’s peace accord, or a govern-
ment willing to buck orthodoxy like Bolivia’s. With state presence growing
and coca declining in Colombia, traffickers may target Peru. By the 2020s,
even without dramatic increases in cultivation, Peru’s coca crop could sur-
pass Colombia’s.
Even in the optimistic scenario, coca persists in the Andes due to orga-
nized crime-related state corruption. Still, throughout the Andes, illicit coca
production declines. Though the countryside lags behind cities, small farm-
ers see better governance and prosperity.
The cocaine trade does not give up, of course: it starts expanding beyond
the Andes. Venezuela’s cash-strapped government proves unable, or too cor-
rupted, to eradicate new plantings. Honduras and Chiapas, Mexico, where
discoveries of coca plantings shocked authorities in 2014 and 2017, see
more of the crop (El Heraldo 2017; Reporte Indigo 2014).
The U.S. government, responding to a clamor from communities hit by
opioids, increases funding for drug treatment. An important portion of
heavy users of cocaine turns its lives around, and demand for the drug stays
flat or declines. In any case, drug users come to prefer synthetics, instead of
products that must travel from South America.
168  Adam Isacson
In either scenario, the U.S. role diminishes. This is not just because aid levels
are likely to decline: the U.S. share of world coca consumption—­estimated at
barely over 30 percent in 2012—may continue to fall (Vargas Meza 2017).
Markets elsewhere—Brazil and other Latin American states, Russia, the
United Kingdom, Asia—have been growing more quickly. As these govern-
ments adopt what are likely to be divergent strategies, it is impossible to
predict what their net effect on the Andean coca economy might be.
Even the optimistic scenario for the next few years is far from perfect.
There is no such thing as a mix of policies that can eradicate all harms
caused by drug abuse. Still, improvement from the status quo is possible
with shifts in mindset and resource allocation.
The effect on the Andes’ illicit coca-growing zones would be dramatic.
Here, policies must build the presence of a state that provides public goods
and is held accountable for its actions. This, of course, is the urgent response
to more than coca in rural zones. It is fundamental for democracy, stability,
and international competitiveness. When farmers turn to illicit crops, it is
because they lack these.
If the goal is a functioning state in the countryside, progress is not mea-
sured in hectares of coca. That may be one indicator, but more important
ones focus on socioeconomic conditions in rural zones. Are children nour-
ished? Are schools making them proficient at relevant skills? Can farmers
transport products to market? Do they have clear titles and access to credit?
Do residents feel safe? Are emblematic cases of corruption and rights abuse
brought to justice?
The perennial dilemma is that such state-building strategies take a genera-
tion or more. They do not yield results in a single presidential or gubernato-
rial period—and rural zones have fewer voters anyway. They do not promise
the quick but temporary payoff that massive forced eradication provides.
The best way out of this dilemma runs through civil society. It takes orga­
nized, persistent citizens to force political leaders to stay engaged and resist
the hardline reflex when progress is slow. This means people at the local
level doing the hard work of making demands, reaching agreements, and
ensuring their compliance. It often means working with national and inter-
national accompaniment to stave off threats. Colombia offers a wealth of
civil-society experience from which to learn—though not to emulate exactly,
as local conditions vary and mistakes can be more instructive than successes.
The experience of members of the REDPRODEPAZ network in Colom-
bia during this century deserves study, particularly that of the Magdalena
Medio Peace and Development Program since the 1990s. At the local level,
examples of independent grassroots development abound, from the Diocese
of Quibdó in Chocó to the women of the Tejedores de Vida in Putumayo to
the producers’ associations of CISCA in Catatumbo.
With these ingredients, and over a long-term commitment, illicit coca
could cease to be an “alarmed and alarming” crisis. Rather than asking
“how could this happen,” U.S. policymakers could be celebrating remark-
able progress. It begins with resisting the temptation to take shortcuts.
 Epilogue  169
Note
1 The Catatumbo, Chocó, and Caquetá river examples come from fieldwork by the
author in February 2014, July 2016, and October 2016.

References
Andean Information Network. 2010. “Unresolved Equilibrium: Yungas Coca
Growers Test Bolivian Government.” Andean Information Network web-
site, October 19. Accessed June 25, 2017. http://ain-bolivia.org/2010/10/
unresolved-equilibrium-yungas-coca-growers-test-bolivian-government/.
Andean Information Network. 2017. “Bolivia’s New Coca Law.” Andean Infor-
mation Network website, March 3. Accessed June 25, 2017. http://ain-bolivia.
org/2017/03/ain-infographic-bolivias-new-coca-law/.
