Citation
Outsmarting the state

Material Information

Title:
Outsmarting the state a comparative case study of the learning capacity of Colombian drug trafficking organizations and government drug enforcement agencies
Portion of title:
Comparative case study of the learning capacity of Colombian drug trafficking organizations and government drug enforcement agencies
Creator:
Kenney, Michael C
Publication Date:
Language:
English
Physical Description:
xiv, 332 leaves : ill. ; 29 cm.

Subjects

Subjects / Keywords:
Cocaine ( jstor )
Criminals ( jstor )
Drug trafficking ( jstor )
Law enforcement ( jstor )
Learning ( jstor )
Marijuana ( jstor )
Narcotics ( jstor )
Police ( jstor )
Smuggling ( jstor )
Terminal operations ( jstor )
Dissertations, Academic -- Political Science -- UF ( lcsh )
Drug control ( fast )
Drug dealers ( fast )
Drug traffic ( fast )
Political Science thesis, Ph. D ( lcsh )
Colombia ( fast )
City of Miami ( local )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Thesis:
Thesis (Ph. D.)--University of Florida, 2002.
Bibliography:
Includes bibliographical references (leaves 269-231).
Additional Physical Form:
Also available online.
General Note:
Printout.
General Note:
Vita.
Statement of Responsibility:
by Michael C. Kenney.

Record Information

Source Institution:
University of Florida
Holding Location:
University of Florida
Rights Management:
Copyright [name of dissertation author]. Permission granted to the University of Florida to digitize, archive and distribute this item for non-profit research and educational purposes. Any reuse of this item in excess of fair use or other copyright exemptions requires permission of the copyright holder.
Resource Identifier:
028813052 ( ALEPH )
50949508 ( OCLC )

Downloads

This item has the following downloads:


Full Text










OUTSMARTING THE STATE:
A COMPARATIVE CASE STUDY OF THE LEARNING CAPACITY OF COLOMBIAN DRUG TRAFFICKING ORGANIZATIONS AND
GOVERNMENT DRUG ENFORCEMENT AGENCIES













By

MICHAEL C. KENNEY














A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE REQUIREMENT FOR THE DEGREE OF DOCTOR OF PHILOSOPHY UNIVERSITY OF FLORIDA 2002














ACKNOWLEDGMENTS

As with the production, transportation, and distribution of psychoactive

substances, writing a dissertation is a collective endeavor. This study was completed with the support of many individuals and institutions. Fellowships from the National Science Foundation, the Tinker Foundation, and the Division of Sponsored Research at the University of Florida supported field work in Bogoti and various locations in the United States. A residential fellowship at the Center for International Security and Cooperation at Stanford University supported data analysis and writing.

While conducting research in Bogotd, I benefitted enormously from the Institute of Political Studies and International Relations (LEPRI) at the National University of Colombia. The director of the Institute, William Ramirez, provided a place to hang my hat between interviews, and numerous JEPRI scholars offered constructive criticism and friendship. For their assistance during this period, I thank Alvaro Carnacho, Nathanial Christie, Fernando Cubides, Francisco Guti~rrez, Andr~s L6pez, Gregorio Perez, Alvaro Valencia, and Carlos Veli.squez. During research trips to Washington, D.C. and Miami, Fiona Wright and Juan Carlos Valencia provided greater hospitality than I deserved.

Dozens of U.S. and Colombian officials gave generously of their time during

interviews. Without their willingness to share their knowledge, this study could not have been written. I owe a similar debt of gratitude to the former traffickers that spoke with me. For reasons of discretion, many of my respondents must remain anonymous. They





ii









know who they are and I thank them. I also thank Gabriela Rovillon Acosta, Maxine Downs, and Emilia Gioreva for their professional transcriptions of dozens of interviews.

The Center for International Security and Cooperation (CISAC) and Stanford University provided a stimulating and supportive environment to write most of this dissertation. For their support and friendship during my stay in Palo Alto, I thank Herb Abrams, Chris Chyba, Nisha Fazal, Karen Guttieri, Carole Hyde, Jacques Hymans, Wu Jun, James March, Dinshaw Mistry, Barry O'Neill, Barbara Platt, Woody Powell, Scott Sagan, Steve Stedman, and Jeremi Suri. Special thanks go to Lynn Eden, who served as my mentor at Stanford. Lynn scrutinized all of the chapters I wrote at CISAC, always tendering her enthusiastic support insightful criticism. Robert Axelrod, Martha Feldman, and Richard Valcourt were kind enough to share their perceptive comments on separate chapters.

At the University of Florida I have beneffited enormously from my association with professors and fellow students over the past several years. My dissertation committee chair, Philip Williams, and co-chair, Terry McCoy, guided me through graduate school and this study. I am enormously grateful for their friendship and support. The other members of my committee, Leann Brown, Joseph Spillane, and Larry Dodd, provided assistance throughout the project. For additional support at the University of Florida, I thank Leslie Anderson, Ralph DiMuccio, Errol Henderson, Lalitha Henderson, Peter Hildebrand, Goran Hyden, Michael Martinez, Geraldine Nichols, Richard Nolan, Hazel Phillips, Sandra Russo, Steve Sanderson, Marty Swilley, Les Thiele, and Debbie Wallen. For their friendship during my years in Gainesville, I thank Shawn Bird, Parakh Hoon, Ajent Shriar, Juan Carlos Valencia, Eric Cooper, Fabiano Toni, Scott Richards,



iii









Vilma Fuentes, Josh Gordon, Ed Greaves, Brian Gridley, Tom Nisley, Liz Oldmixon, Jim Conley, Lee Walker, and Fiona Wright.

I extend my deepest gratitude to my family for their love and support during the years that I have been working on this study. My parents, Peter, Chris, and Margaret Kenney, provided unwavering affection in life and graduate school. My biggest debt of gratitude goes to my wife, Emilia, who sacrificed as much as 1, if not more, to see this project through to its completion.







































iv














TABLE OF CONTENTS

pM,e

ACKNOWLEDGMENTS ......................................................................... ii

LIST OF TABLES .................................................................................. ix

LIST OF FIGURES ................................................................................. x

KEY TO ABBREVIATIONS ..................................................................... xi

A B STR A C T ........................................................................................ xiii

CHAPTER

I INTRODUCTION ............................................................................... I

Persistence of Illicit Drug Production in Colombia ......................................... 2
Explaining the Colombian Drug Dilemma ................................................... 6
Toward an Alternative Explanation: Organizational Leaming by Trafficking
E nterp rises ................................................................................... 8
Significance of Research ....................................................................... 10
R esearch D esign ................................................................................ 12
Overview of Study .............................................................................. 15

2 HISTORICAL OVERVIEW OF THE ILLICIT DRUG INDUSTRY IN
C O L O M B IA ..................................................................................... 17

Colombian Smuggling Tradition .............................................................. 18
Phase One- 1930s: Colombia as Transit Point ............................................. 20
Government Drug Enforcement Efforts During Phase One ...................... 21
Phase Two-Late 1930s to early 1960s: From Transit Point to Producer .............. 22
Herran Olazaga Brothers Enterprise ................................................. 24
Other Phase Two Trafficking Groups ................................................ 25
Government Drug Enforcement Efforts During Phase Two ...................... 26
Phase Three-Mid-1960s to late 1970s: Expansion of Colombian Drug Industry.... 27
Herrera Enterprise ...................................................................... 29
B ravo Enterprise ........................................................................ 31
Government Drug Enforcement Efforts During Phase Three .................... 32
Phase Four- 1980s to mid 1990s: Rise and Fall of the "Cartels .......................... 36



v









Emergence of the Core Organizations ............................................... 38
Smuggling Methods of Core Organizations ......................................... 40
Diversifying to Heroin ................................................................. 43
Government Drug Enforcement Efforts During Phase Four ..................... 44
Phase Five-Mid- I 990s to 2000: Atomization of Colombian Narcotics Industry.... 49 Government Drug Enforcement Efforts During Phase Five ...................... 52
C onclusion ....................................................................................... 54

3 TOWARD A THEORETICAL UNDERSTANDING OF ORGANIZATIONS
AND ORGANIZATIONAL LEARNING .................................................. 58

O rganizations ................................................................................... 60
Organizational Tasks .................................................................. 60
Environm ents ............................................................................ 61
Hostile environments .................................................................. 63
Organizational Structure ............................................................... 63
R outines ................................................................................. 64
Participants in Organizations ......................................................... 67
T echnology ............................................................................... 68
Defining Organizational Learning ........................................................... 69
Acquiring Information ................................................................. 70
Interpreting Information ............................................................... 71
Applying Information .................................................................. 73
Research Propositions .................................................................. 75
Alternative Explanations for Organizational Change ..................................... 75
Learning Ecologies ............................................................................. 78
"Productive" Learning ......................................................................... 80
Levels of Learning: Tactical vs. Strategic Adaptation .................................... 81
Learning under Ambiguity .................................................................... 85
Properties of Organizational Learning ...................................................... 90
Envirom-nental Hostility vs. Benevolence ........................................... 90
Organization Size and Decision-Making Hierarchies .............................. 91
C onclusion ...................................................................................... 93

4 ORGANIZATIONAL ANALYSIS OF THE ILLICIT DRUG INDUSTRY IN
C O LO M B IA .................................................................................... 95

Tasks of Colombian Trafficking Enterprises ............................................... 96
Understanding Tasks in Colombian Trafficking Enterprises ..................... 100
Environments of Colombian Trafficking Enterprises ..................................... 101
Trafficking Enterprises and Law Enforcement Agencies ......................... 101
Other Organizations in the Task Environment ..................................... 105
Structure of Colombian Trafficking Enterprises .......................................... 106
Centralized Decision-making Hierarchies .......................................... 107
"Flat" Decision-making Hierarchies ................................................ 109
R ole Specialization .................................................................... 112


vi









Compartmentalization ................................................................. 114
Routines in Trafficking Enterprises .................................................. 115
R outine Failure ......................................................................... 118
Participants in Colombian Trafficking Enterprises ....................................... 120
Participants as a Source of Risk in Trafficking Enterprises ....................... 122
Turnover in Trafficking Enterprises ................................................. 124
Promotion in Trafficking Enterprises ................................................ 127
Technologies in Colombian Trafficking Enterprises ..................................... 130
Practical vs. Scientific Knowledge .................................................. 133
C onclusion ..................................................................................... 136

5 ORGANIZATIONAL LEARNING BY COLOMBIAN TRAFFICKING
EN TER PR ISES ............................................................................... 138

Re-Statement of Organizational Learning Proposition ................................... 139
A First Cut: Quantifying Interview Results ................................................ 139
Acquiring Knowledge and Experience ..................................................... 143
Counter-Surveillance Activities ...................................................... 145
Intelligence Capabilities of Rodriguez Orejuela, Organization ................... 147
Informal Exchange of Information .................................................. 148
Drug Smuggling U: Acquiring Trafficking Knowledge in Prison .............. 150
Recording and Storing Information through Organizational Memories ....... 152
Interpreting Knowledge and Experience ................................................... 156
D iscovery Process ...................................................................... 157
Applying Knowledge and Experience ...................................................... 159
Adaptations in Drug Processing Routines .......................................... 160
Adaptations in Transportation Routines ............................................ 162
Adaptations in Distribution Cells .................................................... 167
Strategic Leaming in Colombian Trafficking Organizations ............................ 170
Diversifying into New Products and Markets ...................................... 171
Changing Organizational Structures: "Smaller is Better ........................ 173
Obstacles to Learning by Colombian Trafficking Organizations ........................ 175
Restricted Information Flows ........................................................ 175
Organizational Memories and Participant Turnover ............................... 177
Additional Barriers to Learning ...................................................... 180
C onclusion ...................................................................................... 182

6 ORGANIZATIONAL LEARNING BY DRUG ENFORCEMENT AGENCIES ... 184

A First Cut II: Quantifying Interview Results ............................................. 185
Drug Enforcement Intelligence Activities ................................................. 187
DEA's Role in Drug Enforcement Intelligence .................................... 192
CNP Drug Enforcement Intelligence ................................................ 194
Gathering Intelligence through Criminal Investigations ................................. 195
Criminal Informants ................................................................... 196
Surveillance and Undercover Operations ............................................ 197


vii









Manufacturing Drug Enforcement Intelligence ............................................ 198
From Intelligence to Operations ............................................................. 200
Changes in U.S. Drug Enforcement ......................................................... 201
Structural Adaptation Case #1: Toward an Administrative Merger of the
D EA and FB I ......................................................................... 202
Other DEA Structural Adaptations .................................................. 205
Reforming the Colombian National Police ................................................ 206
Structural Adaptation Case #2: Cultural Transformation of the CNP ........... 207
Adaptations in Drug Enforcement Operations ............................................. 211
Banshee or Buy-Bust? ................................................................. 211
Innovations in Conspiracy Investigations ............................................ 213
Innovations in Undercover Operations ............................................... 215
Competitive Learning Games ................................................................ 217
Factors that Affect Outcomes in Narco-Narc Learning Games ......................... 222
Playing Catch Up ....................................................................... 223
The Survival Imperative ............................................................... 224
The Enforcement Dilemma ........................................................... 226
Smaller in Size, Flatter in Structure .................................................. 228
The Red Tape Trap .................................................................... 232
C onclusion ..................................................................................... 236

7 CONCLUSION ................................................................................ 238

What Have We Learned? ..................................................................... 239
Alternative Explanations of Routine Change ....................................... 244
Implications for Counter-Drug Policy: Or How I Learned to Stop Worrying
and Love the Drug Trade ................................................................. 248
Diffusion of Trafficking Technologies .............................................. 251
What is to be Done? ........................................................................... 252
Legalization and Hann Reduction ................................................... 253
A Modest Proposal ..................................................................... 255
C onclusion ...................................................................................... 260

APPENDIX

A RESPONDENT DATA ....................................................................... 262

B QUESTIONNAIRE ............................................................................ 266

BIBLIOGRAPHY ................................................................................. 269

BIOGRAPHICAL SKETCH ..................................................................... 332







viii















LIST OF TABLES

Table Page

2-1 Colombian Drug Enforcement Indicators, 1974-1979 ................................. 33

2-2 Colombian Drug Enforcement Indicators, 1980-1995 ................................. 49

2-3 Colombian Drug Enforcement Indicators, 1996-1999 ................................. 54

2-4 Evolution of Illicit Drug Industry in Colombia ......................................... 56

4-1 National Government Agencies Involved in Drug Enforcement ..................... 104

4-2 Role Specialization in Colombian Trafficking Enterprise ............................ 113

4-3 Risk Reduction Routines in Colombian Trafficking Enterprises .................... 117

4-4 Number of Drug-Related Arrests by Colombian National Police, 19741998 (even years) .......................................................................... 125

5-1 Professional Affiliation and Nationality of Respondents ............................. 140

5-2 Do Colombian Trafficking Organizations Learn? Results from Interviews ........ 142

5-3 Transportation Innovations in Drug Smuggling Operations .......................... 166

6-1 Does the State Learn? Results from Interviews ........................................ 186

6-2 U.S. Drug Enforcement Intelligence Agencies, Types of Intelligence
and Scope of A ctivities .................................................................... 191

6-3 DEA Domestic Arrests by Class Level, Fiscal Years 1979 and 1982 .............. 213

6-4 Number of DEA Electronic Surveillance Orders Conducted and Facilities
Covered, FY 1990-1998 ................................................................... 215

6-5 Authorized DEA Personnel, Fiscal Year 2000 ......................................... 229





ix














LIST OF FIGURES

Figure pae

3-1 Simple Process Model of Organizational Learning ..................................... 70

3-2 Alternative Sources of Change in Organizational Routines ............................ 77

3-3 Complex Process Model of Organizational Learning .................................. 85

4-1 ")Wheel" Structure of Illicit Drug Network .............................................. 99

4-2 Management Levels within Large Trafficking Enterprise ............................ Ill

4-3 Stash House Inventory Routine .......................................................... 119

6-1 Organization of DEA Intelligence Division, circa 1996 .............................. 193

6-2 PHVA Management Cycle ................................................................ 208

6-3 Example of DEA Management Levels .................................................. 231

7-1 Sources of Change in Organizational Routines, Amended ........................... 245

7-2 Phased Evolution of Illicit Drug Industry .............................................. 248




















x













KEY TO ABBREVIATIONS AUC Autodefensas Unidas de Colombia (United Self-Defense Forces of
Colombia)

BINLEA Bureau of International Narcotics and Law Enforcement Affairs

BINM Bureau of International Narcotics Matters

CIA Central Intelligence Agency

CNN Cable News Network

CNNespafiol Cable News Network Espafiol CNP Policia Nacional de Colombia (Colombian National Police)

DAS Departamento de Seguridad Administrativo (Department of
Administrative Security)

DEA Drug Enforcement Administration

DOJ Department of Justice

ELN Ejercito de Liberaci6n Nacional (Army of National Liberation)

EPIC El Paso Intelligence Center

FAA Federal Aviation Administration

FARC Fuerzas Armadas Revolucionarias de Colombia (Revolutionary Armed
Forces of Colombia)

Fiscalia Fiscalia General de la Naci6n (National Prosecutor General)

FBI Federal Bureau of Investigation

FY Fiscal Year

GAO General Accounting Office



xi









G-DEP Geographic-Drug Enforcement Program

IEPRI Institute of Political Studies and International Relations

INTERPOL International Criminal Police Organization

IRS Internal Revenue Service

NADDIS Narcotics and Dangerous Drugs Information System

NAS Narcotics Assistance Section

NDIC National Drug Intelligence Center

NSC National Security Council

ONDCP Office of National Drug Control Policy

PCOC President's Commission on Organized Crime

Pseud Pseudonym

RCN Radio Cadena Nacional de Colombia (National Radio Network of
Colombia)

WOLA Washington Office on Latin America























xii














Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the Requirements for the Degree of Doctor of Philosophy OUTSMARTING THE STATE:
A COMPARATIVE CASE STUDY OF THE LEARNING CAPACITY OF COLOMBIAN DRUG TRAFFICKING ORGANIZATIONS AND
GOVERNMENT DRUG ENFORCEMENT AGENCIES By

Michael C. Kenney

May 2002

Chairpersons: Philip J. Williams, Terry L. McCoy Major Department: Political Science

Over the past several decades, Colombian drug trafficking enterprises have

transported large quantities of cocaine, heroin, and other psychoactive drugs to the United States in spite of government efforts to disrupt this illicit commerce. Contrary to much existing scholarship, this study argues that the persistence of the Colombian drug dilemma is, in part, due to the ability of criminal enterprises to alter their organizational structures and behavior in response to information and experience, store this knowledge in practices, procedures and performance programs, and select and retain innovations that produce satisfactory outcomes. Drawing on a multidisciplinary body of literature on organizational learning, and interviews with U.S. and Colombian officials, researchers, and former drug traffickers, the study demonstrates that smuggling organizations "learn," and in the process become increasingly difficult for drug enforcement agencies to identify and dismantle.


xiii









Drug trafficking enterprises are not the only organizational learners of interest in this study. Law enforcement agencies also alter their practices and procedures in response to knowledge and experience, thereby improving task performance and facilitating their survival as bureaucratic organizations. To understand interactions between these interdependent, rationally bounded actors, the study introduces the concept of competitive leading games, and discusses numerous organizational and environmental factors that tilt the playing field in favor of criminal enterprises. The study also includes an historical analysis of the Colombian drug trade dating back to the 1930s. Outsmarting the State is of broad theoretical and empirical interest to scholars and policyrnakers. The study contributes to diverse bodies of literature in political science, sociology, economics, organization theory, and criminology. The research has significant implications for existing U.S. drug and crime control policies in Colombia, Mexico, and elsewhere.


























xiv














CHAPTER 1
INTRODUCTION

Over the past two decades, the governments of Colombia and the United States identified drug trafficking as a national security threat to their respective countries; a threat they sought to eliminate by attacking individuals and groups engaged in production, transportation, and distribution of controlled substances. Despite devoting considerable resources to drug enforcement policies since the early 1980s, the two governments failed to reduce illicit drug flows between their countries. Indeed, according to the best available estimates, in Colombia the production of cocaine and heroin has increased substantially in recent years.

What accounts for this intractable dilemma? Why have the Colombian and U.S. governments been unable to impede the production of psychoactive drugs? Why are these governments unable to resolve a national security threat that emanates not from sovereign neighbors but from presumably weaker non-state actors? This study answers these questions through a comparative case study of Colombian trafficking enterprises and U.S. and Colombian drug enforcement agencies. Drawing on organization theory and extensive primary and secondary-source data, the study develops a learning-based explanation for the persistence of drug production in Colombia.

In this chapter, my aim is to set the stage for what follows. I begin with a

description of illicit drug dilemma in Colombia and conventional explanations offered for this conundrum. I continue with a brief discussion of the alternative explanation that


1





2


forms the heart of this study: the organizational leading capacity of Colombian trafficking enterprises. After emphasizing the contribution of the research to existing bodies of literature in political science, organization theory, and criminology, I describe the research design in detail. The chapter concludes with an overview of the remainder of the study.

Persistence of Illicit Drug Production in Colombia

Over the past two decades, the Colombian government has sought to eliminate the production and transit of illicit narcotics in its national territory. Working closely with the U.S. and other members of the inter-American narcotics control regime, the Colombian government has implemented "supply-reduction" programs that eradicate drug plantings, destroy drug processing laboratories, intercept the transportation of narcotics and the chemicals used to make them, and apprehend suspected drug traffickers and confiscate their illicit profits.

The costs of these programs, in terms of budget allocations and human personnel, are significant. Since the early 1980s, the Colombian government has spent several billion US dollars to implement supply-reduction initiatives within its national territory. While the Colombian government has received considerable anti-narcotics assistance from the U.S. and other foreign governments over the years, it has also invested a substantial portion of its own resources in the "war on drugs." Moreover, in recent years, the Colombian government's anti-drug expenditures have increased significantly. In the 1980s, Colombia's anti-narcotics budget varied between US$ 20 and 25 million per year, with the U.S. providing half this amount. In 1995 the Colombian government devoted US$ 900 million of its own funds to anti-drug efforts, and in 1996 this amount increased





3


to over US$ 1.3 billion. In 1997, the Colombian government allocated US$ 1.1 billion for counter-narcotics efforts, which represented 4.8% of the government's budget for that year (CNN, 1998a; Lee, 1989, p. 202; Republic of Colombia, 1997).

The human costs of the Colombian government's counter-narcotics efforts are even greater. Every year thousands of Colombian civilian and military officials participate in various phases of planning and/or implementing supply-reduction policies. The danger inherent in this line of work in Colombia is illustrated by the deaths of hundreds of government officials over the past 15 years. During one gruesome stretch from 1985 to 1988, Colombian traffickers, particularly those associated with the so-called Medellin cartel, killed over 400 police and military officers, dozens ofjudges and journalists, and several high-ranking government officials. While the number of government officials killed by drug traffickers has declined in recent years, death remains a persistent reality for those covering the anti-drug beat in Colombia. Since 1994, 31 counter-narcotics Colombian agents have been killed and 68 have been wounded in the line of duty (CNN 1998a; Lee, 1989, p. 202).

Notwithstanding the considerable financial and human costs of Colombian antidrug efforts, in recent years the government has enjoyed limited successes in implementing supply-reduction policies. One example of success has to do with recent increases in the eradication of illicit drug plantings. Over an 8-year period, beginning in 1986 and ending in 1993, the Colombian government eradicated an estimated 5,714 hectares of coca plantings. In 1996 alone, drug enforcement units eradicated 5,600 hectares of coca plantings and 6,028 hectares of opium poppy plantings. Four years later, they eradicated an estimated 47,000 hectares of coca plantings and 9,2542 hectares of





4


opium plantings.' Moreover, since 1991 Colombian anti-drug units, working closely with their U.S. counterparts, have captured the leaders of the Medellin, Cali and the North Valley drug trafficking organizations. These arrests are part of a government strategy aimed at dismantling the communications, transportation, and financial infrastructures of the larger trafficking organizations by targeting their senior managers. The goal of the socalled "kingpin strategy" is to disrupt the production of narcotics in Colombia by capturing, incarcerating, and confiscating the wealth of those individuals who direct the illicit drug industry (CNN, 1998b; Nieves, 1997).

Unfortunately, the kingpin strategy and increased eradication efforts have failed to resolve the Colombian narcotics dilemma. Indeed, according to estimates contained in the most recent International Narcotics Control Strategy Report, the estimated amount of potential coca production in Colombia increased from 50,900 hectares in 1995 to 169,800 hectares in 2001. 2 Corresponding with the increase in coca production has been increases in estimated cocaine production from 230 metric tons in 1995 to more than 580 metric tons in 2001. The estimated cultivation of opium poppies in Colombia, used to make heroin, has also increased in recent years. In 1995 an estimated 10,300 hectares of Colombian farmland were devoted to opium production. By 1997, the last year for which



1 These figures are contained in the International Narcotics Control Strategy Report, which is published annually by the Bureau of International Narcotics and Law Enforcement Affairs (BINLEA and BINM, various years, and ONDCP 2002). A caveat is warranted in presenting these and other drug-related data in this study. Generating estimates for illicit drug production and drug performance outputs remains, at best, an iniprecise science. The clandestine nature of narcotics trafficking and the politicized nature of drug enforcement make it difficult to produce accurate approximations for these activities. Throughout this study, figures regarding drug production and drug enforcement indicators are meant to illustrate trends rather than offer precise estimates.
2 Although this report contains what many specialists regard as the most accurate production estimates of illicit drugs available, several researchers who have conducted field studies argue that the report underestimates the amount of cocaine and heroin produced in Colombia (Clawson & Lee, 1996; Uribe Ramirez, 1997). A hectare is equivalent to 2.47 acres.





5


figures are available, the estimate increased to 13,572 hectares. The estimated production of opium gum, which is used to make heroin, increased slightly from 65 metric tons in 1995 to 66 tons in 1997. Over the past several years, the number of hectares in Colombia devoted to marijuana cultivation and the estimated production of marijuana itself have stabilized at approximately 5,000 hectares and 4,150 metric tons respectively (BINLEA, 1998, 2001).

VVhile the trafficking of illicit narcotics is more difficult to estimate than its

production, there are indications that the supply of the Colombian-produced cocaine to the U.S. has stabilized in recent years. For example, the U.S. Drug Enforcement Administration (DEA) estimates that the price and availability of cocaine in the U.S. have remained relatively constant. For the past several years, wholesale (per kilogram) prices for cocaine have remained within a $10,500-36,000 price band, while cocaine purity at the retail level (per gram) has stabilized at approximately 60% (DEA, various years). Meanwhile, the flow of Colombian-produced heroin to the U.S. appears to be rising. The DEA reports that the nationwide average purity for heroin at the retail level has increased from 27% in 1991 to 36% in 1996. Spearheading this trend is South American heroinmost of which originates in Colombia-which has a retail purity of 50.3% (DEA 1997). In addition, the amount of Colombian-produced heroin seized by U.S. and Colombian anti-drug authorities has also increased in recent years (BINLEA, 1998).

These developments have led the State Department to observe that in spite of

improved anti-drug efforts on the part of the Colombian government, the country remains "the world's leading producer and distributor of cocaine and an important supplier of heroin to the U.S." (BINLEA, 1998).





6


Explaining the Colombian Drug Dilemma

What accounts for the persistence of illicit drug production in Colombia at a time when the Colombian government has devoted more financial and human resources to its anti-drug efforts than ever before? Why has the Colombian government been unable to reduce the production and transit of illicit narcotics within its national territory? Over the past decade, scholars and policy makers have produced a cottage industry of literature that attempts to answer these questions. Several prevalent explanations have emerged from these studies. For example, many argue that the Colombian government is unable to resolve the narcotics dilemma because it lacks sufficient anti-drug resources. In an article that was published in the influential journal Foreign Affairs, Bagley argued that the Colombian government lacks the necessary resources to effectively counter the drug trade. In order for "real progress" to be made he insisted that the U.S. should provide Colombia and the other Andean governments "with the $2-3 billion annually that would be required to establish effective control over drug producing areas" (Bagley, 1988, pp. 87, 89). In a more recent study of the cocaine industry, Clawson and Lee argued that, in spite of increased international assistance over the past ten years, Colombia and the other Andean governments still lack the necessary resources to stop drug production and trafficking in their countries (Clawson & Lee, 1996, p. viii).

Similarly, U.S. and Colombian officials charged with implementing counternarcotics programs in Colombia maintain that they do not receive adequate resources to effectively carry out their mandates. The U.S. and Colombian officials interviewed in this research emphasize that insufficient resources were a significant impediment to achieving greater counter-narcotics success in Colombia. The Director of the Narcotics





7


Assistance Section (NAS) of the U.S. Embassy in Colombia complained of recent budget cuts to his program, wondering aloud, "How am I going to wipe it out?" (Moreno, 1997). Several officials from the Anti-Narcotics Directorate of the Colombian National Police (CNP) underscored the difficulties in covering an area larger than the state of Texas with only 2,500 agents (Buitrago, 1997).

A second prevalent explanation of the Colombian narcotics dilemma is that the Colombian government's will to implement effective policy has been seriously compromised by drug-related corruption. This charge has been leveled at the leaders of the political system and the government agencies responsible for carrying out anti-drug programs. In another article, Bagley argued that even if the Colombian government were to receive sufficient aid, it would still face significant obstacles to implementing more effective policies, principal among them being the significant level of drug-related corruption among government agencies and institutions (Bagley, 1989-1990).' Nadelmann claims that during the 1970s and 1980s, drug traffickers succeeded in corrupting "all levels of government, including a significant share of the legislature and cabinet-level ministers.... Even the Colombian military, which has remained a powerful and relatively corruption-free organization, could not resist the inducements offered by the drug traffickers" (Nadelmann, 1993, p. 279).

In the 1990s, drug-related corruption continued to flourish. Ernesto Samper's presidency (1994-1998) was severely weakened by credible allegations that his presidential campaign had received a US$ 6 million donation from several cocaine entrepreneurs based in Cali. In 1998, U.S. authorities seized a Colombian military plane

3 Andreas et al. (1991-1992) make a similar argument.






8


in Fort Lauderdale carrying 1,639 pounds of cocaine and 13 pounds of heroin, tarnishing the drug-fighting image of Colombian president, Andr~s Pastrana. The incident, which came a week after Pastrana's state visit to the U.S., prompted Colombia's Defense Minister to concede that the country's air force has been "seriously infiltrated" by trafficking organizations. U.S. and Colombian officials that interviewed in this research also stress the role of drug-related corruption in preventing more effective counternarcotics efforts on the part of the Colombian government (CNN, 1998d).

Toward an Alternative Explanation:
Organizational Learning by Trafficking Enterprises

Although these explanations help account for the persistence of drug production and trafficking in Colombia, they do not provide a fully satisfactory explanation of the drug dilemma. One reason for this is that they do not give adequate attention to the nonstate actors that manage the production and distribution of cocaine and heroin within the Western Hemisphere. Since the 1970s, Colombian trafficking enterprises have provided a steady flow of narcotics to the U.S. and Europe. In the process, these criminal enterprises have proven to be innovative and highly adaptive. These skills enabled them to maintain the profitability of their enterprises and survive persistent government efforts to destroy them.

The literature on the illicit drug industry in Colombia includes government reports journalistic accounts,5 and academic studies.6 Several recent studies have 4~ See the Drug Enforcement Administration (1994, 1995, 1996, 1997), FBI (1993), White House (1998), and Zabludoff (1997).
5 See Cai6n (1994), Castillo (1987, 1991, 1996), Castro (1996), Duzin (1994), Eddy et al. (1988), Garcia (1991), Giraldo (1992), Gugliotta and Leen (1989), Mermeistein (1990), Rinc6n (1990, 1994), and Shannon (1988).
6 Betancour and Garcia (1994), Clawson and Lee (1996), Domnbois (1998), Geopolitical Drug Watch (1998), Griffith (1997, 1998), Krauthausen and Sarmiento (199 1), and Tokatlin (1995).






9


highlighted the adaptability of the organizations that coordinate this industry. Griffith (1997) noted that "[t]raffickers are not only creative, they are also adaptive, changing methods and operatives depending on the success of counter-narcotics measures" (p. 84). Clawson and Lee argued that trafficking organizations have adapted to government efforts to reduce their profit margins by increasing production efficiency, improving and diversifying smuggling methods, increasing their "downstream" penetration of U.S. drug markets, and diversifying products and markets (Clawson & Lee 1996, pp. 10-12). Krauthausen and Sarmiento argued that pressure from drug enforcers compels traffickers to invent new methods and strategies for carrying out their illicit activities make the point that in the face of continuous law enforcement pressure (Krauthausen & Sarmiento, 1991, p. 37).

Government agencies and officials also recognize the adaptability of transnational trafficking groups and the threat they pose to U.S. interests. The 1998 National Security Council's International Crime Control Strategy stated that, internationalnl criminal organizations are adaptive and resilient, responding quickly to effective law enforcement pressure" (NSC, 1998). A recent General Accounting Office report links traffickers' adaptability to the failure of government drug enforcement efforts:

A key reason for U.S. countemnarcotics programs' lack of success is that international drug trafficking organizations have become sophisticated,
multibillion-dollar industries that quickly adapt to new U.S. drug control efforts.
As success is achieved in one area, the drug-trafficking organizations change
tactics, thwarting U.S. efforts (GAO, 1997, p. 3).

Although these studies suggest that the trafficking enterprises adapt their behavior in response to drug enforcement activities, they do not provide a conceptual framework for understanding this process. However, there is a multidisciplinary body of literature on






10


organizational learning that provides some clues as to how and why drug trafficking organizations may learn. This literature examines the behavior of firms in competitive markets and attempts to explain how they

build, supplement and organize knowledge and routines around their activities and
within their cultures, and adapt and develop organizational efficiency by
improving the use of the broad skills of their work forces (Dodgson, 1993, p.
377).

Significance of Research

This study is of broad theoretical and empirical interest to scholars and policy makers. The research contributes to diverse bodies of literature in political science, sociology, economics, organization theory, and criminology. Within political science, the research transcends traditional sub-disciplinary boundaries between international relations, comparative politics, and public policy. It focuses on actors (criminal organizations and law enforcement agencies) and issue areas (transnational organized crime and drug trafficking) often neglected in studies of international security, foreign policy, globalization, and state making.7 While political scientists have long been interested in organizational learning, the law enforcement agency and criminal enterprise have rarely been topics of interest. Instead, they have directed their attention towards more accessible subject matter, including government institutions, intergovernmental policy-making bodies, and military organizations. 8 These organizations are recognized by


7 Of course there are exceptions. Examples include Andreas (2000), Arquilla & Ronfeldt (1993, 2001), Desch et al. (1998), Farer (1999), Friman & Andreas (1999), Griffith (1997), Nadelmann (1993), and Williams (1994, 1998).
8 Examples of this literature include Allison (1971), Breslauer and Tetlock (1991), Brown (2000), Deutsch (1966), Dodd (1994), Eden (Forthcoming), Etheredge (1985), Evangelista (1988), Feldman (1989), Golant (1998), Goldgeier (1994), Haas (1990), Haas (1992), Hall (1993), Heclo (1974), Jervis (1976), Khong (1992), Le Prestre (1995), Leng (1992), Leeuw et al. (1994), Levy (1994), McCoy (2000), Mendelson (1993), Moltz (1993), Nagl (1999), Neustadt & May (1986), Nye (1987), Posen (1984), Reiter (1996), Sagan (1993), Sikkink (1997), Snyder (199 1), Stein (1994), and Weir and Skocpol (1985).








states and possess formal legal standing. Organizations that operate outside the rule of law-such as criminal enterprises and terrorist networks-have received scant attention from students of organizational learning. My goal is to develop a general theoretical explanation for how the learning capacities of these sovereignty free actors facilitate their defiance of state prerogatives, undermining government's ability to provide stability, order, and security for its citizens.

This research is also relevant to U.S. security interests, broadly defined.9 The

research has significant implications for existing U.S. drug and crime control policies in Colombia and other countries where transnational criminal organizations have become institutionalized, including Mexico, Russia, Italy, and Nigeria. The ability of criminal enterprises to alter their organizational structures and operations in response to law enforcement suggests that headhunting approaches to drug control, as exemplified in the Kingpin strategy, are unlikely to achieve satisfactory outcomes. This is not to suggest that such programs fail to impact trafficking organizations. Indeed, the Kingpin program demonstrated that when U.S. law enforcement and intelligence agencies coordinated effectively with their Andean counterparts cartel leaders could be apprehended and their


9 Some observers have legitimately questioned the trend by international relations scholars of labeling social problenLs such as the drug trade and crime "security" issues. As Michael Desch notes, "Broadening the definition of security in recent years was the result of the desire of many scholars and practitioners to direct more attention and resources to serious problems that previously had not received adequate consideration or money. As a rhetorical device for energizing scholarly and governmental discussion of, and directing attention to, the economic, environmental, social, and political problems affecting the post-Cold War Caribbean, this broader definition of security seems reasonable. But whether this rhetorical strategy will actually lead to a concerted effort to deal with these problems is unclear" (Desch 1998, p. 146). This point is well taken. However, two recent developments in Colombia and the U.S., have heightened the importance of this study to U.S. security concerns. The first development refers to the changing nature of the long-standing social conflict in Colombia, in which armed insurgents and paramilitaries have increased their participation in the drug trade in order to finance the escalation of their violent confrontation with each other and the Colombian state. This development produced alarm among Washington policymakers and led to substantial U.S. assistance for the Pastrana administration's controversial supply-reduction strategy, Plan





12


smuggling operations disrupted-at least temporarily. However, no sooner were these underworld luminaries removed than replacements emerged, eager to pick up where their predecessors left off. These post-cartel enterprises have learned from the mistakes of the groups that went before them, downsizing their operations to reduce their vulnerability to government anti-drug efforts. To the chagrin of state officials, existing groups have proven more difficult to investigate and dismantle, hampering law enforcers' ability to repeat the successes of the Kingpin strategy.

Research Design

This is a comparative case study of the criminal enterprises that coordinate the illicit drug trade and the state drug enforcement agencies charged with disrupting these illegal activities. The basic objective of this research is to determine whether, and to what extent, these sovereignty-free and sovereignty-bound actors learn as organizations. Organizational teaming refers to the process by which participants acquire, interpret, and apply knowledge to practices, rules, and procedures that guide organizational behavior. Organizational teaming theory offers a framework for understanding the behavior of trafficking enterprises and law enforcement and intelligence agencies. This processoriented approach illustrates how these state and non-state collectivities gather, analyze, and apply information to change their structures and operations in order to achieve satisfactory outcomes. Organizational teaming also helps explain why trafficking enterprises are "better" learners than state agencies, and how the illicit drug industry persists in Colombia in spite of government successes in dismantling a number of leading cartels.


Colombia. The second development refers to the tragic events in New York City of September I I th and





13


This case study is what Eckstein refers to as a "plausibility probe." Plausibility probes are preliminary tests of candidate-theories, or in this case candidate-hypotheses, which are undertaken before more rigorous-and costly--cross-national tests are attempted (Eckstein, 1992, pp. 147-149). This probe is undertaken to determine whether the organizational leading proposition is worth further consideration in future studies of the international drug trade and transnational organized crime. If the proposition does not receive empirical support in the case of Colombian trafficking organizations, then it is unlikely to have explanatory value in other cases. This is because many Colombian enterprises are among the most sophisticated, highly organized criminal groups in the world. Moreover, the learning ecology within which they operate is intensely competitive, largely due to anti-drug pressure exerted by the U.S. and Colombian governments. If Colombian trafficking groups are not learning in this environment and if this process does not help explain their ability to undermine government efforts to stop them, then it is unlikely that the explanation would be valid in other cases where criminal organizations are less sophisticated.

To verify learning by trafficking enterprises and drug enforcement agencies, I must demonstrate that changes in practices, procedures, and performance programs are due to information processing, rather alternative considerations, such as power and environmental selection. The process-oriented model of learning used in this study necessitates an in-depth examination of how these collectivities fiction, rather than superficial "measurements" of improved task performance. Learning cases should




their aftennath.





14


produce thick descriptions of how organizations acquire and manipulate information and experience.

To evaluate the explanatory power of organizational learning I have systematically gathered primary and secondary source data from a variety of sources in Colombia and the U.S. The theory of organizational learning and the method of structured, focused comparison guided data collection and analysis. I identify Colombian trafficking enterprises and U.S. and Colombian drug enforcement agencies as primary units of analysis that can be fruitfully studied through intra-case comparison. 10 From the leading literature I developed a semi-structured instrument for conducting interviews." Questions address the acquisition, interpretation, and application of knowledge and experience by Colombian trafficking enterprises and government counter-narcotics agencies. Other questions focus on their organizational structures, routines, and practices. Using (non-probability) snowball-sampling techniques, I interviewed seventy-six U.S. and Colombian government officials, researchers, and former drug traffickers 1 2

1 collected additional primary and secondary source data from government documents located at the National Archives in College Park, Maryland, the U.S. District Court in Miami, and several government offices and collections in Bogoti, Colombia. Secondary sources of information include government documents, transcripts '0 For discussion of the method of structured focused comparison, see George (1979), George & McKeown (1985), and King, Keohane & Verba (1994), especially p. 45. For discussion of intra-case comparison, see Lebow (2001).
" I chose the semi-structured inter-view format for their two principal strengths: structure and flexibility. The first ensures that all relevant topics will be covered during the course of interviews; the second allows researchers to purue leads into other topics as they arise. For discussion of these issues, see Bernard (1995).
12 The sample contained fifty-one government officials, many of whom worked in law enforcement agencies such as the Drug Enforcement Administration and the Colombian National Police, fourteen academic





15


and other court documents from the U.S. trials of prominent traffickers, memoirs of former traffickers and anti-drug agents, journalistic and academic studies, and newspaper accounts. These research methods have been applied in a number of studies on organized crime 13 and Colombian drug trafficking. 14 While there are an abundance of data available on the Colombian drug trade, these data are not always reliable. To minimize this problem, I cross-check information against independent, alternative accounts wherever possible and my own knowledge of the drug trade. 15 Overview of Stud

The remainder of this study proceeds as follows. Chapter 2 lays the empirical foundation for additional analysis by tracing the evolution of the illicit drug industry in Colombia over the past seven decades. This historical narrative centers on the organizational forms that coordinate the Colombian drug trade, and the efforts by U.S. and Colombian drug enforcers to stop them. Chapter 3 provides the theoretical framework for the research. Following an extended treatment of this routine-based, process-oriented approach to organizational learning, I discuss alternative explanations for organizational change, including power and environmental selection. The chapter also emphasizes psychological, organizational, and environmental impediments to organizational leading, suggesting that learning cannot be presumed to exist a priori but requires sustained application to the units of analysis.




researchers, six professional journalists, and five former participants that worked in different areas of the Colombian drug trade.
13 See, for example, Abadinsky (1983), Adler (1993), and Reuter (1983). 14 Arango (1988), Castillo (1991, 1996), Gugliotta and Leen (1989), and Krauthausen and Sarmiento (1991).
15 For a discussion of crosschecking procedures used in conducting this type of research, see Adler (1993).





16


This application begins with an organizational analysis of the illicit drug industry in Colombia. Using an open-systems perspective of organization theory, Chapter 4 examines the tasks, environments, structures, participants, and technologies that together form Colombian trafficking enterprises. Having delineated the structural and environmental conditions of these criminal firms, Chapter 5 examines the extent to which they gather, interpret, and apply information to practices, procedures, and performance programs that guide collective behavior. In recognition that trafficking enterprises are not the only organizational learners of interest, Chapter 6 applies the process-oriented learning model to U.S. and Colombian drug enforcement agencies, in particular the Drug Enforcement Administration and the Colombian National Police. The chapter describes how these bureaucratic institutions develop, manufacture, and act on tactical and strategic intelligence regarding trafficking enterprises. The chapter also develops the notion of competitive learning games to describe interactions between these sovereignty-free and sovereignty-bound actors. Chapter 7 concludes the study by summarizing the theoretical and empirical findings of the research, and discussing the implications of competitive leading games for U.S. and Colombian drug control policies and programs.














CHAPTER 2
HISTORICAL OVERVIEW OF THE
ILLICIT DRUG INDUSTRY IN COLOM131A Conventional wisdom maintains that the Colombian drug industry is a recent

phenomenon. A number of government reports, Congressional hearings, and journalistic and academic studies date the beginning of the Colombia's drug trade in the 1960s, or even the early 1970s. 1 Contrary to these accounts, Colombian smugglers have been involved in the transnational narcotics traffic since well before the 1960s. Indeed, criminal entrepreneurs have been smuggling cocaine and heroin from Colombia almost since national governments and international conventions formally banned these commodities. Since the 1930s, Colombian traffickers have built on their country's vibrant history in import and export contraband smuggling to organize for the purpose of participating in this illicit commerce.