Asamblea Constitutiva de la Coordinadora de Cultivadores de Coca, Amapola y
Marihuana. 2017. “Comunicado de COCCAM – 28 de enero de 2017.” INDE-
PAZ website, January 28. Accessed June 25, 2017. www.indepaz.org.co/7145/
comunicado-de-coccam/.
Brownfield, William R. 2017. “Prepared Statement of: Ambassador William R.
Brownfield Assistant Secretary of State for International Narcotics and Law
Enforcement Affairs Hearing before the: Senate Foreign Relations Subcommittee
on Western Hemisphere.” U.S. Senate Committee on Foreign Relations website,
August 2. Accessed August 16, 2017. https://www.foreign.senate.gov/imo/media/
doc/080217%20Brownfield_Testimony.pdf.
Center for International Policy. 2017. “Colombia Military Aid.” Security Assistance
Monitor website. Accessed June 25, 2017. http://securityassistance.org/data/
program/military/Colombia/2000/2018/all/Global//.
CNN Español. 2017. “William Brownfield: ‘Hay una relación simbiótica entre la
producción de cocaína en Colombia y el consumo en EE.UU.’” CNN Español
website, May 29. Accessed June 25, 2017. http://cnnespanol.cnn.com/video/
cnnee-dusa-intvw-william-brownfield-lucha-contra-el-narcotrafico/.
Domínguez, Agustín E. 2015. “After the Negotiations: How Reconstruction Teams
Can Build a Stronger Peace in Colombia.” PRISM 5(4). Center for Complex
Operations website. Accessed August 13, 2017. http://cco.ndu.edu/Portals/96/
Documents/prism/prism_5-4/After%20Negotiations.pdf.
El Heraldo. 2017. “Incautan plantación Clandestina de coca en Esquipulas del
Norte.” El Heraldo website, May 1. Accessed June 25, 2017. www.elheraldo.hn/
inicio/1067087-465/honduras-incautan-plantaci%F3n-clandestina-de-coca-en-
esquipulas-del-norte.
Fajardo, M. Darío. 2014. “Estudio sobre los orígenes del conflicto social armado,
razones de su persistencia y sus efectos más profundos en la sociedad colombi-
ana.” Centro Nacional de Memoria Histórica website. Accessed August 13, 2017.
https://www.centrodememoriahistorica.gov.co/descargas/comisionPaz2015/
FajardoDario.pdf.
García Villegas, Mauricio, José Rafael Espinosa Restrepo, and Felipe Jiménez Ángel.
2013. “Instituciones y Narcotráfico: La geografía judicial de los delitos de drogas
en Colombia.” Dejusticia website, July. Accessed August 13, 2017. www.dejusti-
cia.org/files/r2_actividades_recursos/fi_name_recurso.590.pdf.
Garibello, Andrés. 2004. “Viaje hacia el foro de los olvidados.” El Tiempo web-
site, December 19. Accessed August 16, 2017. www.eltiempo.com/archivo/
documento/MAM-1551100.
170  Adam Isacson
Garzón, V. Juan Carlos, and Jose Luis Bernal. 2017. “¿En qué va la sustitución de
cultivos ilícitos? Principales avances, desafíos y propuestas para hacerles frente.”
Fundacion Ideas por la Paz website, July 4. Accessed August 13, 2017. www.
ideaspaz.org/publications/posts/1536.
Gomez Muñoz, Natalia. 2016. “El Desaparecido Plan Nacional de Rehabil-
itación (PNR) y el Esperado Posconflicto.” La Silla Vacía website, April 1.
Accessed August 13, 2017. http://lasillavacia.com/silla-llena/red-rural/historia/
el-desaparecido-plan-nacional-de-rehabilitaci-n-pnr-y-el-esperado.
González Posso, Darío. 2017. “Descriminalización y Tratamiento Penal Diferencial:
Articulación de los Puntos 1 y 4 de los Acuerdos de la Habana.” INDEPAZ
website, March 8. Accessed June 25, 2017. www.indepaz.org.co/wp-content/
uploads/2017/03/Implementacion-Punto-4.pdf.
Government of Colombia. 1999. “Plan Colombia.” Adam Isacson website. Accessed
June 25, 2017. https://adamisacson.com/files/old_cip_colombia/plancolombia.
pdf.
Government of Peru. 2015. “Proceso peruano de erradicación de coca es inédito
y capta atención de varios países.” Andina website, December 16. Accessed
June 24, 2017. www.andina.com.pe/agencia/noticia-proceso-peruano-erradica-
cion-coca-es-inedito-y-capta-atencion-varios-paises-589731.aspx.
Guyton, Kathryn Z., Dana Loomis, Yann Grosse, Fatiha El Ghissassi, Lamia Benbrahim-​
Tallaa, Neela Guha, Chiara Scoccianti, Heidi Mattock, and Kurt Straif. 2015. “Car-
cinogenicity of tetrachlorvinphos, parathion, malathion, diazinon, and glypho-
sate.” The Lancet Oncology 16(5). The Lancet website, March 20. Accessed June
24, 2017. www.thelancet.com/journals/lanonc/article/PIIS1470-2045(15)70134-8/
abstract.
High Commissioner for Peace. 2016. “Final Agreement to End the Armed Conflict
and Build a Stable and Lasting Peace.” Presidency of Colombia website, Novem-
ber. Accessed June 24, 2017. www.altocomisionadoparalapaz.gov.co/Prensa/
Paginas/2017/Mayo/El-Acuerdo-de-paz-en-ingles.aspx.
Isacson, Adam. 2013. “Consolidating ‘Consolidation’: Colombia’s Plan to Govern
Neglected Territories Stumbles.” Washington Office on Latin America website,
January 10. Accessed June 24, 2017. https://www.wola.org/analysis/consolidat-
ing-consolidation-colombias-plan-to-govern-neglected-territories-stumbles/.
Isacson, Adam. 2015. “Even if Glyphosate Were Safe, Fumigation in Colombia
Would Be a Bad Policy. Here’s Why.” Washington Office on Latin America
website, April 29. Accessed July 3, 2017. https://www.wola.org/analysis/even-
if-glyphosate-were-safe-fumigation-in-colombia-would-be-a-bad-policy-heres-
why/.
Machado, C Absalón. 2009. La reforma rural, una deuda social y política. Bogotá:
Universidad Nacional website. Accessed August 13, 2017. www.cid.unal.edu.co/
cidnews/archivos/ReformaRural.pdf.
Mercado, Bibiana Marcela. 1993. “En qué anda el PNR.” El Tiempo website,
October 17. Accessed August 13, 2017. www.eltiempo.com/archivo/documento/
MAM-243600.
Ministry of Defense of Colombia. 2015. “EE. UU. ratificó respaldo con el proceso
de paz y el posconflicto: Mindefensa.” Ministry of Defense of Colombia web-
site, November 20. Accessed June 25, 2017. www.mindefensa.gov.co/irj/portal/
Mindefensa/contenido/noticiamdn?idXml=/pccdesign/PortalMDN/Espanol/
NoticiasFecha/Noticias/Shared%20Content/Plantillas/wpc_plantillaNoti-
cias_1448038627440&date=20102015&guest_user=Guest_MDN.
Molano, Alfredo. 2013. Dignidad campesina, entre la realidad y la esperanza.
Bogotá: Ícono.
 Epilogue  171
Office of National Drug Control Policy. 2017. “ONDCP Releases Data on Cocaine
Cultivation and Production in Colombia.” Office of National Drug Control Pol-
icy website, March 14. Accessed June 25, 2017. https://defenseoversight.wola.
org/primarydocs/170314_colombia_coca.pdf.
Reporte Indigo. 2014. “Hallan cultivo de coca en Chiapas.” Reporte Indigo website,
September 11. Accessed June 25, 2017. www.reporteindigo.com/reporte/mexico/
hallan-cultivo-de-coca-en-chiapas.
Reyes Posada, Alejandro. 2017. “Es la paz, estúpido.” El Espectador website,
August 12. Accessed August 16, 2017. www.elespectador.com/opinion/es-la-paz-
estupido-columna-707688.
Rico, Daniel, Juan Carlos Garzón, and Julián Wilches. 2017. “Como afectar
el narcotrafico sin concentrarse en la mata de coca.” La Silla Vacia web-
site, March 7, 2017. Accessed August 13, 2017. http://lasillavacia.com/blogs/
como-afectar-el-narcotrafico-sin-concentrarse-en-la-mata-de-coca-60032.
Semana. 2017a. “No veo la estrategia para enfrentar los cultivos de coca.” Semana
website, March 3. Accessed June 24, 2017. www.semana.com/nacion/articulo/
experto-en-narcotrafico-daniel-rico-critica-politica-de-cultivos/517393.
Semana. 2017b. “¿Se volverán a ‘narcotizar’ las relaciones entre Colombia y Estados
Unidos?” Semana website, May 17. Accessed June 24, 2017. www.semana.com/
nacion/multimedia/juan-manuel-santos-y-donald-trump-hablaran-sobre-el-te-
ma-drogas-en-colombia/525553.
Serrano, Alfonso. 2016. “Amid Heroin Crisis, GOP Contenders Reframe Addiction
as a Health Crisis.” Al Jazeera America website, February 1. Accessed June 25,
2017. http://america.aljazeera.com/multimedia/2016/2/amid-heroin-crisis-gop-
contenders-reframe-addiction-as-a-health-crisis.html.