This chapter presents an historical overview of the illicit drug industry in Colombia, focusing on the social organizations that produce, process, and export marijuana, cocaine, and heroin in Colombia. The chapter highlights the dynamic and fluid nature of these criminal enterprises, and lays the empirical foundation for the


' In 1980, a Drug Enforcement Administration official testified before the U.S. Senate Subcommittee on Investigations that prior to the mid- I 960s drug trafficking in Colombia was largely confined to importing cocaine for the domestic market and exporting small quantities of marijuana to neighboring countries (Clifford, 1980, p. 474). Several years later, the President's Commission on Organized Crime presented an influential report that traced Colombian involvement in the U.S. cocaine trade to the early 1960s (PCOC, 1986, p. 77). These views are accepted by a number of researchers that have written about the development of the Colombian drug trade, including MacDonald (1988, p. 27), Carney (2000, p. 8) and Chepesiuk (1999a, p. 56). In a recent edited volume on cocaine based on archival research, the contributor to the Colombia case study inexplicably dates the "formal appearance" of the country's cocaine industry to 1972 (Roldin, 1999, p. 176; also see Gootenberg, 1999).


17





18


chapters that follow. This chapter frames the evolution of the Colombian drug industry in five distinct, but overlapping, phases of development. Over the past seventy years, trafficking enterprises in Colombia have become increasingly sophisticated, increasing the challenge for U.S. and Colombian drug enforcement efforts. The chapter begins with a brief discussion of the country's long-standing tradition in contraband smuggling, and then proceeds through the different phases of the illicit drug industry. Within each period, attention is given to Colombia's position in the transnational drug trade and government efforts to dismantle trafficking enterprises.

Colombian Smuggling Tradition

Transnational. drug trafficking in Colombia has its roots in a long-tradition of contraband smuggling tradition dating back to Colonialism, when Spanish authorities sought to regulate trade within their Latin American dependencies. During the 17 th and 18h centuries, contraband smuggling was common throughout Nueva Granada, the area that encompasses contemporary Colombia. To avoid government duties and satisfy consumer demand, enterprising smugglers transported food, licor, cigarettes, machinery, and weapons across Riohacha, Santa Marta, and Cartagena. They also developed a number of maritime smuggling routes through Caribbean sea lanes (Dye, 1998; Grahn, 1997; Junguito & Caballero, 1982; L6pez Restrepo & Camacho Guizado, 2001; Useche, 1997). In Nueva Granada, smugglers developed a number of practices to evade or co-opt law enforcement authorities.

they [smugglers] utilized specific strategies and schemes that corresponded to local conditions. Along unguarded coastlines, for example, smugglers sailed
close to shore and deposited their goods at prearranged sites where buyers were waiting on the beach. In ports and near guarded anchorages, they used Spanish
intermediaries to arrange deals and bribe local officials while they waited offshore... When local administrators threatened this illicit coastal trade,






19


smugglers transacted their business at sea, out of sight of roving ships and
lookouts (Grahn, 1997, pp. 28-29).

Whiile many countries in Latin America and the Caribbean have distinguished traditions in smuggling contraband imports, Colombia is one of the few countries in the region to have a long history in contraband exports as well. For centuries, criminal entrepreneurs based in Colombia have smuggled sugar, manufactured goods and livestock to other countries in Latin America, including Brazil, Ecuador, Peru and Venezuela. In recent decades, Colombian smugglers have established a flourishing trade in contraband coffee with numerous countries in order to bypass quotas established by the International Coffee Agreement.

However, the most significant drug-free contraband export has been illegal emeralds. For years, workers in loosely-controlled government mines have stolen emeralds, often under cover of nightfall, and sold them to emerald traders. These traders work within close-knit, clan-based organizations that rely on secrecy, loyalty, and coercion to run their illicit operations, attributes that would prove useful to those who later expanded into other illicit industries. Emerald smugglers also learned the mechanics of selling commodities on national and international black markets, laundering foreign exchange, and the importance of hiring pistoleros to provide security to their operations. Over time, a number of leading esmeralderos, such as Gonzalo Rodriguez Gacha, used the knowledge and capital they acquired in the emerald trade to expand into the production and exportation of marijuana and cocaine (Mrango & Child, 1984, pp. 188189; Krauthausen, 1998, pp. 138-139; L6pez Restrepo 2000; Thoumni, 1995, p. 173; Thoumni, 1995, p. 173; Thoumni, 1992, p. 51; Betancourt & Garcia, 1994, p. 106).






20


Emerald traders were not the only Colombian smugglers to expand into the illicit drug industry. In the 1960s and 1970s, a number of black marketeers in cigarrettes, whiskey, and domestic appliances established themselves in marijuana and cocaine exports. Smugglers drew on contacts, customs, and capital they acquired in smuggling these commodities to diversify into marijuana and subsequently cocaine. The subsequent development of the marijuana and cocaine industry in Colombia was heavily influenced by the informal practices and procedures of these contrabandistas (Arango & Child, 1987, p. 130; Betancourt & Garcia, 1994, p. 107; L6pez Restrepo, 2000; Torres, 2000; Vald~s, 2000; Valencia, 2000).

The re-emergence of the Colombian heroin trade in the 1990s was stimulated by well-established cocaine traffickers seeking to diversify their illicit activities (FBI 1993; Farah 1997; Semana 1999; Zabludoff interview). In all three cases, criminal entrepreneurs drew on previously acquired contacts, customs, and capital to establish themselves in more lucrative endeavors. This knowledge and experience was critical for smugglers operating within highly uncertain and even hazardous business environments.

Phase One-1930s: Colombia as Transit Point2

In the early 1930s reports of captured Colombian drug smugglers began to appear in U.S. government documents and Colombian press accounts. In 1932 a smuggler was captured in the Panama Canal Zone carrying 25 grams of cocaine hydrochloride hidden inside a cartridge belt. The individual claimed that he had acquired the drug from a group of traffickers operating in Cartagena, Colombia. The following year, a different smuggler was arrested in the Canal Zone carrying 100 grams of cocaine hydrochloride,


2 This section draws on the work of Eduardo Sienz Rovner (1996, 1997), a Colombian economic historian that conducted archival research at the National Archives in Washington, D.C.






21


also allegedly obtained in Cartagena. The same year, El Espectador, a leading Colombian newspaper, reported that Colombian police officials confiscated a large consignment of illicit drugs in the home of a prominent Bogotano. A confidential U.S. State Department document from 1936 claimed that the Colombian islands of Providencia and San Andr~s were being used by smugglers for exporting illicit drugs and alcohol to the U.S. (Sdenz Rovner, 1996, pp. 69-70).

Though sketchy in the details, these government documents and press reports suggest that a number of smugglers based in Colombia were involved in illicit transnational networks linking European drug producers with Caribbean and North American consumers. They also indicate that Colombian ports located on the Caribbean sea, including Barranquilla, Cartagena, and Santa Marta, were transshipment points for opiates and cocaine produced in Europe. Given the paucity of data, I cannot say much about the organizational forms involved in these criminal actitivities. Nor is it possible to draw direct linkages between these early pioneers and smuggling groups in later decades. However, it is clear that at least some trafficking was taking place in Colombia as early as the 1930s, during which time the country served as a significant transit point for international drug flows (Mrango & Child, 1984, p. 176; L6pez Restrepo & Camnacho Guizado, 2001, pp. 3-4; Walker, 1989, pp. 75-76). Government Drug Enforcement Efforts During Phase One

In the 1 920s and 1 930s the government of Colombia came to view the

consumption of coca leaves by its citizens as a social problem (Walker, 1989, p. 73). Increased concern among national policy makers was reflected in a number of measures during this period. According to L6pez Restrepo and Camacho Guizado, Law I11 of





22


September 15, 1920 was the first piece of national legislation outlawing the consumption, trade, and production of certain narcotics. Eighteen years later the Directorate of National Hygiene enacted Decree 95 to regulate the sale of coca leaves to licensed pharmacies. In July 1938, the Colombian government established a new penal code that increased the criminal penalties for transacting in opiates. One month later the Ministry of Work, Hygiene, and Welfare was created with the partial purpose of applying nationallevel regulations concerning the drug traffic that Colombia had accepted at international narcotics control conferences sponsored by the League of Nations (L6pez Restrepo, 2000, p. 8; L6pez Restrepo & Camacho Guizado, 2001, p. 3; Sienz Rovner, 1996, p. 72).

In the mid- I 930s, drug interdiction emerged as an active component of the U.S. government's counter-narcotics strategy. Prior to this U.S. drug enforcement largely consisted of local officials intercepting drugs in port cities, sometimes with the help of federal customs and counter-drug agents. The first, tentative steps towards the internationalization of American drug enforcement were sporadic Coast Guard patrols in Caribbean sea lanes and the Gulf of Mexico. As William Walker reports, additional efforts were made to conduct aerial surveillance near the U.S.-Mexican border, apparently without much success. In general, early drug enforcement activities conducted by the Colombian and U.S. governments were haphazard, sporadic, and failed to impede smugglers' access to morphine and cocaine supplies (Walker, 1989, pp. 57, 140-141; Walker, 1992, p. 269).

Phase Two-Late 1930s to early 1960s: From Transit Point to Producer

Colombia's role as a transit point in the international drug trade was not destined to last for long. As early as 1939 the U.S. Treasury Department reported that opium was






23


being grown on large extensions of Colombian territory and that cocaine was being processed in-country and exported to Panama. According to this confidential report, Colombians of German extraction processed cocaine for export and transported the product by train to Cartagena and then by automobile or pack animals to the Caribbean ports of Tolfi, Cispatd, or Acandi, where it was hidden in banana boats and smuggled to Panama. Once in Panama, the drugs were transferred from steamship to small launches and transported to Puerto Pil6n, where they were guarded by a group of German smugglers until it was feasible to transport the cargo by automobile to the Canal Zone (Sienz Rovner, 1996, pp. 77-78). During World War II, contrabandistas from the Urabd region of Colombia formed smuggling networks that transported illicit drugs, whiskey, and cigarrettes through Panama and other Central American and Caribbean countries (L6pez Restrepo, 2000; Walker, 1989).

Colombia's shift from transit point to drug producer was facilitated by Cuba's emergence in the international drug trade following World War 11. In post-war Havana criminal organizations coordinated transnational networks composed of Latin American cocaine suppliers and French-Corsican and Italian heroin brokers. The networks used couriers, including seamen, stewards, passengers, and pilots, to smuggle drugs from South America to Cuba, sometimes by way of Central America. These human mules transported small quantities of coca paste in suitcases with specially modified compartments, a practice that remains popular among cocaine and heroin couriers today. In Cuba and Colombia processing groups refined the coca paste into cocaine hydrochloride, and sent along the finished product to the United States (MacDonald,



3 However, U.S. and Colombian officials disputed this report (Sienz Rovner, 1996, pp. 78-83).






24


1988, pp. 27-28; Sdenz Rovner, 1996, p. 89; Treasury Department, "The Illicit Narcotic Traffic in Cuba"; Treasury Department, May 27, 1954). Herran Olazaga Brothers Enterprise

In 1957 U.S. Federal Bureau of Investigation agents and Colombian police officials discovered a cocaine and heroin processing laboratory in Medellin. The processing lab was owned by a pair of Colombian twins that had been involved in the narcotics trade at least since 1948, over time developing into significant suppliers for Cuban trafficking groups. Rafael and Tomds Herran Olazaga ran their clandestine operation out of a furniture workshop that served as a front for their drug processing lab. The Herran Olazaga brothers purchased coca leaves from the Colombian department of Cauca and opium gum from Ecuador. They obtained the precursor chemicals used in the refining process through a nearby commercial drug lab that also served as cover. After processing, the brothers or their associates would smuggle the finished narcotics to Havana, where they would sell them to independent Cuban traffickers that would transport the drugs to the United States, Mexico, and other countries (Arango & Child, 1984, pp. 166-169; Montoya, 1959a, p. 3; Sdenz Rovner, 1996, p. 90).

The arrest of the Herran Olazaga brothers was important for law enforcement officials because it confirmed long-standing rumors that Colombian traffickers were smuggling narcotics into Cuba (Treasury Department). The case also indicates that Colombian traffickers were increasing their role in transnational drug networks. No longer were Colombian intermediaries merely purchasing cocaine and heroin from French and Italian traffickers and passing it onto to Italian-American and Cuban






25


traffickers located in Havana. At least a few Colombian traffickers had expanded into drug production.

Other Phase Two Trafficking Groups

The Herran Olazaga brothers were not the only drug smugglers operating in the Antioquia region during this period. According to Mrango and Child, by the mid-1I950s a number of Colombian con trabandistas decided to enter the drug trade. In making the shift from black market cigarettes, alcohol, and domestic appliances to drugs, these illicit entrepreneurs drew on existing resources, including their practical knowledge of smuggling methods and Caribbean maritime routes, investment capital, Cuban-based mafiaa contacts," and Colombian-based drug chemists (Mrango & Child, 1984, pp. 165166).

In the early 1 960s, the Venezuelan press began reporting about an "alarming invasion" of cocaine, heroin, and marijuana, much of it originating in Colombia. A number of smuggling rings transported drugs from Colombia into Venezuela by means of go-fast motor boats. At least one group of Colombian smugglers was affiliated with the well-known Italian mafiosi, Lucky Luciano. Unlike other rings, Luciano's group preferred to move its contraband into Venezuela through ground transportation. The cross border traffic was facilitated by numerous well-equipped processing laboratories in Medellin run by "German technicians" (El Espectador, 196 1la, p. 3).

While the Venezuelan press was reporting about the invasion of Colombian drugs, El Espectador published several articles detailing the growing traffic and consumption of marijuana in Colombia. Although marijuana had been produced in Colombia since the times of the Spanish conquest, consumption of the drug was generally concentrated






26


among socially marginal groups located in port cities and the sugar growing regions of Valle del Cauca. According to Thoumni, increases in domestic demand in the late 1950s and 1960s stimulated greater production of la mala hierba. A series of arrests by Colombian police officials in the summer of 1961 demonstrated the existence of an organized, city-wide network of marijuana traffickers in Bogotd (El Espectador, 1961ib, p. 9; El Espectador, 196 1c, p. 9; El Espectador, 196 1d, p. 9; El Espectador, 196 1e, p. 9; Thoumi, 1995, p. 126).

Government Drug Enforcement Efforts During Phase Two

Due to the lack of data regarding drug enforcement outputs during this period, such as numbers of participants arrested and kilograms of drugs seized, it is difficult to assess the effectiveness of government counter-narcotics efforts. By the 1 960s, officials from both countries expressed concern at Colombia's growing role in supplying transnational drug networks. However, drug enforcement programs in Colombia during this period were not particularly successful. In the aftermath of la Violencia, Colombian police and military authorities had more pressing concerns than stopping the drug trade, such as reducing political violence in the countryside (Walker, 1999, p. 146; Arango & Child, 1988, pp. 169-170). When drug-related arrests were made, the perpetrators were often low-level street dealers and drug addicts, ineffective targets for dismantling growing trafficking networks.

The fate of the Herran Olazaga brothers following their arrest in Cuba is

indicative of the failure of drug enforcement efforts during these years. Rafael Herran Olazaga fled Cuba immediately after posting bond from his 1956 arrest. While brother Tomi.s was eventually found guilty of drug trafficking, he served only one year in a






27


Cuban penintenciary before returning to Medellin. In subsequent years, law enforcement authorities suspected the two brothers of continuing their illicit drug activities from Colombia. However, they apparently avoided further prosecution. According to Arango and Child, the Antioquian trafficker died a successful businessman, owner of motels and several food establishments (Mrango & Child, 1984, p. 169; Treasury Department).

Phase Three-Mid-1I960s to late 1970s: Expansion of Colombian Drug Industry

In the mid and late 1 960s the involvement of Colombian smuggling groups in the international drug trade continued to expand. After the fall of the Batista regime in Cuba a number of Cuban and Italian-American organized crime figures based in Havana fled the island for the United States. While some of these traffickers remained active in the drug trade, Cuba's value as a transit point was much diminshed. Colombian smugglers with strong connections to Cuban trafficking networks found themselves well positioned to take advantage of this development. By 1965, Colombian trafficking enterprises were providing almost the entire cocaine supply for U.S.-based Cuban smugglers. Also during this period, Colombian smugglers supplied cocaine and marijuana to American, Chilean, and Mexican traffickers (Betancourt & Garcia, 1994, p. 57; Cervantes, 1980; Clifford, 1980, p. 474; Gugliotta & Leen, 1989, p. 22; Jonnes, 1996, p. 338; MacDonald, 1988, p. 28; President's Commission on Organized Crime [PCOC], 1986, pp. 77-78; Ruiz Hernindez, 1979; Sabbag, 1990; Thoumi, 1995, p. 126; U.S. House, 1977, pp. 3, 19;).

In the late 1 960s and early 1 970s, Colombian traffickers established their own transportation routes and wholesale distribution networks in the U.S. and Europe. Smugglers drew on their previous experiences and knowledge in trafficking illicit drugs and other contraband goods. A number of whiskey, cigarrette and marijuana smugglers





28


from the Antioquia and Guajira regions used their skills, contacts, and capital to enter the cocaine business in the 1970s. Enterprising marimberos and contrabandistas discovered that the smuggling infrastructures and methods used to transport marijuana to the United States worked well for cocaine. Marijuana distribution networks converted easily to cocaine. A number of marimberos made the switch gradually by pigbacking small quantities of cocaine on their marijuana loads and obligating their U.S. distributors to sell both products. Transportation methods used by marijuana smugglers, such as flying private planes of various sizes to clandestine airstrips along the Altantic Colombian coast, loading them with drugs, and returning to the U.S. with the illicit cargo, also adapted easily to cocaine trafficking. Indeed, a number of smuggling innovations pioneered by marimberos later became popular among cocaine traffickers, such as air dropping specially wrapped packages of marijuana in the Caribbean sea, and using large seafaring freighters, known as "mother ships," to transport huge quantities (upwards of 100 tons) of marijuana to prearranged locations 200 miles off the U.S. coastline, where private yachts and go-fast motor boats would converge on the mother ship, receive and transport smaller qualities into the U.S. (Cervantes, 1980, pp. 47-48; Krauthausen & Sarmiento, 1991, p. 147; MacDonald, 1988, p. 28; Nieves, 1997, p. 3; Passic 2000; PCOC, 1986, p. 78VelAsquez Romero, 2000; Vald~s 2000).

By the mid- 1 970s, Colombian trafficking enterprises were recognized as

preeminent players in the inter-American cocaine industry. In a short period of time, they transformed themselves from mere intermediaries for trafficking groups based in other countries to vertically integrated, production-transportation-wholesale distribution networks. Along the way Colombian enterprises developed practices, procedures, and





29


technologies for producing and transporting illicit drugs that would be used over the next two decades. Small, makeshift laboratories processed coca paste into cocaine hydrochloride. Human couriers, private aviation aircraft, and maritime vessels transported drugs across the Caribbean by sea or air. Navigation technology located drug drop-off points in the high seas. Sophisticated communications equipment, including Bearcat scanners, CB, VHF and single side band radios, marine radios, and telephone paging systems, were used to communicate within smuggling groups and to monitor law enforcement activities (Cervantes, 1980, pp. 80-81; Martin, 1978, p. 61; Battard, 1978, p. 75; U.S. House, 1977, p. 9; Wille, 1978, p. 88).

The development of Colombian trafficking enterprises did not go unnoticed by state authorities. By the mid-1970s, U.S. drug enforcers were alarmed at the growth of Colombian wholesale distribution rings operating in New York, Miami, and other large urban markets. In New York City, several Colombian organizations were believed to control the cocaine trade in Queens and Manhattan. These groups, led by such mythical figures as Benjamin "the Black Pope" Herrera, Griselda "the Black widow" Blanco, and Veronica "the queen of cocaine" Rivera, developed many of the smuggling practices and procedures that the cocaine cartels would refine and make famous ten years later (Nieves, 1997, p. 5; Semana 1987). The following sections briefly describe the structure and modus operandi of two of these enterprises. Herrera EnteEprise

The Herrera trafficking network was based in Cali, Colombia, and led by

Benjamin Herrera, the son of an alleged cocaine chemist who had been involved in the drug trade for a number of years. This family enterprise featured numerous brothers,





30


sisters, cousins, and in-laws as participants. However, the Herreras also recruited outside the family for professionally trained participants. In all, the network contained more than ninety members and exported approximately forty kilograms of cocaine a month to New York and Miami. The Herreras acquired coca paste in Bolivia, Ecuador, and Peru. Using human couriers carrying ordinary suitcases, the enterprise transported it to international airports in Cali and BogotA, where the couriers whisked past compliant customs officials "on the organization's payroll." The coca paste was then transported to one of several processing laboratories owned by the organization, where it would be processed into cocaine hydrochloride. One processing lab captured by Colombian authorities outside of Cali contained a twenty-five ton mechanical press for packing cocaine into fine sheets and other equipment, materials and precursor chemicals worth approximately US$800,000. Among the eight persons arrested at this lab was a chemistry professor from Santiago University in Cali (Gage, 1975a, p. 26; Gugliotta, 1987; Moreau, 1976; Semana, 1987).

Fully refined cocaine was distributed to one of three Herreras operating in

Barranquilla, Bogoti, or Medellin. These transporters would export the drug to the U.S., again relying on drug-running couriers holding small quantities, generally between two and four kilos. Couriers sometimes posed as students, carrying books that contained cocaine secreted sheets. These "students" were paid between US$ 500 and US$ 1,000 per trip, plus expenses and a new suit of clothes to be used when transporting their pedagogical materials to New York. False documents, including student visas, were obtained by a family member that also served as a "liaison" with several of the network's U.S.-based buyers. In addition to customs officials, the enterprise also received





31


protection from officials in the Colombian police and judiciary, as well as "leading members of Colombian society" that invested in cocaine shipments (Gage, 1975a, p. 26). Bravo EntpMris

The Bravo network contained several large and loosely connected drug rings that smuggled large quantities of cocaine, heroin, and marijuana from Colombia to the U.S. From 1972 to 1974, the Bravo enterprise distributed drug shipments throughout North America, including Manhattan, Queens, Miami, Los Angeles, San Antonio, Texas and Toronto. Large cocaine shipments, which in those days meant anything greater than four kilos, were transported in containerized cargo, speed boats, and private planes using a variety of routes, including Panama, Mexico, Puerto Rico, Canada, and even Munich, West Germany. Smaller quantities of cocaine and heroin were hidden inside luggage or clothes and transported by couriers. The Bravo network developed a number of creative smuggling methods, including soaking clothes in a liquid solution of cocaine hydrochloride, swallowing drug-filled prophylactic condoms, and hiding cocaine in falsebottomed wine bottles, painting frames, hollow ski poles and wooden hangars, specially modified bras and girdles, and a dog cage, complete with live canine. On reaching the U.S., the cocaine was distributed by members of the Bravo network, which included approximately 150 couriers, brokers, and distributors working in the New York metropolitan area alone. Although U.S. police officials eventually captured, prosecuted and convicted over a dozen members of what federal prosecutors described as "the biggest Colombian narcotics organization ever uncovered," several entrepreneurs, including Alberto Bravo, Carlos Bravo and Griselda Blanco, were able to continue their





32


operations from Colombia (Gage, 1975b, p. 24; Hudson, 1974, pp. I & 18; Lubasch, 1976, p. 21; New York Times, 1976, p. 20; Nieves, 1997, p. 5). Government Drug Enforcement Efforts During Phase Three

In response to the growth of Colombian trafficking enterprises, the U.S. and

Colombian governments increased their drug enforcement activities, achieving a number of successes in the latter half of the 1970s. Police agencies from both countries conducted criminal investigations that disrupted several trafficking enterprises. Dozens of participants from the Herrera and Bravo networks were captured, prosecuted, and convicted of drug-related offenses. In 1975, in what amounted to the largest cocaine seizure to date, Colombian police officials captured 600 kilograms of cocaine in a small plane at the Cali airport (Frontline, 2000).

Notwithstanding these achievements, drug enforcement in Colombia was

inconsistent during this period, as reflected in the wide disparity of annual performance indicators (see Table 2-1 below). For the entire year following the Cali airport bust, Colombian officials managed to capture only 138 kilograms of cocaine. In 1977, the amount of cocaine captured by Colombian authorities sunk to a dismal 32 kilograms. In the U.S., complex conspiracy cases targeting cocaine operations, such as the Herrera and Bravo networks, were few and far between. In the 1970s, many drug enforcement managers were wary of these resource-heavy investigations, preferring low-cost "buybust" operations targeting street dealers (CNP, 2000, p. 540; Nieves, 1997, p. 5).4

Drug enforcers mediocre counter-cocaine performance during the 1970s reflected a widespread belief among U.S. and Colombian policy makers and police officials that marijuana, rather than cocaine, was the primary drug threat facing their respective
4 Buy-bust operations are discussed in detail in Chapter 6.





33


countries. Consequently, drug enforcement resources were directed towards eradicating Colombia's booming marijuana industry. The year of the Cali cocaine bust, Colombian authorities captured 78,000 kilograms of cannabis and destroyed almost 1.5 million marijuana plants. In 1977, the year Colombian officials netted 32 kilograms of cocaine, they also captured 187,077 kilograms of marijuana (see Table 2-1).

Table 2-1 Colombian Drug Enforcement Indicators, 1974-1979
Year Persons Cocaine Marijuana Coca Marijuana Druglabs
apprehended interdicted interdicted plants plants destroyed (kilograms) (kilograms) destroyed destroyed
1974 1,305 164 90,000 0 37,500 6
1975 1,484 699 78,000 0 1,494,000 10
1976 769 138 27,000 0 25,000 15
1977 945 32 187,077 1,000 805,700 14
1978 555 194 158,272 4,195 431,614 34
1979 457 1,252 325,656 185,700 398,255 30
Source: CNP (2000)

Even regarding marijuana there was ambivalence among Colombian and U.S.

policy makers during much of the 1970s. In neither country did large groups of citizens view marijuana consumption as a significant public health problem, and in Colombia at least political elites did not see the growing drug industry as a threat to the country's national security interests. In the mid-1970s, when officials from the Ford and Carter administrations pressured their Colombian counterparts about drug trafficking within their national borders, Colombian President Alfonso L6pez Michelsen was quick to remind them that U.S. consumers rather than Colombian suppliers represented the source of the problem. Even within the U.S. policy community officials disagreed over the degree to which the drug trade, in particular marijuana, threated U.S. interests. President Nixon's fight against crime agenda produced some changes in federal drug enforcement policy (see Chapter 6), but the widespread moral fervor associated with the cocaine wars






34


remained several years away (Massing, 1998; Randall, 1992, p. 246; Tokatlian, 1988, p. 139; Tokatlian, 1990, pp. 293-294; Walker, 1999, pp. 148-149).

Notwithstanding this ambivalence, the 1970s witnessed an increase in the U.S. government's drug enforcement presence in Colombia. Prior to the Nixon Administration, U.S. drug enforcers, including the FBI and the Bureau of Narcotics and Dangerous Drugs, operated sporadically, on a case-by-case basis, in South America. With the creation of the Drug Enforcement Administration in 1973, the Nixon Administration ushered in a new era of the internationalization of U.S. drug enforcement policy. By the late 1970s, the new "super" drug enforcement bureau was active in dozens of countries, and established a permanent presence in Colombia, working closely with the Colombian National Police (CNP) (Nadelmann, 1993).

With the change of Colombian presidential administrations in August 1978, the Colombian government's attitude towards drug trafficking shifted. President Julio Turbay implemented a more aggressive counter-narcotics policy than his predecessors following allegations in the U.S. media that two of his ministers and his own family were connected to the illicit drug industry. Following his inauguration, President Turbay placed the Guajira Peninsula, which had long served as a major production center for marijuana and transit point for marijuana and cocaine, under the jurisdiction of Colombia's armed forces. By the end of 1978, 10,000 army troops were in the region engaged in the manual eradication of marijuana. Operation Fulminant, as the anti-drug mission was called, produced immediate results: 3,500 tons of marijuana were seized, which represented an estimated loss of US$ 70 million to Colombian smugglers, approximately seventy drug running airplanes and seventy seafaring vessels were






35


captured, and 1,000 Colombians and Americans were arrested (Krause, 1979; Thoumni, 1995, p. 210; Walker, 1999, p. 150).

Following on the heels of the DEA's Operation Stopgap, which targeted

Colombian and American marijuana smugglers in the Caribbean, Fulminant temporarily disrupted the marijuana industry in Colombia. However, the operation was short lived, in part due to a lack of support from the Colombian military. The generals were concerned that too many resources were being shifted away from more pressing anti-guerrilla operations. They also worried about the debilitating impact of drug-related corruption on field officers involved in Fulminant (Orjuela, 1990, p. 218).

However, the most significant impact of the marijuana eradication and

interdiction campaigns undertaken by U.S. and Colombian drug enforcers in the late

1 970s was an unintended one. In successfully targeting a number of leading marimberos, Operations Fulminant and Stopgap substantially increased the risks and costs associated with marijuana trafficking. Given limited state resources, cocaine did not receive as much attention from drug enforcers as marijuana, providing an additional incentive for smugglers to diversify into cocaine.5 Many of those with sufficient capital, contacts, and knowledge did, transferring their marijuana smuggling expertise to the more lucrative cocaine trade. In this respect, short-sighted government drug enforcement policies and programs that focused on marijuana trafficking helped set the stage for the dramatic increase in cocaine trafficking over the next decade (Betancourt & Garcia, 1994; Cannon, 1979; Krauthausen & Sarmiento, 1991; Orjuela, 1990, p. 218; O'Toole, 1978).


5Apart from increased marijuana drug enforcement, there were other incentives to switch to cocaine trafficking. For one thing, cocaine was less aromatic than marijuana, making it easier to hide and transport. Also, per unit profits for cocaine were substantially greater than for marijuana, reducing the need to transport such large quantities to ensure adequate profits.





36


Phase Four-1980s-mid 1990s: Rise and Fall of the "Cartels"

By the early 1980s a number of Colombian trafficking organizations were

organizing large-scale cocaine shipments from South America to the United States and Europe. These criminal enterprises provided a measure of coordination to what had been a highly fragmented industry. At the zenith of this transnational productiontransportation-marketing structure stood several vertically integrated "core" organizations that, according to one estimate, supplied over sixty percent of the cocaine reaching U.S. and European drug markets. Size varied considerably among different core organizations. Small core enterprises were composed of anywhere from two dozen to a hundred members, while large organizations contained several hundred members. Participants were compartmentalized into discrete units, organized along functional lines, such as drug processing, transportation, and wholesale distribution. Core organizations generally contained three or more levels of management, with at least two layers insulating leaders from the activities of rank-and-file workers (PCOC, 1986, pp. 100-101; Thompson, 2000; Vargas, 2000; Zabludoff, 1997, p. 24).

The core organizations transformed the Colombian drug industry by coordinating international cocaine shipments that measured in metric tons rather than kilograms. To do so, they relied on networks of hundreds of individuals and groups that provided specialized goods and services. Independent suppliers from Bolivia and Peru provided coca paste, a semi-processed form of cocaine, which was often transported to Colombia for further processing. In completing the cocaine refining process, some Colombian core organizations owned their own processing laboratories, while others subcontracted processing services to independent groups of chemists that operated hundreds of





37


"kitchen" labs outside of Medellin, Cali, or Bogotd. On occasion different organizations pooled resources to develop large-scale cocaine refining operations, such as the infamous Tranquilandia complex that was discovered and destroyed by Colombian authorities in 1984. In certain regions of Colombia left-wing guerrilla "fronts" and paramilitary groups provided protection for cocaine processing labs. Small trafficking groups and individuals were recruited to help finance large-scale shipments and share the risk associated with government interdiction efforts. American pilots were often hired to transport the finished product to consumer markets in the U.S. and Europe. Independent money launderers were contracted to repatriate illicit profits. In this manner, core organizations 'outsourced' numerous activities to independent groups linked in ad hoc support networks. Participation in these networks was rather fluid: Depending on the needs and circumstances involved in each shipment, core organizations could choose from among different suppliers, processors, transporters, and financiers (Castillo, 1996; Diehl, 1982; Morganthau, 1989, p. 23; Juan David Ochoa, 2000; Thourni, 1999; Zabludoff, 1997, p. 25).

Core enterprises focused on enforcing agreements among network participants, providing security for trafficking operations, gathering intelligence on government drug
6
enforcement efforts, and protecting leaders' political and economic interests. Several core organizations also maintained distribution "cells" in the U.S. responsible for receiving drug shipments, storing merchandise in warehouses or stash houses, distributing cocaine at the wholesale level, and shipping revenues back to Colombia. However, the core organizations ftmctioned more as voluntary export associations and



6 To perform these functions core enterprises relied on the use of intimidation and violence.





38


interest groups than monolithic firms (Clawson & Lee, 1996, pp. 39-40; Krauthausen & Sarmiento, 1991; Lee, 1991, p. 14; Zabludoff, 1997, pp. 23-26; Zabludoff, 2000). Emergence of the Core Organizations

Core trafficking organizations and their support networks did not spring up overnight. Rather, they developed over time as their leaders acquired experience, contacts, and knowledge in a variety of criminal endeavors, including smuggling contraband goods, robbing cars, even kidnapping. The largest, and most well-known, core organizations were led by entrepreneurs based in Medellin and Cali. During the 1980s and 1990s, such figures as Pablo Escobar Gaviria, Carlos Lehder Rivas, Gonzalo Rodriguez Gacha, Jos6 Santacruz Londofio, and the Ochoa and Rodriguez Orejuela brothers became household names as Colombian and American journalists assiduously chronicled their criminal exploits.7

Many of the leaders of the Medellin and Cali core enterprises began their delictive careers working as low-level members for established criminal gangs. In the early 1970s, three of the future leaders of the Cali-based cocaine network, Gilberto and Miguel Rodriguez Orejuela and Jos6 Santacruz Londofio, worked as foot soldiers in a Cali criminal gang known as Los Chemas. This gang was primarily involved in kidnapping and counterfeiting but gradually expanded into smuggling cocaine base from Bolivia and Peru and converting it into cocaine hydrochloride in Colombia (Castillo, 1987, pp. 41-42; DEA, 1994b, p. 1).


7 Due to the fascination of the Colombian and U.S. media and public with the Medellin and Cali core organizations an enormous body of literature emerged in the 1980s and 1990s, much of it written by journalists more interested in dramatization and hyperbole than scientific analysis. Some of the more wellknown journalistic accounts, in both English and Spanish, include: Caf16n (1994), Castillo (1987, 1991, 1996), Castro-Caycedo (1 996b), Cervantes (1980), Cort6s (1993), Duzin (1994), Eddy et al. (1988), Freemantle (1985), Giraldo (1992), Gugliotta and Leen (1989), Mermelstein (1990), Mills (1986), Reyes (1999), Rice (1989), Rinc6n (1990, 1994), Shannon (1988), Torres (1995), Torres & Sarmiento (1998).





39


Also in the early 1 970s Pablo Escobar, who would emerge a decade later as one of the leaders of the Medellin cocaine network, worked as an enforcer for a contraband smuggler that specialized in whiskey, cigarrettes, watches, and second-hand pianos. In an interview with a Colombian journalist, Escobar referred to this contrabandista as his "maestro," from whom he learned the smuggling business (Castro Caycedo, 1 996a, p. 284). Around 1975 the ambitious Escobar became involved in cocaine trafficking, organizing small-scale smuggling ventures. Through contacts and intermediaries he arranged for the purchase of kilo quantities of cocaine hydrochloride from an Ecudorian supplier, and transported the drug to the United States through human couriers. In 1976 he was arrested by Colombian authorities near Medellin for transporting thirty-nine kilograms of cocaine hidden inside the spare tire of a truck. Escobar was never successfully prosecuted for this charge, and later in the decade convinced Fabio Ochoa, an experienced contraband smuggler, "to use his well-established and well-connected smuggling routes for the more profitable drug business" (Chepesiuk, 1999a, p. 142). The old-time con trabandista apparently agreed and a business partnership gradually emerged between the senior Ochoa, his three sons, and Escobar. This joint venture endured for a number of years and, along with the participation of other leading traffickers, formed the basis of what later became known as the Medellin "cartel" (Caft6n, 1994, pp. 59-60; Castillo, 1987, p. 60; Kraar, 1988, p. 34).

However, this partnership never functioned as a cartel in the strict sense that

economists use the term. Rather Escobar and the Ochoas pooled their criminal resources, divided responsibilities, and coordinated large-scale cocaine shipments to American and






40


European markets. 8 As the size of their trafficking venture grew they gradually absorbed or co-opted the smuggling operations of several of the "pioneering" trafficking groups operating in the 1970s (Gugliotta, 1987; PCOC, 1986, pp. 101-103; Vald~s, 2000).

Similar arrangements developed between different core enterprises in the Calibased network, including the Rodriguez-Orejuela, Santacruz Londofto, and Herrera Buitrago groups. Although the leaders of these separate groups ran their own smuggling operations with their own personnel, they collaborated extensively on strategic matters affecting the entire network, such as infiltrating the different security agencies of the Colombian state or corrupting well-placed Colombian congressmen sympathetic to their political and economic interests, such as outlawing the extradition of Colombian nationals. Moreover, on occasion, low-level workers for one core organization, such as cocaine chemists, truck drivers, or merchandise off-loaders, would be put to work for another core group, without the workers' knowledge that the ultimate locus of authority had changed (Torres & Sarmiento, 1998; Veldsquez Romero, 2000). Smuggling Methods of Core Organizations

Core enterprises used a variety of methods for transporting cocaine from South America to U.S. and European drug markets. A number of core organizations continued the practice of using human couriers to smuggle cocaine. Although the quantities smuggled by individual couriers remained small, generally less than four kilograms, core enterprises had large numbers of participants at their disposal. In an effort to overwhelm law enforcement authorities, now wise to this time-honored smuggling method, some

8 Medellin core organizations also absorbed the smuggling operations of several of the "pioneering" trafficking groups operating in the 1 970s. When Manuel Garc~s was sentenced to jail in Colombia, the Escobar organization took over his smuggling routes and operations (Vald~s interview). When Benjamin Herrera was paroled from a U.S. federal prison in 1975, he returned to South America and collaborated with the Medellin core organizations (Gugliotta, 1987).





41


commercial Rights contained as many as many as seventeen drug-running mules. Organizers overloaded the planes with mules with the expectation that drug enforcers would identify and apprehend several, which was considered acceptable as long as the majority cleared Customs. Couriers continued to use false-bottomed suitcases to smuggle drugs, as well as dog cages, hollow coat hangars, and specially designed bras and girdles. In addition to these 'external body' carries, large numbers of Colombian couriers ingested balloons or condoms containing small amounts of cocaine (DEA 1993, p. 18; Raab, 1984; Ramos, 1985).

However, the bulk of cocaine transported by the core organizations was through general aviation aircraft and maritime vessels. A number of core enterprises outsourced air transportation. services drug-running American pilots and support personnel. These transportation rings used light twin engine aircraft, such as Gulfstream Aero Commanders, Beechcraft King Air 300s, Piper Navajos, Aztecs and Cheyennes, and Cessna 400s and Conquest Us, to transport several hundred kilos of cocaine over a range of approximately 1,800 miles. Many planes were equipped with rubber fuel "bladders" for traveling long distances without the need to refuel, along with radar detectors and radio scanners for tracking government planes and seafaring vessels on drug patrol. In the 1990s, some Cali core organizations were using "large, long-range, multi-engine" aircraft, such as Convair 580s and Boeing 727s, to smuggle several tons of cocaine into Mexico and Canada. For smaller planes that required refueling, smuggling networks relied on a number of transhipment points in Mexico, Central America, and the Caribbean where drug-running pilots could land and, for a hefty commission, refuel.9


9 One of the leaders of the Medellin-based smuggling network, Carlos Lehder, maintained a transshipment point on Norman's Cay, a small island in the Bahamas where he owned several properties. In exchange for






42


While some transshipment points were merely refueling stations, others contained small crews of workers that off-loaded and repackaged cocaine shipments. In Mexico a number of trafficking groups with marijuana and heroin smuggling experience provided this service, breaking down the cocaine into smaller shipments and transporting it through Mexico in autmobiles and trucks (DEA, 1993, p. 17; DEA, 1994b, p. 7; Mermeistein, 1990; Pallomari, 1997; PCOC, 1986, pp. 88-89, 98; Raab, 1984; Rice, 1989, pp. 145-146).

Maritime smuggling was conducted by seafaring vessels of various types and transportation capacities, including high-speed "cigarette" or go-fast boats, sport fisherman's boats, yachts, medium-sized commercial fishing boats, large, dilapidated freighters, and even what the DEA identified to as "submarine-like semi-submersible vessels" capable of tranporting hundred kilogram quantities of cocaine through the Carribean (DEA, 1994b, p. 7). Maritime vessels often departed from commercial ports in Colombia, such as Buenaventura, Barranquilla and Cartagena, as well as hundreds of clandestine debarkation points located along the country's northern coast. On smaller boats, cocaine was hidden in customized storage compartments, with the vessels frequently modified for this purpose. On large freight ships, the drug was hidden within containerized cargo shipments, including blue jeans, ceramic floor tiles, fresh cut flowers, fruits and vegetables, fish, coffee, lumber, concrete posts, blue jeans, and dozens of other exports from South America (Maitland, 198 1; Griffith, 1997, p. 8 1; Castillo, 1996, p. 5 1; Mermelstein, 1990, p. 146; N~stor, 2000).



a commission, Lehder allowed independent cocaine and marijuana smugglers as well as those associated with the Medellin network to land and refuel before continuing to the United States. For a description of Lehder's operation, see Gugliotta & Leen (1989) and Shannon (1988). Also, see the interview granted by Jorge Luis Ochoa (2000) to Frontline.





43


In reflection of the smuggling capacity of the core enterprises, in the 1980s U.S. and Colombian authorities began to discover containerized shipments and stash houses containing several tons of cocaine hydrochloride. As early as March 1982 U.S. Customs authorities discovered 3,096 pounds (or 1.38 tons) of cocaine mixed in mixed in a shipment of hundreds of cartons of clothing apparel. The cocaine was wrapped with yellow plastic in one-kilo packages bearing different coded markings. According to DEA investigators the load belonged to fifteen different trafficking organizations, including the Ochoa and Escobar groups. In 1989, DEA and Customs agents discovered approximately 12,000 pounds (5.35 tons) of cocaine hidden inside drums of powdered lye. The cocaine, wrapped in red, yellow and blue plastic and marked with the code name Baby I, was discovered at a warehouse in Queens, New York used by a trafficking ring associated with the Cali-based trafficking network. Also in 1989, U.S. authorities confiscated cocaine shipments of 11,200 pounds (5 tons) on a Panamanian freighter in the Gulf of Mexico, 20,160 pounds (9 tons) hidden in a stash house in Texas, and 44,800 pounds (20 tons) in an unguarded warehouse near Los Angeles. The latter seizure remains the largest ever recorded in the United States. Two years later, DEA and Customs officials discovered 26,880 pounds (12 tons) of cocaine hidden inside concrete fence posts and cornerstones. In 1994, federal authorities uncovered 8,000 pounds (3.57 tons) of cocaine hidden inside empty barrels in a tractor-trailor truck in Texas (Dillow, 1994; Gugliotta & Leen, 1989, pp. 71-73; Isikoff, 1991; McKinley, 1989). Diversif)jng to Heroin

At some point during the 1980s several core enterprises expanded into the heroin business. It is believed that this development was driven by the same factors that caused





44


numerous marimberos to diversify into cocaine in the preceding decade, including the growing risks associated with cocaine trafficking due to successful drug enforcement efforts, and the greater per unit profitability of heroin. In addition, traffickers may have been attracted by profitable European drug markets, where heroin distribution networks were well-established, and law enforcement efforts posed fewer risks to drug smugglers. In 1994, the DEA reported that core organizations led by the Ochoa brothers and Leonidas Vargas Vargas were using their cocaine networks to smuggle heroin. The same report claimed that the Rodriguez Orejuela brothers and Helmer Herrera Buitrago had also invested in the heroin industry. Two other core organizations associated with the Cali network, the Santacruz Londofio and the Urdinola Graj ales groups, began experimenting with heroin production in the early 1980s, and supplied Colombian farmers with the necessary inputs necessary to cultivate opium. According to a former director of the Colombian National Police (CNP), Santacruz Londofto brought Afghani and Pakistani opium cultivators to Colombia in order to learn how to cultivate opium poppies. Since then Colombia has become a major source of heroin to the U.S., supplying a higher grade product that many U.S. consumers favor over Southeast Asian heroin (DEA, 1994a, p. 6; Farah, 1997; Farah, 1999; FBI, 1993, p. 2 1; Moore & Farah, 1998; Revista Cambic 16, 1999; Semana, 1999; Serrano Cadena, 1999, pp. 50-5 1; "Statement of Witness," 1989, p. 11; Zabludoff, 1999). Government Drug Enforcement Efforts During Phase Four

Throughout the 1980s and the 1990s the U.S. and Colombian governments directed considerable resources towards dismantling the core cocaine organizations. While bilateral cooperation between the two countries was occasionally marred by






45


substantive disagreement over a number of policy issues, particularly extradition, the two governments worked together to implement a variety of supply-reduction programs targeting the Medellin and Cali-based trafficking networks.