Stone, Hannah. 2012. “Back to Business as Usual as Peru Loses Progressive Drug Czar.”
InsightCrime website, January 15. Accessed June 24, 2017. www.insightcrime.
org/news-analysis/back-to-business-as-usual-as-peru-loses-progressive-drug-czar.
Trumka, Richard Louis, Luis Alejandro Pedraza, and Luis Miguel Morantes. 2014.
“Three Years of Non-Compliance with the Obama-Santos Colombia Labor
Action Plan.” Washington Office on Latin America website, April 7. Accessed
June 24, 2017. https://www.wola.org/analysis/three-years-of-non-compliance-
with-the-obama-santos-colombia-labor-action-plan/.
U.S. Department of State. 2000–2017. “Narcotics Control Reports.” U.S. Department
of State website. Accessed March 15, 2017. https://www.state.gov/j/inl/rls/nrcrpt/.
U.S. Department of State. 2016. “International Narcotics Control Strategy Report.”
Bureau of International Narcotics and Law Enforcement Affairs, U.S. Depart-
ment of State website, March 2. Accessed June 25, 2017. www.state.gov/j/inl/rls/
nrcrpt/2016/index.htm.
U.S. Department of State. 2017a. “International Narcotics Control Strategy Report.”
Bureau of International Narcotics and Law Enforcement Affairs, U.S. Depart-
ment of State website, March 2. Accessed June 25, 2017. https://www.state.gov/j/
inl/rls/nrcrpt/2017/index.htm.
U.S. Department of State. 2017b. “FY 2018 Congressional Budget Justification –
Department of State, Foreign Operations, and Related Programs.” Department
of State website, May 23. Accessed June 25, 2017. https://www.state.gov/s/d/rm/
rls/ebs/2018/pdf/index.htm.
U.S. House Committee on Appropriations. 2016. “Budget Hearing – United States
Customs and Border Protection.” U.S. House of Representatives website, March 1.
http://appropriations.house.gov/calendar/eventsingle.aspx?EventID=394403.
172  Adam Isacson
U.S. House Committee on Appropriations. 2017. “Department of State – Budget
Hearing.” U.S. House of Representatives website, June 14. Accessed June 25, 2017.
https://appropriations.house.gov/calendar/eventsingle.aspx?EventID=394899.
U.S. Senate Appropriations Subcommittee on Homeland Security. 2017. “Review
of the FY2018 Budget Request for the U.S. Department of Homeland Security.”
U.S. Senate website, May 25. Accessed June 25, 2017. https://www.appropri-
ations.senate.gov/hearings/review-of-the-fy2018-budget-request-for-the-us-de-
partment-of-homeland-security.
U.S. Senate Appropriations Subcommittee on State, Foreign Operations, and Related
Programs. 2017. “Hearing to Review U.S. Assistance for Central America.” U.S.
Senate website, May 23. Accessed June 24, 2017. https://www.appropriations.
senate.gov/hearings/hearing-to-review-us-assistance-for-central-america.
U.S. Senate Committee on Armed Services. 2017. “United States Southern Command
and United States Northern Command.” U.S. Senate website, April 6. Accessed
June 24, 2017. https://www.armed-services.senate.gov/hearings/17-04-06-unit-
ed-states-southern-command-and-united-states-northern-command.
U.S. Senate Committee on Foreign Relations. 2017. “Review of the FY 2018 State
Department Budget Request.” U.S. Senate website, June 13. Accessed June 25, 2017.
https://www.foreign.senate.gov/hearings/review-of-the-fy-2018-state-department-
budget-request-061317r.
UN Office on Drugs and Crime. 2002–2016. “UNODC and Illicit Crop Monitor-
ing.” UN Office on Drugs and Crime website. Accessed August 16, 2017. https://
www.unodc.org/unodc/en/crop-monitoring/index.html.
UN Office on Drugs and Crime. 2017. “Estado Plurinacional de Bolivia: Moni-
toreo de Cultivos de Coca 2016.” UNODC website, July 2017. Accessed August
12, 2017. https://www.unodc.org/documents/crop-monitoring/Bolivia/2016_
Bolivia_Informe_Monitoreo_Coca.pdf.
Valencia, León, and Camila Obando. 2015. “Los retos nacionales y territoriales del
posconflicto.” Fundación Paz y Reconciliacion website. Accessed June 24, 2017.
www.pares.com.co/wp-content/uploads/2015/02/Descargue-Informe-Completo.
pdf.
Vargas Meza, Ricardo. 2017. “Colombia y el mercado mundial de la cocaína.”