During this period the Colombian government engaged in a series of

"crackdowns" directed at the leaders of the Medellin-based trafficking enterprises and subsequently the Cali organizations. Several crackdowns were undertaken in response to specific criminal acts attributed to Medellin trafficking groups, such as the assassination of Rodrigo Lara Bonilla, the Colombian Minister of Justice, in 1984.10 Each time Colombian authorities responded by carrying out dozens of raids against the properties of leading Medellin traffickers, destroying processing laboratories, seizing cocaine and money, an apprehending low and mid-level participants in these criminal enterprises. However, the leaders remained beyond the reach of drug enforcers, developing sophisticated security arrangements allowing them to evade government efforts. After several weeks, the crackdowns would subside, and traffickers could return to business as usual (Gugliotta, 1992, pp. 122-123; Matthiesen, 2000).

In 1989 Colombian president Virgilio Barco launched a fourth crackdown in response to the assassination the leading presidential candidate for the 1990 general elections, by sicarios (paid assassins) working for the Medellin traffickers. By this point, the Colombian president and other political elites were convinced that the Medellin organizations presented a legitimate threat to the country's political system. In a televised address, President Barco declared a nationwide state of siege against the traffickers. At the behest of President Bush, the U.S. offered their beleagued drug war 10 Bonilla's assassination was carried out in retaliation for his role in planning a police raid on the Tranquilandia cocaine refining complex owned by several leaders of the Medellin smuggling network (Gugliotta & Leen, 1989, pp. 132-137; Lee, 1989, p. 171; Tokatlidn, 1990: 317).





46


ally an emergency assistance package of US$ 90 million, and the Drug Enforcement Administration undertook Operation Bolivar, specifically targeting the Medellin organizations.

The fourth crackdown proved more effective than previous efforts because lasted over a year, as opposed to six weeks, and possessed a more coherent strategy that included extradition. The relentless pressure exerted by Colombian and U.S. authorities forced the leaders of the Medellin organizations into hiding, reducing their ability to coordinate large-scale cocaine shipments. Several leading traffickers were captured or killed during police raids. Moreover, the crackdown produced ripple effects throughout the Colombian drug industry. The inability of processing groups associated with the Medellin network to buy the precursor chemicals needed for cocaine processing on a regular basis reduced the price for Peruvian and Bolivian coca leaf to less than half of production costs. The scarcity of cocaine supply affected consumer markets, causing wholesale and retail cocaine prices in the United States to rise between fifty and one hundred percent. However, this price spike proved short-lived, lasting approximately eighteen months (Bagley, 1989-1990, p. 155; Caulkins & Reuter, 1998, p. 603; Farah 1990a, 1990b; Gugliotta 1992, p. 124; Krauss, 1995, p. A 12; Nieves, 1997, p. 11; Velisquez, 1993; Whynes, 1992, p. 333).

Nevertheless, by the early 1990s many of the Medellin core organizations had

been severely damaged by Colombian and U.S. anti-drug efforts. Prompted by effective police pressure and a decree by a new Colombian president, several leading traffickers, most notably the three Ochoa brothers, turned themselves in to the authorities. In June 1991, when it became clear that the new Colombian constitution would ban extradition of





47


Colombian nationals, Pablo Escobar also turned himself in to Colombian authorities. However, Escobar's surrender turned out to be largely meaningless as he continued his criminal activities in a luxurious correctional facility of his own construction, from which he escaped after a year of "imprisonment." Over the next two and a half years, Escobar managed to elude Colombian and U.S. authorities and former business associates that turned against him. But the effort to stay alive required all of his resources, eliminating his role as a major trafficker."' Meanwhile, competing trafficking enterprises based in Cali took advantage of drug enforcers' single-minded focus on capturing Escobar by expanding their own transnational operations (Cafl6n, 1994; Nieves, 1997, p. 12).

With the death of Escobar at the hands of an elite Colombian police unit in December 1993, the Colombian government could now turn its attention to the Cali enterprises. However, it would be another fourteen months before Colombian authorities systematically cracked down against the leading Cali traffickers. The impetus for the 1995 offensive was not an act of violence by the Caleijos, who relied more on bribery than intimidation to protect their interests, but political pressure from Washington. Disgusted with Colombian president Ernesto Samper's ability to stay in power despite a campaign finance scandal in which he was believed to have knowlingly accepted contributions from the Rodriguez Orejuela brothers and other traffickers, the U.S. Senate threatened to impose economic sanctions on Colombia if the government failed to achieve certain drug enforcement benchmarks in 1995. These included capturing the leaders of the Cali core organizations, confiscating their illicit assets, and dismantling their criminal operations (Cafi6n 1994; Economist Intelligence Unit, 1995, p. 4; Latin


" For further discussion of these events see Luis Caf16n El Patrdn: Vida y muerte de Pablo Escobar (Bogoti, Colombia: Planeta, 1994).






48


American Weekly Report, 1995a; 1995d; Mattiesen, 2000, pp. 309-3 10; Nieves, 1997, p. 11; Semana, 1995, p. 28; Thounii 1995, pp. 226-228).

In February 1995, President Samper launched a comprehensive assault on the leaders of the Cali trafficking network. Over the next several months, the Colombian National Police, led by General Rosso Jos6 Serrano, captured eighteen tons of coca leaves and two tons of cocaine hydrochloride, destroyed 195 processing laboratories, eradicated 6,000 hectares of coca plantings, and detained 616 suspected traffickers. While impressive, the results did not dramatically exceed the drug enforcement indicators in proceeding years (see Table 2-2 below).

In June the Samper government received a major boost when a special joint

military-police unit captured Gilberto Rodriguez Orejuela, the alleged maximum leader of Cali traffickers. Over the next year, the Colombian National Police, with the assistance of the DEA and the CIA, exerted enormous pressure on the remaining Cali leaders, all of whom were eventually either captured or surrendered to government authorities. In combination with the Peruvian government's controversial "shoot-down" policy in the coca paste air bridge between Peru and Colombia, the Cali crackdown produced ripples throughout the cocaine industry. By June 1995, the price of coca paste in Colombia had dropped 50%. By mid-September of the same year, wholesale and retail cocaine prices in New York City, the Cali network's most important U.S. market, increased 50% and 30% respectively. Similar increases were noted in other U.S. cities that received substantial cocaine supplies through the Cali transportation and distribution network (Caulkins & Reuter, 1998, p. 603; Geopolitical Drug Dispatch, 1995; Krauss,






49


1995; Latin American Weekly Report, 1995b, 1995c, 1995e, 1995f, 1995g, 1995h, 1995i;

Serrano Cadena, 1999).

However, like earlier crackdowns against the Medellin enterprises, these price

increases disappeared completely within a year, reflecting the drug industry's ability to

absorb even the most intense drug enforcement efforts in the long term. Indeed,

following a brief period of regeneration, new and revitalized trafficking groups emerged

in Colombia to produce greater amounts of cocaine and heroin than before.

Table 2-2 Colombian Drug Enforcement Indicators, 1980-1995
Year Persons Cocaine Cocaine Heroin Marijuana Druglabs
apprehended interdicted base Interdicted interdicted destroyed
(kilos) interdicted (kilos) (kilos) (kilos)
1980 358 748 0 0 192,422 29
1981 798 339 0 0 3,302,242 31
1982 1,139 651 0 0 3,283,878 165
1983 1,073 2,083 0 0 3,537,387 113
1984 5,251 19,582 9,448 0 4,301,263 137
1985 1,951 4,239 3,674 0 1,021,046 696
1986 3,699 3,039 4,070 2 846,000 572
1987 4,732 8,326 6,712 2 1,287,272 1,359
1988 4,929 12,047 2,554 0 842,994 655
1989 5,217 24,668 9,601 0 617,925 389
1990 6,253 16,000 3,429 850 659,047 268
1991 6,349 59,347 8,223 0 381,157 235
1992 6,770 28,016 5,289 36 206,934 95
1993 8,136 19,137 6,945 42 505,274 241
1994 7,221 28,145 22,580 94 161,322 334
1995 8,053 34,577 15,375 184 171,347 331
Source: CNP (2000)


Phase Five-Mid-1990s to 2000: Decentralization of Colombian Drug Trad

With the fall of the Medellin and Cali cartels the Colombian drug trade has

decentralized. Instead of a handful of vertically integrated core organizations that

dominate the industry, hundreds of independent trafficking groups operate throughout






50


Colombia, conducting their activities on a smaller, more discrete scale.'2 Some organizations, particularly several based in the Northern Cauca Valley and Atlantic Coast areas, are long-standing enterprises that escaped the drug enforcement net as the CNP and DEA pursued more prominent core organizations.' 3 In addition to these medium-sized survivors, hundreds of small trafficking operations have emerged in recent years. There is variation in size among existing enterprises, with medium-sized organizations ranging from twenty to one hundred participants, and smaller groups containing ten to twenty associates. Contemporary enterprises tend to be "flatter" than the core organizations, and more circumspect in carrying out their activities. Small groups are frequently led by a single figure that exerts substantial decision-making authority, while co-equal subordinates perform the work. Medium-sized organizations often contain different leaders that occupy discrete, functionally-specific managerial positions, similar to the core enterprises (Aero, 2000; DEA, 2000; FBI, 1993; Gallego Castrill6n, 2000; Nicosa, 2000; Semana, 2000a).

Instead of coordinating several phases of trafficking activities like the core enterprises, many trafficking operations today specialize in single-phases of drug production, processing or transportation. While different groups, including former 12 It is not clear how many trafficking enterprises actually exist in Colombia, which is reflected in the wide range of expert estimates. General Serrano, the former head of the Colombian National Police, estimates that are between 350 and 400 'minor' trafficking organizations operating in Colombia. Alejandro Reyes Posada, a leading Colombian drug researcher, is even more precise. According to Reyes Posada there are 438 narcotics organizations in Colombia, 256 of which are re-structured off-shoots of the former cartels, while 182 represent new groups. A recent report in Semana magazine estimates there are between 80 and 100 groups in Colombia that specialize solely in the exportation of cocaine or heroin (Reyes Posada, 1999, p. 2; Semana, 2000a, p. 7 1; Serrano Cadena, 1999, p. 237). My own research suggests that the clandestine nature of drug trafficker merits greater discretion in formulating estimates. While DEA and CNP intelligence officials confirm that there are "hundreds" of trafficking enterprises in Colombia, they also stress the difficulty of providing exact estimates. The highest-ranking DEA official in Colombia told me that while he believes there are several hundred organizations operating in Colombia he does not know the exact number (Arreguin, 2000).
13 Indeed, a number of these groups provided transportation services to the Medellin and Cali core organizations during the 1980s and 1990s.






51


competitors, cooperate and pool resources, there appears to be no centralizing coordination mechanism that provides direction and contract enforcement. Another development is that many Colombian enterprises now avoid establishing distribution cells in the U.S., often selling their loads to Mexican trafficking enterprises that complete the international transportation and distribution process.

Trafficking enterprises in Colombia today are frequently led by people with considerable criminal experience. A number of post-cartel start-ups were founded by former mid-level managers of the Medellin and Cali core organizations. Other prominent traffickers have formally "retired" from the day-to-day business of drug trafficking but continue to invest in shipments and offer their advice when solicited. This suggests that the knowledge and experience of the former cartels has not been lost. Traffickers from the core organizations draw on contacts and experience in conducting their present operations, applying previous knowledge to new ventures. Government officials interviewed for this research suggest that one reason why existing operations remain small is that surviving traffickers have learned from experience that they are better off avoiding drug enforcers by purposely limiting the size of their operations.

Contemporary trafficking enterprises rely on basic smuggling routines developed by their predecessors, while exploiting advances in communications and transportation technologies when modifying these practices and procedures. Traffickers continue to use a variety of maritime vessels to transport multi-ton loads and general aviation aircraft for 400-800 kilogram drug shipments. However, at least one enterprise has upgraded to the next level of submarine technology. In September 2000 Colombian officials discovered a partially finished submarine based on advanced Russian technology, including a






52


computer navigation system and engine room. When completed the submarine would have measured over 100 meters in length and had a transportation capacity of 200 tons, far in excess of the one to three tons of cocaine transported by the "mini-subs" in the Caribbean during the 1990s. Other recent innovations include using calling cards, satellite phones, electronic mail, and encryption technology to communicate information, mixing cocaine hydrochloride with charcoal and iron dust, welding cocaine into cargo ship hulls, and using double-lined latex pellets for internal body carries (Arreguin, 2000; Associated Press, 1998a; Branigan, 1999; Bohning, 2000; Brooks & Farah, 1998; CM&, 2000; CNN, 2000; DEA, 2000; El Tiempo, 2000d; Francis, 2000; Johnson, 2000; Markowitz, 2000; Smith, 2000a).

Government Drug Enforcement Efforts During Phase Five

Colombian and U.S. drug enforcers continued to target the largest, most wellknown trafficking organizations in the late 1990s. In the post-cartel era, this meant the medium-sized enterprises operating in the Northern Cauca Valley and along the Atlantic Coast. Between 1997 and 2000, the leaders of several enterprises surrendered or were captured by government authorities. In August 1997, CNP officials arrested Julio Cesar Nasser David, one of the architects behind the Atlantic Coast trafficking network. 14 The following month, Orlando Henao Montoya, a leader in the North Valley network, surrendered to Colombian officials in Bogotd. In February 1998, CNP forces apprehended Jos6 Nelson Urrego, another leader of the North Valley network, at his country estate outside of Medellin. Four months later, Colombian authorities captured 14 Nasser David's participation in the Colombian narcotics industry dates back to the 1970s, when his organization smuggled several tons of marijuana to Southern Florida using 'mother ships' and speedboats. In the mid-1980s, Nasser David's organization moved into the cocaine trade, transporting and distributing multi-hundred kilogram shipments of cocaine to the U.S. through containerized cargo and general aviation aircraft (Constantine, 1998; FBI, 1993, p. 3 1; Chepesiuk, 1999b, p. 153; Reuters, 1998c).






53


Alberto Orlandez Gamboa, the leader of a large Atlantic Coast trafficking enterprise (Associated Press, 1998; CNNespahiol, 1998; Chepesiuk, 1999b, pp. 92, 153, 253; Constantine, 1998; DEA, 2000; FBI, 1993, p. 31; Guzman, 1998; Los Angeles Times, 1997; Reuters, 1998a, 1998c).15

In October 1999, CNP and DEA officials implemented Operation Millennium,

targeting a highly sophisticated smuggling network exporting between ten and thirty tons of cocaine per month. In a series of carefully orchestrated raids in BogotA, Medellin and Cali, CNP agents arrested thirty-three members of the network, including the alleged leader, Bernal Madrigal, and Fabio Ochoa, one of the leaders of the old Medellin cartel (Bajak, 1999; Johnson & Davies, 1999; El Tiempo, 1999).16 Also during this period, CNP drug enforcement units captured a number of traffickers from smaller smuggling groups and disrupted their cocaine and heroin operations (El Tiempo, 1997a; El Tiempo, 1997b; El Tiempo, 1997c; Caracol, 2000a; Caracol, 2000b; DAS, 1997; RCN, 2000; Reuters, 1998b).

These drug enforcement operations indicate that CNP and DEA officials continue to target trafficking organizations in Colombia. However, in spite of these successes, a number of important traffickers remain at large. According to the DEA, Diego Montoya Sanchez, once a mid-level figure in the North Cauca Valley network now leads one of the major organizations in the area. Montoya Sanchez was the principal cocaine supplier for the Madrigal operation and continues to coordinate multi-ton shipments of cocaine to 15 During the early 1990s, Orlandez Gamboa's organization was associated with the Cali-based trafficking network. Following the capture of the leaders of Cali core organizations in 1995 and 1996, Orlandez Gamboa branched out on his own, exploiting maritime and air routes through Mexico and the Caribbean to smuggle multi-ton shipments of cocaine and marijuana to the U.S. 16 Since their arrest Ochoa and Madrigal have been extradited to the U.S., where they are now awaiting trial for numerous trafficking-related criminal charges (El Tiempo, 2001d, 2001e; Forero, 2001b; U.S. Attorneys Office, 2001a, 2001b).






54


Mexico and the U.S. The DEA and CNP have identified other traffickers they believe now lead the Atlantic Coast and North Cauca Valley networks. Moreover, drug enforcers fear that surviving remnants of the Medellin and Cali cartels are reorganizing, led by former mid-level managers. In addition to these medium-sized enterprises there are now hundreds of smaller independent trafficking groups operating in Colombia. Some of these groups are led by notorious traffickers that continue to elude drug enforcers, others are run by individuals unknown to Colombian and U.S. authorities. With so many criminal enterprises determined to produce and transport illicit drugs, Colombia remains poised to continue its leading position in the international drug trade.

Table 2-3 Colombian Drug Enforcement Indicators, 1996-1999
Year Persons Cocaine Cocaine Heroin Marijuana Drug labs
apprehended interdicted base Interdicted interdicted destroyed (kilos) interdicted (kilos) (kilos)
(kilos)
1996 5,703 17,808 11,142 88 101,519 436
1997 10,711 35,792 14,906 176 117,880 228
1998 18,276 46,256 11,346 341 73,085 198
1999 21,168 21,423 9,621 541 71,369 100
Source: CNP (2000)

Conclusion

Over the past seventy years, the Colombian drug trade, and the illicit enterprises that coordinate it, have undergone a variety of changes. While the empirical record of this clandestine industry remains sketchy, particularly in the early decades, this chapter draws on the available materials to craft a credible, coherent narrative. In the beginning, drug trafficking in Colombia was sporadic. Smugglers, often working alone, transported minute quantities of European produced cocaine and heroin through Colombia and Panama on the way to North American markets. Sometime in the 1940s or 1950s, a number of pioneers, perhaps of German extraction, developed cocaine and heroin






55


processing laboratories on Colombian soil, transforming the country from mere transhipment point to drug producer. Early producers sold their illegal wares to international smugglers based in Cuba who transported the drugs to the U.S. During the 1950s and 1960s, many smugglers based in Colombia were content to serve as suppliers and transhippers for trafficking enterprises based in other countries, including Cuba, Chile, Mexico, and the United States. The primary psychoactive drugs produced in or passing through Colombia at this time were marijuana and cocaine.

In the late 1960s and early 1970s a fimdamental change occurred in the

Colombian drug trade. Several enterprises expanded into U.S. wholesale distribution markets. These networks developed infrastructures and technologies that laid the foundation for core enterprises in the 1980s. With the rise of the Medellin and Cali "cartels," and their multi-tonnage trafficking capacities, Colombia's dominant position in the international cocaine trade solidified. Driven by growing U.S. appetites for marijuana and cocaine, the amount of psychoactive drugs coming out of Colombian in the 1970s and 1980s increased substantially, while the organizations coordinating this traffic also grew and became increasingly bureaucratic.

By the mid- I 980s, a number of Colombian trafficking enterprises had become too large, their leaders too ambitious. Over the next decade, the Colombian and U.S. governments waged an on-again, off-again campaign to dismantle the leading core organizations. While drug enforcers achieved remarkable victories, dismantling several Medellin and Cali cartels, their ability to disrupt the industry was limited by the fluid nature of trafficking enterprises. Within months of even the most effective supplyreduction programs, drug smugglers would bounce back, revitalizing existing operations






56


or creating new ones, driven by the strong demand for their illicit commodities in U.S.

and European drug markets.

Table 2-4 Evolution of Illicit Drug Industry in Colombia
Phase Time Period Characteristics of Colombian Drug Trade Organizational Exemplar
One 1930s Individual smugglers transport cocaine and N/A
heroin to Panama
Two Late 1930s- Smugglers transport cocaine and heroin in Herran
early 1960s Caribbean, supply criminal organizations Olazaga based in Havana, development of cocaine brothers and heroin processing labs in Antioquia Three Late 1960s- Marijuana and cocaine trafficking groups Herrera
1970s supply Cuban, American, Chilean & network,
Mexican networks; establishment of Bravo
Colombian distribution networks in U.S. network,

Four Late 1970s- Rise of core organizations and support Core
early 1990s networks; spread of cocaine production organizations throughout Colombia; participation of in Medellin guerrillas and paramilitary groups; and Cali
dismantling of Medellin and Cali core organizations
Five Mid 1990s- Decentralization of Colombian drug trade; North Valley
present small and medium-sized groups; growth of network,
cocaine & heroin production in Colombia Atlantic Coast network,
Madrigal
network


Today, many trafficking enterprises in Colombia bear a greater resemblance to the

pre-cartel groups of the 1960s and 1970s than to the core organizations that followed.

They are small and contain flat decision-making hierarchies. They specialize in drug

production or transportation, but not both. Their leaders avoid the limelight, managing

their operations with the utmost discretion. Yet, in one significant respect existing

groups remain similar to their larger predecessors. They continue to modify their

trafficking practices by exploiting advances in communications and transportation






57


technologies, ensuring that their smaller size does not equate with lesser sophistication. The remainder of this study is devoted to understanding how these criminal enterprises, and their state adversaries, draw on information and experience to alter their form and function over time.














CHAPTER 3
TOWARD A THEORETICAL UNDERSTANDING OF
ORGANIZATIONS AND ORGANIZATIONAL LEARNING

Organizational learning is a slippery concept. It contains multiple meanings that are open to a variety of interpretations. Learning is difficult to define, let alone measure and apply to empirical settings. Reflecting such difficulties, the academic literature lacks a generally agreed upon theory of organizational learning (Levy, 1994, p. 280; Fiol & Lyles, 1985, p. 803; Keohane & Nye, 1989, p. 264; Stein, 1994, p. 224). In spite of the theoretical and methodological challenges associated with organizational learning, over the past several decades social scientists have drawn on a number of basic concepts to develop a cumulative body of research that demonstrating learning to be a valid explanation for certain types of organizational change. Dating back to Herbert Simon's seminal formulation, first published in 1945, that organizations "intelligently" adjust routines in response to external stimuli and experience, political scientists have applied notions of learning to a variety of subject matter, including government institutions, intergovernmental bodies, military organizations, transnational epistemic communities, and elite decision-makers. 1


' Two decades after the appearance of the first edition Administrative Behavior (Simon, [ 1945] 1997), Karl Deutsch applied the concept of cybernetics to communications systems within governments and other social organizations (Deutsch, 1966). In the early 1970s several political scientists and diplomatic historians analyzed how policy makers draw on "lessons of the past" when making decisions (Lowenthal, 1972; May, 1973). In 1976, Robert Jervis drew on social and cognitive psychology to explain how decision makers reason and learn from previous experience (Jervis, 1976). In recent years the research agenda on leading in political science has gathered momentum as a growing number of scholars have applied learning concepts to individual decision-making and organizational behavior. Examples of this literature include-but are not limited to: Breslauer and Tetlock (199 1), Dodd (1994), Eden (forthcoming), Etheredge (1985), Evangelista (1988), Goldgeier (1994), Goldstein & Keohane (1993), Haas (1990), Haas 58






59


In this literature, criminal organizations and law enforcement agencies have rarely been topics of empirical interest. This research builds on existing studies by examining organizations that stand at the intersection of the rule of law and national security. The lack of attention provided trafficking enterprises and drug enforcement agencies in the organizational learning literature raises a number of questions. Most basically, what is a drug trafficking organization that it may learn, and undertake other collective actions? Do police agencies and trafficking enterprises learn from experience and information? If these state and non-state collectives alter their practices and procedures, how can we be sure that such changes are attributable to learning rather than alternative considerations? What are the individual and organizational obstacles to learning? Assuming that trafficking enterprises and law enforcement agencies do on occasion learn, what conditions facilitate this process?

I do not propose to answer these demanding questions in this chapter. That challenge unfolds later in this study. Here, my aim is more modest: to develop the theoretical framework that guides subsequent analysis. I begin by discussing a concept that subsumes organizational learning: organization. In order to evaluate whether trafficking enterprises and drug enforcement agencies learn as organizations, I am compelled to explain the latter concept. I do so by drawing on the open-systems perspective of organization theory. Following this discussion, I proceed with a definition of organizational learning that disaggregates the process into separate but overlapping



(1992), Haas & Haas (1995), Hall (1993), Heclo (1974), Khong (1992), Leng (1992), Le Prestre (1995), Mann (1992), McCoy (2000), Mendelson (1993), Moltz (1993), Nagi (1999), Ndurnbaro, (1998), Nolan (1994), Nye (1987), Posen (1984), Reiter (1996), Sabatier (1988), Sagan (1993), Sikkink (1997), Snyder (1991), Stein (1994), Weir and Skocpol (1985), Welch Larson (1985), Zarkin (2000), and Zeng (1999). For recent overviews of the political science literature on learning see Bennett & Howlett (1992) and Levy (1994).





60


phases of information acquisition, interpretation, and application. In recognition that leading is not the only source of change in organizational routines, I address alternative sources of change in organizational behavior, including power and envirom-nental selection. I follow this with brief analyses of several different learning concepts that are pertinent to this research, including competitive leading ecologies, "productive" learning, tactical adaptation, and strategic learning. The chapter continues with a detailed discussion of the numerous psychological, organizational, and environmental obstacles to organizational leading, including bounded rationality, "tall" hierarchies, and benevolent environments.

As fascinating as the concept of organizational learning may be, my purpose in these pages is not to theorize for its own sake, but to develop a series of propositions that can be applied to the units of analysis in this research. Throughout this chapter I intersperse the discussion with falsifiable propositions that I apply to trafficking enterprises and drug enforcement agencies in subsequent chapters.

Organizations

At the most basic level, organizations are simply collections of participants that coordinate their activities through patterned behaviors, and draw on resources and technologies to achieve specific goals. Organizations are social systems that contain five inter-related components: tasks, environments, structures, participants, and technologies (Scott, 1998).

Organizational Tasks

Organizations exist to achieve aims shared amongst participants. Tasks are embodied in goals that drive organization behavior, and objectives to achieve these





61


desired ends. Organizational goals include creating goods and services, attaining satisfactory profits and other economic, social and political benefits, and ensuring the survival of the enterprise. Objectives include constructing production facilities, searching for new markets, devising innovative production and marketing methods, recruiting people to work for the organization, and reducing internal and external threats to the integrity of the enterprise. The determination of "satisfactory" benefits is highly subjective and varies among organizations. However, autonomous organizations do not exist for long unless the value of their products exceeds the resources needed to make them (Leavitt et aL, 1973, pp. 10-11; Scott, 1998, pp. 20-21).

Tasks may be clearly defined and widely appreciated within an organization, or they may be vaguely understood and unique to sub-parts of the collective (Leavitt et aL, 1973, pp. 23, 27). As Bamard and Simon recognized decades ago, individuals within organizations frequently do not share a common conception of organizational goals (Bamard, 1938; Simon, [ 1945] 1997). Moreover, organizations typically pursue multiple objectives simultaneously, which can create competition and conflict among different parts of the enterprise.

Environments

Organizations develop within environmental contexts. Environments provide

opportunities for participants to coordinate their behavior, and the material and symbolic resources necessary to survive. In organization analysis, the environment is a residual concept. It refers to anything outside the organization. Dill's notion of the task environment provides an element of precision to this broad concept. The task envirom-nent refers to those parts of the environment that are relevant to the





62


organization's aim or purpose. Dill (195 8), and subsequently Thompson (1967), disaggregated the task environment into four areas: (1) customers or clients of the organization (including distributors and consumers); (2) suppliers of labor, capital and technologies; (3) competitors for market share and resources; and (4) regulatory groups, which include government agencies. The organizations that fulfill these roles comprise an organization set. An organization set is composed of the customers, suppliers, competitors and regulators with which an enterprise interacts in order to achieve its purpose (Blau & Scott, 1962; Dill, 1958; Evan, 1966; Kauftnan, 1985, p. 41; Posen, 1984, p. 43; Scott, 1998, p. 147; Scott, 1998, pp. 124, 147; Smith, 1980, p. 376; Thompson, 1967, pp. 27-28).

Organizations draw on task environments for a wide variety of physical,

technological, and symbolic inputs. Organizations obtain necessary inflows of capital, participants, and technology from external suppliers. Capital provides the enterprise with the financial resources necessary to acquire other material and informational resources. Participants bring knowledge, skills and experience that organizations use to perform tasks. Technologies include the equipment, machines, instruments, and information necessary for the organization to accomplish work. Of course, interactions between enterprises and task environments are not unidirectional: organizations trade outputs with customers or clients in return for more inputs. In order to receive the necessary inputs, organizations must produce goods or services valued by their customers or clients. This is true for both legal and illegal organizations (Posen, 1984, p. 43; Scott, 1998, pp. 21-22, 229; Smith, 1975, pp. 340-341; Thompson, 1967, p. 28).





63


Hostile environments

Environments are not only a source of opportunities and resources for

organizations; they are also a source of uncertainty and risk. The ongoing flows of personnel, resources and information between organizations and their task environments are frequently erratic and unpredictable. While a number of factors affect the degree of risk and uncertainty within an environment, few are as salient as the degree of environmental hostility, as opposed to benevolence. A hostile task environment contains two or more interconnected adversaries, at least one of which seeks to dismantle other actors within the system. While a benevolent environment is not necessarily devoid of competition, it is "friendly" in the sense that the ultimate goal is to out-perform but not undermine the other units, so that all may survive to play another day. Organizational Structure

Structure refers to relationship patterns among participants in organizational

settings. These patterns emerge from values, norms, roles and routines that characterize how organizations function. Values refer to criteria for selecting desired goals of organizations, such as increasing profits and ensuring organizational survival. Norms are general rules that specify appropriate means for pursuing goals, such as exploiting new markets and investing in research and development. Roles are behavioral expectations for organization participants, often dealing with task performance. Roles define the division of labor within organizations, assigning participants to tasks that facilitate collective aims. Structure is not arranged randomly but ordered into coherent "beliefs and prescriptions governing the behavior of participants" (Scott, 1998, pp. 17-18). These patterns provide enduring, but not necessarily static, frameworks for structuring





64


authority, work, and communication within organizations (Argyris & Sch6n, 1996, pp. 89; Donald & Wilson, 2000, p. 194; Leavitt, 1965, p. 1144; Leavitt et aL, 1973, p. 4; McCluskey & Wardle, 2000, p. 252).

Routines

Routines are an essential part of organization structure. They include rules, procedures, conventions, and practices that guide individual behavior within organizations. Routines "interlock" behavior among multiple participants by matching action to informational cues. Routines provide "rules of thumb" for participants to respond to incoming stimuli in ways that make sense. In this manner, routines coordinate the behavior of different individuals into coherent and predictable patterns of behavior. This coordination allows organizations to accomplish work reliably and consistently (Cohen & Bacdayan, 1996, p. 348; Cyert & March, 1963; Feldman, 2000, p. 611; March & Simon, 1958; Weick, 1979, pp. 112-115).

In addition to organizing individual action into collective behavior, routines

influence participants by providing rules for conduct, along with an incentive system that matches rewards and punishments for specific action. Through training and socialization, participants learn the rules and procedures of their organization and receive rewards (or punishments) according to their ability to follow them. Successful participants generally move up the ranks of the organization and tend to apply the same criteria to their subordinates. In this manner, routines frequently become institutionalized within organizations, outlasting individual participants and sometimes even their original usefulness (Levitt & March, 1988, p. 320; March & Simon, 1958, p. 145; Posen, 1984, p. 44).





65


Organizations use routines to structure authority relations, allocate resources, make plans, perform tasks, assign roles, communicate information, build culture, and solve problems. Routines describe how organizations structure authority among different layers of decision-making hierarchy. They describe the functional responsibilities of different roles within the collective. Routines explain how organizations create the goods or services that they produce. They show how organizations maintain records of at least certain aspects of their operations. Routines determine how information is shared within the collective and who has access to what data. They express important values and provide a means of constructing organizational culture. Finally, routines describe how organizations respond to existing problems and plan future endeavors (Cyert & March, 1963, pp. 103-104; Decker et al., 1998, p. 407; Levitt & March, 1988, p. 320).

Routines are nested in several levels of organizational behavior. Standard

operating procedures are simple rules of thumb for accomplishing basic tasks, as when a participant responds to a stimulus by performing a specific action. Basic procedures are combined to create more complicated action-sequences called performance programs and repertoires. A performance program is simply a cluster of procedures for dealing with a specific situation. A repertoire is the sum of the organization's performance programs for organizing behavior in an area of activity (March & Simon, 1958, p. 141; Posen, 1984, p. 44). 2

Like their building blocks, performance programs are triggered by stimuli. The sounding of an alarm bell in a fire station provokes a fairly elaborate program of activity carried out by numerous participants. Two or more procedures can be combined to form 2 This is similar to Swidler's notion of cultural tool kits. According to Swidler, culture influences action by providing a repertoire or "tool kit" of habits, skills, and styles from which people construct "strategies of action." See Swidler (1986) and Kier (1997).





66


a performance program. Posen provides a simple but illustrative example in his organizational analysis of military doctrine. In most Western armies, it is standard operating procedure for tanks to fire from the "hull-down" position as this provides greater protection from enemy fire. Another standard procedure calls for tanks to advance by exploiting topographical irregularities in the surrounding terrain. Military organizations often combine these two procedures into an effective performance program: while some tanks advance on the enemy by exploiting the terrain, others remain in defensive firing position and provide cover for the advancing units. An organizational repertoire refers to the combination of performance programs available to an organization when undertaking an activity. The performance program described by Posen is one of many available in a military organization's repertoire for engaging in land-based warfare (March & Simon, 1958, p. 141; Posen, 1984, p. 44).

Organizations vary in the degree to which their routines are formalized. Rules

and procedures may be considered more or less formal to the extent that they are codified in documents. "Formal" routines are explicitly documented in manuals, files, data bases, financial accounts, and other forms of organizational memory. "Informal" routines are not recorded in documents but exist in the inter-subjective mind of organization participants. These customs, norms, rules and procedures are shared verbally and reinforced amongst participants through socialization processes. Certain understandings of "how things are done around here" are reinforced over others as participants engage in their daily activities. Vvlile some organizations may contain mostly formal or informal routines, many contain both types, with some rules and procedures formally recorded in





67


organizational memories and other routines found in inter-subjective understandings among participants.

Participants in Organizations

Organizations pursue their tasks through participants that make contributions in return for inducements. Participants supply the human resources that allow organizations to accomplish work. They define goals and objectives, form routines, perform tasks, make decisions, share information, plan future actions, and address problems. Participants join organizations for different reasons. Moreover, their assessments of satisfactory inducements are subjective. Even wages are not a sufficient inducement for all potential recruits. Human beings seek authority, status, power, fellowship, and a host of other entitlements besides money. Complicating matters, participants frequently belong to more than one organization, and their loyalty to the different collectives varies considerably. Participants flow from one organization to another, carrying with them values, norms, and beliefs about themselves and the roles they play (Barnard, 1938; Cohen et al., 1988, p. 295; Feldman, 2000, p. 614; Leavitt, 1965, p. 1144; Posen, 1984, p. 43; Scott, 1998, p. 19).

In addition to their myriad contributions, participants are also a source of great uncertainty for organizations. To accomplish their tasks, organizations require coordination and control, and this is not easily achieved among participants with fluid participation, wavenng loyalties, and different conceptions of satisfactory inducements. Moreover, even the most loyal and satisfied participants are limited in their ability to pursue organizational aims. Like all human beings, they are bound by cognitive, corporeal, and cathectic constraints. Of course, organizations are designed with these





68


limitations in mind, and much of their social structure can be understood as mechanisms by which organizational leaders and managers seek to channel participants' competing aspirations, energies and activities into a coherent and consistent pattern of behavior that facilitates the organization's purpose. However, people are not mindless automatons, and their aspirations, hopes, and fears cannot be calculated and manipulated like threshing gears and rotating gyroscopes (Feldman, 2000, p. 614; Posen, 1984, p. 43; Simon, 1997). Technology

Organizations perform work. Some organizations manufacture materials or

machines; others provide services, information or other products. Technology refers to the material and symbolic inputs that organizations use to complete work. Technology includes the equipment, instruments, and information that participants use to complete their activities, as well as the technical knowledge and skills of participants themselves (Scott, 1998, p. 2 1; Leavitt, 1965, p. 1144; Leavitt et al., 1973, p. 4).

Whether the purpose of an organization is to produce legal or illegal products, it draws on necessary technological resources. If the purpose of an organization is to provide education, it will make use of material and symbolic resources that allow it to accomplish this aim. Such resources include buildings for housing classrooms and administrative offices, computers, projectors and books for developing and delivering course materials, and the accumulated knowledge and skills of educators and their support staff. Likewise, an organization that provides escort services will require an office and/or other physical location(s) to coordinate and implement operations; telephones, beepers and other equipment for communicating among participants, customers, suppliers, and regulators; computers or other instruments for keeping track of





69


organizational inputs and outputs; support materials and services that allow participants to carry out their activities; and the knowledge, skills and experience of sex workers themselves.

Depending on the degree of hostility in the task environment, criminal

organizations may be required to devote a considerable portion of their technological resources to conceal and otherwise protect the physical integrity of their operations. However, some illicit activities, including illegal gambling and prostitution, are more accepted among law enforcement and government officials than others, such as narcotics and weapons trafficking (Reuter, 1983; Smith, 1975). For this reason, it is not surprising that many contemporary narcotics organizations devote a substantial portion of their technological resources towards reducing uncertainty and risk emanating from the task environment. Trafficking enterprises use these material and symbolic inputs to hide narcotics shipments, corrupt government officials, encipher communications, gather counter-intelligence, stockpile inputs, and maintain warehouse inventories.

Defining Organizational Learnin

Organizations learn when participants receive, interpret, and apply information to behavior through organizational routines. Organizational learning is a form of organizational action. When an organization "acts," such that participants fulfill roles and perform functions on behalf of the collective, then it may be said to "learn" when participants gather, analyze, and apply information (Argyris, 1993, p. 8; Argyris & Sch6n, 1996, p. 10; Farkas, 1998; Hedberg, 198 1, p. 3; Simon, 1996, p. 176). Throughout this study, organizational learning refers to a certain type of behavior performed by individuals acting on behalf of the collective.





70


The process of organizational learning can be conceptually disaggregated into three steps: acquiring, interpreting, and acting on information (see Figure 3-1 below) 3 Organizational learning unfolds as information is discovered, shared, and applied by participants that coordinate their activities through formal and informal routines. The three phases of organizational learning are disaggregated for heuristic purposes; in reality, information can be acquired, interpreted, and applied simultaneously. The different steps are connected through feedback loops that reinforce certain interpretations and behaviors while disregarding others. There are a number of impediments to organizational learning, and the process may be blocked at any phase. I discuss obstacles to learning in detail later in the chapter.


Acquiring Interpreting Acting on
information 10 information P- information


T ---I


Figure 3-1 Simple Process Model of Organizational Learning Source: Adapted from Daft & Weick (1984), Garvin (2000). Acquiring Information

The first step in the organizational learning process is gathering information.

Information is the lifeline of organizational learning. Information includes know-how, understanding, techniques or practices and is useful to the extent that it helps organizations achieve targets. Organizations acquire information through their own trial Although the language varies, a number of analysts conceptualize organizational learning as a process composed of three or four discrete stages. Daft and Weick describe learning according to three analogous stages, including scanning (data collection), interpretation (data given meaning), and learning (action taken) (Daft & Weick, 1984, p. 286). Kim describes learning as composed of four stages: observing, assessing, designing, and implementing (Kim, 1993, p. 44). See Garvin for discussion on this point (Garvin, 2000, endnote 4, p. 223).





71


and error experience and exogenous search. Trial and error experience refers to the process by which organizations interact with their environment, using a variety of tactics and strategies designed to achieve targets. The experience that emerges from this activity provides a valuable guide for future behavior. Search refers to the process by which organizations gather information from sources exogenous to the organization but endogenous to the task environment. Search provides the means to draw on the knowledge and experience of others, including customers, consultants and competitors, in making decisions (Argyris & Sch6n, 1996, p. 3; Weick, 1979).

Gathering information is costly. It involves an expenditure of organizational resources, including time, attention, personnel, and money. Because resources are limited, organizations are more likely to acquire knowledge and experience when they have a reason to do so, such as the need to find a solution to a problem. As targetoriented entities, the failure to meet targets or objectives is a common source of problems in organizations. The failure to match expected outcomes with actual outcomes produces emotional arousal on the part of participants that triggers a search for information, including alternative courses of action to resolve the problem (Allison, 1971, p. 84; Argyris & Sch6n, 1996, p. xxiii; Garvin, 2000, p. 23; March, 1988, p. 3; March & Olsen, 1989, p. 60; Weick, 1995, pp. 45-46).

Interpreting Information

For the learning process to proceed, participants must interpret information. Daft and Weick provide an illuminating description of this complex process:

Organizations must make interpretations. Managers literally must wade into the ocean of events that surround the organization and actively try to make sense of
them. Organizational participants physically act on these events, attending to
some of them, ignoring most of them, and talking to other people to see what they





72


are doing.... Interpretation is the process of translating these events, of
developing models for understanding, of bringing out meaning, and of assembling
conceptual schemes among managers (Daft & Weick, 1984, p. 286).

Interpretation is a social process through which organizational participants share their perceptions and construct inter-subjective understandings that allow them to make sense of relevant events. When engaged in interpretation, or "sensemaking," participants communicate, share knowledge and experience, and construct linkages between the present situation and prior experiences, often with the assistance of analogical reasoning (Daft & Weick, 1984, p. 286; Khong, 1992; Neustadt & May, 1986; Weick, 1995, pp. 4546).

Organization managers and other participants are individual cognitive agents; they draw on "schemas" when analyzing information.4 Schemas are knowledge structures that individuals use to "lend order to an otherwise incomprehensible confusion of information and experience" (Reiter, 1996, p. 21). Schemas help individuals make sense of what has gone before them and what transpires around them by providing cognitive shortcuts or frames of reference. As personalized frames, they can be an impediment to organizational learning, preventing decision-makers from interpreting a problem or solution in the same light.

Within organizations, participants share, record, and interpret knowledge and

experience through formal and informal organizational memories. Formal organizational memories include files, manuals, correspondence, databases, financial accounts, and physical objects that record and store knowledge, such as monthly accounting statements, minutes from meetings, electronic mail, and computers (Arygris & Sch~n, 1996, p. 16; 4~ "Schemas" are knowledge structures that individuals use to "lend order to an otherwise incomprehensible confusion of information and experience" (Reiter, 1996, p. 21). Schemas help individuals make sense of experience and their environment by providing cognitive shortcuts or frames of reference.





73


Simon 1997, p. 218). Although recent advances in information technology have reduced the costs of recording and retrieving information, much organizational experience remains undocumented. Undocumented knowledge is often transmitted through informal organizational memories. These include the conversations, stories, and myths through which participants construct and conserve inter-subjective understandings regarding appropriate practices and procedures (Levitt & March, 1988, p. 327; Argyris & Sch6n, 1996, p. 16).

Organizations interpret only a small portion of their experience at a time.

Organizational memories are evoked at certain times by particular parts of the collective. Depending on how they are organized, some aspects of organizational memories may be more accessible than others. In particular, the availability of memories is associated with their recency and frequency of use, as well as their location within the organization. Recently and frequently used memories are easier to retrieve than those that are used only sparingly or have not been accessed over a considerable period of time. Information stored in electronic documents that forrn part of computerized data bases, or paper documents located in well-organized filing systems are more likely to be used than isolated documents stored in obscure locations (Levitt & March, 1988, pp. 328-329). Appl3jng Information

The process of organizational leading is not complete until knowledge that has been acquired and interpreted by participants is applied to collective behavior (Garvin, 2000, p. 26). Organizations enact knowledge and experience through routines. Routines include rules, procedures, conventions, and practices that guide individual behavior within organizational settings. Routines provide the mechanism through which the





74


behavior of multiple participants is "interlocked" (Feldman, 1989, p. 136; Hutchins, 1996, p. 50; Weick, 1979, p. 3). Routines are transmitted over time through socialization, training and personnel movement. They are recorded in formal and informal organizational memories. Moreover, they are "independent of the individual actors who execute them and are capable of surviving considerable turnover in individual actors" (Levitt & March, 1988, p. 320). Members may come and go and the organization's leadership may change, but when an organization learns, knowledge is preserved in formal and informal memories and routines that outlast individual participants. Routines that are associated with desirable consequences, such as improved task performance, tend to be selected and retained by organizations If participants can be shown to alter a routine or series of routines as a result of acquiring and interpreting experience, then the organization may be said to have learned from that experience (Daft & Weick, 1984, p. 285; Fiol & Lyles, 1985, p. 804; Hedberg, 1981, p. 6; Levitt & March, 1988, p. 326; Nye, 1987, p. 381; Zeng, 1999, p. 39).

Organizations differ with respect to the extent that their behavior is determined by formal or informal routines. Large government bureaucracies with extensive institutional histories tend to document experience through codified rules and procedures. Small craft-based enterprises rely more on informal understandings transmitted verbally among participants. Organizations that face complex uncertainties or hostile task envirom-nents rely more on informally shared understandings than organizations that fiction with relatively simple, stable environments (Levitt & March, 1988, p. 327; Ouchi, 1980).





5 However, this is not always the case, as we shall see below when discussing learning under ambiguity.





75


Research Propositions

Organizational learning refers to the process by which participants gather, interpret, and apply information on behalf of an organization. The application of knowledge and experience occurs through changes in informal and formal routines that guide participants' behavior. This study applies a process-oriented model of organizational learning to two units of analysis: Colombian drug trafficking enterprises and U.S. and Colombian drug enforcement agencies.