Razón Pública website, July 23. Accessed August 16, 2017. http://razonpublica.
com/index.php/conflicto-drogas-y-paz-temas-30/10422-colombia-y-el-mercado-
mundial-de-la-coca%C3%ADna.html.
Vizcaíno, John, and Melissa Fares. 2016. “Coca as Currency in Colombia.” Reuters
website, June 3. Accessed August 16, 2017. https://widerimage.reuters.com/story/
coca-as-currency-in-colombia.
The White House. 2017a. “Remarks by President Trump and President Santos of
Colombia in Joint Press Conference.” The White House website, May 18. Accessed
June 25, 2017. https://www.whitehouse.gov/the-press-office/2017/05/18/remarks-
president-trump-and-president-santos-colombia-joint-press.
The White House. 2017b. “Remarks by Vice President Pence and President Santos
of Colombia in Joint Press Conference.” The White House website, August 13. Accessed
August 16, 2017. https://www.whitehouse.gov/the-press-office/2017/08/13/remarks-
vice-president-pence-and-president-santos-colombia-joint-press.
Index
Afghanistan 13; see also Helmand
Province
agrarian reform 2, 8–9, 12, 54–9, 61–3,
73, 77, 133–5, 141–3, 149
agriculture 2, 7, 19–22, 24–5, 28, 43,
58, 61–2, 77–8, 116, 122, 125–7,
139, 142, 149; agricultural census
145: agricultural credit 72, 78;
agricultural development 10, 30, 77;
agricultural mechanization 68–9, 71
agricultural modernization 53, 55,
57; agriculturalists 21–2; see also
credit
Agrogüejar 153
Alliance for Progress 2, 9, 60–1, 64, 77,
134, 136, 141–2
alternative development 7, 12, 14,
154–5
Amazonia 7, 9, 11, 19–20, 25, 30, 42,
54–6, 64
Amazonian forest 7, 20, 119; see also
rainforest
Anslinger, Harry J. 113
Antioquía 14
Apurímac 41–2, 60
Ariari 9, 41–2, 133–41, 148–51, 154–5
armed conflict 1, 21, 30; see also
terrorism; warfare
autodefensas 76; see also rondas
campesinas
Avichure 139; see also Canaguaro
Ayacucho 15, 42
Belaúnde Terry, President Fernando 1,
56, 59–65, 72
Beni 42, 85–7, 92–6, 108
BID (Banco Inter-American de
Desarrollo) 1; see also IDB, Inter-
American Development Bank
biodiversity 7, 14, 20, 27, 111
Bogotá 16–7, 81, 136, 138, 150, 156
bonanza marimbera 13, 125; see also
marimbero boom
Brazil 13, 22, 50, 61
breadbasket 2; see also food security
cacao 14, 125, 128
Caja de Crédito Agrario, Industrial y
Minero 139; see also rehabilitation
program
Calamar 36, 40, 150
Cali 4
campesinos 11, 23, 39, 41, 43–4, 74,
76, 123–4, 126, 128, 143, 163;
Andean campesino 7, 43
Canaguaro 139–41, 149
cannabis 1, 13, 150; see also marijuana
capitalism 133, 152; capitalists 14, 62,
133; guerrilla capitalists, 150, 152
Caquetá department 9, 41–2, 114–15,
121–4, 126, 128, 153
Carretera Marginal de la Selva 1, 41,
87; see also Marginal de la Selva;
Marginal Highway
cartels 10–11; see also Cali; Medellín
cattle ranching 13, 25, 34, 36, 38, 69,
75, 125, 149
Cauca 114–15, 149, 151, 155
Ceja de Montaña, 56–8; see also
montaña
CENCIRA, 73, 75
census 73, 90, 133, 151
CEPAL 11; see also ECLAC
Chachapoyas 74
Chapare 1, 7–9, 11, 26, 42, 84–8, 92–6,
104–8
Chimoré 107–8; see also Chapare
Chinchao 75
174 Index
CIDA (Comité Inter-Americano para
Desarrollo Agropecuario) 56, 62, 67;
see also institutions
coca: coca cultivation 7, 19, 22–30, 41–3,
53, 76, 93, 95, 108, 115, 125–6, 127,
128, 150, 152; coca deforestation, 19,
23, 43; coca economy 53, 160, 168;
coca plantations 75, 93–4, 107, 114,
127–8, 167; coca planting, 4, 128, 167
coca paste 93, 115, 151,154; see also
pasta básica de cocaína (PBC)
coca/cocaine boom, 1, 4–5, 11, 55,
75–6, 84, 92, 95, 125, 128, 135, 149,
152, 163
cocaine 1, 7–8, 10–14, 43, 53–5, 61,
75–7, 84–5, 92–3, 95, 106–8, 115,
122, 128, 133, 135, 149–51, 155
cocaine factories 100–1, 107; cocaine
processing, 3, 5, 115, 122
cocaine trafficking 1, 3–4, 6, 11–12, 25,
54–5, 90–2, 100–1, 107–8, 124, 128,
149; see also narcotráfico
cocaleros 1, 7–8, 10, 14, 53, 55, 76,
84–5, 87–8, 93, 95, 105, 107–8, 124,
126, 152–3
Cochabamba 7–8, 84–7, 93–5, 102–8
coffee bonanza 116–18
coffee-growing regions 139
Cold War 2, 8–11, 15, 54, 59, 77, 84,
111, 133–5, 141
colonists (colonizers) 2, 9, 28–9, 41,
43, 55, 58, 72–6, 107–8, 124–5,
127, 135, 139–41, 149–51; colonist
dependence 55, 72–4, 77; colonos
2, 55, 135, 139–40, 151–2; see also
settlers
colonization: colonization front 29–30,
150–1; colonization projects 41–3,
55, 61, 63–5, 73–5, 77, 114, 136,
139, 142; colonization wedge 7,
33–4, 40; colonization zones 59, 73,
85, 87, 94, 96, 124; colonizing efforts
2, 9, 53, 56–8, 61, 127
Comisariato 139, 141
communists 2, 8, 61, 77, 134–6, 138,
150, 155
cooperativas 70–1; cooperatives, 3, 65,
70–1, 74–5
Corona project 9, 84, 87–8, 96, 108
counterinsurgency 11, 16, 133, 135,
150, 154
credit 2, 41, 43, 61, 64, 72, 75, 116,
125, 128, 134; see also loans
Cundinamarca department 144–5, 147,
151
debt 3, 70–1, 116, 139, 141; see also
loans
deforestation: dynamics 7; deforestation
front 7, 50; deforestation rate 6, 21,
24–30, 33–7, 44; deforestation wedge
20, 43; see also colonization wedge
democratization 3, 63, 155, 168
demographic effects 21, 41, 55, 72–3,
108, 125; see also Malthusian
development: agencies 53, 55, 63–4,
72, 74, 77; developmental concepts
3–4, 8, 10–12, 14, 17; development
plans 23, 55, 124–5; 12, 54; funds 39
developmentalism (desarrollismo) 54;
developmentalists
directed colonization 125, 141
displaced people 10, 115, 135, 155
drug trafficking 53–5, 76, 108, 124,
149; see also cocaine trafficking
ECLAC 8, 54–5; see also CEPAL
ecological impacts 7, 72, 75, 84, 93,
106, 154; ecosystem 20
economic development 9, 11, 21–2, 47,
95, 114–15, 121–3, 125, 127; see
also development
economists 9, 12, 14, 54, 150
Ecuador 21–2, 26, 41–3
El Retorno 125, 150; see also Guaviare
department
employment 73, 125, 127, 154
enganche 64; see also partija
Entre Ríos 89–90; see also Chapare
environmental degradation 21
eradication of coca 4, 6–7, 10, 19, 29,
95, 99, 126, 153, 160–1, 164–6
Espíritu 84, 87, 95, 107
ethnography 9, 11
extraction 14, 20, 37
FARC (Fuerzas Armadas Revolucionarias
de Colombia) 2, 7, 11, 14, 121, 126,
128, 135, 150–3, 155
farmable land 63; see also land use
fertilizers 68–9, 139, 146
food: food demand 117, 128, 153–4;
production 39, 56, 117; food
security, 43
Fordlandia 15
forest loss 20, 22–6, 28–9; see also coca
deforestation; deforestation rate
frontier colonization 44, 73, 133, 135,
146–9; frontier deforestation 22–3,
37–9; frontier roads 4, 39–40, 43–4;
see also deforestation wedge

frontiers 3–5, 9, 19, 23, 27–8, 30, 38, 75,
84–5, 87, 89, 95–7, 99, 108, 136, 144
fumigation of coca 10, 126, 153
Gaviria, President César 116–17, 162
geography 20–1, 84, 93–4
Guaviare department 9, 11, 25–6,
30, 41–2, 114–15, 121–6, 150, 153;
see also Calamar; El Retorno; San
José (del Guaviare)
Guayabero 134–5, 150–1; see also
Meta department
guerrillas 2, 11, 76, 124, 135–6, 151–3,
155; see also FARC, MIR, Sendero
Luminoso
hacendados 94, 115
haciendas 2, 54, 56–8, 84, 93–4, 150
Helmand Province, 13; see also
Afghanistan
high modernists 14; see also
modernization
highways 39, 56–7, 60–1, 64–5, 73, 87,