If participants within either organization alter their routines as a result of

acquiring and interpreting knowledge and experience, then the enterprise in question may be said to have learned. In trafficking enterprises, changes in organizational practices and procedures may involve the production, transportation, and distribution of illicit drugs, or money laundering activities. For law enforcement agencies, changes in organizational practices and procedures may involve criminal investigations or drug enforcement operations. For both organizations, changes in routines may improve organizational task performance, but they are not required to do so.

These ideas yield the following empirically verifiable propositions.

Pl: If participants in Colombian drug trafficking enterprises gather, interpret, and apply information to collective behavior by changing existing routines or creating
new ones, then organizational learning has occurred.

P2: If participants in U.S. and Colombian drug enforcement agencies gather, interpret, and apply information to collective behavior by changing existing
routines or creating new ones, then organizational learning has occurred.

Alternative Explanations for Organizational Change

Learning is not the only source of change in organizations, and it is important to clarify alternative explanations for this phenomenon. Perhaps the most common source





76


of non-leaming change is the imposition of new or modified routines by more powerful external actors on the unit of analysis. In the late 1960s and early 1970s, U.S. federal drug enforcement agencies underwent a variety of organizational changes. However, these changes were due to the imposition of new organizational structures and procedures by policyrnakers in the Johnson and Nixon administrations, rather than information processing by decision-makers within the drug enforcement agencies. 6

It is possible to argue that learning takes place in such circumstances, but it is the more powerful actor that learns and passes lessons on to the subordinate agency in the form of routines that encode experience. While this may be true in some cases, the explanation is largely unsatisfactory because it allows the analyst to change the unit of analysis expostfacto. To reliably evaluate the research propositions, it is important to maintain a consistent unit of analysis during the design and implementation of the research. In this study, the units of analysis are Colombian drug trafficking enterprises and U.S. and Colombian drug enforcement agencies. If change in routines is not due to the information processing activities of these organizations, then learning, according to the parameters of this research design, has not occurred.

P3: Organizational leading does not occur when changes in organizational
routines are due to the imposed preferences of more powerful political or
bureaucratic actors.

Another source of non-leaming change is environmental selection. This occurs when changes in organizational structures or operations appear as one set of actors is selected out of the system and replaced by another. Because they operate in hostile task environments it is not uncommon for trafficking enterprises to be selected out of the illicit drug industry (i.e. identified and dismantled) by drug enforcement agencies. In this
6 For analysis of these refonns, see Rachal (1983) and Wilson (1978, 1989).





77


manner, successful drug enforcement yields a process of "natural selection" that favors some criminal enterprises over others. Trafficking organizations that survive drug enforcement efforts may contain more effective combinations of decision-making hierarchies, leadership styles, and operating procedures (Alchian, 1950; Katifinan, 1985; Moe, 1984, p. 746; Nelson & Winter, 1981).

Over time, organizational forms better suited to survive hostile task environments may populate the illicit drug industry (see Chapter 2). However, if survivors do not change their routines as a result of information processing, then learning has not taken place. Surviving enterprises may be more effective than their predecessors, at least in terms of their ability to elude the state, but their ability is not due to organizational learning. On the other hand, if participants in trafficking enterprises recognize that they must change their organizational routines to survive hostile environments, and they do so as a result of this recognition, then leading has transpired.

P4: Changes in the illicit drug industry may be due to natural selection rather than
organizational learning by individual enterprises.






Change in
Learning organizational
routines


Environmental
selection


Figure 3-2 Alternative Sources of Change in Organizational Routines





78


Organizational learning is a cognitive activity performed by participants that observe and interpret events and make decisions based on their perceptions of reality. While organizations are open systems, sensitive to exogenous pressures, the source of learning is ultimately endogenous to the enterprise. This does not require participants with access to complete information or unlimited computational abilities (see discussion below). It does require participants that recognize they belong to a goal-oriented collective, and act and manipulate information on behalf of this entity. To qualify as organizational learning, change in routines must be the result of information acquisition and processing activities on the part of participants. Otherwise, it is not certain that organizational change is due to leading or alternative explanatory variables, such as
7
power or environmental selection.

Learning Ecologies

Organizations interact with other organizations that share common task

environments. Organizations draw on task environments for resources, including personnel, equipment, capital, and information. In common task environments, one organization's knowledge and experience can be can be captured by others through the diffusion of technology. Technology, understood broadly to include material equipment, routines, and organizational memories, spreads through environments as organizations copy or mimic one another. Often, targets of technology transfer are those organizations that are seen as successful by their peers. These model organizations use superior


7 By distinguishing learning ftom other sources of organizational change, this study avoids a common tautology in the literature. However, by making the distinction between learning and change so explicit, I have increased the difficulty of demonstrating learning in subsequent chapters. My challenge is to show that changes in routines by drug trafficking enterprises and drug enforcement agencies are due to information acquisition, processing, and application, rather than other factors, such as power and environmental selection.





79


technologies believed to be worthy of emulation. Agents of difftision include participants that communicate with inter-connected organizations, including competitors, and independent consultants that contain the requisite knowledge and experience (DiMaggio & Powell, 1983, pp. 150-153; Levitt & March, 1988, p. 329; March, 1999, p. 45; Scott, 1998, p. 213).

"Ecologies of teaming" develop when multiple organizations learn from each other within a shared environment. The classic type is a "collection of competitors," in which rivals are closely linked through the diffusion of experience and the effects of their activities on one another. In this type of teaming ecology, organizations compete not only for resources, customers, and profits but for political power and social legitimacy. Organizations learn from their competitors by analyzing their competitor's reactions to their own actions (DiMaggio & Powell 1983, p. 150; Hedberg 1981, p. 13; Levitt and March 1988, pp. 329, 331-332).

P5: Colombian drug trafficking enterprises and U.S. and Colombian drug
enforcement agencies learn from each other within common competitive teaming
ecologies.

Competitive teaming ecologies complicate the task of analysis. Because organizations in these environments draw on their competitors for knowledge and experience, teaming by the one cannot be fully understood without reference to teaming by the other. In recognition of this complication, this study analyzes interactions between trafficking enterprises and drug enforcement agencies to determine whether they team from each other (see Chapter 6).





80


"Productive" Learning

Among organization theorists, learning is commonly equated with improved task performance. 8A number of analysts have drawn on this assumption to define learning as the organization's ability to improve task performance over time. According to Fiol and Lyles, learning is "the process of improving actions through better knowledge and understanding" (Fiol & Lyles, 1985, p. 803). Dodgson describes learning as "the ways firms build, supplement and organize knowledge and routines around their activities... and develop organizational efficiency by improving the use of the broad skills of their workforces" (Dodgson, 1993, p. 377). In their early collaborative research, Argyris and Schdn equate learning with increasing organizational effectiveness, however, in their more recent work they distinguish "productive" learning, which improves performance, from "generic" learning that does not. The former variety remains the focus of their theoretical and empirical work. Other analysts have chosen different labels for learning that improves performance, including "instrumental, ". .effective," and "efficiency" learning (Argyris & Sch~Sn, 1978; Argyris & Sch6n, 1996, p. 4; Hedberg, 1981, p. 3; Ndumbaro, 1998, p. 39 Tetlock, 1991, p. 35).

The classic example of efficiency learning is found in early studies of airplane manufacturing. Beginning in the 1920 and 1930s, a number of industrial economists discovered that the costs of airplane manufacturing fell predictably with increased production volumes (Wright, 1936; Arrow, 1962). Increases in production were seen as proxies for greater skill and knowledge accumulated through repetition of performance tasks (Garvin, 2000, p. 94). This phenomenon-labeled the "learning" or "experience" 8 The assumption that learning results in improved organizational performance derives from the broader assumption in "behavioral" studies of organizations that organizational learning-and behavior more generally-is fundamentally geared towards "targets" or goals.






81


curve-was subsequently found in a number of manufacturing industries and other production tasks involving repetition, although learning rates varied considerably across industries, products and time (Levitt & March, 1988, p. 321; Epple, Argote & Devadas, 1996; Yelle, 1979).9

One problem with the efficiency conception of learning is that it disregards instances whereby changes in routines based on the acquisition and interpretation of information fail to improve task performance. As numerous organization theorists have recognized, learning does not always increase effectiveness (Cook & Yanow, 1996; Hedberg, 1991, p. 89). Like individuals, organizations may learn dysfunctional behavior, as when they adopt routines that hinder task performance or require more organizational resources to achieve the same outcome. Productive learning is an important subspecies of organizational learning. However, analysts should distinguish clearly between the two varieties and avoid defining the latter by the former (Argyris & Sch~Sn, 1996). Organizational learning may improve task performance, and it may not.

Levels of Learning: Tactical vs. Strategic Adaptation

A number of analysts have identified different "degrees" or "levels" of

organizational learning. While a variety of terminology has been developed to explain this process the basic idea revolves around the distinction between the goals and objectives of organizational behavior. "Simple," "low" or "lower," "single-loop" learning and "adaptation" describe the process by which organizations adjust the means






9 Moreover, in some industries production costs have actually increased over time and accumulated experience, thus demonstrating little capacity for learning. The commercial nuclear reactor industry is one notable-and firightening-example (Perrow, 1984, pp. 34-35; Bupp & Derian, 1978).





82


of behavior in response to environmental feedback. 10 "Complex ... .. high" or "higherlevel ... .. meta-level ... .. double-loop" learning, and just plain "learning" all refer to the

process by which organizations respond to goal-changing feedback by altering the fundamental values and basic goals that drive organizational behavior." However, the dividing line between simple and complex learning is inherently fuzzy. Within organizations, goals and objectives are ambiguous and frequently change over time. Moreover, the meaning of goals and objectives varies in different parts of the organization. An objective for one part of an organization represents a goal for another. Even worse, different parts of the same organization may have conflicting goals and objectives (Levitt & March, 1988, p. 325; Tetlock, 1991, p. 46).

Distinguishing between the goals and objectives of organizational behavior is

tricky business. Yet, it is also true that not all learning is the same. There is a difference between ad using the day-to-day activities of an organizational subunit and a fundamental transformation in the organization's raison-d'etat. Rather than focus on the nebulous distinction between goals and objectives, an alternative approach is to distinguish between tactical and strategic changes in organizational behavior.

In the discourse on warfare, tactics refer to "small-scale" actions that occur during the course of battle. Strategy refers to the immediate or long-range planning of tactical maneuvers that takes place prior to battle. According to the economic historian, Alfred Chandler, strategy determines "basic long-range goals and objectives of an enterprise, and the adoption of courses of action and the allocation of resources necessary for


'0 Deutsch (1966) and Nye (1987), Fiol and Lyles (1985) and Hedberg (198 1), and Arygris and Sch6n (1978, 1996), and Haas (1990) use these different terms in their respective studies. 11 Again, Deutsch (1966) and Nye (1987), Fiol and Lyles (1985), Hedberg (198 1), Arygris and Sch6n (1978, 1996), and Haas (1990) use the different terminology.





83


carrying out these goals" (Chandler, 1962; Scott, 1998, p. 287). Tactics are the procedural and technological means by which organizations implement courses of action. VA-lile strategy is the overarching concept, military strategists have long-recognized the causal relationship is not unidirectional. Tactics can influence strategy, as when technological innovations developed on the field of battle lead to changes in strategic planning.

In this study, I am concerned not with annies waging warfare but criminal

enterprises that produce, process, transport, and distribute illicit drugs, and the state drug enforcement agencies charged with stopping them. Therefore, the warfare analogy is not without merit. As seen in Chapter 2, narco-narc interactions have developed increasing degrees of intensity during the last two decades of the war on drugs. In this context, tactical routines refer to the rules, procedures, conventions, technologies, and intersubjective understandings used by trafficking enterprises and drug enforcement agencies in carrying out their day-to-day tasks and meeting their immediate objectives in the ongoing "battles" of the drug war. Strategic routines refer to those rules and procedures governing long-range goals and objectives, strategies for achieving these goals and objectives, and basic organizational structures. The latter may include decision-making hierarchies, management layers, and organization subunits.

Most organizational learning takes place at the level of tactical routines. Organizations tinker with tactics in response to-frequently negative-feedback regarding task performance. The information gathering and interpretation costs of tactical learning are relatively minor. If the organization contains a number of alternative routines in its performance program, learning may consist of merely matching the most





84


appropriate routine to fit the existing contingency in "cybernetic" fashion. Still, even cybernetic adjustment requires human agents (rather than thermostats) capable of recognizing when a routine is no longer appropriate, searching organizational memories to find alternative procedures, and selecting the one with the best chance of producing favorable outcomes. 12

Organizations that engage in successful tactical adaptations have little need for strategic learning. 1 3 However, tactical adjustments do not always yield intended results. When organizations experience repeated failure with tactical adjustments, they are more likely to engage in strategic learning (Levy, 1994, p. 286; Tetlock, 1991, p. p. 28). Strategic learning involves changes in existing routines or the creation of new ones dealing with long-range goals and objectives, strategic action plans, decision-making hierarchies, organization subunits, and resource allocation. Strategic learning is more complex than tactical adaptation and less common. Ideas for strategic adaptations may come from within the organization or through search within the ecology of learning that surrounds it. Strategic learning may also include changes in the fundamental assumptions and worldviews of decision-makers and the structure of the organization itself. 14






12 In their treatment of adaptation or simple leading, a number of organization theorists downplay the cognitive effort involved in these actions.
13 1 am grateful to James G. March for clarifying this point to me. 14 By including changes in "fundamental" values and worldviews under the rubric of strategic learning I am "lumping" where others have split. In particular, Hall distinguishes "second-order" and "third-order" learning and Tetlock distinguishes between "strategic" and "fundamental" learning (Hall, 1993, pp. 278-9; Tetlock, 1991, pp. 29-30). The main reason for lumping the two types has to do with the methodological and theoretical problems in distinguishing between ftmdamental goals and strategic objectives discussed above.





85

Learniniz under Ambiguity

To this point, the discussion has been deceptively simple. Organizations are described as purposive agents that intentionally create and modify routines through search and trial and error experimentation. However, in the real world of organizational behavior, leading often occurs under conditions of profound ambiguity and uncertainty. At each phase of the process, individuals and organizations confront a variety of psychological and organizational factors that complicate leading: bounded rationality, incomplete and imperfect information, biased inferences, ambiguous experience, and organizational inertia. Recognition of these factors increases our understanding of organizational learning and yields a more accurate and satisfying model of the process.



Acquiring Interpreting Acting on
information information information





Figure 3-3 Complex Process Model of Organizational Learning

Some difficulties are due to basic limitations in human cognition and

coordination. All individuals and organizations operate under conditions of information uncertainty and bounded rationality (Simon, 1997, 1955). Individuals and organizations lack access to perfect information (March, 1999, p. 2). Much about past experience remains unknown, and it is difficult to produce future anticipations based on present actions. Individuals face significant limitations in their ability to analyze feedback from complex environments. They use cognitive short cuts to make sense of the





86


overwhelming abundance of incoming stimuli. However, these cognitive schemes are based on prior beliefs that bias interpretations in subtle ways.

Within organizations, attention is a scarce resource; the time and effort these

collectivities direct towards information acquisition and interpretation is constrained by other needs (March, 1988; March & Olsen, 1988). Because problems influence priorities, attention is often forthcoming only in times of crisis, which means that information acquisition and interpretation are often made under conditions of stress. A number of these general difficulties manifest themselves as organizations acquire, interpret, and apply information and experience to routines.

As they acquire and interpret feedback from their environment organizations are vulnerable to a host of "learning disabilities," including blind spots, filtering, flawed interpretation, and secrecy. Blind spots occur when search is narrow or misdirected, leading managers to miss or misinterpret important feedback signals. Filtering occurs when critical information is ignored or downplayed because it is inconsistent with decision-makers prior beliefs or powerful interests within the organization (Garvin, 2000, pp. 28-29). Organization leaders tend to accept information that attributes success to their own actions and failure to the actions of others or to exogenous sources (Levitt & March, 1988, p. 324). As Scott Sagan notes, problematic situations "are not simply internal data points to be used to improve organizational performance. They are also political events for which credit and blame must be assigned." Interpretation is easily politicized within organizations. Sometimes it may be easier to "kill the messenger" (i. e. suppress or distort information) than assign blame to powerfid coalitions or individuals (Sagan, 1993, p. 208; Huber, 199 1, p. 95). While organizational learning depends on the




Full Text
148
prosecutors in Washington (Farah, 1999a; Kerry, 1997, p. 77; Pallomari, 1997, V. 39,
pp. 6312, 6319).
During the 1990s, the Rodrguez Orejuela organization established an elaborate
intelligence network in Cali using local cab drivers that provided information regarding
suspicious activity, in exchange for receiving cars through interest-free loans and free
mechanical assistance. Before such technologies were widely available in Colombia, the
taxis were equipped with beepers, communications radios, and cellular telephones to help
them carry out their counter-surveillance mission. Back in the U.S., the enterprise hired
attorneys to research government regulations regarding wiretaps, and gather information
about law enforcement tactics in criminal cases targeting their distribution cells (Derian,
1999; Evans, 1999; Francis, 2000; McNamara, 1999; Kerry, 1997, p. 78; Serrano Cadena,
1999, p. 154).
Informal Exchange of Information
The diffusion of knowledge and experience among participants in trafficking
enterprises is often fairly informal. Participants meet in informal settings, such as
restaurants, bars, and dance clubs, to discuss impending activities, business problems, and
recent developments in the industry. Nstor offers an example from his trafficking days,
Yeah, sure we sit down at a restaurant. I tell them what to do. Nothing like a corporate
meeting. Sometimes we meet half way from one location to another. Very informal
meeting, no set of written instructions (Nstor, 2000). Private social gatherings,
including birthday parties, baptisms, and weddings, provide participants additional
opportunities for swapping stories and exchanging trade rumors. In this manner, less
experienced participants are socialized to the norms and practices of their organization,


20
Emerald traders were not the only Colombian smugglers to expand into the illicit
drug industry. In the 1960s and 1970s, a number of black marketeers in cigarrettes,
whiskey, and domestic appliances established themselves in marijuana and cocaine
exports. Smugglers drew on contacts, customs, and capital they acquired in smuggling
these commodities to diversify into marijuana and subsequently cocaine. The subsequent
development of the marijuana and cocaine industry in Colombia was heavily influenced
by the informal practices and procedures of these contrabandistas (Arango & Child,
1987, p. 130; Betancourt & Garcia, 1994, p. 107; Lpez Restrepo, 2000; Torres, 2000;
Valds, 2000; Valencia, 2000).
The re-emergence of the Colombian heroin trade in the 1990s was stimulated by
well-established cocaine traffickers seeking to diversify their illicit activities (FBI 1993;
Farah 1997; Semana 1999; Zabludoff interview). In all three cases, criminal
entrepreneurs drew on previously acquired contacts, customs, and capital to establish
themselves in more lucrative endeavors. This knowledge and experience was critical for
smugglers operating within highly uncertain and even hazardous business environments.
# i
Phase One1930s: Colombia as Transit Point
In the early 1930s reports of captured Colombian drug smugglers began to appear
in U.S. government documents and Colombian press accounts. In 1932 a smuggler was
captured in the Panama Canal Zone carrying 25 grams of cocaine hydrochloride hidden
inside a cartridge belt. The individual claimed that he had acquired the drug from a
group of traffickers operating in Cartagena, Colombia. The following year, a different
smuggler was arrested in the Canal Zone carrying 100 grams of cocaine hydrochloride,
2 This section draws on the work of Eduardo Senz Rovner (1996, 1997), a Colombian economic historian
that conducted archival research at the National Archives in Washington, D.C.


49
1995; Latin American Weekly Report, 1995b, 1995c, 1995e, 1995f, 1995g, 1995h, 1995i;
Serrano Cadena, 1999).
However, like earlier crackdowns against the Medellin enterprises, these price
increases disappeared completely within a year, reflecting the drug industrys ability to
absorb even the most intense drug enforcement efforts in the long term. Indeed,
following a brief period of regeneration, new and revitalized trafficking groups emerged
in Colombia to produce greater amounts of cocaine and heroin than before.
Table 2-2 Colombian Drug Enforcement Indicators, 1980-1995
Year
Persons
apprehended
Cocaine
interdicted
(kilos)
Cocaine
base
interdicted
(kilos)
Heroin
Interdicted
(kilos)
Marijuana
interdicted
(kilos)
Drug labs
destroyed
1980
358
748
0
0
192,422
29
1981
798
339
0
0
3,302,242
31
1982
1,139
651
0
0
3,283,878
165
1983
1,073
2,083
0
0
3,537,387
113
1984
5,251
19,582
9,448
0
4,301,263
137
1985
1,951
4,239
3,674
0
1,021,046
696
1986
3,699
3,039
4,070
2
846,000
572
1987
4,732
8,326
6,712
2
1,287,272
1,359
1988
4,929
12,047
2,554
0
842,994
655
1989
5,217
24,668
9,601
0
617,925
389
1990
6,253
16,000
3,429
850
659,047
268
1991
6,349
59,347
8,223
0
381,157
235
1992
6,770
28,016
5,289
36
206,934
95
1993
8,136
19,137
6,945
42
505,274
241
1994
7,221
28,145
22,580
94
161,322
334
1995
8,053
34,577
15,375
184
171,347
331
Source: CNP (2000)
Phase FiveMid-1990s to 2000: Decentralization of Colombian Drug Trade
With the fall of the Medellin and Cali cartels the Colombian drug trade has
decentralized. Instead of a handful of vertically integrated core organizations that
dominate the industry, hundreds of independent trafficking groups operate throughout


192
enforcement intelligence. With the creation of NDIC, policy makers reaffirmed the El
Paso Intelligence Centers long-standing role as the primary provider of tactical and
strategic drug intelligence (GAO, 1993, pp. 8-15; also, see GAO 1992).4
DEAs Role in Counter-Drug Intelligence
As the primary drug enforcement agency in the U.S. federal government, the
DEA directs the production and dissemination of counter-narcotics intelligence. An
Intelligence Division, headed by the Assistant Administrator for Intelligence, oversees
the agencys intelligence activities. This division produces a variety of intelligence
products for decision-makers, including strategic assessments of production and
consumption trends and trafficking patterns, tactical and operational analyses of the
structure and membership of transnational and domestic trafficking organizations, and
financial reports on money-laundering techniques and the macroeconomic impact of the
illicit drug trade. As of 1996, the Intelligence Division contained three offices and one
center (see Figure 6-1 below).5 The Office of Investigative Intelligence designs,
organizes, and executes DEA investigative intelligence programs. This office includes
separate cocaine and heroin investigative units responsible for identifying major criminal
enterprises involved in the production, transportation, and distribution of these
psychoactive drugs. The Office of Intelligence Liaison and Policy plans strategy and
policy for DEA intelligence programs and coordinates their execution. The office
includes Foreign Strategic Intelligence Units for Latin America, Europe, and Asia and
4 The effectiveness of the latter reform is questionable. Some observers suggest that the NDIC, located in
out-of-the way Johnstown, PA, is more a creature of pork-barrel politics than drug enforcement needs.
Others see NDIC as an inferior analog of the highly sophisticated El Paso Intelligence Center (EPIC).
5 The following description and diagram are based on Richelson (1999). Because his discussion draws on
primary documents that are several years old, including a 1996 memorandum from the DEA Administrator
to the Assistant Administrator for the Intelligence Division, some of this information may be dated.


aptitude and hard work. Homero describes how he benefited from turnover in his
trafficking group:
128
Homero: Initially, I was a mule, but after I traveled to the U.S... they needed
someone they could trust to stay here to deliver [the mules] to others, and that was
how I became involved in that aspect of the business.
Author: So it functioned as a form of promotion?
Homero: Yes, of course, because what happens is that you go and return a couple
of times, and you are a competent person, so possibly in the U.S. or whatever
country, they lose someone... for whatever reason, he goes to prison or he doesnt
want to work anymore, so they need someone of confidence in order to secure the
distribution, that the production line goes to the consumers. So I began to receive
mules that arrived first in Miami, then I went to New York, to obtain mules. Then
I was contracting cars, etc. (Homero, 2000, authors translation).
When turnover is high, promotion opportunities are numerous, and competent
associates may find themselves rising rapidly through the managerial ranks. According
to Fuentes, a member of the core organization led by Helmer Herrera quickly rose from
errand boy to section leader in charge of all collection, counting, and packaging of
drug monies generated by the M7 cell (Fuentes, 1998, pp. 87-88). In some enterprises,
imprisonment can lead to promotion for the convicted associate, provided the person
serves the sentence without cooperating with the authorities. Jos Patio, a member of
the Rodrguez Orejuela organization, was arrested in New York in 1979 for trafficking
activities. While all of his co-defendants fled the jurisdiction, Patio dutifully served his
sentence and was promoted upon his release, eventually becoming the head of the
organizations European operations (Minority Staff Statement, 1989, pp. 125-126).
Promotion is complicated when leaders are the ones to be replaced. Who
succeeds the entrepreneur? In small, clan-based trafficking enterprises, where authority
is centered on a single figure, the leader may be irreplaceable. In these groups, removal


224
how well nares play the game, they are always a step behind their adaptive counterparts
(Brandy, 2000; Crank, 2000; Fernandez, 2000; Frechette, 1997; Gale & Rodriguez, 2000;
Uribe Ramirez, 2000).
The Survival Imperative
Outside of terrorism and warfare it is difficult to imagine a more hostile
environment than the one shared by nares and narcos. Drug enforcement agencies exist
to dismantle trafficking enterprises, and traffickers only survive to the extent that they
can avoid or undermine the states intentions. Environmental hostility can impede or
facilitate organizational learning (see Chapter 3). Hostility impedes learning by
increasing stress and anxiety for decision-makers, causing them to ignore or misinterpret
critical feedback. Hostility can also impede learning by inducing agents to make rapid,
maladaptive responses to spur of the moment crises. However, hostility also facilitates
learning by providing a clear and unambiguous signal: Change or die.
Task environments for trafficking enterprises are as hostile as any in the
organizational world. Colombian trafficking groups face danger from competitors,
suppliers, customers, and, of course, law enforcement agencies (see Chapter 4). They
compete for market share and resources with hundreds of criminal enterprises, guerrilla
groups, and paramilitary organizations within and outside of Colombia (Arreguin, 2000;
Francis, 2000). With many of their opponents operating outside the rule of law,
trafficking groups face intense competition. Competitors frequently use deceptive and
violent means to defeat them. They are vulnerable to robbery from suppliers, customers,
corrupt police officials, even their own participants. Facing potential deception and
violence at every turn, traffickers must contend with enormous levels of stress.


243
As Chapter 6 makes clear, trafficking enterprises are not the only learners of
interest in this study. Similar to the narcos they seek to dismantle, law enforcement
nares acquire, interpret, and apply knowledge and experience to their activities. Drug
enforcement agencies draw on a variety of investigative practices and procedures to
gather information about trafficking enterprises and the illicit drug trade. This tactical,
operational, and strategic intelligence is analyzed, reported, and disseminated among
police officials, prosecutors, and policy makers. Drug enforcers in Colombia and the
U.S. use this knowledge to plan and carry out their everyday activities. Moreover,
developments in the Colombian drug trade and pressure from other government
institutions have led drug enforcers to make a number of structural and operational
adaptations. Structural adaptations include the creation of new enforcement
organizations, such as the DEA, and the formation of new sub-units within existing
agencies. Operational adaptations include innovations in investigative routines, such as
reverse undercover stings and controlled deliveries. One outcome of these structural and
operational adaptations is that drug enforcers have improved their ability to target major
trafficking organizations through complex conspiracy investigations, electronic
surveillance technologies, undercover operations, and financial investigations.
However, organizational learning by U.S. and Colombian drug enforcement
agencies has failed to yield satisfactory policy outcomes. Colombian trafficking
enterprises continue to produce and transport greater quantities of psychoactive drugs
than ever before. To explain this dilemma, Chapter 6 offers the concept of competitive
learning games, a metaphorical device to describe the accumulation of knowledge and
experience among drug enforcers and drug traffickers as they engage in repeated


BIBLIOGRAPHY
Abadinsky, Howard. The Criminal Elite: Professional and Organized Crime. Westport:
Greenwood Press, 1983.
Abadinsky, Howard. Organized Crime. 5th edition. Chicago: Nelson-Hall Publishers,
1997.
Adams, David. Nation's biggest cocaine case goes to trial today. St. Petersburg Times
(16 October 1995). Http://web.lexis-nexis.com/universe [Accessed 19 March
2001],
Adler, Patricia A. Wheeling and Dealing: An Ethnography of an Upper-Level Drug
Dealing and Smuggling Community. 2nd edition. New York: Columbia University
Press, 1993.
Aero, Jorge [pseud], Tecnical professional, Judicial Police Directorate, Colombian
National Police. Bogot. 21 March 2000.
Akers, Ronald L. Social Learning and Social Structure: A General Theory of Crime
and Deviance. Boston: Northeastern University Press, 1998.
Akers, Ronald L. and Gang Lee. Age, Social Learning, and Social Bonding in
Adolescent Substance Abuse. Deviant Behavior 19 (1999): 1-25.
Alchian, Armen A. Uncertainty, Evolution, and Economic Theory. Journal of Political
Economy 58.3 (1950): 211-221.
Allison, Graham T. Essence of Decision: Explaining the Cuban Missile Crisis. Boston:
Little, Brown & Company, 1971.
Alvarado, Luis Eduardo. Algunos resultados del combate del Estado colombiano frente
al lavado de activos. Coloquio: Revista de la Direccin Nacional de
Estupefacientes 6.6 (1998): 88-99.
Ambrose, Myles. Former head, Office of Drug Abuse Law Enforcement. Frontline: Drug
Wars. 2000.
Http://www.pbs.orgAvgbh/pages/frontline/shows/drugs/interviews/ambrose.html
[Accessed 21 January 2002].
269


258
Elite units should include both drug enforcement agents and intelligence analysts.
However, both types of participants should receive sufficient training in each area. When
conducting investigations agents and analysts should closely coordinate their activities.
Intelligence analysts should accompany enforcement specialists on all raids and begin
gathering and processing information on site. With the assistance of computer-mediated
technologies, analysts should integrate field data with the organizational memories
maintained at headquarters. Analysts and agents should take a broad view towards
intelligence gathering, extrapolating information even from seemingly innocuous sources,
such as the personal libraries of suspected traffickers. The retrieval of tactical and
operational intelligence should be a priority in all enforcement activities. Units should
carefully study their criminal adversaries and pay close attention to their sources of
knowledge and experience.
For these elite units to achieve maximum success in disrupting trafficking
networks it is critical that their agents enjoy substantial operational discretion and not be
overburdened by the need to follow myriad organizational procedures. Agents should
have the authority to conduct repeated transactions with criminal enterprises, without
conducting any enforcement activities, until they have infiltrated the entire network.
Undercover agents should be allowed to engage in minor infractions, such consuming
psychoactive drugs, in order to earn the trust of unsuspecting targets. However, care
must be taken to protect the rights of civilians and perpetrators. Drug enforcement units
that trample on the rule of law defeat the purpose for which they were created.
Of course, the focus on elite drug enforcement units is not new. Over the past
several decades, the DEA and CNP have created a number of elite units and task forces to


70
The process of organizational learning can be conceptually disaggregated into
three steps: acquiring, interpreting, and acting on information (see Figure 3-1 below).
Organizational learning unfolds as information is discovered, shared, and applied by
participants that coordinate their activities through formal and informal routines. The
three phases of organizational learning are disaggregated for heuristic purposes; in
reality, information can be acquired, interpreted, and applied simultaneously. The
different steps are connected through feedback loops that reinforce certain interpretations
and behaviors while disregarding others. There are a number of impediments to
organizational learning, and the process may be blocked at any phase. I discuss obstacles
to learning in detail later in the chapter.
Figure 3-1 Simple Process Model of Organizational Learning
Source: Adapted from Daft & Weick (1984), Garvin (2000).
Acquiring Information
The first step in the organizational learning process is gathering information.
Information is the lifeline of organizational learning. Information includes know-how,
understanding, techniques or practices and is useful to the extent that it helps
organizations achieve targets. Organizations acquire information through their own trial
3 Although the language varies, a number of analysts conceptualize organizational learning as a process
composed of three or four discrete stages. Daft and Weick describe learning according to three analogous
stages, including scanning (data collection), interpretation (data given meaning), and learning (action taken)
(Daft & Weick, 1984, p. 286). Kim describes learning as composed of four stages: observing, assessing,
designing, and implementing (Kim, 1993, p. 44). See Garvin for discussion on this point (Garvin, 2000,
endnote 4, p. 223).


153
addition to monitoring inputs and outputs, record keeping allows leaders and managers to
Q
supervise participant behavior and prevent intra-organizational theft.
Due to the illegal nature of their business dealings, record keeping is risky for
participants. Drug enforcers may capture them during raids, in which case they become
invaluable material evidence for criminal prosecutions (Cash, 1999; also, see Lupallo,
2000). Many trafficking enterprises seek to reduce these risks by minimizing their data
collection activities, often relying on the unaided memories of participants or storing bits
of coded information on scraps of paper that can be destroyed quickly as the need arises.
Nstor claims that he avoided any and all written records during his several years in the
business: Nothing. Not even for monetary transactions. I found out right away you dont
need to do this (Nstor, 2000). Homero stresses that participants in his trafficking group
refrained from writing things down and relied on their own memories (Homero, 2000).
While Nstor, Homero and other traffickers are undoubtedly capable of recalling
a great deal of information, human memories are inherently limited vessels for storing
data. Individuals and organizations involved in multiple, large-scale transactions often
require some formal documentation for keeping track of their dealings. Depending on
their roles, participants may be expected to memorize an assortment of details, including
quantities, prices, bank accounts, addresses, telephone and pager numbers, and code
words and aliases. They are also obliged to remember numerous practices and
procedures, such as processing formulas, transportation routes, warehousing and
distribution routines, and communication rules. Former DEA official Tom Cash and
other drug enforcers acknowledge that participants in trafficking enterprises document
8 For discussion of this point regarding a drug retailing street gang in the U.S., see Levitt & Venkatesh
(1998).


108
and prices, money laundering schemes, and a host of other business-related decisions.
Lower-ranking participants carry out these decisions, their actions carefully managed
through regular communication with their supervisors and elaborate routines (more on
this below), and the presence of other participants. While mid-level managers in may
suggest improvements in tactical routines, they are generally careful to clear any changes
with the higher ups back in Colombia. Some leaders monitor cell managers by assigning
trusted participants to provide assistance, when their real function is to keep an eye on
things and report back to the entrepreneur (Reyes, 2000).
In some organizations, leaders make tactical decisions, managing day-to-day
activities and problems that develop in the course of clandestine operations. Some
leaders have a tendency to micro-manage their operations, communicating with their
overseas representatives on a daily basis regarding the more mundane details of
international drug trafficking.9 Not surprisingly, the tendency to micro-manage and
centralize decision-making is more pronounced in activities that are most vulnerable to
robbery and disruption, such as the repatriation of drug profits (Arreguin, 2000; Evans,
2000; Francis, 2000; Sanchez, 2000; Reyes, 2000; Thompson, 2000).
However, not all Colombian-based entrepreneurs are detail-oriented managers. In
some enterprises, upper-level participants, such exportation managers or chief overseas
representatives, enjoy considerable decision-making authority regarding day-to-day
9 One of the leaders of a Cali-based core organization, Miguel Rodrguez Orejuela, was well known among
drug enforcers for his tendency to micro-manage every aspect of his vast criminal operations, including
such minutia as the best comer of a certain vegetable box for hiding cocaine. Several DEA officials
interviewed stress that Rodriguez Orejuelas decision-making style came back to haunt him as they were
able to record incriminating phone conversations between the entrepreneur and his U.S.-based subordinates
(Arreguin, 2000; Evans, 2000; Francis, 2000; Sanchez, 2000).


TABLE OF CONTENTS
page
ACKNOWLEDGMENTS ii
LIST OF TABLES ix
LIST OF FIGURES x
KEY TO ABBREVIATIONS xi
ABSTRACT xiii
CHAPTER
1 INTRODUCTION 1
Persistence of Illicit Drug Production in Colombia 2
Explaining the Colombian Drug Dilemma 6
Toward an Alternative Explanation: Organizational Learning by Trafficking
Enterprises 8
Significance of Research 10
Research Design 12
Overview of Study 15
2 HISTORICAL OVERVIEW OF THE ILLICIT DRUG INDUSTRY IN
COLOMBIA 17
Colombian Smuggling Tradition 18
Phase One1930s: Colombia as Transit Point 20
Government Drug Enforcement Efforts During Phase One 21
Phase TwoLate 1930s to early 1960s: From Transit Point to Producer 22
Herran Olazaga Brothers Enterprise 24
Other Phase Two Trafficking Groups 25
Government Drug Enforcement Efforts During Phase Two 26
Phase ThreeMid-1960s to late 1970s: Expansion of Colombian Drug Industry.... 27
Herrera Enterprise 29
Bravo Enterprise 31
Government Drug Enforcement Efforts During Phase Three 32
Phase Four1980s to mid 1990s: Rise and Fall of the Cartels 36
v


176
Compartmentalization leads to loss of tactical knowledge in trafficking
enterprises. Moreover, it affects traffickers ability to learn from their own and others
experience. In multi-task enterprises, such as the core networks, compartmentalization
hampers information sharing and interpretation between different cells. One implication
is that effective practices and procedures developed in one cell are unlikely to diffuse
across the enterprise, unless it captures the attention of a cell manager or workers from
different cells communicate on a regular basis (which defeats the purpose of
compartmentalization). In highly compartmentalized operations, different cells are
unable to benefit from the innovations of their colleagues. Moreover, compartmentalized
enterprises may experience the same problems repeatedly because different sub-units
cannot access the experience of their peers.
In compartmentalized trafficking enterprises, intelligence regarding drug
enforcement activities may also remain localized, reducing participants ability to
recognize patterns in law enforcement practices. More generally, compartmentalization
produces participants that are ill informed about the overall operations of their enterprise.
One of my former participant respondents, Emil, had been a low-level member of a core-
affiliated transportation cell. While he was able to discuss his own activities in detail,
throughout the two hour interview it became increasingly clear that his knowledge of the
overall operation was limited (Emil, 2000). Lacking general knowledge of the enterprise,
workers such as Emil are limited in their ability to suggest improvements in their
organizations strategic routines.
However, it is important not to overemphasize the negative implications of
compartmentalization on organizational learning. Although numerous respondents


186
five respondents addressed the topic, all of them replying in the affirmative. Consistent
with the samples bias toward U.S. and Colombian law enforcement officials, the
overwhelming majority of responses came from this group (N=17). In addition, two
Colombian scholars and one former participant affirmed state learning, providing
examples to support their contentions. Table 6-1 classifies respondents according to their
responses to the question of drug enforcement learning.
Table 6-1 Does the State Learn? Results from Interviews
Type of Respondent
Affirmative
Negative
Not Discussed
Colombian government officials
12
0
14
U.S. government officials
9
0
15
IGO official
1
0
0
Colombian journalists
0
0
4
U.S. journalists
0
0
2
Colombian scholars
2
0
9
U.S. scholars
0
0
3
Former participants
1
0
4
TOTAL
25
0
51
The interview results indicate strong support among respondents for the
proposition of state learning. However, only one-third of those interviewed actually
addressed the issue of adaptation by law enforcement agencies. It is possible, though
admittedly unlikely given the existing results, that as many as two-thirds of the
respondents do not support the notion of learning by drug enforcement agencies. Of
greater potential significance is the samples bias toward government officials. Eighty-
eight percent (N=22) of the respondents that addressed drug enforcement learning were
active employees of the U.S. or Colombian governments at the time of their interview. It
is not surprising that these officials, many of whom design and carry out drug
enforcement policies and programs, would be eager to affirm the notion of state learning.


299
Krauthausen, Ciro. Padrinos y Mercaderes: Crimen organizado en Italia y Colombia.
Bogot: Planeta Colombiana Editorial, 1998.
Krauthausen, Ciro and Luis Femando Sarmiento. Cocana y Co.: Un mercado ilegal por
dentro. Bogot: Tercer Mundo, 1991.
Krogh, Egil Bud Jr. Former Deputy, Domestic Affairs, White House. Frontline: Drug
Wars. 2000.
Http://www.pbs.orgAvgbh/pages/frontline/shows/drugs/interviews/krogh.html
[Accessed 21 January 2002].
La Porte, Todd R. Organized Social Complexity: Explication of a Concept. Organized
Social Complexity: Challenge to Politics and Policy. Ed. Todd R. La Porte.
Princeton: Princeton University Press, (1975): 3-39.
Labaton, Stephen. Administration Drops Plan to Merge Justice Agencies. New York
Times (22 October 1993). Http://web.lexis-nexis.com/universe [Accessed 24 June
2001],
Latin American Weekly Report. Cali cartel bosses manage to escape as anti-drug force
tightens the noose. (6 April 1995a): 145.
Latin American Weekly Report. Samper Unscathed by Accusations: U.S. Envoy Gives
Backing to Campaign Against Cali Bosses. (20 April 1995b): 173.
Latin American Weekly Report. Anti-Drug Successes. (4 May 1995c): 183.
Latin American Weekly Report. Gelbard Does the Rounds in Bogot. (8 June 1995d):
245.
Latin American Weekly Report. Body Blow to Cali Cartel Claimed: Rodriguez
Orejuela Estates Occupied by Commandos. (15 June 1995e): 255.
Latin American Weekly Report. Capture of Rodrguez Orejuela is Good News but it is
Too Soon to Celebrate. (22 June 1995f): 265.
Latin American Weekly Report. Colombia draws the Scorpions sting but expects others
to take his place. (29 June 1995g): 277.
Latin American Weekly Report. Third Cali boss gives up the fight. (6 July 1995h): 297.
Latin American Weekly Report. Two more Cali bosses arrested. (20 July 1995i): 314.
Lawrence, Paul R. and Jay W. Lorsch. Organization and Environment: Managing
Differentiation and Integration. Boston: Graduate School of Business
Administration, Harvard University, 1967.


Manufacturing Drug Enforcement Intelligence 198
From Intelligence to Operations 200
Changes in U.S. Drug Enforcement 201
Structural Adaptation Case #1: Toward an Administrative Merger of the
DEA and FBI 202
Other DEA Structural Adaptations 205
Reforming the Colombian National Police 206
Structural Adaptation Case #2: Cultural Transformation of the CNP 207
Adaptations in Drug Enforcement Operations 211
Banshee or Buy-Bust? 211
Innovations in Conspiracy Investigations 213
Innovations in Undercover Operations 215
Competitive Learning Games 217
Factors that Affect Outcomes in Narco-Narc Learning Games 222
Playing Catch Up 223
The Survival Imperative 224
The Enforcement Dilemma 226
Smaller in Size, Flatter in Structure 228
The Red Tape Trap 232
Conclusion 236
7 CONCLUSION 238
What Have We Learned? 239
Alternative Explanations of Routine Change 244
Implications for Counter-Drug Policy: Or How I Learned to Stop Worrying
and Love the Drug Trade 248
Diffusion of Trafficking Technologies 251
What is to be Done? 252
Legalization and Harm Reduction 253
A Modest Proposal 255
Conclusion 260
APPENDIX
A RESPONDENT DATA 262
B QUESTIONNAIRE 266
BIBLIOGRAPHY 269
BIOGRAPHICAL SKETCH 332
vni


330
Wilson, James Q. Thinking about Reorganization, U.S. Intelligence at the Crossroads:
Agendas for Reform. Eds. Roy Godson, Ernest R. May, and Gary Schmitt.
Washington: Brasseys, (1995): 28-35.
Wirpsa, Leslie. Colombian drug baron is shot dead. Houston Chronicle (3 December
1993). Http://web.lexis-nexis.com/universe [Accessed 28 June 2001].
Wirpsa, Leslie. Cali Cartel Leaders Promise to Surrender Come January. Houston
Chronicle (4 December 1993): 26A. Http://web.lexis-nexis.com/universe
[Accessed 22 February 1999].
Wisotsky, Steven. Beyond the War on Drugs: Overcoming a Failed Public Policy.
Buffalo: Prometheus Books, 1990.
Wolff, Kay and Sybil Taylor. The Last Run: An American Woman's Years Inside a
Colombian Drug FamilyAnd Her Dramatic Escape. New York: Viking, 1989.
Woodiwiss, Michael. Crime, Crusades, and Corruption: Prohibitions in the United
States, 1900-1987. London: Pinter Publishers, 1988.
Wright, Theodore P. Factors Affecting the Costs of Airplanes. Journal of Aeronautical
Science 3 (1936): 122-128.
Wrobleski, Ann B. Testimony of Assistant Secretary Ann B. Wrobleski, Bureau of
International Narcotics Matters, Department of State. Colombian Drug
Trafficking and Control. Hearing before the Select Committee on Narcotics
Abuse and Control, U.S. House of Representatives. One Hundredth Congress,
first session (6 May 1987). Washington: U.S. Government Printing Office.
Yelle, L. E. The Learning Curve: Historical Review and Comprehensive Survey.
Decision Science 10 (1979): 302-328.
Yin, Robert K. Case Study Research: Design and Methods. 2nd edition. London: Sage
Publications, 1994.
Zabludoff, Sidney Jay. Colombian Narcotics Organizations as Business Enterprises.
Transnational Organized Crime 3.2 (1997): 20-49.
Zabludoff, Sidney. Former economist, Central Intelligence Agency. Washington. 19 July
1999.
Zarkin, Michael J. The Transformation of US Telecommunications Policy: Political
Learning and Policy Change, 1900-1996. Unpublished Ph.D. Dissertation,
University of Florida, 2000.