92–5, 104, 107
Huallaga region 4, 11, 38, 54, 60–1,
72, 75–6; Huallaga River 2, 36,
39, 56, 72; Huallaga Valley 1–2,
36, 39; see also VAH (Valle de Alto
Huallaga)
Huánuco 2, 54, 57, 60, 64–5, 75
ICAD (Inter-American Committee
for Agricultural Development)
62; see under CIDA, (Comité
Inter-Americano para Desarrollo
Agropecuario)
IDB (Inter-American Development
Bank) 1, 8, 60, 65–70; see also
development agencies; institutions
IICA (Inter-American Institute for
Cooperation on Agriculture) 67;
see also institutions
illicit crops 1, 19, 24–6, 29, 44–5,
50, 135, 154, 168; see also coca
cultivation
illicit economies 11, 53, 122, 124
immiseration 1, 22; see also marginal
lands
INCORA (Instituto Colombiano de la
Reforma Agraria) 41, 43, 156
indigenous groups 30, 52, 58, 114,
124–5, 133
industrialization 54, 67–8
inequality 9, 11–2, 30, 41, 56, 114, 129,
141, 161
Index  175
institutions 5, 13, 21, 54–5, 58, 63, 66,
69, 72–4, 81, 108, 139; see also
ECLAC; IDB, Inter-American
Development Bank; Organization of
American States (OAS); World Bank
insurgencies 10, 13, 57, 76, 135, 151,
155; see also guerrillas
intendancies (intendencias) 122, 126,
128; see also precincts (comisarías)
international development agencies 7,
53–5, 63–4, 66–9, 71–2, 77; see also
IDB, Inter-American Development
Bank; World Bank
irrigation 13, 57, 59, 63, 68, 163
jungle 2, 6, 53–4, 56–7, 60, 65, 73–4,
77; see also Oriente; Selva Alta;
Yungas
Kennedy, President John F. 2, 60–1, 66,
77, 134, 141–2; see also Alliance for
Progress
La Libertad 92, 150; see also Meta
department
La Violencia 9, 117, 134–7, 140–1, 144,
147; see also rehabilitation program
land: distribution 41, 43, 54, 57, 59,
62, 67, 117, 144–5, 149; tenure 56,
60, 62, 123–4, 142, 149; titles 2,
142, 144, 146–7, 149, 153, 162, 163,
166–8
land use 20–1, 24–8, 30, 96; see also
deforestation front; deforestation
wedge; frontier deforestation
landholders 57, 62, 124, 136, 142,
146; landholding elites 55, 61–2, 78;
landholdings 55, 61–2, 68, 77, 124,
144
landless campesinos 3, 23, 39, 61, 134,
142
Leoncio Prado 54, 74
Lleras Camargo, President Alberto 136,
138
Lleras Restrepo, President Carlos 116
loans 39, 57, 61, 64–6, 69–72, 139–42;
see also debt
López Michelsen, Senator Alfonso 142–3
Macarena National Park 33, 136, 147,
149–50, 153; Macarena colonists,
150; Serranía de la Macarena 141,
150–1
Malthusian 21; see also demographic
effects
176 Index
Marginal de la Selva 1, 41–3, 87; see also
Carretera Marginal de la Selva;
Marginal Highway
Marginal Highway 60, 64–5, 73; see
also Carretera Marginal de la Selva;
Marginal de la Selva
marginal lands 22; see also
immiseration
marijuana 12–3, 16, 115, 125, 150;
see also bonanza marimbera;
marimbero boom
marimbero boom 150; see also bonanza
marimbera
Medellín 4, 13, 115, 118; see also
Antioquia
Meta department 1, 9–10, 30, 41–2, 52,
114, 121, 123, 125–6, 133, 135–6,
138, 149–51, 153, 156; see also
Villavicencio
methodologies 5–6, 8, 55, 99
Mexico 11, 13
Michoacán 12
migration 10–11, 13, 27, 41, 64–5,
72–3, 75, 78, 84, 114, 121, 126–7,
134, 136; see also colonists; settlers
MIR (Movimiento Insurgente
Revolucionario) 57; see also
insurgencies; guerrillas
MNR (Movimiento Nacionalista
Revolucionario) 5
modernization 1–2, 8, 10–11, 13,
53, 55, 57–8, 77, 107–8, 134–5;
as modernity 13–14; modernizing
effects, 1, 3, 14, 39, 44, 104, 135;
modernizing efforts 59–61, 74, 125,
142
montaña 18, 56–8, 78, 81; see also
Selva Alta
Morales, President Evo 161, 166
narcotráfico 16, 82; see also cocaine
trafficking
National Agrarian Society 57; see under
SNA (Sociedad Nacional Agraria)
National Front 134, 136–8, 142–3
national security 3, 35, 99, 114, 128,
135
oil palm 47, 154
Organization of American States (OAS)
61, 135
Oriente 6, 43
Oxapampa 42
pacification 136, 139
palm oil 125, 149; palm plantations 14;
see also oil palm
Paraguay 4, 13
paramilitaries 124, 128–9, 154
partija 146–7; see also enganche
pasta básica de cocaína (PBC) 17, 55;
see also cocaine processing
peasantry 13, 61, 77, 104; see also
campesinos
peasants 1–2, 8–12, 14, 45, 53–8, 63,
72–3, 75–7, 115, 135–6, 139, 141–3,
148–50, 152–4; see also campesinos
photogrammetry 9; photography 9, 87,
96; see also Corona project
Pichis-Palcazú 25, 41, 52
Pojo Trópico 89–90; see also
Cochabamba
population growth 21, 23, 30, 74, 127;
see also Malthusian
Prado Ugarteche, President Manuel
57–8
precincts (comisarías) 122, 126; see also
intendancies(intendencias)
productivity (agricultural) 2, 7, 14, 21,
28, 72, 74, 125
Pucallpa 57
Puerto Asís 41–2, 127; see also
Putumayo department
Puerto Villarroel 98, 107; see also
Chapare
Putumayo department 9, 41–3, 114,
119, 121–4, 126–7; see also Puerto
Asís
Radio Sutatenza 125
rainforest 7–8, 14, 136, 150; see also
Amazonian forest
reforma agraria 2; see also agrarian
reform
refugees 3, 15, 72, 108
rehabilitation program 134, 136,
138–41, 144
revolutionaries 2, 53, 57, 63, 74–5,
133, 142, 150, 152; see also FARC
(Fuerzas Armadas Revolucionarias
de Colombia); guerrillas;
MIR (Movimiento Insurgente
Revolucionario); Sendero Luminoso
road construction 23, 64, 69, 72, 118,
139, 163
rondas campesinas 76; see also
autodefensas

San José (del Guaviare) 36, 38, 40, 125,
150; see also Guaviare department
Santa Cruz 1, 13, 27, 41–2, 85–7, 92–5,
102–4, 107–8
Santos, President Juan Manuel 163,
164, 166
security forces 153, 160
Selva Alta 53, 74–5; see also montaña;
Oriente
Sendero Luminoso 11, 76
settlers 8, 40, 43, 85, 89, 93, 97, 135,
139–41, 146–7; see also colonists
sharecropping 146; see also partija
Shinaota 89–90, 100; see also Chapare
smallholders 7, 14, 23, 152;
smallholdings, 142, 145, 161
SNA (Sociedad Nacional Agraria) 57
spontaneous colonization 65, 72–3,
95–6, 139–41
Tarapoto 39, 64
technology 46, 124–5; see also
agricultural mechanization
Territorios Nacionales 122; see under
intendancies(intendencias); precincts
(comisarías)
terrorism 11, 18, 76–7; see also warfare
Tingo María 8, 41, 56–7, 61, 64–5, 72,
74–5; see also Tocache
Tocache 4, 8, 15, 41–2, 61, 64–5, 70–2,
74, 77–80, 83; see also Huallaga
Valley; Tingo María
Tolima department 140, 151, 155
Tranquilandia 156; see also cocaine
factories
Index  177
Tulumayo 75
TVA (Tennessee Valley Authority) 60
Ucayali 57, 60, 65
underdevelopment 54, 81, 142
UNODC (United Nations Office on
Drugs and Crime) 22, 24–30, 41–2,
93, 128; UNODCP 154
unsustainability 25, 27, 128
urbanization 21, 35
Urubamba 60
VAH (Valle de Alto Huallaga) 4, 8,
53–5, 63–5, 72–7; see also Huallaga
Valley
Velasco Alvarado, General Juan 56, 63,
73–4
Viejo Caldas 139; see also coffee-
growing regions
Villa Tunari 84, 87, 95, 107; see also
Chapare
Villavicencio 40, 125, 150; see also
Meta department
VRAEM (Valle de los Ríos Apurímac,
Ene y Mantaro) 4
warfare 135–6; see also armed
conflict
World Bank 1, 45, 54–5, 57, 79, 92;
see also development agencies
yungas 6, 90, 93–4, 99, 107; see also
jungle; Oriente; Selva Alta
Yungas de La Paz 7, 9, 38, 42, 93–4,
99–100, 106, 161, 167