264
Name of
Date of
Type of
Trafficker
State learning?:
respondent
interview
respondent
learning?:
Affirmative,
negative, not
discussed
Affirmative,
negative, not
discussed
Juan Molano
[pseud]
7/21/00
Colombian
government
Affirmative
Affirmative
Carlos Montoya
[pseud]
4/4/00
Colombian
government
Affirmative
Affirmative
Pedro Morales
[pseud]
7/11/00
Colombian
government
Affirmative
Not discussed
Eduardo Nicosa
[pseud]
6/14/00
Colombian
government
Affirmative
Not discussed
Luis Oporto
[pseud]
5/3/00
Colombian
government
Affirmative
Affirmative
Antonia Puglisi
[pseud]
7/4/00
Colombian
government
Not discussed
Affirmative
Andrs Soto
Velasquez
6/27/00
Colombian
government
Not discussed
Affirmative
Nelson Vargas
[pseud]
7/13/00
Colombian
government
Affirmative
Not discussed
Paula Villega
[pseud]
3/3/00
Colombian
government
Affirmative
Not discussed
Leo Arreguin
6/5/00
U.S. government
Affirmative
Affirmative
Thomas Cash
8/3/99
U.S. government
Affirmative
Not discussed
Avrum Bird
[pseud]
5/2/00 &
5/8/00
U.S. government
Affirmative
Not discussed
Jerome Crank
[pseud]
3/15/00
U.S. government
Affirmative
Affirmative
Frank Derian
[pseud]
7/23/99
U.S. government
Affirmative
Not discussed
Samuel Evans
[pseud]
7/22/99
U.S. government
Affirmative
Not discussed
John Francis
[pseud]
6/5/00
U.S. government
Affirmative
Affirmative
Myles Frechette
6/25/97
U.S. government
Affirmative
Affirmative
John Fernandez
1/27/00
U.S. government
Affirmative
Affirmative
Marvin Hatch
[pseud]
3/1/00
U.S. government
Affirmative
Not discussed
Sherman Hinson
7/3/97
U.S. government
Affirmative
Affirmative
Larry Lyons
6/12/97
U.S. government
Affirmative
Not discussed
Katherine
McNamara [pseud]
7/19/00
U.S. government
Affirmative
Not discussed
Luis Moreno
6/17/97
U.S. government
Affirmative
Affirmative
Jeremy Nolan
[pseud]
6/12/97
U.S. government
Affirmative
Not discussed


62
organizations aim or purpose. Dill (1958), and subsequently Thompson (1967),
disaggregate the task environment into four areas: (1) customers or clients of the
organization (including distributors and consumers); (2) suppliers of labor, capital and
technologies; (3) competitors for market share and resources; and (4) regulatory groups,
which include government agencies. The organizations that fulfill these roles comprise
an organization set. An organization set is composed of the customers, suppliers,
competitors and regulators with which an enterprise interacts in order to achieve its
purpose (Blau & Scott, 1962; Dill, 1958; Evan, 1966; Kaufman, 1985, p. 41; Posen,
1984, p. 43; Scott, 1998, p. 147; Scott, 1998, pp. 124, 147; Smith, 1980, p. 376;
Thompson, 1967, pp. 27-28).
Organizations draw on task environments for a wide variety of physical,
technological, and symbolic inputs. Organizations obtain necessary inflows of capital,
participants, and technology from external suppliers. Capital provides the enterprise with
the financial resources necessary to acquire other material and informational resources.
Participants bring knowledge, skills and experience that organizations use to perform
tasks. Technologies include the equipment, machines, instruments, and information
necessary for the organization to accomplish work. Of course, interactions between
enterprises and task environments are not unidirectional: organizations trade outputs
with customers or clients in return for more inputs. In order to receive the necessary
inputs, organizations must produce goods or services valued by their customers or clients.
This is true for both legal and illegal organizations (Posen, 1984, p. 43; Scott, 1998, pp.
21-22, 229; Smith, 1975, pp. 340-341; Thompson, 1967, p. 28).


CHAPTER 4
ORGANIZATIONAL ANALYSIS OF THE
ILLICIT DRUG INDUSTRY IN COLOMBIA
The production, transportation, and distribution of illicit drugs are social
activities. Processing cocaine base into cocaine hydrochloride or opium gum into heroin;
shipping refined narcotics from one country to another; selling drugs to independent
wholesalers; collecting, laundering and repatriating proceeds from drug sales: all of
these activities require planning and coordination among various interdependent
conspirators. Although the scope of social organization varies by activity and the amount
of product under exchange, even the most rudimentary transactions in the transnational
drug trade require the coordination of numerous like-minded individuals. Of course,
coordination alone does not a criminal organization make. While drug trafficking
involves coordination among participants, less clear are the organizational forms such
coordination takes.
The purpose of this chapter is to examine Colombian trafficking enterprises from
the perspective of organization theory. The task is an important one, and stands prior to
the inquiry that drives this study. We cannot consider whether trafficking enterprises
learn in an organizational sense until we first understand how they are structured and
function. Unlike the large drug enforcement bureaucracies maintained by contemporary
states, trafficking enterprises cannot be assumed to operate as formal organizations.
Smuggling operations often exist on an ad hoc basis, with participants going their
separate ways following completion of a single transaction. Even when they develop
95


283
Donald, Ian and Angela Wilson. Ram Raiding: Criminals Working in Groups. The
Social Psychology of Crime: Groups, Teams and Networks. Eds. David Canter
and Laurence Alison. Aldershot: Dartmouth, (2000): 191-246.
Dodgson, Mark. Organizational Learning: A Review of Some Literatures.
Organizational Studies 14.3 (1993): 375-94.
Dorn, Nicholas and Nigel South. Drug Markets and Law Enforcement. British Journal
of Criminology 30.2 (1990): 171-188.
Drug Enforcement Administration (DEA), U.S. Department of Justice. The National
Narcotics Intelligence Consumers Committee Report (August 1984). Publications
Unit, Intelligence Division, DEA Headquarters.
Drug Enforcement Administration (DEA), U.S. Department of Justice. The National
Narcotics Intelligence Consumers Committee Report (August 1985-86).
Publications Unit, Intelligence Division, DEA Headquarters.
Drug Enforcement Administration (DEA). Drug Enforcement Handbook (1987a).
Publications Unit, DEA Headquarters.
Drug Enforcement Administration (DEA), U.S. Department of Justice. The National
Narcotics Intelligence Consumers Committee Report (August 1987b).
Publications Unit, Intelligence Division, DEA Headquarters.
Drug Enforcement Administration (DEA), U.S. Department of Justice. The Illicit Drug
Situation in Colombia: Drug Intelligence Report (November 1993). Publications
Unit, Intelligence Division, DEA Headquarters.
Drug Enforcement Administration (DEA), U.S. Department of Justice. ColombianOpiate
Assessment (June 1994a). Publications Unit, Intelligence Division, DEA
Headquarters.
Drug Enforcement Administration (DEA), U.S. Department of Justice. The National
Narcotics Intelligence Consumers Committee Report (August 1994b).
Publications Unit, Intelligence Division, DEA Headquarters.
Drug Enforcement Administration (DEA), U.S. Department of Justice. The Cali Cartel:
The New Kings of Cocaine (November 1994c). Publications Unit, Intelligence
Division, DEA Headquarters.
Drug Enforcement Administration (DEA), U.S. Department of Justice. The Cocaine
Threat to the United States (March 1995). Publications Unit, Intelligence
Division, DEA Headquarters.


22
September 15, 1920 was the first piece of national legislation outlawing the consumption,
trade, and production of certain narcotics. Eighteen years later the Directorate of
National Hygiene enacted Decree 95 to regulate the sale of coca leaves to licensed
pharmacies. In July 1938, the Colombian government established a new penal code that
increased the criminal penalties for transacting in opiates. One month later the Ministry
of Work, Hygiene, and Welfare was created with the partial purpose of applying national-
level regulations concerning the drug traffic that Colombia had accepted at international
narcotics control conferences sponsored by the League of Nations (Lpez Restrepo,
2000, p. 8; Lpez Restrepo & Camacho Guizado, 2001, p. 3; Senz Rovner, 1996, p. 72).
In the mid-1930s, drug interdiction emerged as an active component of the U.S.
governments counter-narcotics strategy. Prior to this U.S. drug enforcement largely
consisted of local officials intercepting drugs in port cities, sometimes with the help of
federal customs and counter-drug agents. The first, tentative steps towards the
internationalization of American drug enforcement were sporadic Coast Guard patrols in
Caribbean sea lanes and the Gulf of Mexico. As William Walker reports, additional
efforts were made to conduct aerial surveillance near the U.S.-Mexican border,
apparently without much success. In general, early drug enforcement activities
conducted by the Colombian and U.S. governments were haphazard, sporadic, and failed
to impede smugglers access to morphine and cocaine supplies (Walker, 1989, pp. 57,
140-141; Walker, 1992, p. 269).
Phase TwoLate 1930s to early 1960s: From Transit Point to Producer
Colombias role as a transit point in the international drug trade was not destined
to last for long. As early as 1939 the U.S. Treasury Department reported that opium was


38
interest groups than monolithic firms (Clawson & Lee, 1996, pp. 39-40; Krauthausen &
Sarmiento, 1991; Lee, 1991, p. 14; Zabludoff, 1997, pp. 23-26; Zabludoff, 2000).
Emergence of the Core Organizations
Core trafficking organizations and their support networks did not spring up
overnight. Rather, they developed over time as their leaders acquired experience,
contacts, and knowledge in a variety of criminal endeavors, including smuggling
contraband goods, robbing cars, even kidnapping. The largest, and most well-known,
core organizations were led by entrepreneurs based in Medellin and Cali. During the
1980s and 1990s, such figures as Pablo Escobar Gaviria, Carlos Lehder Rivas, Gonzalo
Rodrguez Gacha, Jos Santacruz Londoo, and the Ochoa and Rodrguez Orejuela
brothers became household names as Colombian and American journalists assiduously
chronicled their criminal exploits.7
Many of the leaders of the Medellin and Cali core enterprises began their delictive
careers working as low-level members for established criminal gangs. In the early 1970s,
three of the future leaders of the Cali-based cocaine network, Gilberto and Miguel
Rodrguez Orejuela and Jos Santacruz Londoo, worked as foot soldiers in a Cali
criminal gang known as Los Chemas. This gang was primarily involved in kidnapping
and counterfeiting but gradually expanded into smuggling cocaine base from Bolivia and
Peru and converting it into cocaine hydrochloride in Colombia (Castillo, 1987, pp. 41-42;
DEA, 1994b, p. 1).
7 Due to the fascination of the Colombian and U.S. media and public with the Medellin and Cali core
organizations an enormous body of literature emerged in the 1980s and 1990s, much of it written by
journalists more interested in dramatization and hyperbole than scientific analysis. Some of the more well-
known journalistic accounts, in both English and Spanish, include: Can (1994), Castillo (1987, 1991,
1996), Castro-Caycedo (1996b), Cervantes (1980), Corts (1993), Duzn (1994), Eddy et al. (1988),
Freemantle (1985), Giraldo (1992), Gugliotta and Leen (1989), Mermelstein (1990), Mills (1986), Reyes
(1999), Rice (1989), Rincn (1990, 1994), Shannon (1988), Torres (1995), Torres & Sarmiento (1998).


146
obtain access to vital tactical intelligence, including the identities of confidential
government informants. Testifying before the U.S. Senate Subcommittee on Security and
Terrorism, DEA Administrator Francis Mullen argued that over eighty percent of the
FOIA requests received by his agency were from criminals, or their associates, trying to
learn how they ended up in prison and who was involved in putting them there (Mullen,
1984, p. 26). In a separate Congressional hearing, he claimed that traffickers also collect
information on how to avoid detection by the DEA from paid advertisements in certain
narcotics-oriented and motorcycle publications (Mullen, 1983, p. 27).
Drug enforcers themselves provide useful intelligence to trafficking groups. In
their quest for budget allocations and organizational legitimacy, law enforcement
agencies such as the DEA and Customs often promote their latest activities in the war on
drugs. However, as Nstor points out, the publicity causes traffickers to change their
operations:
They [law enforcement authorities] start talking, publicizing their activities. You
find out they had some boat in Miami. So you ask yourself what the hell do they
want with this boat? Then the smugglers know something is going on, so you
switch modes. So when the government came out in the media saying, We got
helicopters, were doing this and this, then the smugglers would say, Okay, lets
switch the operations (Nstor, 2000).
Many enterprises invest considerable financial and human resources in
developing sophisticated counter-intelligence operations. During the 1980s, several
cocaine transportation rings established listening posts in Miami and other coastal areas
to monitor the radio frequencies of Customs interdiction flights and local narcotics
agents. Some groups also stationed observers at federal facilities in South Florida,
including the Homestead Air Force Base and Boca Chica Naval Air Station, in order to
monitor interdiction flights. Trafficking organizations acquired lists of the radio


Frankel, Glenn. U.S. War on Drugs Yields Few Victories. Washington Post (8 June
1997). Http://web.lexis-nexis.com/universe [Accessed 16 July 2001].
290
Frechette, Myles. U.S. Ambassador to Colombia. Bogot. 25 June 1997.
Freemantle, Brian. The Fix: Inside the World Drug Trade. New York: Tor Books, 1985.
Frontline. Drug Wars, Thirty Years of Americas Drug Wards: A Chronology. (2000).
Http://www.pbs.org/wgbh/pages/frontline/shows/drugs/cron/index.html#l 1
[Accessed 11 July 2001],
Fuentes, Joseph R. The Life of a Cell: Managerial Practices and Strategy in Colombian
Cocaine Distribution in the United States. Ph.D. dissertation submitted to the
Graduate Faculty in Criminal Justice, City University of New York, 1998.
Fuentes, Thomas V. Phil Williams View of Criminal Organizations and Drug
Trafficking: Another Perspective. Transnational Threats: Blending Law
Enforcement and Military Strategies. Ed. Carolyn W. Pumphrey. Carlisle:
Strategic Studies Institute, (2000): 203-212.
Furden, Jos [pseud]. Head, Judicial Police unit, Administrative Security
Department. Bogot. 2 June 2000.
Gage, Nicholas. Latins now Leaders of Hard-Drug Trade. New York Times (21 April
1975a): 1 & 26.
Gage, Nicholas. Drug-Smuggling Logistics Bizarre and Often Fatal. New York Times
(22 April 1975b): 1 & 24.
Gale, Lanny [pseud] and Helmer Rodriguez [pseud]. Special agents, Administrative
Security Department. Bogot. 6 June 2000.
Gallego Castrilln, Leonardo. Director, Judicial Police Directorate, Colombian National
Police. Bogot. 21 March 2000.
Gallie, W. B. Essentially Contested Concepts. Proceedings of the Aristotelian Society
56. London: Compton Press.
Garca Mrquez, Gabriel. News of a Kidnapping. New York: Penguin Books, 1996.
Garcia, Miguel. Los barones de la cocana. Mexico City: Planeta, 1991.
Garvin, David. Learning in Action: A Guide to Putting the Learning Organization to
Work. Boston: Harvard Business School Press, 2000.


102
commerce.6 To survive, these competitors rely on contributions from other actors in their
respective organization sets. Trafficking enterprises stay in business by attaining
sufficient profits from buyers of their illicit products (and avoiding government efforts to
dismember them). Drug enforcement agencies stay in business by receiving annual
budgets sufficient to implement their policies and programs.
In Colombia and the U.S. hundreds of government agencies seek to identify and
dismantle trafficking enterprises. These include drug enforcement bureaus, military
organizations, prosecutors offices, and other government agencies that spray drug
plantings, destroy processing laboratories, seize psychoactive drugs and precursor
chemicals, interdict drug shipments, confiscate illicit earnings, and investigate,
apprehend, prosecute, and incarcerate participants. In Colombia, several independent
directorates of the Colombian National Police (CNP), the Department of Administrative
Security (DAS), and joint military-CNP units specifically target trafficking enterprises.
The National Prosecutors office (Fiscala General), the Attorney Generals office
{Procuradora General), the National Directorate of Dangerous Drugs, and various
branches of the Colombian military also participate in drug enforcement (see Table 4-1).
In the U.S., there are 17,000 federal, regional, state, local, and tribal law
enforcement agencies, many of which are directly involved in drug enforcement. At the
federal level, government agencies with a primary role in counter-narcotics efforts
include the Drug Enforcement Administration, the Office of National Drug Control
Policy, the Federal Bureau of Investigations, the Central Intelligence Agency, the
6 In a recent statement to the U.S. Senate Caucus on International Narcotics Control, the director of the
Drug Enforcement Administration stated that his agencys mission is to identify, target, and dismantle the
most powerful chug syndicates operating around the world who are responsible for supplying drugs to
American communities (Marshall, 2001a).


239
of reforms that would increase their ability to target and apprehend major trafficking
enterprises.
What Have We Learned?
This is a study about organizational learning. Learning by the good guys, the
drug enforcement nares that identify, apprehend, and dismantle trafficking networks; and
learning by the bad guys, the narcos that manufacture, transport, and distribute
psychoactive drugs in the U.S. and Colombia. In preceding chapters, I demonstrate how
these state and non-state actors learn. The theoretical framework for this endeavor comes
from a multidisciplinary body of literature on organizational learning. In Chapter 2,1
draw on a number of theoretical studies, and the broader literature on organization
analysis, to develop a routine-based, process-oriented notion of learning. According to
this formulation, organizations learn when their participants gather, interpret, and apply
information by changing practices, procedures, and performance programs that guide
collective behavior. While individuals perform the cognitive labors of learning, this
process is greater than the sum of participants information processing activities.
Learning is not organizational until knowledge is distributed in collective memories
and embedded in organizational routines. Information may be disseminated through
written manuals, accounting logs, and computer files, or it may be transmitted through
conversations, stories, and myths that participants use to construct inter-subjective
understandings of organizational norms and practices.
Much of the empirical work in the learning literature, such as it is, deals with
legally sanctioned organizations, including business firms and government bureaucracies.
Given the dearth of studies on criminal enterprises, it is reasonable to question whether


225
Moreover, decision-makers must be able to distinguish signals from noise, discern truth
from lies and hyperbole. These demands tax the cognitive abilities of even the most cool-
headed decision-makers and make learning difficult.
When drug enforcers identify traffickers the environmental hostility they face
increases exponentially. Drug enforcement task forces, elite units, and joint military-
police operations are capable of exerting tremendous pressure on identified enterprises.
When investigations move into full swing drug enforcement pressure becomes cascading.
Initial raids and arrests lead to the discovery of information that leads to more operations
producing more arrests that leads to more information leading to more raids, and so on.
When the Colombian government directed its substantial drug enforcement capabilities,
supplemented with DEA and CIA assistance, against the Medellin and Cali core
organizations in the 1990s, extraordinary pressure was brought to bear on the leaders of
these enterprises and their families. As the number of raids rose from the hundreds to the
thousands, leaders were compelled to cease their trafficking activities and devote their
resources to protecting themselves and their families. Their ability to respond effectively
to the feverish crescendo of drug enforcement is limited, and it is usually only a question
of time before their trafficking operations are significantly disrupted (Crank, 2000; Gale
& Rodriguez, 2000; Montoya, 2000; Pineda, 2000; and Smith, 2000c).
However, the hostile environment confronting trafficking enterprises also
facilitates learning. For traffickers the stakes in competitive learning games are high.
Every lab destroyed, every shipment interdicted, every stash house raided, and every
participant apprehended creates a potentially fatal vulnerability. This provides traffickers
a paramount incentive to maintain operational secrecy and change their activities as


179
productivity. While organizational forgetting is a common problem among trafficking
enterprises, it is exacerbated in organizations where participants lack mechanisms to
store, share, and retrieve collective memories. Without accounting ledgers, manuals and
oral traditions to remind them of the innumerable details involved in their daily activities,
participants require more time to relearn their roles and responsibilities following periods
of extended inactivity.
However, the loss of accumulated knowledge and experience in trafficking
organizations due to personnel turnover is not always a bad thing. A number of
organization theorists stress that turnover can assist organizations by forcing them to
unlearn unsatisfactory practices and procedures that hamper performance (Hedberg,
1981, p. 18; Nystrom & Starbuck, 1984, p. 58; Virany et al., 1996). Changing personnel
is a fast, effective method for removing practices and procedures that led the trafficking
enterprise to an undesirable state of affairs. By bringing in new people that possess novel
solutions to problems that plagued former participants, turnover also increases
organizational flexibility and competitiveness. In this respect, limited personnel turnover
may actually help some trafficking organizations survive dynamic and hostile task
environments. By increasing the amount of turnover among Colombian trafficking
enterprises in the 1990s, drug enforcement agencies may have unintentionally helped
these criminal organizations develop structural characteristics essential for long-term
survival (Hedberg, 1981, p. 18; Morrison, 1997, p. 18; Nystrom & Starbuck, 1984, p. 58;
Virany et al., 1996).


152
and there was a president of a bank in there and he pumped him constantly about
the banking system in America and how one can launder money, and he kept files
and files on everything. He kept notes constantly. He never stopped. He was
obsessed with it (Jung, 2000).
Following their release from prison, the two convicts used their knowledge to
establish themselves as leading transporters and distributors for several Medellin core
enterprises.
Lehder and Jung were certainly not the only small-time drug smugglers to parlay
their prison time into lucrative criminal careers. Numerous former participants
interviewed by Reuter and Haaga note that the personal contacts they made in prison
allowed them to increase their participation in the narcotics trade significantly. In his
autobiography, Jorge Valds notes, In prison I established relationships with some
powerful individuals, contacts that I knew would come in handy once we were released.
Indeed, Valds used one contact, a former colleague who was visiting another inmate, to
establish a new distribution route (Valds, 1999, pp. 148, 183). These observations are
echoed in personal interviews conducted with DEA officials, and indicate that the states
strategy of arresting and incarcerating large numbers of participants has the unintended
effect of producing a more astute and better connected class of narcotics smuggler
(Passas, 2000; Reuter & Haaga, 1989, p. 38; Smith, 2000a).
Recording and Storing Information through Organizational Memories
Central to the process of organizational learning are the mechanisms by which
organizations document and store knowledge and experience. Colombian trafficking
enterprises record and store information through files, manuals, ledgers, notebooks, and
correspondence that document practices and procedures and track resource flows. In


LIST OF FIGURES
Figure P*gg
3-1 Simple Process Model of Organizational Learning 70
3-2 Alternative Sources of Change in Organizational Routines 77
3-3 Complex Process Model of Organizational Learning 85
4-1 Wheel Structure of Illicit Drug Network 99
4-2 Management Levels within Large Trafficking Enterprise Ill
4-3 Stash House Inventory Routine 119
6-1 Organization of DEA Intelligence Division, circa 1996 193
6-2 PHVA Management Cycle 208
6-3 Example of DEA Management Levels 231
7-1 Sources of Change in Organizational Routines, Amended 245
7-2 Phased Evolution of Illicit Drug Industry 248
x


256
possible bureaucratic and legalistic constraints to action. In other words, drug
enforcement agencies, or at least select units, should function more like the fluid, ad hoc
networks they are designed to dismantle. The units should be designed to accelerate
information flows, shorten decision cycles, and reduce bureaucratic protocols, improving
their ability to respond rapidly to trafficking enterprises. However, in making these
modifications, policy makers should preserve drug enforcers advantages over their illicit
rivals in the critical areas of training, intelligence, and material assets.
A number of specific improvements are feasible. Elite drug enforcement units in
Colombia and the U.S. should be smaller and flatter than existing agencies. Units should
contain fewer than two dozen participants and no more than three layers of management
(manager, assistant managers, field agents). They should direct their activities towards
investigations of specific enterprises. Yet, they should avoid diluting their resources by
undertaking only one or two major investigations at a time. In recognition of the
transnational, polydrug character of many trafficking enterprises, units should be
organized on a regional, rather than a drug-specific, basis. Each region should contain
several independent units, with the exact number depending on existing trafficking
patterns. Administrative management and oversight of the units should be directed from
a single, centralized location, rather than separate regional divisions. Individual units
should enjoy unfettered access to the computerized intelligence databases, electronic
surveillance technologies, satellite imagery systems, and other technological resources
needed to conduct even the most complex conspiracy investigations. Units should
function as semi-autonomous agencies but they should have the flexibility to outsource


174
organizations accept the risk of direct participation (Aero, 2000; Arreguin, 2000;
Caldern, 2000; Camacho Guizado, 2000; Contreras & Ambrus, 2000; Darling, 1996;
Farah, 1997; Francis, 2000; Ledwith, 2000; Lyons, 1997; Morales, 2000; Nicosa, 2000;
Oporto, 2000; Oppenheimer, 1997; Rabasa & Chalk, 2001; Semana, 2000a).
Many of these changes emerged as a direct result of drug enforcers ability to
penetrate and dismantle the big cartels, and the reactivation of Colombias extradition
treaty in 1997. Collectively, they represent a significant shift in the marketing strategies
of Colombian traffickers. For the first time since the early 1970s, these criminal
enterprises no longer endeavor to dominate wholesale distribution in the lucrative U.S.
market.
The opportunity costs associated with this change are considerable. As
economists point out, most value added in the international drug trade occurs at the
wholesale and retail distribution stages in consumer markets, where risks to suppliers are
highest. By removing themselves from importation and wholesale distribution in the
U.S., Colombian traffickers sacrificed market share and profits to competing
organizations in exchange for greater security from drug enforcers. Interviews and other
sources suggest that these adaptations were part of a deliberate strategy by Colombian
traffickers, indicating that at least some organizations dispel the profit motive to ensure
the integrity of their operations (Arreguin, 2000; Crank, 2000; DEA, 2001a; Francis,
2000; Oporto, 2000; Smith, 2000a, 2000b; Ledwith, 1999, pp. 123-124; Marshall,
2001b; Reuter, 1985; Thoumi, 1995, p. 135).12
12 When asked how they knew that this strategic shift was, in fact, part of an intentional decision-making
process, several officials responded that their knowledge was based on intelligence of these organizations
gathered through both eavesdropping technologies (i.e. wiretaps & other recordings) and human resources
(i.e. confidential informants).


135
# ic ... .
scientific training. Even more cerebral activities, such as planning drug shipments and
coordinating the activities of numerous participants, do not require considerable scientific
knowledge. While it may be desirable to have cell managers that possess some formal
training in business administration, just as it may be desirable to have drug processors
with formal training in chemistry, it is not essential. More important than formal
educational training is a sufficient degree of practical knowledge and common sense.
The lack of scientific knowledge required in most trafficking activities has
significant implications for the development of the illicit drug trade in Colombia and
elsewhere. Because trafficking knowledge is largely practical rather than scientific,
relevant information can be disseminated widely, even within so-called developing
countries that lack strong educational systems. In Colombia, the information involved in
drug trafficking has spread throughout the country over the last several decades. This is
one reason why new and re-organized enterprises are quick to step into the trafficking
void left in the wake of drug enforcement operations. With hundreds, perhaps thousands,
of individuals that possess a working knowledge of drug trafficking still on the streets
even the most successful interventions by law enforcers are likely to cause, at best, a
temporary disruption of trafficking patterns. Individuals with the knowledge, resources,
and desire to participate in these criminal activities will create organizational forms that
allow them to do so. In the process, the practical knowledge involved in drug trafficking
spreads further still, as new entrants that lack previous experience in these activities are
recruited into trafficking enterprises. Ancillary services that require more scientific
25 As one respondent who has conducted extensive research on cocaine processing in Colombia recalls, I
had a chemist once who told me, I do not know how to read and write but I am damn good chemist,
[respondent laughs]. Actually, he said, I only have [a] 3rd grade education but [Im] a chemist (Uribe
Ramirez, 2000).


36
Phase Four1980s-mid 1990s: Rise and Fall of the Cartels
By the early 1980s a number of Colombian trafficking organizations were
organizing large-scale cocaine shipments from South America to the United States and
Europe. These criminal enterprises provided a measure of coordination to what had been
a highly fragmented industry. At the zenith of this transnational production-
transportation-marketing structure stood several vertically integrated core organizations
that, according to one estimate, supplied over sixty percent of the cocaine reaching U.S.
and European drug markets. Size varied considerably among different core
organizations. Small core enterprises were composed of anywhere from two dozen to a
hundred members, while large organizations contained several hundred members.
Participants were compartmentalized into discrete units, organized along functional lines,
such as drug processing, transportation, and wholesale distribution. Core organizations
generally contained three or more levels of management, with at least two layers
insulating leaders from the activities of rank-and-file workers (PCOC, 1986, pp. 100-101;
Thompson, 2000; Vargas, 2000; Zabludoff, 1997, p. 24).
The core organizations transformed the Colombian drug industry by coordinating
international cocaine shipments that measured in metric tons rather than kilograms. To
do so, they relied on networks of hundreds of individuals and groups that provided
specialized goods and services. Independent suppliers from Bolivia and Peru provided
coca paste, a semi-processed form of cocaine, which was often transported to Colombia
for further processing. In completing the cocaine refining process, some Colombian core
organizations owned their own processing laboratories, while others subcontracted
processing services to independent groups of chemists that operated hundreds of


331
Zeng, Menghua. An Interactive Perspective of Chinese Aid Policy: A Case Study of
Chinese Aid to Tanzania. Unpublished Ph.D. Dissertation, University of Florida,
1999.


285
Drug Enforcement Administration (DEA). OCDETF: Organized Crime Drug
Enforcement Task Forces [No date].
Http://www.usdoj.gov/dea/programs/ocdetf.htm [Accessed 23 July 2001].
Drug Enforcement Administration (DEA). DEA History, 1994 1998 [No date].
Http://www.usdoj.gov:80/dea/deamuseum/1994 1998.htm [Accessed 7 February
2002].
Drug Enforcement Administration (DEA). DEA Staffing & Budget [No date].
Http://www.usdoi.gov:80/dea/agencv/staffmg.htm [Accessed 11 February 2002].
Drug Enforcement Administration (DEA). DEA Organizational Chart [No date].
Http://www.usdoi.gov:80/dea/agency/staffing.htm [Accessed 11 February 2002],
Derian, Frank [pseud]. Former staffer, Committee on Intelligence, United States Senate.
Washington. 23 July 1999.
Durfey, Robert W. Prepared Statement of Rear Admiral Robert W. Durfey, U.S. Coast
Guard, Department of Transportation. Cocaine and Marijuana Trafficking in
Southeastern United States. Hearings before the Select Committee on Narcotics
Abuse and Control, United States House of Representatives. Ninety-fifth
Congress, second session (9-10 June 1978). Washington: U.S. Government
Printing Office.
Durkheim, Emile. The Division of Labor in Society. Translated by W. D. Halls.
New York: Free Press, 1997.
Duzn, Maria Jimena. Death Beat: A Colombian Journalists Life Inside the Cocaine
Wars. Translated and edited by Peter Eisner. New York: HarperCollins, 1994.
Dye, Roy Thomas. A Social History of Drug Smuggling in Florida. Unpublished
Ph.D. Dissertation, Florida State University, 1998.
Economist Intelligence Unit. Country Report: Colombia. 2nd Quarter. London: The Unit,
1995.
Eckstein, Harry. Case Study and Theory in Political Science. Strategies of Inquiry. Eds.
Fred I. Greenstein and Nelson W. Polsby. Reading: Addison-Wesley (1975).
Eddy, Paul, Hugo Sabogol and Sara Walden. The Cocaine Wars. New York: W.W.
Norton & Co, 1988.
Eden, Lynn. The Hypothetical Organization: Organizational Learning and Interpretation
in U.S. Strategic Nuclear Targeting. Paper delivered at the 1990 American


47
Colombian nationals, Pablo Escobar also turned himself in to Colombian authorities.
However, Escobars surrender turned out to be largely meaningless as he continued his
criminal activities in a luxurious correctional facility of his own construction, from which
he escaped after a year of imprisonment. Over the next two and a half years, Escobar
managed to elude Colombian and U.S. authorities and former business associates that
turned against him. But the effort to stay alive required all of his resources, eliminating
his role as a major trafficker.11 Meanwhile, competing trafficking enterprises based in
Cali took advantage of drug enforcers single-minded focus on capturing Escobar by
expanding their own transnational operations (Can, 1994; Nieves, 1997, p. 12).
With the death of Escobar at the hands of an elite Colombian police unit in
December 1993, the Colombian government could now turn its attention to the Cali
enterprises. However, it would be another fourteen months before Colombian authorities
systematically cracked down against the leading Cali traffickers. The impetus for the
1995 offensive was not an act of violence by the Caleos, who relied more on bribery
than intimidation to protect their interests, but political pressure from Washington.
Disgusted with Colombian president Ernesto Sampers ability to stay in power despite a
campaign finance scandal in which he was believed to have knowlingly accepted
contributions from the Rodrguez Orejuela brothers and other traffickers, the U.S. Senate
threatened to impose economic sanctions on Colombia if the government failed to
achieve certain drug enforcement benchmarks in 1995. These included capturing the
leaders of the Cali core organizations, confiscating their illicit assets, and dismantling
their criminal operations (Can 1994; Economist Intelligence Unit, 1995, p. 4; Latin
11 For further discussion of these events see Luis Can El Patrn: Vida y muerte de Pablo Escobar
(Bogot, Colombia: Planeta, 1994).


323
Sterling, Claire. Octopus: The Long Reach of the International Sicilian Mafia. New
York: W.W. Norton & Company, 1990.
Stem, Jessica. The Ultimate Terrorists. Cambridge, MA: Harvard University Press, 1999.
Steve. Former drug trafficker and money launderer. Frontline: Drug Wars. 2000.
Http://www.pbs.org/wgbh/pages/frontline/shows/drugs/interviews/steve.html
[Accessed 6 March 2001].
Strange, Susan. The Retreat of the State: The Diffusion of Power in the World Economy.
Cambridge: Cambridge University Press, 1996.
Stutman, Robert. Former Special Agent in Charge, New York Division, Drug
Enforcement Administration. Frontline: Drug Wars. 2000.
Http://www.pbs.org/wgbh/pages/frontline/shows/drugs/interviews/stutman.html
[Accessed 13 June 2001],
Sutherland, Edwin H. The Professional Thief by a Professional Thief. Annotated and
interpreted by Edwin H. Sutherland. Chicago: University of Chicago Press, 1937.
Sutherland, Edwin H. On Analyzing Crime. Ed. Karl Schuessler. Chicago: University of
Chicago Press, 1973.
Swidler, Ann. Culture in Action: Symbols and Strategies. American Sociological
Review 51.2 (1986): 273-286.
Tamayo, Juan C. U.S. officials tie Colombian guerrillas to drug exports: Group likened
to 'cartel'. Miami Herald (13 December 2000).
Http://www.herald.com/americas/ [Accessed 18 December 2000].
Tetlock, Philip E. Learning in U.S. and Soviet Foreign Policy: In Search of an Elusive
Concept. Learning in U.S. and Soviet Foreign Policy. Eds. George W. Breslauer
and Philip E. Tetlock. Boulder: Westview Press, (1991): 20-61.
Third Superceding Indictment, United States of America vs. Michael Abbell, William
Moran, Luis Grojales, Eddy Martinez, Ramon Martinez, J.L. Pereira-Salas, et al,
Defendants, Case No. 93 470-CR-WMH.
Thompson, Jack [pseud]. Group Supervisor, Miami Division, Drug Enforcement
Administration. Miami. 15 February 2000.
Thompson, James D. Organizations in Action: Social Science Bases of Administrative
Theory. New York: McGraw-Hill, 1967.
Thornton, Mary. FBI to Take Major Role in Drug Enforcement. Washington Post (22
January 1982). Http://web.lexis-nexis.com/universe [Accessed 24 June 2001].


126
enrichment goals, or receive an opportunity to make money in the legitimate economy.19
Finally, retirement may be driven by a combination of factors. An unwelcome brush with
the law may cause a participant to reconsider his priorities, making him less tolerant of
work-related stresses and more open to gainful employment in dull but less
troublesome legitimate occupations (Fuentes, 1998, p. 164; Mermelstein, 1990; Montoya,
2000; Valds, 1999).
Another source of turnover in trafficking enterprises is involuntary expulsion.
Participants in trafficking groups may be fired for failing to adequately perform their
work-related activities or engaging in behavior that puts the enterprise or leaders at risk.
Former participants and law enforcement officials interviewed by Fuentes identify
instances where participants in Colombian organizations were fired for consuming
cocaine or abusing alcohol, exposing themselves to unnecessary risk through their
personal activities, continually showing up late for scheduled appointments, and
repeatedly getting into troublesome situations. In addition, participants may be dismissed
if the enterprise is experiencing difficult times. When trafficking enterprises face intense
pressure from drug enforcers, they may downsize their operations and lay off unessential
personnel (Emil, 2000; Fuentes, 1998, pp. 165-167, 191; Mermelstein, 1990).
In some circumstances, participants are subject to more drastic sanctions than loss
of employment. If a worker is believed to have willingly cooperated with law
enforcement authorities, stolen product or proceeds from the entrepreneur, or engaged in
reckless behavior that threatens the existence of the enterprise, he or she may be subject
to physical intimidation and violence, including assassination. However, acts of physical
19 However, greed has a tendency to raise the retirement bar for many, making deliberate withdrawal an
elusive option (Fuentes, 1998, p. 164).


164
Policy identifies several changes in transportation vessels by Colombian trafficking
enterprises:
When we started going after the small planes they moved to the big jets. When
we had figured that they had moved to the big jets then they started moving more
by maritime. And when we started paying more attention to the maritime stuff...
now you see a lot more fast boat stuff. When we started chasing the fast boats
then every once in a while youd see this stuff about the semi-submersibles
(McNamara, 1999).
Changes in transportation vessels are not limited to the exportation of
psychoactive drugs stage. At each phase of the transnational commodity chain,
trafficking groups switch transportation modes in response to state interdiction efforts.
When the Peruvian government implemented a policy to shoot down suspected drug
planes in the early 1990s, transshipment groups shifted to rivervine vessels to move coca
paste and cocaine base into Colombia for further refining. U.S.-based transportation
rings have used a variety of overland vehicles, including tractor trailers, trains, pick-up
trucks, minivans, and sedans to ship cocaine across the country, switching among
different types as necessary in order to avoid law enforcement efforts (Bajak, 1997;
CNN, 1999; DEA, 1993a, p. 19; Derian, 1999; Emil, 2000; Faiola & Wilson, 2001;
Frechette, 1997; Furden, 2000; Hinson, 1997; Johnson, 2000; Thompson 2000).
Trafficking enterprises also alter conveyance methods. Containerized cargo are
particularly useful for smugglers because they provide dozens of options to hide illicit
drugs within shipments of legitimate Central and South American exports to the U.S. and
Europe, including vegetables and flowers, fruits and fruit juices, coffee and seafood,
mechanical equipment, and an assortment of other commodities. When drug enforcers
discover a particular scheme, enterprises quickly change to other containerized cargo
routes already in operation, or develop new ones, sometimes by purchasing established


CHAPTER 3
TOWARD A THEORETICAL UNDERSTANDING OF
ORGANIZATIONS AND ORGANIZATIONAL LEARNING
Organizational learning is a slippery concept. It contains multiple meanings that
are open to a variety of interpretations. Learning is difficult to define, let alone measure
and apply to empirical settings. Reflecting such difficulties, the academic literature lacks
a generally agreed upon theory of organizational learning (Levy, 1994, p. 280; Fiol &
Lyles, 1985, p. 803; Keohane & Nye, 1989, p. 264; Stein, 1994, p. 224). In spite of the
theoretical and methodological challenges associated with organizational learning, over
the past several decades social scientists have drawn on a number of basic concepts to
develop a cumulative body of research that demonstrating learning to be a valid
explanation for certain types of organizational change. Dating back to Herbert Simons
seminal formulation, first published in 1945, that organizations intelligently adjust
routines in response to external stimuli and experience, political scientists have applied
notions of learning to a variety of subject matter, including government institutions,
intergovernmental bodies, military organizations, transnational epistemic communities,
and elite decision-makers.1
1 Two decades after the appearance of the first edition Administrative Behavior (Simon, [1945] 1997), Karl
Deutsch applied the concept of cybernetics to communications systems within governments and other
social organizations (Deutsch, 1966). In the early 1970s several political scientists and diplomatic
historians analyzed how policy makers draw on lessons of the past when making decisions (Lowenthal,
1972; May, 1973). In 1976, Robert Jervis drew on social and cognitive psychology to explain how
decision makers reason and leam from previous experience (Jervis, 1976). In recent years the research
agenda on learning in political science has gathered momentum as a growing number of scholars have
applied learning concepts to individual decision-making and organizational behavior. Examples of this
literature includebut are not limited to: Breslauer and Tetlock (1991), Dodd (1994), Eden (forthcoming),
Etheredge (1985), Evangelista (1988), Goldgeier (1994), Goldstein & Keohane (1993), Haas (1990), Haas
58


21
also allegedly obtained in Cartagena. The same year, El Espectador, a leading
Colombian newspaper, reported that Colombian police officials confiscated a large
consignment of illicit drugs in the home of a prominent Bogotano. A confidential U.S.
State Department document from 1936 claimed that the Colombian islands of
Providencia and San Andrs were being used by smugglers for exporting illicit drugs and
alcohol to the U.S. (Senz Rovner, 1996, pp. 69-70).
Though sketchy in the details, these government documents and press reports
suggest that a number of smugglers based in Colombia were involved in illicit
transnational networks linking European drug producers with Caribbean and North
American consumers. They also indicate that Colombian ports located on the Caribbean
sea, including Barranquilla, Cartagena, and Santa Marta, were transshipment points for
opiates and cocaine produced in Europe. Given the paucity of data, I cannot say much
about the organizational forms involved in these criminal actitivities. Nor is it possible to
draw direct linkages between these early pioneers and smuggling groups in later decades.
However, it is clear that at least some trafficking was taking place in Colombia as early
as the 1930s, during which time the country served as a significant transit point for
international drug flows (Arango & Child, 1984, p. 176; Lpez Restrepo & Camacho
Guizado, 2001, pp. 3-4; Walker, 1989, pp. 75-76).
Government Drug Enforcement Efforts During Phase One
In the 1920s and 1930s the government of Colombia came to view the
consumption of coca leaves by its citizens as a social problem (Walker, 1989, p. 73).
Increased concern among national policy makers was reflected in a number of measures
during this period. According to Lpez Restrepo and Camacho Guizado, Law 11 of


55
processing laboratories on Colombian soil, transforming the country from mere
transhipment point to drug producer. Early producers sold their illegal wares to
international smugglers based in Cuba who transported the drugs to the U.S. During the
1950s and 1960s, many smugglers based in Colombia were content to serve as suppliers
and transhippers for trafficking enterprises based in other countries, including Cuba,
Chile, Mexico, and the United States. The primary psychoactive drugs produced in or
passing through Colombia at this time were marijuana and cocaine.
In the late 1960s and early 1970s a fundamental change occurred in the
Colombian drug trade. Several enterprises expanded into U.S. wholesale distribution
markets. These networks developed infrastructures and technologies that laid the
foundation for core enterprises in the 1980s. With the rise of the Medellin and Cali
cartels, and their multi-tonnage trafficking capacities, Colombias dominant position in
the international cocaine trade solidified. Driven by growing U.S. appetites for marijuana
and cocaine, the amount of psychactive drugs coming out of Colombian in the 1970s and
1980s increased substantially, while the organizations coordinating this traffic also grew
and became increasingly bureaucratic.
By the mid-1980s, a number of Colombian trafficking enterprises had become too
large, their leaders too ambitious. Over the next decade, the Colombian and U.S.
governments waged an on-again, off-again campaign to dismantle the leading core
organizations. While drug enforcers achieved remarkable victories, dismantling several
Medellin and Cali cartels, their ability to disrupt the industry was limited by the fluid
nature of trafficking enterprises. Within months of even the most effective supply-
reduction programs, drug smugglers would bounce back, revitalizing existing operations


270
Ambrus, Steven. Changing Lifestyles; A New Mood in Medellin. Los Angeles Times (8
November 1994). Http://web.lexis-nexis.com/universe [Accessed 16 February
1999],
Anderson, Annelise G. The Business of Organized Crime: A Cosa Nostra Family.
Stanford: Hoover Institution Press, 1979.
Anderson, Annelise G. Organised crime, mafia and governments. The Economics of
Organised Crime. Eds. Gianluca Fiorentini and Sam Peltzman. Cambridge:
Cambridge University Press, (1995): 33-54.
Andreas, Peter. Border Games: Policing the U.S.-Mexico Divide. Ithaca: Cornell
University Press, 2000.
Andreas, Peter R., Eva C. Bertram, Morris J. Blachman, and Kenneth E. Sharpe. Dead
End Drug Wars. Foreign Policy 85 (1991-1992): 106-128.
Anonymous. El Cultivo de la Coca y la Elaboracin de la Cocana: Informacin General
(February 1991).
Arango, Mario. Impacto del Narcotrfico en Antioquia. 3rd edition. Medellin: Editorial J.
M. Arango, 1988.
Arango, Mario and Jorge Child. Narcotrfico: Imperio de la Cocana. Medellin: Editorial
Percepcin, 1984.
Arenas, Femando. Former drug trafficker. Frontline: Drug Wars. 2000.
Http:/Avww.pbs.orgAvgbh/pages/frontline/shows/drugs/interviews/arenas.html
[Accessed 29 March 2001].
Argyris, Chris. On Organizational Learning. Cambridge: Blackwell Publishers, 1993.
Argyris, Chris and Donald A. Schon. Organizational Learning. Reading: Addison
Wesley, 1978.
Argyris, Chris and Donald A. Schon. Organizational Learning II: Theory, Method, and
Practice. Reading: Addison-Wesley, 1996.
Arlacchi, Pino. Mafia Business: The Mafia Ethic and the Spirit of Capitalism. Translated
by Martin Ryle. London: Verso, 1986.
Arquilla, John and David Ronfeldt. The Advent of Netwar: Analytic Background.
Studies in Conflict & Terrorism 22 (1999): 193-206.
Arreguin, Leo. Country Attach, Drug Enforcement Administration, Colombia. Bogot.
15 June 2000.


73
Simon 1997, p. 218). Although recent advances in information technology have reduced
the costs of recording and retrieving information, much organizational experience
remains undocumented. Undocumented knowledge is often transmitted through informal
organizational memories. These include the conversations, stories, and myths through
which participants construct and conserve inter-subjective understandings regarding
appropriate practices and procedures (Levitt & March, 1988, p. 327; Argyris & Schon,
1996, p. 16).
Organizations interpret only a small portion of their experience at a time.
Organizational memories are evoked at certain times by particular parts of the collective.
Depending on how they are organized, some aspects of organizational memories may be
more accessible than others. In particular, the availability of memories is associated with
their recency and frequency of use, as well as their location within the organization.
Recently and frequently used memories are easier to retrieve than those that are used only
sparingly or have not been accessed over a considerable period of time. Information
stored in electronic documents that form part of computerized data bases, or paper
documents located in well-organized filing systems are more likely to be used than
isolated documents stored in obscure locations (Levitt & March, 1988, pp. 328-329).
Applying Information
The process of organizational learning is not complete until knowledge that has
been acquired and interpreted by participants is applied to collective behavior (Garvin,
2000, p. 26). Organizations enact knowledge and experience through routines. Routines
include rules, procedures, conventions, and practices that guide individual behavior
within organizational settings. Routines provide the mechanism through which the


259
combat major narcotics organizations. In the 1970s, the DEA created the CENT AC
program composed of quasi-autonomous drug enforcement units that conducted
conspiracy investigations of, as one observer notes, the biggest of the big-time
trafficking enterprises (Glionna, 1986). In the 1980s and 1990s, the CNP established
rural and urban elite units that sprayed illicit drug plantings, destroyed processing
laboratories, interdicted precursor chemicals, raided properties, seized drugs, confiscated
assets, and apprehended the leaders of the cocaine cartels. In recent years, the two
agencies have collaborated to create four CNP Special Investigative Units (SIUs) that
focus on drug trafficking, money laundering, and the supply of precursor chemicals.
Participants in the program undergo a rigorous screening process that includes
computerized background investigations, security questionnaires, background interviews,
medical and psychological testing, polygraph examinations, and urinalysis testing.
Successfully vetted recruits attend a four-to-five week training course in the U.S. and
receive ongoing training when they return to Colombia to join their units. SIUs have
access to sophisticated weaponry, electronic surveillance equipment, and transportation
and communications technologies (BINM, 1988, p. 89; BINM, 1991, p. 97; Camacho
Leyva, 1993, p. 263; CNP, 2000, p. 365; GAO, 1999, pp. 56, 58; Nadelmann, 1993, p.
291-292).
The failure of these elite units to reduce aggregate drug production levels speaks
to the difficulty of the task and the limits of the reform efforts. For all their emphasis on
recruitment, vetting procedures, and training programs, the bureaucracies that created
them ultimately bound these DEA and CNP units. Within the units, flows of information
remained subject to taller (and slower) decision-making hierarchies of the bureaucracy,


4
opium plantings.1 Moreover, since 1991 Colombian anti-drug units, working closely with
their U.S. counterparts, have captured the leaders of the Medellin, Cali and the North
Valley drug trafficking organizations. These arrests are part of a government strategy
aimed at dismantling the communications, transportation, and financial infrastructures of
the larger trafficking organizations by targeting their senior managers. The goal of the so-
called kingpin strategy is to disrupt the production of narcotics in Colombia by
capturing, incarcerating, and confiscating the wealth of those individuals who direct the
illicit drug industry (CNN, 1998b; Nieves, 1997).
Unfortunately, the kingpin strategy and increased eradication efforts have failed
to resolve the Colombian narcotics dilemma. Indeed, according to estimates contained in
the most recent International Narcotics Control Strategy Report, the estimated amount of
potential coca production in Colombia increased from 50,900 hectares in 1995 to 169,800
hectares in 2001.2 Corresponding with the increase in coca production has been increases
in estimated cocaine production from 230 metric tons in 1995 to more than 580 metric
tons in 2001. The estimated cultivation of opium poppies in Colombia, used to make
heroin, has also increased in recent years. In 1995 an estimated 10,300 hectares of
Colombian farmland were devoted to opium production. By 1997, the last year for which
1 These figures are contained in the International Narcotics Control Strategy Report, which is published
annually by the Bureau of International Narcotics and Law Enforcement Affairs (BINLEA and BINM,
various years, and ONDCP 2002). A caveat is warranted in presenting these and other drug-related data in
this study. Generating estimates for illicit drug production and chug performance outputs remains, at best,
an imprecise science. The clandestine nature of narcotics trafficking and the politicized nature of chug
enforcement make it difficult to produce accurate approximations for these activities. Throughout this
study, figures regarding drug production and drug enforcement indicators are meant to illustrate trends
rather than offer precise estimates.
2 Although this report contains what many specialists regard as the most accurate production estimates of
illicit chugs available, several researchers who have conducted field studies argue that the report
underestimates the amount of cocaine and heroin produced in Colombia (Clawson & Lee, 1996; Uribe
Ramirez, 1997). A hectare is equivalent to 2.47 acres.


140
preliminary assessment of the primary research proposition, I begin by classifying the
negative and affirmative responses to this question.
The interview sample includes seventy-six Colombian and U.S. respondents. At
the time of their interviews, sixty-seven percent (N=51) of the respondents were
government officials. Sixty percent of these officials (N=31) worked in drug
enforcement agencies, such as the Colombian National Police (CNP) and the Drug
Enforcement Administration (DEA). In addition, I interviewed fourteen academic
researchers, six professional journalists, and five former participants. The former
participants were involved in different trafficking enterprises and activities, including:
Coca farmer/coca paste processor from southern Colombia (Putumayo
province) (Carlos);
Independent (i.e. non-core organization affiliated) broker of marijuana and
cocaine (Nestor);
Member of an independent, U.S.-based polydrug distribution ring (Homero);
Member of a core-affiliated international transportation cell (Emil); and
Head of U.S. operations for a core affiliated narcotics organization (Jorge
Valds);2
Table 5-1 Professional Affiliation & Nationality of Respondents
Government
Academic
Professional
Former direct
official*
researcher
journalist
participant
Colombian
26
11
4
4
United States
24
3
2
1
Note: One government official worked for the United Nations.
In the course of conducting interviews, I asked seventy respondents if Colombian
trafficking enterprises adjust their behavior in response to experience or information.3
Ninety-six percent of these respondents (N=67) replied in the affirmative. Forty-seven
2 With the exception of Jorge Valds, pseudonyms are used throughout this study to protect the identities of
these respondents. Valds, who has written an autobiography of his former criminal career (Valds, 1999),
granted permission to use his real name.
3 In six interviews, four with Colombian government officials and two with Colombian academics, the
topic of organizational learning by narcotics enterprises was not addressed. For a complete list of all the
respondents, see Appendix A.


303
MacDonald, Scott B. Dancing on a Volcano: The Latin American Drug Trade. New
York: Praeger, 1988.
MacDonald, Scott B. Mountain High, White Avalanche: Cocaine and Power in the
Andean States and Panama. New York: Praeger, 1989.
MacDonald, Scott B. Colombia. International Handbook on Drug Control. Eds. Scott
B. MacDonald and Bruce Zagaris. Westport: Greenwood Press, (1992): 157-170.
Maitland, Leslie. Through Luck and Dogged Work, Agents Disrupt a Top Cocaine
Ring. New York Times (14 June 1981): A1 Http:/Aveb.lexis-nexis.com/universe
[Accessed 29 November 1999],
McNamara, Katherine [pseud]. Senior Policy Analyst, Office of National Drug Control
Policy. Washington. 19 July 1999.
Mann, Dean E. Environmental Learning in a Decentralized Political World. Journal of
International Affairs 44.2 (1992): 301-37.
March, James G. Introduction: A Chronicle of Speculations About Decision-Making in
Organizations, Decisions and Organizations. Ed. James G. March. New York:
Basil Blackwell, (1988): 1-21.
March, James G. A Primer on Decision Making: How Decisions Happen. New York:
Free Press, 1994.
March, James G. Introduction, The Pursuit of Organizational Intelligence. Ed.
James G. March. Malden: Blackwell Business, (1999): 1-10.
March, James G. Understanding How Decisions Happen in Organizations, The
Pursuit of Organizational Intelligence. Ed. James G. March. Malden: Blackwell
Business, (1999): 13-38.
March, James G. Continuity and Change in Theories of Organizational Action, The
Pursuit of Organizational Intelligence, by James G. March. Malden: Blackwell
Business, (1999): 39-51.
March, James G. and Johan P. Olsen. The Uncertainty of the Past: Organizational
Learning under Ambiguity, Decisions and Organizations. Ed. by James G.
March. New York: Basil Blackwell, (1988): 335-358.
March, James G. and Johan P. Olsen. Rediscovering Institutions: The Organizational
Basis of Politics. New York: Free Press, 1989.
March, James G. and Herbert A. Simon. Organizations. New York: John Wiley & Sons,
1958.


88
members, errors in recording experience, conflicting interpretations of experience, and
the challenge of drawing consistent inferences from changing realities (Levitt & March,
1988, pp. 327-328). Even when organizational memories produce clear signals and
interpretation is relatively unambiguous, their behavioral implications may not be clear to
decision-makers (Garvin, 2000, p. 26).
The failure to share information exacerbates problems of information acquisition
and interpretation. Rapid and robust information flows are essential if participants
located in different parts of the organization are to make decisions based on the best
available data. While poorly organized memories may contribute to weak information
flows, many organizations intentionally compartmentalize knowledge in order to protect
proprietary information from outsiders. The need to maintain secrecy may be useful if it
prevents competitors from obtaining access to production methods and innovations,
however, it can also limit the organizations ability to learn from its own and others
experience. If one part of the organization is unaware of the trial and error experience of
other subunits, it may be more likely to make similar mistakes; similarly, secrecy
between organizations limits vicarious learning (Sagan, 1993, p. 43; Huber, 1991, p. 95).
Difficulties also plague the final phase of the learning process. Even when
organizations acquire and interpret information and experience, this knowledge may fail
to impact organizational behavior in desired ways. Structural inertia may prevent the
organization from adopting innovative technologies and routines that could improve task
performance. Organizations often face large sunk costs with existing technologies, and
therefore may be hesitant to embrace new ways of doing things. Organizations also face
internal political constraints to adaptation. Powerful interests within organizations may


274
Bowden, Mark. Killing Pablo: The Hunt for the Worlds Greatest Outlaw. New York:
Atlantic Monthly Press, 2001.
Brandy, Guillermo [pseud]. Head, Special Unit of Financial Investigations,
Administrative Security Department. Bogot. 2 May 2000.
Branigin, William. Cartels Shipping 'Black' Cocaine. Washington Post (28 April 1999):
A02. Http://web.lexis-nexis.com/universe [Accessed 16 April 2001].
Breslauer, George and Philip Tetlock. Eds. Learning in U.S. and Soviet Foreign Policy.
Boulder: Westview Press, 1991.
Brinkley, Joel. 4-Year Fight in Florida Just Cant Stop Drugs. New York Times (4
September 1986). Http://web.lexis-nexis.com/universe [Accessed 26 March
2001],
British Broadcasting Company (BBC News). My Story: The Drug Smuggler. (8 June
2000). Http://news.bbc.co.uk/hi/english/world/newsid 779000/779728.stm
[Accessed 19 February 2001].
Brooks, Laura. Drug Lords' Art Doesn't Pass the Taste Test. Washington Post (15 June
1998). Http://www.washingtonpost.com/wp-srv/WPlate/1998-06/15/0631-061598-
idx.html [Accessed 27 January 1999].
Brooks, Laura and Douglas Farah. New Breed of Trafficker Replacing Drug Cartels:
Small Groups, Shipments Are Trademarks. Washington Post (22 February
1998). Http://www.washingtonpost.com/wp-srv/WPlate/1998-02/22/1611-022298
idx.html [Accessed 27 February 1998].
Brown, M. Leann. Scientific Uncertainty and Learning in European Policymaking.
Policy Studies Journal 28.3 (2000).
Buell, Barbara. The Forgetting Phenomenon Can Decrease Productivity, Researcher
Discovers. Stanford Report (20 September 2000): 7.
Cook, Jonathan [pseud]. Head, International Liaison Office, Anti-Narcotics Directorate,
Colombian National Police. Bogot. 18 June 1997.
Bupp, Irvin C. and Jean-Claude Derian. Light Water: How the Nuclear Dream Dissolved.
New York: Basic Books, 1978.
Bureau of International Narcotics and Law Enforcement Affairs (BINLEA), U.S.
Department of State. International Narcotics Control Strategy Report, 1996.
Washington (March 1997).


185
games to describe how these state and non-state competitors learn through their repeated
interactions. I then examine the organizational and environmental conditions that tilt the
playing field in favor of trafficking enterprises.
A First Cut II: Quantifying Interview Results
Do drug enforcement agencies change their practices, procedures, and programs
as a result of acquiring and analyzing information? The original survey instrument did
not address the issue of state learning.1 Therefore, unlike the questions discussed in
Chapter 5, those on state learning were not put to all of the respondents. Fortunately, the
semi-structured format of my interviews provided the flexibility to discuss the issue when
it emerged, which was surprisingly often. On numerous occasions, government officials
pointed out that their agencies learn from experience. In one interview, a high-ranking
official within the Colombian National Police (CNP), stressed that his organization
learned by studying the narcotraficantes and their methods of operations (Gallego
Castnlln, 2000). Similar comments were made by other knowledgeable officials,
including the U.S. Ambassador to Colombia, the director of the State Departments
Narcotics Assistance Section in Colombia, the DEA deputy director for Colombia, the
head of a CNP investigative unit, the director of the CNPs aerial eradication program,
and the head of the United Nations Drug Control Program in Colombia. In all, twenty-
1 An explanation is in order. Originally, the empirical scope of the research was limited to Colombian drug
trafficking organizations. Drug enforcement agencies were not part of the original research design or
dissertation prospectus. However, in the course of undertaking the research, I realized that the DEA and
CNP were also changing practices, procedures, and performance programs in response to environmental
feedback, at which point I began gathering data on state learning. Unfortunately, fewer interviews
systematically addressed the topic of state learning. This chapter examines the relevant materials.
2 Similar comments were made by the U.S. Ambassador to Colombia, the director of the State
Departments Narcotics Assistance Section in Colombia, the deputy director of the DEA in Colombia, the
head of an elite CNP investigative unit, the director of the CNPs aerial eradication program, the head of
the United Nations Drug Control Program in Colombia, and other knowledgeable officials (Bandera, 2000;
Frechette, 1997; Hinson, 1997; Lupallo, 2000; Molano, 2000; Moreno, 1997; Nyholm, 2000; Oporto, 2000;
Pineda, 2000).


ACKNOWLEDGMENTS
As with the production, transportation, and distribution of psychoactive
substances, writing a dissertation is a collective endeavor. This study was completed
with the support of many individuals and institutions. Fellowships from the National
Science Foundation, the Tinker Foundation, and the Division of Sponsored Research at
the University of Florida supported field work in Bogot and various locations in the
United States. A residential fellowship at the Center for International Security and
Cooperation at Stanford University supported data analysis and writing.
While conducting research in Bogot, I benefitted enormously from the Institute
of Political Studies and International Relations (IEPRI) at the National University of
Colombia. The director of the Institute, William Ramirez, provided a place to hang my
hat between interviews, and numerous IEPRI scholars offered constructive criticism and
friendship. For their assistance during this period, I thank Alvaro Camacho, Nathanial
Christie, Fernando Cubides, Francisco Gutirrez, Andrs Lpez, Gregorio Prez, Alvaro
Valencia, and Carlos Velsquez. During research trips to Washington, D.C. and Miami,
Fiona Wright and Juan Carlos Valencia provided greater hospitality than I deserved.
Dozens of U.S. and Colombian officials gave generously of their time during
interviews. Without their willingness to share their knowledge, this study could not have
been written. I owe a similar debt of gratitude to the former traffickers that spoke with
me. For reasons of discretion, many of my respondents must remain anonymous. They
11


71
and error experience and exogenous search. Trial and error experience refers to the
process by which organizations interact with their environment, using a variety of tactics
and strategies designed to achieve targets. The experience that emerges from this activity
provides a valuable guide for future behavior. Search refers to the process by which
organizations gather information from sources exogenous to the organization but
endogenous to the task environment. Search provides the means to draw on the
knowledge and experience of others, including customers, consultants and competitors, in
making decisions (Argyris & Schon, 1996, p. 3; Weick, 1979).
Gathering information is costly. It involves an expenditure of organizational
resources, including time, attention, personnel, and money. Because resources are
limited, organizations are more likely to acquire knowledge and experience when they
have a reason to do so, such as the need to find a solution to a problem. As target-
oriented entities, the failure to meet targets or objectives is a common source of problems
in organizations. The failure to match expected outcomes with actual outcomes produces
emotional arousal on the part of participants that triggers a search for information,
including alternative courses of action to resolve the problem (Allison, 1971, p. 84;
Argyris & Schon, 1996, p. xxiii; Garvin, 2000, p. 23; March, 1988, p. 3; March & Olsen,
1989, p. 60; Weick, 1995, pp. 45-46).
Interpreting Information
For the learning process to proceed, participants must interpret information. Daft
and Weick provide an illuminating description of this complex process:
Organizations must make interpretations. Managers literally must wade into the
ocean of events that surround the organization and actively try to make sense of
them. Organizational participants physically act on these events, attending to
some of them, ignoring most of them, and talking to other people to see what they


115
their trafficking operations, and reserve for themselves full information of the criminal
enterprise, including the identities of important suppliers and customers (Fernandez,
2000; Fuentes, 1998; Gale & Rodriguez, 2000; Homero, 2000; Montero, 2000; Olson,
2000; Thompson, 2000; Valencia Tovar, 2000; Zabludoff, 1997, p. 31).13
Routines in Trafficking Enterprises
Routines are a fundamental part of the social structure of Colombian trafficking
organizations. Rules and conventions provide prescriptions and proscriptions for
conduct, while practices and procedures coordinate individual actions into criminal
conspiracies. Trafficking organizations have promulgated a variety of rules for their
participants, many of which reflect commonsensical conventions regarding enterprise
security. In many enterprises, participants are prohibited from discussing business with
non-participants, including members of other trafficking groups. They are expected to
maintain a low profile and avoid attracting unwanted attention, particularly when
conducting criminal activities. If apprehended, participants must refuse to cooperate with
law enforcement authorities, and stealing is strictly forbidden. Although these norms and
rules are not formally codified in documents, they are widely understood among
participants. Some enterprises hold meetings in which leaders or managers inform
workers of the norms and rules of behavior, and warn them of the consequences for
violating these injunctions (Aero, 2000; Arenas, 2000; Emil, 2000; Fuentes, 1998, pp.
155, 161; Henderson, 1992, p. 91; Kacerosky, 1994; Pallomari, 1997 Vols. 36, 40; Third
Superceding Indictment, p. 15; Valds, 2000).
13 The restriction of information flows within Colombian trafficking enterprises contains obvious and
important implications for their ability to learn from previous experience and knowledge. These
implications will be analyzed in the following chapter.


69
organizational inputs and outputs; support materials and services that allow participants
to carry out their activities; and the knowledge, skills and experience of sex workers
themselves.
Depending on the degree of hostility in the task environment, criminal
organizations may be required to devote a considerable portion of their technological
resources to conceal and otherwise protect the physical integrity of their operations.
However, some illicit activities, including illegal gambling and prostitution, are more
accepted among law enforcement and government officials than others, such as narcotics
and weapons trafficking (Reuter, 1983; Smith, 1975). For this reason, it is not surprising
that many contemporary narcotics organizations devote a substantial portion of their
technological resources towards reducing uncertainty and risk emanating from the task
environment. Trafficking enterprises use these material and symbolic inputs to hide
narcotics shipments, corrupt government officials, encipher communications, gather
counter-intelligence, stockpile inputs, and maintain warehouse inventories.
Defining Organizational Learning
Organizations learn when participants receive, interpret, and apply information to
behavior through organizational routines. Organizational learning is a form of
organizational action. When an organization acts, such that participants fulfill roles
and perform functions on behalf of the collective, then it may be said to learn when
participants gather, analyze, and apply information (Argyris, 1993, p. 8; Argyris &
Schon, 1996, p. 10; Farkas, 1998; Hedberg, 1981, p. 3; Simon, 1996, p. 176).
Throughout this study, organizational learning refers to a certain type of behavior
performed by individuals acting on behalf of the collective.


64
authority, work, and communication within organizations (Argyris & Schon, 1996, pp. 8-
9; Donald & Wilson, 2000, p. 194; Leavitt, 1965, p. 1144; Leavitt et al., 1973, p. 4;
McCluskey & Wardle, 2000, p. 252).
Routines
Routines are an essential part of organization structure. They include rules,
procedures, conventions, and practices that guide individual behavior within
organizations. Routines interlock behavior among multiple participants by matching
action to informational cues. Routines provide rules of thumb for participants to
respond to incoming stimuli in ways that make sense. In this manner, routines coordinate
the behavior of different individuals into coherent and predictable patterns of behavior.
This coordination allows organizations to accomplish work reliably and consistently
(Cohen & Bacdayan, 1996, p. 348; Cyert & March, 1963; Feldman, 2000, p. 611; March
& Simon, 1958; Weick, 1979, pp. 112-115).
In addition to organizing individual action into collective behavior, routines
influence participants by providing rules for conduct, along with an incentive system that
matches rewards and punishments for specific action. Through training and socialization,
participants learn the rules and procedures of their organization and receive rewards (or
punishments) according to their ability to follow them. Successful participants generally
move up the ranks of the organization and tend to apply the same criteria to their
subordinates. In this manner, routines frequently become institutionalized within
organizations, outlasting individual participants and sometimes even their original
usefulness (Levitt & March, 1988, p. 320; March & Simon, 1958, p. 145; Posen, 1984, p.
44).


93
able to process information and make changes in organizational practices and procedures
more quickly than larger, taller organizations.
P8: Smaller organizations learn more quickly than larger ones.
P9: Flatter organizations learn more quickly than taller ones.
Conclusion
Organizations leam when participants receive, interpret, and apply knowledge and
experience to practices, procedures, and performance programs that guide their collective
behavior. Organizational memories, including files, correspondence, computers, and
inter-subjective understandings, record knowledge and experience in a manner that is
accessible to participants. Organizations leam through interactions with other actors that
share their task environment, including competitors. In changing organizational routines
through information gathering and analysis, participants encode the lessons of collective
experience.
Learning is not the only source of organizational change. Practices, procedures,
and performance programs may change due to power and environmental selection. When
learning does occur, it is often under conditions of profound ambiguity. Participants
struggle to make sense of ambiguous experience. They make decisions based on
incomplete or inaccurate information. They also face limitations in their computational
abilities, and often make inaccurate and biased inferences. These ambiguities hinder
organizational learning, particularly in large bureaucracies with tall decision-making
hierarchies. Environmental benevolence obstructs learning by reducing the incentives for
innovation and increasing the tendency towards organizational inertia. Even when
organizations manage to overcome these ambiguities and alter their practices and


163
enforcement efforts is less intense. With one sea and two oceans, dozens of international
airports, hundreds of clandestine airstrips, and thousands of miles of coastline from which
to choose in Colombia and neighboring countries, trafficking rings have an almost
unlimited array of transportation routes they can develop. Transportation groups alter
embarkation points among dozens of ports and airstrips in Colombia, Ecuador,
Venezuela, Panama, and Suriname. They shift from traditional central Caribbean routes
(i.e. the Windward passage, Yucatan channel, and the Mona passage) to other passages in
the eastern Caribbean, and back again. They form new maritime routes through the mid-
Atlantic and the Pacific Ocean, and over-land routes through Panama, Central America
Mexico, Brazil, Argentina, and Peru. They change transshipment points from the
Bahamas and Jamaica to Costa Rica, Mexico, Guatemala, Puerto Rico, Haiti, the
Dominican Republic, the archipelago of San Andrs y Providencia, and back again
(Bagley, 1999; Berke, 1989; Brinkley, 1986; Camacho, 2000; Cash, 1999; Durfey, 1978,
p. 77; Dye, 1998, pp. 248-250; GAO, 1999, p. 134; Gugliotta & Leen, 1989, p. 69;
Marshall, 2001c; Mermelstein, 1990, p. 153-154; Rice, 1989, p. 83; U.S. House, 1978).
In addition to changing the geographic locations of their routes, trafficking
enterprises adjust their transportation technologies and practices to avoid government
surveillance and interdiction. One common adaptation is to vary the mode of
transportation. When drug enforcers identify and intercept a particular air, sea, or land-
based vessel used to transport narcotics, transportation rings change to different vessels,
again following a period of re-adjustment in which participants gather and analyze the
requisite information. A senior policy analyst with the Office of National Drug Control


125
Table 4-4 Number of Drug-Related Arrests by
Colombian National Police, 1974-1998 (even years)18
Year
Males arrested for
narcotics violations
Females arrested for
narcotics violations
Foreigners arrested for
narcotics violations
1974
1,065
136
104
1976
740
10
19
1978
500
48
7
1980
334
20
4
1982
1,056
76
7
1984
4,422
802
26
1986
2,974
685
40
1988
4,083
815
31
1990
4,766
1,468
19
1992
5,567
1,151
52
1994
5,789
1,326
106
1996
4,478
1,158
67
1998
15,398
2,799
79
Note: Figures are provided for even years only in order to present data that cover a 24-
year period into a single table.
Source: CNP (2000).
and motivations of participants. Some individuals bum out from the high level of
stress that accompanies this illegal occupation. Others recoil at the steady progression of
criminality that sustained participation brings. Participants that have no qualms about
collecting and laundering drug proceeds or guarding a stash house sometimes balk at the
prospect of engaging in physical intimidation or violence. Some people tire of the
hedonistic lifestyle that prompted their initial participation. Drug addiction, family
problems, depression, and a heavy conscience influence others to cease their criminal
activities. More successful participants may retire because they achieve their personal
that are under suspicion for whatever reason, or that owe the organization money, find it more difficult, if
not impossible, to leave peacefully. In some cases, participants that contain knowledge, skills or
experience that is highly valued by the organization receive pressure not to depart. According to
Mermelstein, his immediate supervisor threatened him when he tried to retire voluntarily (Mermelstein,
1990, p. 125). However, Valds, who occupied a position similar to Mermelstein in a separate enterprise,
argues that he was able to retire without receiving threats his Colombian partners (Valds, 1999, p. 219).
Former participants interviewed by Fuentes (1998, p. 164), Williams (1989, p. 123), and Frontline
(Steve, 2000) confirm that voluntary withdrawal from the narcotics industry is feasible provided the
participant is in good standing with his business associates.
8 Figures are provided for even years only in order to compress data that cover a twenty-four year period
into a single table.


226
frequently and rapidly as circumstances warrant. The need to survive often overrides
other factors that may impede learning in trafficking enterprises, including conflicting
goals among participants and organizational inertia.
By contrast, survival for drug enforcers depends on their ability to secure
sufficient budgetary allocations to carry out their programs and policies. While they have
an incentive to produce satisfactory policy outputs to justify their mission and ensure
additional funding, rarely do they face termination if they fail. Even if their performance
indicators drop precipitously, the DEA and CNP are likely to receive adequate budgets.
Indeed, their funding may actually increase if they can demonstrate that their poor
performance is due to the strength of their illicit adversaries. While traffickers and drug
enforcers operate in hostile environments, the latter do not face the same survival
imperative that confronts their illicit adversaries. For traffickers the stakes of competitive
learning games are higher
The Enforcement Dilemma
Law enforcers have a symbiotic relationship with the traffickers they seek to
dismantle. Drug enforcement organizations, such as the DEA and CNP, exist to identify,
apprehend, and dismantle trafficking enterprises. If they somehow managed to eliminate
all their criminal competitors and eradicate the illicit drug trade, drug enforcers would
also remove their raison dtat. Without trafficking enterprises to disrupt, drug
enforcement agencies have no reason to exist. This prospect may appear slim to some
observers. International police organizations that operate in over fifty countries, such as
the DEA, have little reason to worry about the complete demise of transnational
trafficking. Even if one country wipes out the psychoactive drug industry within its


34
remained several years away (Massing, 1998; Randall, 1992, p. 246; Tokatlian, 1988, p.
139; Tokatlian, 1990, pp. 293-294; Walker, 1999, pp. 148-149).
Notwithstanding this ambivalence, the 1970s witnessed an increase in the U.S.
governments drug enforcement presence in Colombia. Prior to the Nixon
Administration, U.S. drug enforcers, including the FBI and the Bureau of Narcotics and
Dangerous Drugs, operated sporadically, on a case-by-case basis, in South America.
With the creation of the Drug Enforcement Administration in 1973, the Nixon
Administration ushered in a new era of the internationalization of U.S. drug enforcement
policy. By the late 1970s, the new super drug enforcement bureau was active in dozens
of countries, and established a permanent presence in Colombia, working closely with the
Colombian National Police (CNP) (Nadelmann, 1993).
With the change of Colombian presidential administrations in August 1978, the
Colombian governments attitude towards drug trafficking shifted. President Julio
Turbay implemented a more aggressive counter-narcotics policy than his predecessors
following allegations in the U.S. media that two of his ministers and his own family were
connected to the illicit drug industry. Following his inauguration, President Turbay
placed the Guajira Peninsula, which had long served as a major production center for
marijuana and transit point for marijuana and cocaine, under the jurisdiction of
Colombia's armed forces. By the end of 1978, 10,000 army troops were in the region
engaged in the manual eradication of marijuana. Operation Fulminant, as the anti-drug
mission was called, produced immediate results: 3,500 tons of marijuana were seized,
which represented an estimated loss of US$ 70 million to Colombian smugglers,
approximately seventy drug running airplanes and seventy seafaring vessels were


12
smuggling operations disruptedat least temporarily. However, no sooner were these
underworld luminaries removed than replacements emerged, eager to pick up where their
predecessors left off. These post-cartel enterprises have learned from the mistakes of the
groups that went before them, downsizing their operations to reduce their vulnerability to
government anti-drug efforts. To the chagrin of state officials, existing groups have
proven more difficult to investigate and dismantle, hampering law enforcers ability to
repeat the successes of the Kingpin strategy.
Research Design
This is a comparative case study of the criminal enterprises that coordinate the
illicit drug trade and the state drug enforcement agencies charged with disrupting these
illegal activities. The basic objective of this research is to determine whether, and to
what extent, these sovereignty-free and sovereignty-bound actors learn as organizations.
Organizational learning refers to the process by which participants acquire, interpret, and
apply knowledge to practices, rules, and procedures that guide organizational behavior.
Organizational learning theory offers a framework for understanding the behavior of
trafficking enterprises and law enforcement and intelligence agencies. This process-
oriented approach illustrates how these state and non-state collectivities gather, analyze,
and apply information to change their structures and operations in order to achieve
satisfactory outcomes. Organizational learning also helps explain why trafficking
enterprises are better learners than state agencies, and how the illicit drug industry
persists in Colombia in spite of government successes in dismantling a number of leading
cartels.
Colombia. The second development refers to the tragic events in New York City of September 11th and


42
While some transshipment points were merely refueling stations, others contained small
crews of workers that off-loaded and repackaged cocaine shipments. In Mexico a
number of trafficking groups with marijuana and heroin smuggling experience provided
this service, breaking down the cocaine into smaller shipments and transporting it
through Mexico in autmobiles and trucks (DEA, 1993, p. 17; DEA, 1994b, p. 7;
Mermelstein, 1990; Pallomari, 1997; PCOC, 1986, pp. 88-89, 98; Raab, 1984; Rice,
1989, pp. 145-146).
Maritime smuggling was conducted by seafaring vessels of various types and
transportation capacities, including high-speed cigarette or go-fast boats, sport
fishermans boats, yachts, medium-sized commercial fishing boats, large, dilapidated
freighters, and even what the DEA identified to as "submarine-like semi-submersible
vessels capable of tranporting hundred kilogram quantities of cocaine through the
Carribean (DEA, 1994b, p. 7). Maritime vessels often departed from commercial ports in
Colombia, such as Buenaventura, Barranquilla and Cartagena, as well as hundreds of
clandestine debarkation points located along the countrys northern coast. On smaller
boats, cocaine was hidden in customized storage compartments, with the vessels
frequently modified for this purpose. On large freight ships, the drug was hidden within
containerized cargo shipments, including blue jeans, ceramic floor tiles, fresh cut flowers,
fruits and vegetables, fish, coffee, lumber, concrete posts, blue jeans, and dozens of other
exports from South America (Maitland, 1981; Griffith, 1997, p. 81; Castillo, 1996, p. 51;
Mermelstein, 1990, p. 146; Nstor, 2000).
a commission, Lehder allowed independent cocaine and marijuana smugglers as well as those associated
with the Medellin network to land and refuel before continuing to the United States. For a description of
Lehders operation, see Gugliotta & Leen (1989) and Shannon (1988). Also, see the interview granted by
Jorge Luis Ochoa (2000) to Frontline.


2
forms the heart of this study: the organizational learning capacity of Colombian
trafficking enterprises. After emphasizing the contribution of the research to existing
bodies of literature in political science, organization theory, and criminology, I describe
the research design in detail. The chapter concludes with an overview of the remainder of
the study.
Persistence of Illicit Drug Production in Colombia
Over the past two decades, the Colombian government has sought to eliminate the
production and transit of illicit narcotics in its national territory. Working closely with
the U.S. and other members of the inter-American narcotics control regime, the
Colombian government has implemented supply-reduction programs that eradicate
drug plantings, destroy drug processing laboratories, intercept the transportation of
narcotics and the chemicals used to make them, and apprehend suspected drug traffickers
and confiscate their illicit profits.
The costs of these programs, in terms of budget allocations and human personnel,
are significant. Since the early 1980s, the Colombian government has spent several
billion US dollars to implement supply-reduction initiatives within its national territory.
While the Colombian government has received considerable anti-narcotics assistance
from the U.S. and other foreign governments over the years, it has also invested a
substantial portion of its own resources in the war on drugs. Moreover, in recent years,
the Colombian governments anti-drug expenditures have increased significantly. In the
1980s, Colombias anti-narcotics budget varied between USS 20 and 25 million per year,
with the U.S. providing half this amount. In 1995 the Colombian government devoted
US$ 900 million of its own funds to anti-drug efforts, and in 1996 this amount increased


284
Drug Enforcement Administration (DEA), U.S. Department of Justice. The South
American Cocaine Trade: An "Industry" in Transition (June 1996).
Http://www.usdoi.gov/dea/pubs/intel/cocaine.htm [Accessed 10 January 1997],
Drug Enforcement Administration (DEA), U.S. Department of Justice. The Supply of
Illicit Drugs to the United States: The National Narcotics Intelligence Consumers
Committee 1995 Report (August 1996). Http://www.usdoj.gov/dea/pubs/nnicc
95/nnicc-95.htm [Accessed 10 January 1997].
Drug Enforcement Administration (DEA), U.S. Department of Justice. The Supply of
Illicit Drugs to the United States: The National Narcotics Intelligence Consumers
Committee 1996 Report (July 1997).
Http://www.usdoi.gov/dea/pubs/intel/nnicc97.htm [Accessed 24 July 1998],
Drug Enforcement Administration (DEA). Operation Juno Indictment Targets Five
Major Traffickers and $26 Million Worth of Laundered Drug Proceeds. DEA
Press Release (22 December 1999).
Http://www.usdoi.gov/dea/pubs/pressrel/pr991222.htm [Accessed 22 January
2002],
Drug Enforcement Administration (DEA). Operation Journey Dismantles Colombian
Organization that Shipped Cocaine to 12 Nations. DEA Press Release (26
August 2000). Http://www.usdoj.gov/dea/pubs/pressrel/pr082600.htm [Accessed
28 May 2001].
Drug Enforcement Administration (DEA). Drug Trafficking in the United States
(September 2001a). Http://www.dea.gov/pubs/intel/01020/index.html [Accessed 1
April 2002],
Drug Enforcement Administration (DEA). DEA Briefing Book (November 2001b).
Http://www.usdoi.gov/dea/briefingbook/ [Accessed 15 January 2002],
Drug Enforcement Administration (DEA). U.S. Department of Justice, Traffickers from
Colombia [No date]. Http://www.usdoi.gov/dea/traffickers/colombia.htm
[Accessed 16 October 2000].
Drug Enforcement Administration (DEA). U.S. Department of Justice, Traffickers from
Mexico [No date]. Http://www.usdoi.gov/dea/traffickers/mexico.htm [Accessed 8
April 2001].
Drug Enforcement Administration (DEA). Narcotics and Dangerous Drugs Information
System [No date], Http://www.usdoi.gov/dea/foia/naddis.html [Accessed 3 July
2001],
Drug Enforcement Administration (DEA). El Paso Intelligence Center [No date].
Http://www.usdoi.gov/dea/programs/epic.htm [Accessed 3 July 2001],


98
laundering network, organizations complete their tasks for the purpose of attaining
satisfactory profits and other benefits, and avoiding unnecessary risks.4
The core organizations that dominated the Colombian cocaine trade from the late
1970s through the mid-1990s, the so-called fourth phase of the industry discussed in
Chapter 2, were multi-task enterprises that functioned along several nodes in the
transnational commodity network. A number of core organizations purchased cocaine
base, processed base to cocaine hydrochloride, acquired aviation and maritime vessels,
arranged debarkation points, recruited financiers for multi-ton shipments, transported
cocaine (often through transit points) to consumer markets in other countries, and
warehoused and distributed cocaine to independent wholesalers. In addition, core
enterprises procured front companies, laundered and invested their illicit profits,
regulated transactions, provided security and contract enforcement, gathered intelligence
on government drug enforcement efforts, and suborned law enforcement and military
officials and members of the Colombian political class (Zabludoff, 1997, p. 49).
The illicit drug industry in Colombia can be represented as a circular network that
takes the form of a wheel (see Figure 4-1 below). During the fourth phase of the
Colombian drug trade, core organizations served as axles to their respective networks,
coordinating activities among different production, transportation, distribution, and
4 A network is a series of nodes that, in one way or another, are connected together. The nodes can be
individuals, organizations, firms, or even computers, but the critical point is that there are significant
linkages among them. Networks vary in size, shape, membership, cohesion and purpose. Networks can be
large or small, local or global, domestic or transnational, cohesive or diffuse, centrally directed or highly
decentralized, purposeful or directionless. A specific network can be narrowly and tightly focused on one
goal or broadly oriented towards many goals; and it can be either exclusive or encompassing in its
membership (Williams, 1998, p. 74; also see Castells, 1996; Powell 1990; Ronfeldt, 1996; and Arquilla &
Ronfeldt ,1999). In this study, Colombian drug networks are composed of individuals and organizations
interlinked for the purpose of producing, transporting and distributing narcotics and laundering illicit
proceeds. While they vary considerably in size, shape, and structure, they are generally transnational, as
their illegal activities cross national boundaries.


298
Kerry, John. The New War: The Web of Crime That Threatens America's Security. New
York: Simon & Schuster, 1997.
Keohane, Robert O. Ed. Neorealism and Its Critics. New York: Columbia University
Press, 1986.
Keohane, Robert O. and Joseph S. Nye. Eds. Transnational Relations and World Politics.
Cambridge: Harvard University Press, 1972.
Keohane, Robert O. and Joseph S. Nye. Power and Interdependence. 2nd Edition. New
York: HarperCollins Publishers, 1989.
Khong, Yuen Foong. Analogies at War. Princeton: Princeton University Press, 1992.
Kier, Elizabeth. Imagining War: French and British Military Doctrine Between the Wars.
Princeton: Princeton University Press, 1997.
Kim, Daniel H. The Link between Individual and Organizational Learning. Sloan
Management Review 35.1 (1993): 37-50.
King, Gary, Robert O. Keohane, and Sidney Verba. Designing Social Inquiry: Scientific
Inference in Qualitative Research. Princeton: Princeton University Press, 1994.
Kline, Harvey F. The Colombian State in Crisis: Continuity and Change. Democracy in
Latin America: Colombia and Venezuela. Ed. Donald L. Herman. New York:
Praeger, (1988): 87-107.
Kline, Harvey F. Colombia: Building Democracy in the Midst of Violence and Drugs.
Constructing Democratic Governance: South American in the 1990s. Eds. Jorge I.
Dominguez and Abraham F. Lowenthal. Baltimore: John Hopkins University
Press, (1996): 20-41.
Kraar, Louis. Inside the Global Drug Trade. Fortune (20 June 1988): 27-38.
Krasner, Stephen D. Structural Conflict: The Third World Against Global Liberalism.
Berkeley: University of California Press, 1985.
Krause, Charles. Colombia, With U.S. Backing, Mounts Drive on Drug Traffic.
Washington Post (20 June 1979). Http://web.lexis-nexis.com/universe [Accessed
22 February 2000],
Krauss, Clifford. Colombia Arrests Raise Price of Cocaine in New York City. New
York Times (15 September 1995). Http://web.lexis-nexis.comAmiverse [Accessed
4 November 1999].


23
being grown on large extensions of Colombian territory and that cocaine was being
processed in-country and exported to Panama. According to this confidential report,
Colombians of German extraction processed cocaine for export and transported the
product by train to Cartagena and then by automobile or pack animals to the Caribbean
ports of Tol, Cispat, or Acand, where it was hidden in banana boats and smuggled to
Panama. Once in Panama, the drugs were transferred from steamship to small launches
and transported to Puerto Piln, where they were guarded by a group of German
smugglers until it was feasible to transport the cargo by automobile to the Canal Zone
(Senz Rovner, 1996, pp. 77-78).3 During World War II, contrabandistas from the Urab
region of Colombia formed smuggling networks that transported illicit drugs, whiskey,
and cigarrettes through Panama and other Central American and Caribbean countries
(Lpez Restrepo, 2000; Walker, 1989).
Colombias shift from transit point to drug producer was facilitated by Cubas
emergence in the international drug trade following World War II. In post-war Havana
criminal organizations coordinated transnational networks composed of Latin American
cocaine suppliers and French-Corsican and Italian heroin brokers. The networks used
couriers, including seamen, stewards, passengers, and pilots, to smuggle drugs from
South America to Cuba, sometimes by way of Central America. These human mules
transported small quantities of coca paste in suitcases with specially modified
compartments, a practice that remains popular among cocaine and heroin couriers today.
In Cuba and Colombia processing groups refined the coca paste into cocaine
hydrochloride, and sent along the finished product to the United States (MacDonald,
3 However, U.S. and Colombian officials disputed this report (Senz Rovner, 1996, pp. 78-83).


100
Medellin and Cali, medium-sized and smaller groups that avoided the crackdowns
capitalized on the opportunity to increase their market share (see Chapter 2). In the
current phase of the countrys drug industry, independent enterprises, now free of the
overarching authority of a Pablo Escobar or Gilberto Rodrguez Orejuela, continue to
transact with their respective suppliers and buyers. In effect, these transnational networks
have become self-organizing.
Understanding Tasks in Colombian Trafficking Enterprises
While it is difficult to ascertain the degree to which tasks are understood within
clandestine criminal organizations, there are indications that participants in Colombian
trafficking enterprises appreciate the goals and objectives of their illicit activities. One
respondent in this research, a former member of one of the Cali core organizations,
argues that his organization was composed of many little offices, each with their own
purpose, but all of them working for the same objective (Emil, 2000). Emil also pointed
out that the objective of his group was to transport cocaine from Central America and
other transshipment points to the U.S.5
In testimony before the U.S. House Subcommittee on Crime, a former low-level
member of a Medellin enterprise remarked that the main purpose of this organization
was to ship drugs (Mr. Rodriguez, 1997, p. 5). A black market peso broker
interviewed in the Frontline documentary, Drug Wars, pointed out that the
fundamental purpose of his business was to buy money, and sell it for a profit
(David, 2000). Other former direct participants in Colombian trafficking groups offer
detailed descriptions of the production, processing, and transportation tasks undertaken
5 To preserve their anonymity, I provide many respondents with fictitious names. Pseudonyms are used
throughout the text and are listed in the bibliography and Appendix A.


99
money laundering groups. Discrete groups interacted with each other, and the core
enterprise, in completing their assigned tasks. Thus, purchasers of coca paste would
transact with drug farmers that produce this commodity, and processing laboratories that
manufacture cocaine hydrochloride. Processing labs, in turn, would transact with paste
purchasers (backward linkages) and exporters/transportation coordinators (forward
linkages) in carrying out their activities. In this fashion, these transnational networks
turned, symbolized in their circular structures.
Figure 4-1 Wheel Structure of Illicit Drug Network
To the chagrin of government officials, the post-cartel phase of the Colombian
drug industry demonstrates that these transnational networks can function without axles.
Even as drug enforcement agencies dismantled the major core organizations based in


captured, and 1,000 Colombians and Americans were arrested (Krause, 1979; Thoumi,
1995, p. 210; Walker, 1999, p. 150).
35
Following on the heels of the DEAs Operation Stopgap, which targeted
Colombian and American marijuana smugglers in the Caribbean, Fulminant temporarily
disrupted the marijuana industry in Colombia. However, the operation was short lived, in
part due to a lack of support from the Colombian military. The generals were concerned
that too many resources were being shifted away from more pressing anti-guerrilla
operations. They also worried about the debilitating impact of drug-related corruption on
field officers involved in Fulminant (Oijuela, 1990, p. 218).
However, the most significant impact of the marijuana eradication and
interdiction campaigns undertaken by U.S. and Colombian drug enforcers in the late
1970s was an unintended one. In successfully targeting a number of leading marimberos,
Operations Fulminant and Stopgap substantially increased the risks and costs associated
with marijuana trafficking. Given limited state resources, cocaine did not receive as
much attention from drug enforcers as marijuana, providing an additional incentive for
smugglers to diversify into cocaine.5 Many of those with sufficient capital, contacts, and
knowledge did, transferring their marijuana smuggling expertise to the more lucrative
cocaine trade. In this respect, short-sighted government drug enforcement policies and
programs that focused on marijuana trafficking helped set the stage for the dramatic
increase in cocaine trafficking over the next decade (Betancourt & Garcia, 1994; Cannon,
1979; Krauthausen & Sarmiento, 1991; Oijuela, 1990, p. 218; OToole, 1978).
5 Apart from increased marijuana drug enforcement, there were other incentives to switch to cocaine
trafficking. For one thing, cocaine was less aromatic than marijuana, making it easier to hide and transport.
Also, per unit profits for cocaine were substantially greater than for marijuana, reducing the need to
transport such large quantities to ensure adequate profits.


83
carrying out these goals (Chandler, 1962; Scott, 1998, p. 287). Tactics are the
procedural and technological means by which organizations implement courses of action.
While strategy is the overarching concept, military strategists have long-recognized the
causal relationship is not unidirectional. Tactics can influence strategy, as when
technological innovations developed on the field of battle lead to changes in strategic
planning.
In this study, I am concerned not with armies waging warfare but criminal
enterprises that produce, process, transport, and distribute illicit drugs, and the state drug
enforcement agencies charged with stopping them. Therefore, the warfare analogy is not
without merit. As seen in Chapter 2, narco-narc interactions have developed increasing
degrees of intensity during the last two decades of the war on drugs. In this context,
tactical routines refer to the rules, procedures, conventions, technologies, and inter-
subjective understandings used by trafficking enterprises and drug enforcement agencies
in carrying out their day-to-day tasks and meeting their immediate objectives in the
ongoing battles of the drug war. Strategic routines refer to those rules and procedures
governing long-range goals and objectives, strategies for achieving these goals and
objectives, and basic organizational structures. The latter may include decision-making
hierarchies, management layers, and organization subunits.
Most organizational learning takes place at the level of tactical routines.
Organizations tinker with tactics in response tofrequently negativefeedback
regarding task performance. The information gathering and interpretation costs of
tactical learning are relatively minor. If the organization contains a number of alternative
routines in its performance program, learning may consist of merely matching the most


149
the illicit drug industry, and drug enforcement. They draw on this knowledge when
carrying out their tasks for the organization (Caldern, 2000; Emil, 2000; Furden, 2000;
Johnson, 1993; Mermelstein 1990; Valds, 1999; Valds, 2000).
Significantly, this informal exchange of information is not confined to members
of the same enterprise. Diffusion of trafficking knowledge often occurs across
organizations, but within common networks. Individuals that belong to separate
enterprises, yet connected through criminal pursuits and personal contacts may talk and
share know-how at parties and other informal settings. Again, Nstor illuminates from
his own experience:
You know the smugglers, you know who they are, you may not even work with
each other but you got each others back... In restaurants you might meet up with
some guys and you find out how the environment, how the situation is going.
This is all useful information (Nstor, 2000).
These contacts are not restricted to participants from the same country, although
common nationality provides a cultural bond that increases trust. When it suits their
interests, Colombian traffickers exchange information with colleagues from a variety of
countries, including the U.S., Mexico, Russia, Italy, and Japan (Reyes Posada, 1999, p. 4;
Valds, 2000; Zabludoff, 1999).
However, due to the potential for law enforcement penetration, many trafficking
enterprises seek to limit the informal exchange of information. Emil stresses that
participants in his core enterprise were always careful to share knowledge only with the
other members of their circle, even during social gatherings. Fuentes makes a similar
point about the Colombian distribution cells he studied. While social contact between
members of different enterprises was tolerated, participants understood that they were not
to discuss cell activities in open settings (Emil, 2000; Fuentes, 1998, fh. 1, p. 114).


324
Thoumi, Francisco. E. Why the Illegal Psychoactive Drugs Industry Grew in
Colombia. Journal of Inter-American Studies and World Affairs 34.2 (1992): 37
63.
Thoumi, Francisco. E. Political Economy and Illegal Drugs in Colombia. Boulder: Lynne
Rienner Publishers, 1995.
Thoumi, Francisco. E. The Impact of the Illegal Drug Industry on Colombia.
Transnational Crime in the Americas. Ed. Tom Farer. New York: Routledge,
(1999): 117-142.
Thoumi, Francisco E. Independent scholar. Washington. 19 July 1999.
Tofani, Loretta. Departing Attorney General Leaves Question in his Wake. Washington
Post (16 February 1985). Http://web.Iexis-nexis.com/universe [Accessed 24 June
2001],
Tokatlin, Juan G. National Security and Drugs: Their Impact on Colombian-US
Relations. Journal of Interamerican Studies and World Affairs 30.1 (1988): 133-
160.
Tokatlin, Juan G. Evaluacin de la Poltica Anti-narcticos de Colombia frente a
Estados Unidos. Narcotrfico en Colombia: Dimesiones polticas, econmicas,
jurdicas, e internacionales. Eds. Carlos G. Arrieta, Luis J. Oijuela, Eduardo
Sarmiento P., and Juan G. Tokatlin. Bogot, Colombia: Tercer Mundo Editores,
(1990): 277-367.
Tokatlin, Juan Gabriel. "Latin American Reaction to U.S. Policies on Drugs and
Terrorism." Security, Democracy, and Development in U.S.-Latin American
Relations. Eds. Lars Schoultz, William C. Smith, and Augusto Varas. New
Brunswick: Transaction, (1994): 115-132.
Tokatlin, Juan Gabriel. 1995. Drogas, Dilemmas y Dogmas: Estados Unidos y la
Narcocriminalidad en Colombia. Santaf de Bogot: Tercer Mundo Editores.
Toro, Carlos. Former drug trafficker. Frontline: Drug Wars. 2000.
Http://www.pbs.org/wgbh/pages/frontline/shows/drugs/interviews/toro.html
[Accessed 6 March 2001].
Torres, Edgar Arias. Los Mercaderes de la Muerte. Bogot: Intermedio Editores, 1995.
Torres, Edgr. Editor, El Tiempo. Bogot. 15 June 2000.
Torres, Edgar and Armando Sarmiento. Rehenes de la mafia: En las entraas del cartel.
Bogot: Intermedio Editores, 1998.


151
endeavors. Citing 19th Century studies on the French prison system, Michel Foucault
argues, The prison makes possible, even encourages, the organization of a milieu of
delinquents, loyal to one another, hierarchized, ready to aid and abet any future criminal
act (Foucault, 1977, p. 267). More recently, several prominent Colombian scholars
discuss the prison system in their country: It is no secret that the prisons... are schools of
crime in which they [inmates] perfect their methods, increase their dangerousness and
foster the phenomenon of recidivism (Leal Buitrago et al., 1999, p. 90, translated by
author. Also, see McAndrew, 2000, p. 53).
By housing large numbers of convicted drug felons in the same location where
they enjoy close physical proximity to one another, contemporary governments
encourage the formation of trafficking networks. Participants in the Colombian drug
trade have been known to take advantage of their prison time to acquire information and
connections to be exploited following their release from prison. In the early 1970s, a
marijuana smuggler by the name of Carlos Lehder was sentenced to several years in the
Danbury Correctional Institution, where he befriended another marijuana smuggler
named George Jung. This fortuitous pairing led to a productive exchange of knowledge
and experience as the two convicts planned future cocaine deals, poring over maps in the
prison library and designing smuggling routes from Colombia through the Caribbean
(Gugliotta & Leen, 1989, pp. 28-31 Los Angeles Times, 1988; Porter, 1993). As Jung
later recalled in a Frontline interview, his partner also gathered information from other
inmates:
... Carlos never ceased, never stopped. He was like a student is, constantly
pumping peoples brains about money laundering, about this, about that. About
automobiles, about airplanes, about boats. In fact there was a guy in there for
smuggling with boats and he spent hours and hours with him learning navigation,


217
Drug enforcement agencies change their programs and procedures in response to
other organizations, including the illicit trafficking enterprises they seek to dismantle.
The final two sections of this chapter examine how nares and narcos learn from each
other through repeated interactions.
Competitive Learning Games
Trafficking enterprises and drug enforcement agencies have diametrically
opposed aims. Trafficking organizations provide illegal drugs to markets. In so doing,
they seek to evade or otherwise undermine government efforts to stop them. Drug
enforcement agencies aim to identify and dismantle trafficking groups, thereby reducing
illicit drug flows and raising their wholesale and retail prices. Given their conflicting, but
interdependent, goals, these state and non-state organizations share common task
environments.
They also share competitive learning ecologies. In carrying out their respective
tasks they strive to learn from each other. To dismantle trafficking enterprises, drug
enforcement agencies gather, analyze, and act on tactical intelligence regarding
participants identities and the methods they use produce, transport, or distribute illicit
drugs. To evade or undermine drug enforcement efforts, trafficking groups gather,
analyze, and act on information about police agencies and the routines they use to entrap
participants and intercept their communications. Within these learning ecologies,
interactions between rivals can be described as competitive learning games.
The competitive learning game is a theoretical device to conceptualize
interactions between drug enforcement agencies and trafficking enterprises.14 These
14 The device draws on game theoretic concepts and language, sans the abstract mathematical apparatus.
However, the framework is not premised on pure theories of rationality. In these contests, players are only


155
Arreguin, 2000; Bagley, 1999; Francis, 2000; Homero, 2000; Lee, 1999; Morales, 2000;
Zabludoff, 1999).
Miguel Rodrguez Orejuela required his cell managers to maintain diligent
records of their trafficking activities, which were faxed back to the home office in Cali on
a regular basis. Rodrguez Orejuela also hired numerous accountants to maintain records
relating to different areas of his illicit operations, including: income and expenses derived
from narcotics trafficking, monetary payments and personal gifts to government officials
and politicians, phone lines subject to government surveillance, enterprise automobiles
authorized by Colombian law enforcement authorities, and contributions from
prominent traffickers to the organizations numerous political projects. Handwritten
records were stored in accounting ledgers, often with aliases and coded notations.
Electronic records were maintained on laptop computers, with backup copies stored on
computer diskettes. When government anti-drug efforts forced Miguel Rodriguez
Orejuela into hiding, he developed a filing system that allowed him to store important
documents in different brief cases in secret locations throughout Cali. Rodriguez
Orejuela numbered the briefcases and kept lists of the items contained in each one. This
information was stored on a laptop computer used by a personal secretary, who
accompanied his boss in hiding (Arreguin, 2000; Francis, 2000; Velsquez Romero,
2000; Pallomari, 1997, V. 35, 37-38; Reyes, 2000).
The organizational memories of Colombian narcotics enterprises are not restricted
to documenting transactional inputs and outputs. Numerous core organizations have
created manuals that detail rules and procedures regarding specific performance
programs. One Medellin enterprise produced a pamphlet outlining practices for


158
In the 1990s, several Cali enterprises used these procedures to obtain access to
indictments, search warrant affidavits, wiretap affidavits, criminal complaints, law
enforcement intelligence reports, and other government documents that described in
detail how U.S. authorities conducted their criminal investigations. U.S. based lawyers
representing enterprise participants used discovery procedures to obtain these documents
and send them to Colombia, where they were analyzed in meetings between leaders, their
lawyers and high-ranking participants. The purpose of these sensemaking sessions was
to determine how drug enforcers penetrated the organizations operations, identify
confidential informants, learn about the latest tactics in criminal investigations, and
devise strategies to avoid similar mistakes in future operations. These indictments and
affidavits were also made available to mid-level participants who were ordered to study
the materials in order to learn from and avoid the mistakes of their predecessors (Cash,
1999; DEA, 1993a, p. 21; Farah, 1997, 1999a; Fernandez, 2000; Francis, 2000;
Kacerosky, 1994, pp. 36, 53, 67-68; McGee & Duffy, 1997, pp. 65-66; Nieves, 1997, pp.
12-13; Olson, 1999, 2000, pp. 59-60; Smith, 2000a; Steve, 2000; Third Superceding
Indictment, [no date], pp. 36-38; Thompson, 2000).
The practice of sharing information, analyzing problems, and constructing inter-
subjective understandings is not unique to the Cali enterprises. Groups affiliated with the
Medellin organizations also engaged in sensemaking through meetings and
conversations. Moreover, there are indications that post-cartel enterprises also construct
collective interpretations of their task environments, often, but not inevitably, in response
to unforeseen events. Indeed, contemporary trafficking enterprises routinely analyze the
experiences of the Medellin and Cali core associations in order to avoid committing the


105
Other Organizations in the Task Environment
In addition to drug enforcement agencies, trafficking enterprises interact with a
host of independent suppliers, competitors and customers. Suppliers include drug
farmers and processing labs that produce semi-refined narcotics, import-export
companies that supply precursor chemicals to process narcotics, investors that finance
drug shipments, smuggling rings that transport drugs from South America to the U.S. and
other countries, financial service groups, banking institutions, and economic firms that
launder and repatriate drug profits, and law firms that provide legal services to
entrepreneurs and apprehended participants. Competitors include criminal organizations
based in Colombia, Mexico, the Dominican Republic, and other locations that contend
for market share and profits in wholesale drug markets. Other rivals in Colombia include
paramilitary organizations and guerrilla fronts that compete for resources and market
share at the processing and exportation stages of the industry. Customers include
wholesalers that purchase cocaine and heroin in multi-kilogram quantities, and
importation groups in Mexico, the Caribbean, and Canada that complete the international
transportation process.
The task environment confronting Colombian trafficking groups operate is as
hostile as anything in the organizational literature. Trafficking groups face exposure
from drug enforcement agencies, competitors, suppliers, even customers. Government
agencies destroy processing labs, interdict drug shipments, seize illicit profits, and
apprehend participants. Trafficking enterprises compete for market share and profits with
other criminal, guerrilla, and paramilitary organizations in Colombia and elsewhere.
Many of their non-state rivals operate outside the rule of law, making competition


232
is what we will be doing. We have seen them do that. For example, we used to
hit them a lot on the AMTRAK, the train, because what they used to do in New
York, they will take money, they will put in suitcases with wheels, and they will
have old people bringing out their suitcases out, and no one would look at them.
We actually popped some of these guys three and four times... and it took them
like a day to make this adjustment, because [someone] down south got the call
from an attorney, say in a particular state where we made one of the arrests, and
we shared the information we have a narcotics trafficker jabbed in the middle
because he just lost three or four million dollars in a day, and [the kingpin] said,
Look do not put your ass on a train. If anyone is on the train with my papers
[money] [they] will be killed. End of the story. They cannot jump on the train.
You cannot do that in traditional systems (Thompson, 2000).
Other U.S. and Colombian police officials interviewed also recognized that flatter
decision-making hierarchies allow trafficking groups to respond to environmental change
more quickly than drug enforcement bureaucracies (Evans, 1999; Gale & Rodriguez,
2000; Sanchez, 1999; Soto Velasco, 2000; Smith, 2000b; Thompson, 2000; also see
Statement of Witness, 1989, p. 30; Lupsha, 1996, p. 35).18
The Red Tape Trap
Trafficking enterprises function outside the rule of law, law enforcement agencies
within it. This distinction, eloquent in its simplicity, has significant repercussions for
drug enforcers. Like other government organizations, police agencies are sovereignty-
bound actors. They are constrained by legalistic and bureaucratic norms that their
sovereignty-free adversaries lack. Their actions and performance outputs face the
constant regulatory scrutiny of more powerful bureaucratic and political actors.19 When
18 Indeed, several law enforcement officials emphasized that their respective organizations were cognizant
of this dilemma and taking corrective steps, including allowing field agents greater decision-making
authority and forming special review boards to speed up information flows. Notwithstanding these
changes, several officials concede that their criminal competitors maintain a critical advantage on this point
(Pineda, 2000; Soto Velasco, 2000; Smith, 2000b).
19 During separate research trips to Colombia, I became friendly with a contract employee of the State
Departments Narcotics Assistance Section in the U.S. embassy in Bogot. This person spent several years
working within NAS, and he was highly regarded by the NAS director and several high-ranking CNP
officials that I interviewed. During the course of our numerous informal conversations, my friend
mentioned that Congressional and State Department review bodies were constantly inspecting NAS. He


52
computer navigation system and engine room. When completed the submarine would
have measured over 100 meters in length and had a transportation capacity of 200 tons,
far in excess of the one to three tons of cocaine transported by the mini-subs in the
Caribbean during the 1990s. Other recent innovations include using calling cards,
satellite phones, electronic mail, and encryption technology to communicate information,
mixing cocaine hydrochloride with charcoal and iron dust, welding cocaine into cargo
ship hulls, and using double-lined latex pellets for internal body carries (Arreguin, 2000;
Associated Press, 1998a; Branigan, 1999; Bohning, 2000; Brooks & Farah, 1998; CM&,
2000; CNN, 2000; DEA, 2000; El Tiempo, 2000d; Francis, 2000; Johnson, 2000;
Markowitz, 2000; Smith, 2000a).
Government Drug Enforcement Efforts During Phase Five
Colombian and U.S. drug enforcers continued to target the largest, most well-
known trafficking organizations in the late 1990s. In the post-cartel era, this meant the
medium-sized enterprises operating in the Northern Cauca Valley and along the Atlantic
Coast. Between 1997 and 2000, the leaders of several enterprises surrendered or were
captured by government authorities. In August 1997, CNP officials arrested Julio Cesar
Nasser David, one of the architects behind the Atlantic Coast trafficking network.14 The
following month, Orlando Henao Montoya, a leader in the North Valley network,
surrendered to Colombian officials in Bogot. In February 1998, CNP forces
apprehended Jos Nelson Urrego, another leader of the North Valley network, at his
country estate outside of Medellin. Four months later, Colombian authorities captured
14 Nasser Davids participation in the Colombian narcotics industry dates back to the 1970s, when his
organization smuggled several tons of marijuana to Southern Florida using mother ships and speedboats.
In the mid-1980s, Nasser Davids organization moved into the cocaine trade, transporting and distributing
multi-hundred kilogram shipments of cocaine to the U.S. through containerized cargo and general aviation
aircraft (Constantine, 1998; FBI, 1993, p. 31; Chepesiuk, 1999b, p. 153; Reuters, 1998c).


Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
OUTSMARTING THE STATE:
A COMPARATIVE CASE STUDY OF THE LEARNING CAPACITY
OF COLOMBIAN DRUG TRAFFICKING ORGANIZATIONS AND
GOVERNMENT DRUG ENFORCEMENT AGENCIES
By
Michael C. Kenney
May 2002
Chairpersons: Philip J. Williams, Terry L. McCoy
Major Department: Political Science
Over the past several decades, Colombian drug trafficking enterprises have
transported large quantities of cocaine, heroin, and other psychoactive drugs to the United
States in spite of government efforts to disrupt this illicit commerce. Contrary to much
existing scholarship, this study argues that the persistence of the Colombian drug
dilemma is, in part, due to the ability of criminal enterprises to alter their organizational
structures and behavior in response to information and experience, store this knowledge
in practices, procedures and performance programs, and select and retain innovations that
produce satisfactory outcomes. Drawing on a multidisciplinary body of literature on
organizational learning, and interviews with U.S. and Colombian officials, researchers,
and former drug traffickers, the study demonstrates that smuggling organizations learn,
and in the process become increasingly difficult for drug enforcement agencies to identify
xiii
and dismantle.


CHAPTER 6
ORGANIZATIONAL LEARNING BY
DRUG ENFORCEMENT AGENCIES
Similar to the criminal organizations they seek to disrupt, drug enforcement
agencies, represented by such bureaucracies as the U.S. Drug Enforcement
Administration (DEA) and the Colombian National Police (CNP), acquire, analyze, and
apply knowledge and experience to practices, programs, and policies that guide collective
behavior. Over the past several decades, drug enforcers from Colombia and the U.S.
have implemented a variety of administrative and operational reforms in response to
feedback from state institutions and changes in the transnational drug trade.
This chapter examines organizational learning as experienced by Colombian and
U.S. drug enforcement agencies. The chapter begins with a summary of the results of
dissertation interviews that discuss the topic of state learning. I follow this with a
discussion of the intelligence activities of U.S. and Colombian drug enforcement
agencies, including the practices, procedures, and programs they use to gather, interpret,
and apply information. The discussion illustrates how these practices and programs have
changed over time, allowing U.S. and Colombian police officials to improve their
understanding of the illicit drug industry. Next, I analyze numerous adaptations in law
enforcement operations over the past several decades, with specific reference to the DEA
and CNP, the preeminent drug enforcement bureaus in their respective governments.
After demonstrating that numerous innovations by drug enforcers have allowed them to
better target major trafficking organizations, I present the concept of competitive learning
184


231
quickly communicated throughout the enterprise. With quicker information flows and
decision cycles, trafficking groups can also change their operational and procedural
routines more quickly than law enforcement organizations.
DEA Administrator
I
DEA Deputy Administrator
I
Figure 6-3 Example of DEA Management Levels
One DEA official with extensive experience investigating the Cali core
organizations captures the advantage of quicker decision-making for trafficking
enterprises:
With the decision process in the [trafficking] company, they can make this
decision instantaneously and it will be translated in a day by everyone in the chain
of command... It get transmitted immediately and some of the kingpins will
literary jump into the equation... The kingpin can jump into the middle of it,
immediately, and say, Look guys, for the sake of business, for our own sake, this


143
Acquiring Knowledge and Experience
Colombian trafficking organizations obtain information through trial and error
experience and search. Sources of knowledge include their own participants, outside
consultants, other trafficking enterprises, even government officials. Relevant
information includes know-how, techniques and practices regarding the production,
transportation and distribution of narcotics, money laundering, and ancillary activities.
Trafficking enterprises seek information to resolve problematic situations.
Following an unforeseen event, such as the interdiction of a drug shipment, participants
are assigned to gather information about the mismatch from press clippings, public
documents, and other participants that experienced first-hand knowledge of the event.
Organization leaders and managers use the information to determine how the mismatch
occurred and assess culpability among participants. However, a mismatch is not
necessary to stimulate search. Trafficking organizations frequently seek knowledge
regarding technological innovations they can use to stay one step ahead of law
enforcement (Francis, 2000; Mermelstein, 1990, p. 118; Olson, 2000, p. 59; Rice, 1989,
p. 148).
Colombian trafficking enterprises have developed a variety of practices for
gathering information about their trade and socializing members to organizational
practices and procedures. Participants are sent on fact-finding missions to learn about
practices and procedures used in other parts of the enterprise or the narcotics industry in
general. As he began his formal association with a core enterprise, Emil was sent to
Miami to acquire information regarding cell management and maritime shipments.


5
figures are available, the estimate increased to 13,572 hectares. The estimated production
of opium gum, which is used to make heroin, increased slightly from 65 metric tons in
1995 to 66 tons in 1997. Over the past several years, the number of hectares in Colombia
devoted to marijuana cultivation and the estimated production of marijuana itself have
stabilized at approximately 5,000 hectares and 4,150 metric tons respectively (BINLEA,
1998, 2001).
While the trafficking of illicit narcotics is more difficult to estimate than its
production, there are indications that the supply of the Colombian-produced cocaine to
the U.S. has stabilized in recent years. For example, the U.S. Drug Enforcement
Administration (DEA) estimates that the price and availability of cocaine in the U.S. have
remained relatively constant. For the past several years, wholesale (per kilogram) prices
for cocaine have remained within a $10,500-36,000 price band, while cocaine purity at
the retail level (per gram) has stabilized at approximately 60% (DEA, various years).
Meanwhile, the flow of Colombian-produced heroin to the U.S. appears to be rising. The
DEA reports that the nationwide average purity for heroin at the retail level has increased
from 27% in 1991 to 36% in 1996. Spearheading this trend is South American heroin
most of which originates in Colombiawhich has a retail purity of 50.3% (DEA 1997).
In addition, the amount of Colombian-produced heroin seized by U.S. and Colombian
anti-drug authorities has also increased in recent years (BINLEA, 1998).
These developments have led the State Department to observe that in spite of
improved anti-drug efforts on the part of the Colombian government, the country remains
the worlds leading producer and distributor of cocaine and an important supplier of
heroin to the U.S. (BINLEA, 1998).


313
Paret, Peter. Makers of Modern Strategy: From Machiavelli to the Nuclear Age.
Princeton: Princeton University Press, 1986.
Parra, Bernardo. Professor of Economics, National University of Colombia. Bogot. 9
April 2000.
Passas, Monica [pseud]. Head, Intelligence Group, Drug Enforcement Administration,
Colombia. Bogot. 9 March 2000.
Passic, Greg. Former Special Agent, Drug Enforcement Administration. Frontline: Drug
Wars. 2000.
Http://www.pbs.org/webh/pages/frontline/shows/druas/interviews/passic.html
[Accessed 17 April 2001].
Pear, Robert and Philip Shenon. Customs Switches Priority from Drugs to Terrorism.
New York Times (10 October 2001). Http://web.lexis-nexis.com/universe
[Accessed 10 January 2002],
Pcaut, Daniel. Presente, pasado y futuro de la violencia. Anlisis Poltico (30) (1997):
3-36.
Peate, Andrs. El sendero estratgico del ELN: del idealismo guevarista al clientelismo
armado, Reconcer la Guerra para Construir la Paz. Eds. Malcolm Deas and
Mara Victoria Llrente. Bogot: CEREC, (1999): 53-98.
Prez Valds, Jos Gregorio. Correspondent, CM&. Bogot. 5 June 2000a.
Prez Valds, Jos Gregorio. Correspondent, CM&. Bogot. 16 July 2000b.
Perl, Rafael. F. US Congress, International Drug Policy, and the Anti-Drug Abuse Act
of 1988. Journal of Interamerican Studies and World Affairs 30.2-3 (1988): 19-
51.
Perl, Raphael F. United States Andean Drug Policy: Background and Issues for
Decisionmakers. Journal of Interamerican Studies and World Affairs 34.3
(1992): 13-35.
Perl, Raphael F. U.S.-Andean Drug Policy. Drug Trafficking in the Americas. Eds.
Bruce M. Bagley and William O. Walker III. New Brunswick: Transaction,
(1994): 23-41.
Perl, Rafael F. Senior Analyst, Congressional Research Service. Washington. 22 July
1999.


132
Commanders, Beechcraft King Air 300s, and Cessna 400s. Pilots often modified their
planes with technology that allowed them to fly longer and monitor government anti-drug
surveillance. Rubber fuel bladders allowed for flights from Colombia to travel as far
North as Pennsylvania without refueling. Radar detectors, radio scanners and scramblers,
and night vision goggles provided the means to monitor government tracking planes.
Some transportation rings maintained listening posts in the U.S. complete with CB, VHF,
single side band, and marine radios to maintain communication with ongoing flights and
Bearcat scanners and police radios to keep an eye on law enforcement activity in the
landing area. In the 1990s, several core enterprises acquired large cargo aircraft, such as
Convair 580s and Boeing 727s, to transport multi-ton cocaine shipments into Mexico and
Canada (DEA, 1993b, p. 17; Pallomari, 1997).
Colombian trafficking enterprises use a variety of seafaring vessels to transport
cocaine, heroin and marijuana through the Caribbean, including go-fast motor boats,
sport fishermans crafts, yachts, commercial fishing boats, and freighters. Smaller
vessels are frequently modified with secret hiding compartments, extra gasoline tanks,
and large outboard engines. In the early 1990s, a number of enterprises began using low
profile vessels to transport cocaine between Colombia and Puerto Rico. These vessels
are colored dark gray to blend in with the sea and ride only one to two feet above sea
water, making detection by radar extremely difficult. They measure forty-feet in length,
and can transport up to one metric ton of cocaine at a time (DEA, 1993a, p. 16; Dye,
1998, pp. 261-262). Around the same time, DEA intelligence officials reported that some
core organizations were experimenting with semi-submersible vessels used to transport


147
frequencies used by the DEA, FBI and other enforcement agencies from radio shops and
public documents. Some enterprises even purchased flight schedules for U.S. Customs
and Coast Guard planes from corrupt U.S. officials (Dye, 1998, p. 239; Mullen, 1983, pp.
25, 27; Rice 1989, p. 83; Senz Rovner, 2000; Statement of Witness, 1989, p. 10).
In Colombia, core organizations developed sophisticated intelligence networks
with the help of active collaborators from a variety of public and private institutions,
including numerous law enforcement agencies, national and regional legislatures,
prosecution offices, telephone companies, newspapers, taxi cab companies, law firms,
hotels, and other service industries. Through these informants, trafficking enterprises
acquire valuable information regarding government drug enforcement policies and
programs, as well as impending legislation and criminal proceedings.
Intelligence Capabilities of Rodrguez Orejuela Organization
The intelligence gathering capabilities of the Rodrguez Orejuela enterprise were
particularly impressive. According to the organizations chief accountant, the head of the
Regional Prosecutors Office in Cali provided Miguel Rodrguez Orejuela with
confidential government reports, including drug enforcement information sharing
agreements between the U.S. and Colombian governments. The organization also
allegedly bribed dozens of telephone company officials in order to gather information
from unsuspecting government officials, including several at the U.S. Embassy in
Bogot. Senator John Kerry, who conducted several numerous Congressional hearings
on the international drug trade while serving on the Senate Foreign Relations Committee,
claims that the Rodrguez Orejuela enterprise was even monitoring conversations
between the U.S. Drug Enforcement Administration chief in Colombia and senior Justice


173
modifications, to transport marijuana. Marimberos subsequently determined that their
transportation and distribution routines could also be applied to cocaine smuggling.
Cocaleros discovered they many of their cocaine distribution and marketing techniques
worked well for heroin (Homero, 2000).
Changing Organizational Structures: Smaller is Better
From 1989 to 1996, the Colombian government, in close cooperation with the
Drug Enforcement Agency and other U.S. enforcement organizations, dismantled a
number of trafficking enterprise, including several of the leading core organizations from
Medellin and Cali. Since then the CNP and DEA have continued to identify and
dismantle prominent enterprises from the Northern Cauca Valley and Atlantic Coast
areas (see Chapter 2). During this period, many trafficking enterprises were selected out
of the drug industry by successful law enforcement. Many others, however, adapted to
the growing hostility of their task environments by revamping structures and operations.
Surviving remnants of the core organizations downsized and decentralized, creating
hundreds of independent firms that specialize in specific functions. Newly emergent
groups shared these features, adding to the industrys small is beautiful incarnation.
New and reformed enterprises were not only smaller than their predecessors, but flatter,
often comprising no more than two layers of management and a dozen participants.
They also sought to maximize operational secrecy, changing smuggling practices that
had been identified by drug enforcers, and eschewing violence in their business dealings.
Different groups, including former competitors, pooled their resources and cooperated
through transnational networks, often organized on an ad hoc basis. Numerous post
cartel enterprises removed themselves from U.S. drug markets, preferring to let other


50
Colombia, conducting their activities on a smaller, more discrete scale. Some
organizations, particularly several based in the Northern Cauca Valley and Atlantic Coast
areas, are long-standing enterprises that escaped the drug enforcement net as the CNP and
DEA pursued more prominent core organizations.13 In addition to these medium-sized
survivors, hundreds of small trafficking operations have emerged in recent years. There
is variation in size among existing enterprises, with medium-sized organizations ranging
from twenty to one hundred participants, and smaller groups containing ten to twenty
associates. Contemporary enterprises tend to be flatter than the core organizations, and
more circumspect in carrying out their activities. Small groups are frequently led by a
single figure that exerts substantial decision-making authority, while co-equal
subordinates perform the work. Medium-sized organizations often contain different
leaders that occupy discrete, functionally-specific managerial positions, similar to the
core enterprises (Aero, 2000; DEA, 2000; FBI, 1993; Gallego Castrilln, 2000; Nicosa,
2000; Semana, 2000a).
Instead of coordinating several phases of trafficking activities like the core
enterprises, many trafficking operations today specialize in single-phases of drug
production, processing or transportation. While different groups, including former
12 It is not clear how many trafficking enterprises actually exist in Colombia, which is reflected in the wide
range of expert estimates. General Serrano, the former head of the Colombian National Police, estimates
that are between 350 and 400 minor trafficking organizations operating in Colombia. Alejandro Reyes
Posada, a leading Colombian drug researcher, is even more precise. According to Reyes Posada there are
438 narcotics organizations in Colombia, 256 of which are re-structured off-shoots of the former cartels,
while 182 represent new groups. A recent report in Semana magazine estimates there are between 80 and
100 groups in Colombia that specialize solely in the exportation of cocaine or heroin (Reyes Posada, 1999,
p. 2; Semana, 2000a, p. 71; Serrano Cadena, 1999, p. 237). My own research suggests that the clandestine
nature of drug trafficker merits greater discretion in formulating estimates. While DEA and CNP
intelligence officials confirm that there are hundreds of trafficking enterprises in Colombia, they also
stress the difficulty of providing exact estimates. The highest-ranking DEA official in Colombia told me
that while he believes there are several hundred organizations operating in Colombia he does not know the
exact number (Arreguin, 2000).
13 Indeed, a number of these groups provided transportation services to the Medellin and Cali core
organizations during the 1980s and 1990s.


229
Table 6-5 Authorized DEA Personnel, FY 2000
Personnel Type
Number
Special Agents
4,561
Diversion Investigators
523
Intelligence Specialists
686
Chemists
259
Professional/Administrative
1,323
Technical/Clerical
1,780
Total
9,132
Source: DEA, [no date] DEA Staffing & Budget
Officials within these drug enforcement agencies are organized into numerous
layers of management. The top manager in the DEA is the Administrator, who oversees
the entire agency, but, as noted earlier, reports to the FBI Director. Beneath the DEA
Administrator is the Deputy Administrator. Below this office are a number of divisions,
including the Operations Division, the Intelligence Division, and the Financial
Management Division. An Assistant Administrator who oversees a number of offices
leads each division. Each office is further broken down into sections with different areas
of functional responsibility. The top management level in the Colombian National Police
is the General Direction office, led by the Director of the agency. The CNP Director
oversees the General Sub-direction office and several operational directorates, including
Intelligence, Anti-Narcotics, and the Judicial Police. Each directorate is further broken
down into a variety of offices and enforcement units (DEA, [no date] DEA
Organizational Chart; Llrente 1999, p. 414).
These bureaucratic structures produce numerous management layers as each
section and unit and is led by a different official. Within the DEA, a drug enforcement
agent that works in the Cocaine Investigative Section (see Figure 6-1) reports to the head
of this section, who reports to the head of the Investigative Intelligence Section, who
reports to the head of the Office of Intelligence Investigation, who reports to the Assistant


193
Figure 6-1 Organization of DEA Intelligence Division, circa 1996
Source: Adapted from Richelson (1999).
Africa that evaluate drug production, transportation systems, trafficking groups, and drug
interdiction effectiveness in their respective areas. The Office of Intelligence Research


280
Relations Committee, Subcomittee on the Western Hemisphere, Peace Corps,
Narcotics, and Terrorism. (26 February 1998).
Http://www.usdoi.gov/dea/pubs/cngrtest/ct980226.htmv [Accessed 23 October
2000],
Constantine, Thomas A. Prepared Statement of Thomas A. Constantine, Administrator,
Drug Enforcement Administration. Fiscal Budget 1999. Hearing before the
Government Reform and Oversight Committee, U.S. House of Representatives
(19 March 1998). Http://www.usdoi.gov/dea/pubs/cngrtest/ct980319.htm#DEA'S
FY 1999 BUDGET REQUEST [Accessed 4 July 2001],
Contreras, Joseph & Steven Ambrus. Fighting the New Drug Lords. Newsweek,
international edition (21 February 2000): 8-12.
Cook, Scott D. N. and Dvora Yanow. Culture and Organizational Learning,
Organizational Learning. Eds. Michael D. Cohen and Lee S. Sproull. Thousand
Oaks: Sage, (1996): 175-187.
Corts, Fernando. Rodrguez Gacha "ElMexicano. Bogot: Intermedio, 1993.
Craig, Richard B. Colombian Narcotics and United States-Colombian Relations.
Journal of Interamerican Studies and World Affairs 23.3 (1981): 243-270.
Craig, Richard B. Journal of Interamerican Studies and World Affairs 25.3 (1983): 325-
350.
Craig, Richard B. Illicit Drug Traffic: Implications for South American Countries.
Journal of Interamerican Studies and World Affairs 29.2 (1987): 1-34.
Crank, Jerome [pseud]. Special Agent, Drug Enforcement Administration, Colombia.
Bogot. 15 March 2000.
Cressey, Donald R. Methodological Problems in the Study of Organized Crime as a
Social Problem. Annals of the American Academy of Political and Social Science
374 (November) (1967): 101-112.
Cressey, Donald R. Theft of the Nation: The Structure and Operations of Organized
Crime in America. New York: Harper & Row, 1969.
Criminal Justice Newsletter. September 11 Terrorism Brings Changes in Law
Enforcement. 31.18 (2001a): 1-3.
Criminal Justice Newsletter. Ashcroft Announces Plan for DOJ Wartime
Reorganization. 31.20 (2001b): 1-2.


325
Torres Velasco, Javier. La ciudadana pacta con su polica: El proceso de modernizacin
de la polica nacional de Colombia. Orden mundial y seguridad: Nuevos desafos
para Colombia y Amrica Latina. Eds. Francisco Leal Buitrago and Juan Gabriel
Tokatlian. Bogot: Tercer Mundo Editores, 1994.
Treaster, Joseph B. A Top Medellin Drug Trafficker Dies in a Shootout in Colombia
New York Times (16 December 1989). Http://web.lexis-nexis.com/universe
[Accessed 12 January 2000].
Tulls, Lamond. Bolivian and Colombian Policy Options on Illicit Drugs: Constraints
and Opportunities Imposed by International Drug-Control Regimes. Unpublished
paper presented at the Latin American Studies Association conference.
Washington (September 1995).
United Nations. Economic and Social Council. Problems and Dangers Posed by
Organized Transnational Crime in Various Regions of the World: Background
Document. Presented at the World Ministerial Conference on Organized
Transnational Crime. Naples, Italy (21-23 November 1994).
United Press International. 30 Indicted by U.S. on Cocaine Counts. New York Times
(18 June 1985): A-19. Http://web.lexis-nexis.com/universe [Accessed 22
February 1999],
U.S. Attorneys Office, Southern District of Florida. Arrest ofFabio Ochoa. Press
Release (9 September 2001a). Http:/Avww.usdoi.gov/usao/fls/Ochoa.html
[Accessed 10 December 2001].
U.S. Attorneys Office, Southern District of Florida. Extradition of Alejandro Bemal
Madrigal. Press Release (31 October 2001b).
Http://www.usdoi.gov/usao/fls/Bernal.html [Accessed 10 December 2001].
U.S. Attorneys Office, Southern District of Florida. Guy A. Lewis. (No date).
Http://www.usdoi.gov/usao/fls/GuyLewis2.html [Accessed 14 March 2002],
Useche, Mariano. De las rutas del contrabando a las rutas del narcotrfico. Coloquio:
Revista de la Direccin Nacional de Estupefacientes 5 (5) (1997): 38-43.
U.S. House of Representatives. South American Study Mission, August 9-23, 1977.
Report of the Select Committee on Narcotics Abuse and Control. Ninety-Fifth
Congress, first session (November 1977). Washington: U.S. Government Printing
Office.
U.S. House of Representatives. Cocaine and Marijuana Trafficking in Southeastern
United States. Hearings before the Select Committee on Narcotics Abuse and


177
recognized the negative impact of this risk reduction strategy on information flows, none
suggested that it prevented these criminal enterprises from adapting to external pressures.
Compartmentalization does not stop communication within trafficking enterprises but
channels it according to the decision-making hierarchies of the organization.
Compartmentalization restricts communication across functional sub-units, such as
between transportation and distribution cells, not between subordinates and supervisors.
Managers have access to the knowledge and experience of their underlings. The more
decision-making authority an individual wields within the enterprise, the greater his or
her access to the organizations accumulated knowledge base. This suggests that leaders
and high-level managers have the information they need to make effective decisions.
Moreover, informal mechanisms, such as social gatherings, provide opportunities for cell
workers to share and interpret information beyond their own immediate experience
(Fernandez, 2000; Gale & Rodriguez, 2000; Molano, 2000; Olson, 1999; Perl, 1999;
Thompson, 2000; Torres, 2000).
Organizational Memories and Participant Turnover
Many trafficking enterprises find creating and storing records pertaining to their
illicit activities to be a necessary, if perilous, endeavor. To keep track of the myriad
details involved in their transactions, participants document various aspects of their
operations in ledgers, notebooks, and computer files. However, when information is
stored in records it becomes vulnerable to capture and exploitation by regulators,
competitors, and turncoats. Leaders are well aware of the double-edged nature of record
keeping, and seek to minimize their data storage activities, relying instead on the human


204
Controls in order to accommodate the realities of street-level enforcement.
Promotion patterns changed so as to reward agents who participated in complex
investigations that led to the prosecution of major drug conspiracies. At the local
level, interagency committees were established to channel funds into promising
investigations aimed at high-level dealers; the availability of the money enhanced
agency cooperation in sharing information. The FBIs habits, experience, and
technical equipment made it easier and more rewarding to use wiretaps for
acquiring evidence. At the same time, the DEAs pool of drug informants
remained a key resource that the FBI could not duplicate (Wilson, 1989, p. 267).
According to the former director of the DEAs Office of International Operations,
the agency also emerged from the semi-merger better positioned to respond to the growth
of the Colombian cocaine cartels. The drug-specific desk orientation mandated by the
first post-reform Administrator produced a Cocaine Investigations Section that later
became an effective mechanism for coordinating the DEAs numerous investigations of
the core enterprises based in Medellin and Cali (Nieves, 1997, p. 6).
The extent to which the positive results of the 1982 reform were due to policy
makers ability to accurately diagnose the bureaucratic ills of federal drug enforcement is
unclear. Wilson, for one, maintains that the success of the reorganization was more the
result of a happy accident than proper diagnosis (Wilson, 1995, p. 30). Whether the
fruits of policy makers labors were intentional or accidental, it is apparent that power
considerations and learning played a role in this structural adaptation. Initially, changes
in the DEA were due to the imposed preferences of more powerful political actors,
including executive branch policy makers and FBI administrators. However, following a
period of aversion to an unwanted reforms, DEA managers and field agents gradually
adapted to the new rules and procedures, drawing on the expertise of their fellow FBI
agents and adjusting their daily enforcement activities accordingly. As this case
indicates, even when organizational change is due to the imposition of a new set of


120
of code words and aliases to communicate details of complex transactions produces
inevitable misunderstandings among participants. However, many procedures fail not
due to poor design, but rather because the participants that execute them are ill-prepared,
inept, and dishonest. As we shall see in the next section, participants represent a source
of great uncertainty for trafficking enterprises and it is not always possible for leaders and
managers to control their behavior (Emil, 2000; Mastrofski & Potter, 1987; Nstor,
2000).
Participants in Colombian Trafficking Enterprises
Trafficking enterprises pursue their tasks through people. Participants process
drugs, weigh drugs, collect drugs, package drugs, hide drugs, transport drugs, unload
drugs, count drugs, store drugs, distribute drugs, gather drug proceeds, launder drug
proceeds, and perform a host of other drug-related tasks. Participants make decisions,
plan strategy, devise tactics, allocate resources, and communicate with other members.
Due to the illicit nature of their activities and the corresponding need for secrecy, most
Colombian trafficking organizations do not formalize the work-related responsibilities
and rewards of their members in documents. On occasion, participants may contribute to
a criminal conspiracy without recognizing themselves as members of a specific group or
knowing that they are performing an illegal act.14 In such circumstances, the boundaries
of the criminal enterprise become blurred and the identification of participants
problematic.
14 There have been numerous instances over the years where unsuspecting individuals have unknowingly
acted as couriers in the drug trade, transporting small quantities of drugs from Colombia to consumer
countries. These mules are deceived through a variety of mses into carrying luggage or other items, such
as local handcrafts, that contain small quantities of cocaine or heroin. When apprehended by law
enforcement authorities, they are frequently arrested, tried, and found guilty of participating in trafficking
conspiracies. See Molano (1997) and Sabbag (1990) for examples. I am not arguing that all mules fall into
this category. Indeed, interview and secondary source data gathered in this research suggest that many
Colombian mules are aware of the psychoactive nature of their contraband.


145
When knowledge is not available in-house, traffickers draw on the expertise of
outside consultants that, as Peter Lupsha notes, beat a path to their door with new ideas,
technologies, techniques, and investment opportunities (Lupsha, 1996, p. 34).
According to a DEA intelligence analyst, the idea for semi-submersible vessels that glide
along the surface of the Caribbean Sea (see Chapter 4) came from independent folks
hoping to profit from this lucrative commerce by selling their innovation to established
smuggling groups. Russian and U.S. engineers and naval experts were reportedly sub
contracted by Northern Cauca Valley enterprises to construct the submarine discovered
by Colombian authorities in September 2000. When a number of cocaine enterprises
decided to diversify into the heroin market in the 1980s, they hired refining specialists
from Mexico and the Far East to acquire information about processing opium latex into
heroin. More recently, at least one Colombian enterprise hired trained chemists to teach
their participants to process potassium permanganate, an essential precursor chemical for
refining cocaine base (Cambio, 2000, p. 36; Evans, 1999; Farah, 1997; Griffith, 2000;
Restrepo, 2000; Semana, 2000b; Serrano Cadena, 1999, pp. 50-51).
Counter-Surveillance Activities
In reflection of the constant threat they face from drug enforcers, Colombian
trafficking organizations direct many of their information gathering activities towards
government counter-narcotics efforts. Trafficking enterprises gather intelligence about
U.S. and Colombian law enforcement policies and programs from a variety of human,
electronic, and documentary sources, including government officials and records, news
reports, advertisements in magazines, and paid informants. In the 1980s, numerous
organized crime groups regularly exploited the Freedom of Information Act (FOLA) to


46
ally an emergency assistance package of US$ 90 million, and the Drug Enforcement
Administration undertook Operation Bolivar, specifically targeting the Medellin
organizations.
The fourth crackdown proved more effective than previous efforts because lasted
over a year, as opposed to six weeks, and possessed a more coherent strategy that
included extradition. The relentless pressure exerted by Colombian and U.S. authorities
forced the leaders of the Medellin organizations into hiding, reducing their ability to
coordinate large-scale cocaine shipments. Several leading traffickers were captured or
killed during police raids. Moreover, the crackdown produced ripple effects throughout
the Colombian drug industry. The inability of processing groups associated with the
Medellin network to buy the precursor chemicals needed for cocaine processing on a
regular basis reduced the price for Peruvian and Bolivian coca leaf to less than half of
production costs. The scarcity of cocaine supply affected consumer markets, causing
wholesale and retail cocaine prices in the United States to rise between fifty and one
hundred percent. However, this price spike proved short-lived, lasting approximately
eighteen months (Bagley, 1989-1990, p. 155; Caulkins & Reuter, 1998, p. 603; Farah
1990a, 1990b; Gugliotta 1992, p. 124; Krauss, 1995, p. A12; Nieves, 1997, p. 11;
Velsquez, 1993; Whynes, 1992, p. 333).
Nevertheless, by the early 1990s many of the Medellin core organizations had
been severely damaged by Colombian and U.S. anti-drug efforts. Prompted by effective
police pressure and a decree by a new Colombian president, several leading traffickers,
most notably the three Ochoa brothers, turned themselves in to the authorities. In June
1991, when it became clear that the new Colombian constitution would ban extradition of


114
Compartmentalization
The need to avoid drug enforcers and illicit competitors has led many leaders to
compartmentalize their criminal operations into discrete cells. Cells are designed to
protect the enterprise from penetration by law enforcement authorities and other
adversaries, and to limit the damage of infiltration when it does occur. These units are
generally quite small, composed of more than two but fewer than ten participants. The
number of cells in an organization is a function of the size and frequency of transactions
conducted by the enterprise. Core organizations that coordinate numerous multi-ton drug
shipments per year and maintain wholesale distribution networks in consumer markets
contain cells in different parts of the U.S., Europe, Mexico, and Colombia.
Cells operate as quasi-autonomous units. They perform their activities isolated
from other parts of the transnational enterprise. Cells communicate with leaders based in
Colombia, but limit their interaction with other cells. Cell workers may be aware only of
the other members of their unit, and know relatively little about the rest of the enterprise.
Even within cells, information is restricted on a need-to-know basis. Cell workers,
assistant managers, and managers possess the minimal information they need to carry out
their duties. This allows leaders to minimize the damage to their enterprise if workers or
even managers are apprehended by law enforcement and turn states evidence.
In highly compartmentalized operations, only the highest-ranking figures possess
comprehensive knowledge of the structures and functions of the entire enterprise. These
leaders are often located at the home office in Colombia, reducing (but not eliminating)
their exposure to law enforcement. Moreover, in trafficking enterprises knowledge is
power. Leaders understand that their authority stems, in part, from greater knowledge of


326
Control. Ninety-fifth Congress, second session (9-10 June 1978). Washington:
U.S. Government Printing Office.
U.S. House of Representatives. Drug Enforcement Administration Reauthorization for
Fiscal Year 1986. Hearing before the Subcommittee on Crime of the Committee
on the Judiciary. Ninety-ninth Congress, first session (1 May 1985). Washington:
U.S. Government Printing Office.
U.S. House of Representatives. Recent Developments in Colombia. Hearings before the
Committee on Foreign Affairs. One Hundredth Congress, second session (27
January 1988). Washington: U.S. Government Printing Office.
U.S. House of Representatives. Reauthorization of the Drug Enforcement Administration
for Fiscal Year 1988. Hearing before the Subcommittee on Crime of the
Committee on the Judiciary. One Hundredth Congress, first session (2 April
1987). Washington: U.S. Government Printing Office.
U.S. House of Representatives. Overall U.S. Counternarcotics Policy toward Colombia.
Hearing before the Committee on International Relations. One Hundredth Fourth
Congress, second session (11 September 1996). Washington: U.S. Government
Printing Office.
U.S. House of Representatives. International Drug Control Policy: Colombia. Hearing
before the Subcommittee on National Security, International Affairs, and
Criminal Justice of the Committee on Government Reform and Oversight. One
Hundred Fifth Congress, first session (9 July 1997). Washington: U.S.
Government Printing Office.
U.S. House of Representatives. Drug Enforcement Administration. Hearing before the
Subcommittee on Crime of the Committee on the Judiciary. One Hundred Sixth
Congress, first session (29 July 1999). Washington: U.S. Government Printing
Office.
U.S. House of Representatives. The Narcotics Threat from Colombia. Hearing before the
Subcommittee on Criminal Justice, Drug Policy, and Human Resources of the
Committee on Government Reform. One Hundred Sixth Congress, first session (6
August 1999). Washington: U.S. Government Printing Office.
U.S. Senate. DEA Oversight and Authorization. Hearing before the Subcommittee on
Security and Terrorism of the Committee in the Judiciary. Ninety-eighth
Congress, first session (23 February 1983). Washington: U.S. Government
Printing Office.
U.S. Senate. DEA Oversight and Budget Authorization. Hearing before the Subcommittee
on Security and Terrorism of the Committee on the Judiciary. Ninety-eighth


292
General Accounting Office (GAO). Drug Control: An Overview ofU.S. Counterdrug
Intelligence Activities. NSIAD-98-142 (25 June 1998). Http:/Avww.gao.gov/
[Accessed 5 July 2001].
General Accounting Office (GAO). Drug Control: DEA's Strategies and Operations in
the 1990s. GGD-99-108 (21 July 1999). Http://www.gao.gov/ [Accessed 11 July
2001].
General Accounting Office (GAO). Drug Control: U.S. Efforts in Latin America and the
Caribbean. NSIAD-00-90R (18 February 2000). Http://www.gao.gov/ [Accessed
5 July 2001],
Geopolitical Drug Dispatch. Collapse of Coca Prices. (48) (October 1995).
Http://www.ogd.org/gb/48APECTA.html [Accessed 10 January 2000].
George, Alexander L. Case Studies and Theory Development: The Method of
Structured, Focused Comparisons. Diplomacy: New Approaches in History,
Theory, and Policy. Ed. Paul Gordon Lauren. New York: Free Press, (1979): 43-
68.
George, Alexander L. and Timothy J. McKeown. Case Studies and Theories of
Organizational Decision-Making. Advances in Information Processing in
Organizations 2. Ed. Robert F. Coulam and Richard A. Smith. Greenwich: JAI
Press, (1985): 21-58.
Giraldo, Juan Carlos. El retorno de Pablo Escobar. Bogot: Editorial Oveja Negra, 1992.
Glionna, John M. Most drug experts agree with the scenarios Mills describes. San
Diego Union-Tribune (8 October 1986). Http://web.lexis-nexis.com/universe
[Accessed 10 January 1999].
Godson, Roy and William J. Olson. International Organized Crime. Society 32.2
(1995): 18-29.
Golant, Benjamin D. The State, Culture and Organisational Learning in Asia: The
Comparative Case of South Korea and India. Cambridge Review of International
Affairs 11.2 (1998): 238-253.
Goldgeier, James M. Leadership Style and Soviet Foreign Policy: Stalin, Krushchev,
Brezhnev, Gorbachev. Baltimore: John Hopkins University Press, 1994.
Goldstein, Judith and Robert O. Keohane. Eds. Ideas & Foreign Policy: Beliefs,
Institutions, and Political Change. Ithaca: Cornell University Press, 1993.
Gmez, Hernando Jos. La Economa Ilegal en Colombia: Tamao, Evolucin,
Caractersticas e Impacto Econmico. Economa y Poltica del Narcotrfico.


130
Technologies in Colombian Trafficking Enterprises
Colombian trafficking enterprises draw on material and symbolic resources to
perform work. The core technologies of trafficking organizations vary according to the
product or service provided, as well as the amount and frequency of their transactions.
An enterprise that processes small quantities of coca paste into cocaine hydrochloride
requires substantially fewer material and symbolic inputs than an organization that
produces, transports, and distributes several tons of cocaine each month. Yet, even the
smallest operation will contain a core technology composed of the inputs necessary to
perform work. By way of illustrating the technologies involved in these illicit activities,
the following paragraphs discuss the cocaine processing, communications, and
transportations practices of Colombian trafficking enterprises.
Enterprises that specialize in processing cocaine hydrochloride draw on a variety
of technological inputs. A laboratory or other physical installation provides the site to
process the illicit commodity. Chemical solvents, such as potassium permanganate,
ether, acetone, and hydrochloric acid, extract alkaloids and remove impurities from coca
paste and cocaine base. Water dilutes various solvents. Metal recipients, including fifty-
five gallon drums, hold the mixture of cocaine product, solvents, and water during
cooking. Linen sheeting and tables are used to filter the post-cooking substances. Paper
sheets and metal trays are used to dry cocaine hydrochloride. Scales are used for
weighing finished cocaine, and plastic bags for packaging it. Cocaine processing also
requires at least one technician or chemist with knowledge of the formulas involved in
each processing step and workers to provide assistance (Anonymous, 1991; DEA, 1993b,


246
reforms, the DEA and the Customs Service, continued to engage in the kind of
debilitating inter-agency rivalry the reforms were designed to prevent.
Nine years after the creation of the DEA, frustration with federal drug
enforcement efforts led policy makers to implement another administrative
reorganization, the FBI-DEA semi-merger in 1982 discussed in Chapter 6. Again, more
powerful political and bureaucratic actors imposed the reform from above. However, this
reform differed from earlier efforts in one significant respect: drug enforcers learned.
Following a period of aversion to the semi-merger, DEA managers and field agents
gradually adapted to the new rules and procedures, improving their ability to target major
trafficking organizations. The experience of the semi-merger suggests that power does
not necessarily preclude learning. Even when more powerful external actors impose
routines, participants within the targeted agency may accept the changes and modify their
behavior accordingly.
Over the years, a number of changes in the illicit drug industry have been due to
selection pressures rather than organizational learning.1 Trafficking enterprises operate
in extremely hostile environments characterized by a variety of legal and illegal
competitors, including drug enforcement agencies that seek to dismantle them.
Trafficking groups that fail to adjust to the vicissitudes of drug enforcement are often
selected out of the illicit drug industry. During periods of sufficiently vigorous drug
enforcement, some criminal enterprises are favored over others. The ability of trafficking
organizations to survive government crackdowns may have more to do with existing
practices and structures than their ability to alter routines in response to feedback. Some
1 Note that consideration of environmental selection requires that I change the level of analysis from the
individual firm to the industry as a whole.


295
Herrera, Homero [pseud]. Head, Aviation Unit, Anti-Narcotics Directorate,
Colombian National Police. Bogot. 14 July 2000.
Hinson, Sherman. Senior Advisor, Counter-drug strategy, Bureau of International
Narcotics and Law Enforcement Affairs, U.S. Department of State. Bogot. 3 July
1997.
Homero [pseud]. Former drug trafficker. Federal Correctional Complex. Coleman.
19 September 2000.
Howell, William C. Essentials of Industrial and Organizational Psychology. Homewood:
Dorsey, 1976.
Huber, George P. Organizational Learning: The Contributing Processes and the
Literatures. Organization Science: A Journal of the Institute of Management
Sciences 2.1 (1991): 88-115.
Hudson, Edward. Major Drug Raid Under Way Here. New York Times (6 October
1974): 1 & 18.
Hutchins, Edwin. Organizing Work by Adaptation, Organizational Learning. Ed.
Michael D. Cohen and Lee S. Sproull. Thousand Oaks: Sage, (1996): 20-57.
Hutchinson, Asa. Statement of Asa Hutchinson, Administrator, Drug Enforcement
Administration. Hearing before the Subcommittee on Criminal Justice, Drug
Policy and Human Resources, U.S. House of Representatives, One Hundred and
Seventh Congress, first session (13 November 2001). [Accessed 14 March 2002],
Http://www.dea.gov/pubs/cngrtest/ctl 11301 .html
Ianni, Francis A. J. A Family Business: Kinship and Social Control in Organized Crime.
New York: Russell Sage Foundation, 1972.
Internal Revenue Service (IRS). Handbook 9.5: The Investigative Process (2 September
1999). Http:/Avww.irs.gov/prod/bus info/tax pro/irm-part/part09/36221.html
[Accessed 26 July 2001].
Isikoff, Michael. Federal Officials Seize 12-Ton Cocaine Shipment. Washington Post
(3 December 1991): A4. Http://web.lexis-nexis.com/universe [Accessed 6 March
2000],
Jackson, Janet L., Janet C.M. Herbrink, and Robert W.J. Jansen. Examining Criminal
Organizations: Possible Methodologies. Transnational Organized Crime 2.4
(1996): 83-105.


197
traffickers. This practice has been pursued with some success in targeting the leaders of
the Cali and Medellin core organizations in Colombia, allowing police investigators to
penetrate their elaborate security structures (Reyes, 1999, p. 165; Serrano Cadena, 1999,
p. 63).
Surveillance and Undercover Operations
Drug enforcement agents also use surveillance and undercover operations to
gather intelligence about trafficking and money-laundering activities. Surveillance refers
to the clandestine observation of persons, vehicles, places or objects to obtain information
on criminal activities or participants identities. Undercover operations are a form of
surveillance in which investigators or criminal informants assume false or misleading
identities to acquire information. Investigators supplement physical surveillance and
undercover operations with photographic and electronic surveillance equipment that
intercepts communications among participants and records their illicit activities. When
sufficient intelligence has been gathered to execute court-approved search warrants,
police agents raid apartments, offices, and stash houses and capture incriminating
documents, computer files, and other records of criminal activity. All of these documents
and recordings contain useful information that can be exploited by criminal investigators
and government prosecutors (Abadinsky, 1997, pp. 485-491; Furden, 2000; Levine,
1990; Marshall, 1999, p. 25; Marshall, 2001b; Nadelmann, 1993, pp. 207-210, 239;
Nieves, 1997, p. 15; Westrate, 1994, p. 89).
Unfortunately for police officials, the amount of information gathered about
Colombian trafficking enterprises through physical surveillance and undercover
operations is often limited. Unlike street drug sales, most transactions involving


294
Guzman, Hector. Colombia espera que con arresto de Urrego se reconozca su lucha
antidrogas. (19 February 1998).
Http://cnnenespanol.com/latin/COL/l998/02/19/urrego/index.html [Accessed 20
February 1998],
Haas, Ernst. When Knowledge is Power: Three Models of Change in International
Organizations. Berkeley: University of California Press, 1990.
Haas, Peter M. Ed. Knowledge, Power, and International Policy Coordination. Special
issue of International Organization 46 (Winter) (1992).
Haas, Peter M. and Ernst Haas. Learning to Learn: Improving International
Governance. Global Governance 1.3 (1995): 255-284.
Hagedom, John M. Neighborhoods, markets, and gang drug organization. Journal of
Research in Crime and Delinquency 31.3 (1994): 264-294.
Hall, Peter A. Policy Paradigms, Social Learning, and the State. Comparative Politics
25 (April) (1993): 275-296.
Halstead, Boronia. The Use of Models in the Anlisis of Organized Crime and the
Development of Policy. Transnational Organized Crime 4.1 (1998): 1-24.
Hannan, Michael T. and John Freeman. The Population Ecology of Organizations.
American Journal of Sociology 82.5 (1977): 929-964.
Harney, Malachi L. and John C. Cross. The Narcotic Officer's Notebook. 2nd edition.
Springfield: Charles C. Thomas Publisher, 1973.
Harpers. Interview: And Death Shall Have Its Dominion. (October 2000): 15-17.
Hartlyn, Jonathan. The Politics of Coalition Rule in Colombia. New York: Cambridge
University Press, 1988.
Hatch, Marvin [pseud]. Head, U.S. Military Group in Colombia. Bogot. 3 March 2000.
Heclo, Hugh. Modern Social Politics in Britain and Sweden: From Relief to Income
Maintenance. New Haven: Yale University Press, 1974.
Hedberg, Bo. How Organizations Learn and Unlearn, Handbook of Organizational
Design Volume 1: Adapting Organizations to their Environments. Ed. Paul C.
Nystrom and William H. Starbuck. New York: Oxford University Press, (1981):
3-27.
Henderson, Mitchell. Undercover in Colombia: American Merc Infiltrates the Cali
Cartel. Soldier of Fortune 17 (January 1992): 90-93.


191
Table 6-2 U.S. Drug Enforcement Intelligence Agencies,
Types of Intelligence and Scope of Activities3
Agency (Cabinet level
Tactical
Operational
Strategic
Domestic
Foreign
department/organization)
intelligence
intelligence
intelligence
Drug Enforcement Administration
(Justice)
X
X
X
X
X
El Paso Intelligence Center
(Justice)
X
X
X
X
X
Federal Bureau of Investigation
(Justice)
X
X
X
X
X
National Drug Intelligence Center
X
X
X
X
(Justice)
Regional Information Sharing
X
X
X
X
Systems Program (Justice)
Customs Service (Treasury)
X
X
X
X
Customs Domestic Air Interdiction
X
X
X
X
Coordination Center (Treasury)
Financial Crimes Enforcement
X
X
X
X
X
Network (Treasury)
Coast Guard (Transportation)
X
X
X
X
X
Central Intelligence Agency
(Director of Central Intelligence)
X
X
X
X
Crime and Narcotics Center
X
X
X
X
(Director of Central Intelligence)
Defense Intelligence Agency
X
X
X
X
(Defense)
National Security Agency
X
X
X
X
(Defense)
National Imagery & Mapping
X
X
X
X
Agency (Defense)
Joint Interagency Task Force-East
X
X
X
(Defense)
Joint Interagency Task Force-West
X
X
X
(Defense)
Joint Interagency Task Force-
X
X
X
South (Defense)
Joint Task Force Six (Defense)
X
X
X
Office of Naval Intelligence
(Defense)
X
X
X
X
Tactical Analysis Teams (Defense)
X
X
X
Bureau of Intelligence & Research
(State)
X
X
High Intensity Drug Trafficking
Areas Program (Office of National
Drug Control Policy)
X
X
X
X
Note: X indicates type of intelligence produced or range of foreign and domestic intelligence activities.
Source: General Accounting Office (1998).
3 Absent from this list are a number of federal organizations that produce counter-narcotics intelligence as a
byproduct of their primary mission, including the Border Patrol, the Bureau of Alcohol, Tobacco, and
Firearms, the Bureau of Land Management, and the Federal Aviation Administration (GAO, 1998, pp. 5-6).


203
engaged in debilitating rivalries with the Customs Service, FBI, and CIA (Thornton,
1982; Tofani, 1985).
In 1982, Attorney General William French Smith announced an administrative
reform giving the FBI concurrent jurisdiction for federal drug investigations. In
practice this meant that the DEA now fell under the administrative control of the FBI.
While, the DEA nominally remained the lead agency in federal drug enforcement, the
FBI director now had the authority to appoint top-level DEA administrators and
determine policies for hiring, training, and fiscal management. The first DEA
Administrator to serve under the reorganization was a career FBI official who replaced
the DEAs geographic organizational structure (i.e. South American division, European
division, etc.) and with an FBI-inspired section orientation {i.e. heroin desk, cocaine desk,
marijuana desk). The 1982 reform also mandated that the FBI direct its resources and
expertise in the areas of electronic surveillance and financial investigations to assist the
DEAs efforts to target major narcotics violators. As part of this effort, six hundred FBI
agents were shifted to drug investigations (Levine, 1990, p. 78; Mullen, 1983, pp. 31-33;
Nieves, 1997, p. 6; Thornton, 1982; Tofani, 1985; Wilson, 1989, p. 267).
Initially, the semi-merger produced what James Q. Wilson describes as a
profound culture shock in the two agencies, as field agents and managers struggled to
adapt to new decision-making hierarchies and routines. However, with the passage of
time and the implementation of additional procedures and joint training programs,
collaboration between the two enforcement organizations improved, leading Wilson to
conclude his assessment on an optimistic note:
FBI training programs began to produce, even in the eyes of DEA veterans, better
quality agents. The FBI administrators in turn began to loosen their tight central


200
208; NDIC, 2001; ONDCP, 2001; Reuter et al., 1988, p. 42; Richelson, 1999, pp. 142-
143; Westrate, 1994, p. 83).
From Intelligence to Operations
Law enforcement authorities in the U.S. and Colombia use intelligence to plan
and carry out their everyday activities. As the director of the Florida Office of Drug
Control points out, we try to move from intelligence to operations. That is the purpose
of developing an intelligence pictureto act on it (McDonough, 2000, p. 218). To do
so, drug enforcement units hold regular meetings during which managers and field agents
analyze tactical and operational intelligence about specific criminal organizations, discuss
the strengths and weaknesses of ongoing enforcement efforts, and plan upcoming
operations. During an interview, the head of one DAS investigative unit described how
his group met regularly to discuss specific tactical intelligence, such as the contents of an
intercepted telephone call, and determine whether the information was sufficient to merit
further investigation of the enterprise or plan another course of action (Furden, 2000).
Once we terminate a case, the official continued, we have a meeting in which
we all contribute information and we say, Now we are going to work in this form,
because we saw this [criminal activity] (Furden, 2000, authors translation). When
asked whether the diffusion of case-specific information is limited to his unit, the official
answered that he received tactical and operational intelligence regarding different
criminal organizations from DAS units all over the country and that we include in our
knowledge (Furden, 2000, authors translation).


139
Re-Statement of Organizational Learning Proposition
Organizations learn when participants acquire, interpret and apply knowledge and
experience to rules, practices, and procedures that guide collective behavior.
Organizations obtain information through their own trial and error experience and search.
Participants record knowledge and experience in organizational memories, including
manuals, databases, financial accounts, correspondence and computers. Organizations
construct meaning out of knowledge and experience through sensemaking, which refers
to the process by which participants develop inter-subjective understandings through
meetings, conversations, stories and myths. Participants enact information through
practices, procedures and performance programs that coordinate their behavior. When an
organization alters a practice or procedure as a result of acquiring and interpreting
information, it is has engaged in the process of organizational learning.1
Applied to the primary unit of analysis in this study, these ideas yield the
following proposition:
If participants in Colombian trafficking enterprises gather, interpret, and apply
information to collective behavior by changing existing routines or creating new
ones, then organizational learning has occurred.
For learning to explain organizational behavior, as opposed to alternative factors
such as power and environmental selection, participants must alter routines as a result of
acquiring and interpreting knowledge and experience.
A First Cut: Quantifying Interview Results
Do Colombian trafficking organizations learn from knowledge and experience? I
posed this question to several dozen respondents interviewed for this research. As a
1 Unlike some students of organizational behavior, I do not make a strong distinction between learning and
adaptation. Instead, as discussed in Chapter 2,1 distinguish between tactical and strategic learning. In this
study, the terms adaptation and learning are used interchangeably.


27
Cuban penintenciary before returning to Medellin. In subsequent years, law enforcement
authorities suspected the two brothers of continuing their illicit drug activities from
Colombia. However, they apparently avoided further prosecution. According to Arango
and Child, the Antioquian trafficker died a successful businessman, owner of motels and
several food establishments (Arango & Child, 1984, p. 169; Treasury Department).
Phase ThreeMid-1960s to late 1970s: Expansion of Colombian Drug Industry
In the mid and late 1960s the involvement of Colombian smuggling groups in the
international drug trade continued to expand. After the fall of the Batista regime in Cuba
a number of Cuban and Italian-American organized crime figures based in Havana fled
the island for the United States. While some of these traffickers remained active in the
drug trade, Cubas value as a transit point was much diminshed. Colombian smugglers
with strong connections to Cuban trafficking networks found themselves well positioned
to take advantage of this development. By 1965, Colombian trafficking enterprises were
providing almost the entire cocaine supply for U.S.-based Cuban smugglers. Also during
this period, Colombian smugglers supplied cocaine and marijuana to American, Chilean,
and Mexican traffickers (Betancourt & Garcia, 1994, p. 57; Cervantes, 1980; Clifford,
1980, p. 474; Gugliotta & Leen, 1989, p. 22; Jonnes, 1996, p. 338; MacDonald, 1988, p.
28; Presidents Commission on Organized Crime [PCOC], 1986, pp. 77-78; Ruiz
Hernndez, 1979; Sabbag, 1990; Thoumi, 1995, p. 126; U.S. House, 1977, pp. 3,19;).
In the late 1960s and early 1970s, Colombian traffickers established their own
transportation routes and wholesale distribution networks in the U.S. and Europe.
Smugglers drew on their previous experiences and knowledge in trafficking illicit drugs
and other contraband goods. A number of whiskey, cigarrette and marijuana smugglers


44
numerous marimberos to diversify into cocaine in the preceding decade, including the
growing risks associated with cocaine trafficking due to successful drug enforcement
efforts, and the greater per unit profitability of heroin. In addition, traffickers may have
been attracted by profitable European drug markets, where heroin distribution networks
were well-established, and law enforcement efforts posed fewer risks to drug smugglers.
In 1994, the DEA reported that core organizations led by the Ochoa brothers and
Leonidas Vargas Vargas were using their cocaine networks to smuggle heroin. The same
report claimed that the Rodrguez Orejuela brothers and Helmer Herrera Buitrago had
also invested in the heroin industry. Two other core organizations associated with the
Cali network, the Santacruz Londoo and the Urdinola Grajales groups, began
experimenting with heroin production in the early 1980s, and supplied Colombian
farmers with the necessary inputs necessary to cultivate opium. According to a former
director of the Colombian National Police (CNP), Santacruz Londoo brought Afghani
and Pakistani opium cultivators to Colombia in order to learn how to cultivate opium
poppies. Since then Colombia has become a major source of heroin to the U.S.,
supplying a higher grade product that many U.S. consumers favor over Southeast Asian
heroin (DEA, 1994a, p. 6; Farah, 1997; Farah, 1999; FBI, 1993, p. 21; Moore & Farah,
1998; Revista Cambio 16, 1999; Semana, 1999; Serrano Cadena, 1999, pp. 50-51;
Statement of Witness, 1989, p. 11; Zabludoff, 1999).
Government Drug Enforcement Efforts During Phase Four
Throughout the 1980s and the 1990s the U.S. and Colombian governments
directed considerable resources towards dismantling the core cocaine organizations.
While bilateral cooperation between the two countries was occasionally marred by


156
managing a stash house in the U.S., including tips for keeping the lawn trim, not letting
pets loose on the street, and going to the movies every Thursday night. These procedures
were designed to avoid attracting the unwanted attention of neighbors and law
enforcement officials to the stash house (Nordheimer, 1986; Passas, 2000).10 In 1997,
Colombian police officials discovered two manuals in the prison cell of a member the
Rodrguez Orejuela organization. The first manual, brazenly entitled To Successfully
Carry Out an Illegal Flight, Analyze the Following Aspects, outlined numerous
procedures for completing an international drug flight. The document provided radio
frequencies monitored by air traffic controllers, a list of radar installations in several
Latin American countries and the geographic areas covered by each, and suggestions on
how to avoid detection by law enforcement authorities. The second manual, How to
Give Testimony and Receive Judicial Benefits, offered tips for incarcerated members on
which crimes to confess to and other information for receiving the lightest possible
sentence {El Tiempo, 1997; Farah, 1996). As these examples indicate, some Colombian
trafficking organizations maintain sophisticated organizational memories. Participants
draw on the knowledge and experience contained in these documents and manuals, as
well as shared verbal understandings regarding practices and procedures, in completing
their criminal activities.
Interpreting Knowledge and Experience
In addition to gathering, recording and storing information, Colombian trafficking
enterprises give meaning to these data through retrospective sensemaking. This refers to
10 Krauthausen and Sarmiento provide a similar example. According to them, in the 1980s one participant
in Pablo Escobars security contingent prepared a manual teaching the members of the organization how
to infiltrate American society and how to avoid generating suspicion (Krauthausen & Sarmiento, 1991, pp.
169-170, authors translation).


Ill
Leader(s)

i
Exportation manager
Senior advisor/investor
I
Assistant cell manager
Assistant cell manager
1 I I
Key: Single-headed arrows indicate (vertical) flow of decision-making authority; double-headed arrows
indicate (horizontal) flow of information, not authority; dot-lined boxes indicate that position does not
exist in all large trafficking enterprises
Figure 4-2 Management Levels within Large Trafficking Enterprise
Figure 4-2 diagrams the organizational hierarchy of a hypothetical large-scale trafficking
operation. Many core organizations maintain numerous distribution and transportation


187
The samples bias toward drug enforcers and government officials more generally
dictates that I maintain a cautious attitude when interpreting the interview results.
However, the overwhelmingly affirmative responses suggest that I remain open to the
possibility of drug enforcement learning. As in the preceding chapter, a more useful
determination of learning comes from a detailed examination of the primary and
secondary source data gathered in this research, rather than a simple quantification of
responses pro and con.
Drug Enforcement Intelligence Activities
Intelligence refers to the collection, organization, analysis, and dissemination of
information for policy makers and other government officials involved in the formation
and execution of foreign, defense, and economic policy. Drug enforcement intelligence
refers to the collection, organization, analysis, and dissemination of information for
officials that design and implement counter-narcotics policies and programs. Intelligence
is critical to effective drug enforcement. To allocate resources, design programs, and
evaluate results, law enforcers need accurate information about existing trends in the
illicit drug industry. To intercept drug shipments, destroy processing labs, confiscate
illegal proceeds, and dismantle trafficking enterprises, they require timely and reliable
information about participants in these criminal conspiracies and their methods of
operation. Intelligence is also important to drug enforcers ability to learn from
knowledge and experience. To alter practices, procedures, and policies with the purpose
of improving drug enforcement, they need timely and reliable information about specific
trafficking enterprises and their methods of operation (Loch Johnson, 2000; McDonough,


144
Subsequently, he used this knowledge in conducting various activities on behalf of the
organization. The Rodriguez Orejuela brothers sent participants to various parts of their
transnational trafficking operations to observe how different activities were carried out.
One participant was sent to Venezuela to inspect the process by which the enterprise
hid cocaine inside concrete posts and cornerstones and exported them to South Florida.
Another traveled to Los Angeles to show a cell manager and other participants how to
extract the cocaine packages concealed in wood boards. A third went to Guatemala to
receive information about code words and containers for hiding cocaine (Emil, 2000;
Reyes, 1999, pp. 180-181; Third Superceding Indictment, [no date], pp. 9, 17-18).
Another standard practice is to gather information about different ways of
smuggling psychoactive drugs, laundering illicit profits, and manipulating legal systems.
Many trafficking organizations expend considerable resources to gather such knowledge.
Often, they draw on the knowledge and skills of participants to develop innovations, even
assigning participants the task of developing new smuggling routes and methods. When
asked how his trafficking group developed the innovation of hiding cocaine shipments
inside diesel engines, which he claimed was an effective ruse against police sniffer
dogs, Jorge Valds recalled that the idea came from a chemist or law enforcement officer
working on behalf of the enterprise. As the operation grew, Valds and his Colombian
partners hired more participants, adding to the organizations knowledge base (Valds,
2000). To incorporate specialized, in some cases scientific, knowledge regarding
essential activities, trafficking enterprises hire engineers, chemists, lawyers, and former
drug enforcers and military officials (Arreguin, 2000; Brandy, 2000; Francis, 2000;
Oporto, 2000; Thompson, 2000).


127
violence towards participants are generally a measure of last resort, and used sparingly in
most organizations. Debts cannot be recovered from dead customers, and harming
participants disrupts business activity and may cause other members to question their
loyalty to the enterprise (Castillo, 1996, p. 62; Fuentes 1998, p. 198; Gugliotta & Leen,
1989; Mermelstein, 1990; Pallomari, 1997, V. 37, pp. 6060-6061; Sarmiento &
Krauthausen, 1991).
Promotion in Trafficking Enterprises
Irrespective of the cause of turnover, departing participants must be replaced if
the organization is to continue operations. One way of replacing associates is to recruit
new ones from outside the organization, which has already been discussed. Another
practice is to replace participants by promoting from within the enterprise. Promotion is
important because it allows leaders and managers to replace participants with minimal
disruption to the enterprise. Moreover, promotion offers a mechanism for exploiting the
knowledge and skills of the organizations most competent workers.
Internal promotion is common practice in Colombian trafficking enterprises.
Interviews conducted in this research and by Fuentes, as well as court documents,
provide numerous examples of participants working their way up the enterprise,
assuming positions of greater responsibility as they demonstrate their professional
competence and loyalty to leaders and managers. The stimulus for promotion often
comes from effective law enforcement. As members are arrested and incarcerated, or
indicted and forced to flee the U.S., they must be replaced, providing opportunities for
the remaining associates, particularly those that have caught leaders attention through
20 The Third Superceding Indictment in U.S. v. Miguel Rodrguez-Orejuela et. al., Case No. 93-470-CR-
WMH, lists several examples of members of the Rodrguez Orejuela enterprise being promoted. Also, see
Mermelstein (1990, p. 222).


or creating new ones, driven by the strong demand for their illicit commodities in U.S.
and European drug markets.
56
Table 2-4
Evolution of Illicit Drug Industry in Colombia
Phase
Time Period
Characteristics of Colombian Drug Trade
Organizational
Exemplar
One
1930s
Individual smugglers transport cocaine and
heroin to Panama
N/A
Two
Late 193 Os-
early 1960s
Smugglers transport cocaine and heroin in
Caribbean, supply criminal organizations
based in Havana, development of cocaine
and heroin processing labs in Antioquia
Herr an
Olazaga
brothers
Three
Late 1960s-
1970s
Marijuana and cocaine trafficking groups
supply Cuban, American, Chilean &
Mexican networks; establishment of
Colombian distribution networks in U.S.
Herrera
network,
Bravo
network,
Four
Late 1970s-
early 1990s
Rise of core organizations and support
networks; spread of cocaine production
throughout Colombia; participation of
guerrillas and paramilitary groups;
dismantling of Medellin and Cali core
organizations
Core
organizations
in Medellin
and Cali
Five
Mid 19905-
present
Decentralization of Colombian drug trade;
small and medium-sized groups; growth of
cocaine & heroin production in Colombia
North Valley
network,
Atlantic Coast
network,
Madrigal
network
Today, many trafficking enterprises in Colombia bear a greater resemblance to the
pre-cartel groups of the 1960s and 1970s than to the core organizations that followed.
They are small and contain flat decision-making hierarchies. They specialize in drug
production or transportation, but not both. Their leaders avoid the limelight, managing
their operations with the utmost discretion. Yet, in one significant respect existing
groups remain similar to their larger predecessors. They continue to modify their
trafficking practices by exploiting advances in communications and transportation


208
Figure 6-2 PHVA Management Cycle
KEY: Steps 1-4 correspond to plan (planear): (1) Initiate a new plan, (2) define & describe
the process, (3) define the goals, (4) define the action plan. Steps 5 & 6 correspond to do
(hacer): (5) Educate & train those responsible, (6) execute. Steps 7 & 8 correspond to
verify (verificar): (7) verify, (8) control. Steps 9 & 10 correspond to adjust (ajustar): (9)
analyze results, (10) analyze problems and identify corrective action.
Sources: Bandera (2000), CNP (1999), Pinellos (2000). Authors translation.
However, some of the most significant reforms involved the creation and implementation
of new organizational routines. One performance program developed during this period
was the PHVA management cycle; designed to help CNP managers improve their ability
to plan and evaluate programs. The name of the program is an acronym based on the
Spanish verbs for to plan, do, verify, and adjust (emphasis added). It includes procedures
for identifying and analyzing problems, and creating action plans to correct them (see


28
from the Antioquia and Guajira regions used their skills, contacts, and capital to enter the
cocaine business in the 1970s. Enterprising marimberos and contrabandistas discovered
that the smuggling infrastructures and methods used to transport marijuana to the United
States worked well for cocaine. Marijuana distribution networks converted easily to
cocaine. A number of marimberos made the switch gradually by pigbacking small
quantities of cocaine on their marijuana loads and obligating their U.S. distributors to sell
both products. Transportation methods used by marijuana smugglers, such as flying
private planes of various sizes to clandestine airstrips along the Altantic Colombian coast,
loading them with drugs, and returning to the U.S. with the illicit cargo, also adapted
easily to cocaine trafficking. Indeed, a number of smuggling innovations pioneered by
marimberos later became popular among cocaine traffickers, such as air dropping
specially wrapped packages of marijuana in the Caribbean sea, and using large seafaring
freighters, known as mother ships, to transport huge quantities (upwards of 100 tons) of
marijuana to prearranged locations 200 miles off the U.S. coastline, where private yachts
and go-fast motor boats would converge on the mother ship, receive and transport smaller
qualities into the U.S. (Cervantes, 1980, pp. 47-48; Krauthausen & Sarmiento, 1991, p.
147; MacDonald, 1988, p. 28; Nieves, 1997, p. 3; Passic 2000; PCOC, 1986, p.
78Velsquez Romero, 2000; Valds 2000).
By the mid-1970s, Colombian trafficking enterprises were recognized as
preeminent players in the inter-American cocaine industry. In a short period of time, they
transformed themselves from mere intermediaries for trafficking groups based in other
countries to vertically integrated, production-transportation-wholesale distribution
networks. Along the way Colombian enterprises developed practices, procedures, and


252
communities of jailed participants that socialize with fellow traffickers, share knowledge
and experience, and develop new and improved practices and procedures for conducting
their illegal activities. Incarcerating participants has the (presumably) unintended effect
of producing more sophisticated drug traffickers.
What is to be Done?
The scenario depicted in these pages is pessimistic. The supply-reduction
approach to drug control is unlikely to yield satisfactory policy outcomes, in part,
because the flat and fluid enterprises targeted by these efforts will continue to outwit their
slower, more bureaucratic state adversaries. Yet, pessimism need not descend into
despair. While there are no silver bullet solutions to the drug dilemma, policy makers
have a range of options at their disposal, some more feasible than others. These include
legalization, harm reduction, and amending drug enforcement bureaucracies. The first
two prescriptions address the demand-side of counter-narcotics policy; I address these
only sparingly. This is due to the orientation of the study rather than any presumption
that demand factors are less important to drug control than supply considerations.
Indeed, a comprehensive approach to narcotics control includes demand and supply
reduction components, along with ample resources for research. However, because most
of my analysis in the preceding chapters centers on the dilemma of supply-reduction, it
seems pertinent that I direct my policy recommendations to this area. The
recommendations that follow are based on the assumption that drug enforcement, for
better or worse, will continue to be an essential component of U.S. and Colombian
counter-narcotics efforts in the near future.


136
knowledge, such as legal representation and certain forms of money laundering, can be
sub-contracted to independent firms on a needs-only basis.
Conclusion
The illicit drug industry in Colombia is composed of hundreds of collective forms
that engage in a variety of production, transportation, distribution, and money laundering
activities. Many of these enterprises contain the constitutive elements of social
organizations. They are target-oriented entities that provide specific products, from
precursor chemicals to fully refined narcotics, to achieve satisfactory profits and other
benefits. They draw on task environments for material and symbolic resources, including
technology, capital, and participants. They interact with a variety of suppliers,
customers, and competitors that together form a hostile and dynamic environment. They
contain structures that guide participants behavior, including flat decision-making
hierarchies, compartmentalized cells, and routines that minimize organizational
vulnerabilities. They contain participants that make valuable contributions in return for
inducements. While turnover is high in many enterprises, an abundant labor supply and
promotion procedures mitigate its disruptive impact. Finally, trafficking enterprises
exploit a variety of material and symbolic technologies when conducting their criminal
activities. Many organizations have demonstrated the ability to exploit the latest
advances in material technologies to improve the effectiveness, if not necessarily the
efficiency, of their operations. The practical, as opposed to the scientific, orientation of
drug trafficking ensures that the requisite knowledge is easily and widely diffused among
participants, complicating law enforcement strategies to dismantle the drug trade by
targeting a few major narcotics organizations.


Http://www.cnn.com/2000/WQRLD/americas/Q9/07/colombia.sub/index.html
[Accessed 28 May 2001].
277
Cable News Network (CNNespaol). Colombia asesta golpe de gracia al Cartel de la
Costa. (7 June 1998).
Http://cnnenespanol.com/latin/CQL/1998/06/07/orlandez/index.html [Accessed
15 June 1998].
Cable News Network Espaol (CNNespaol) El ejrcito colombiano ocupa un centro de
acopio de drogas de las FARC. (19 February 2001).
Http://cnnenespanol.com/2001/latin/CQL/02/18/farc/index.html [Accessed 19
February 2001).
Caldern, Ricardo. Semana correspondent. Bogot. 1 July 2000.
Camacho Guizado, Alvaro. Droga y sociedad en Colombia: El poder y el estigma.
Bogot: CIDSE, Universidad del Valle, 1988.
Camacho Guizado, Alvaro. La reforma de la Polica: realidades inmediatas y objetivos
estratgicos. Anlisis Poltico (19) (1993): 50-62.
Camacho Guizado, Alvaro. Drug Trafficking and Society in Colombia. Drug
Trafficking in the Americas. Eds. Bruce M. Bagley and William O. Walker III.
New Brunswick: Transaction, (1994): 97-120.
Camacho Guizado, Alvaro. Professor of Sociology, Institute of Political Studies and
International Relations, National University of Colombia. Bogot. 6 March 2000.
Camacho Leyva, Bernardo (Mayor General). Etapa contempornea, Historia de la
Polica Nacional de Colombia. Ed. Alvaro Valencia Tovar (General). Bogot,
Colombia: Planeta Editorial, (1993): 257-284.
Cambio. Made in Colombia. (24 July 2000): 36-37.
Cannon, Lou. Drugs Pour In via "Crazy Colombians." Washington Post (13 August
1979). Http://web.lexis-nexis.comAiniverse [Accessed 16 February 1999].
Can, Luis M. El Patrn: Vida y muerte de Pablo Escobar. Bogot: Planeta, 1994.
Caporaso, James A. Ed. The Elusive State: International and Comparative Perspectives.
Newbury Park: Sage Publications, 1989.
Caracol. Capturado narcotraficante colombiano del Cartel de los Llanos. (11 March
2000a). Http://www.caracol.com.co/titular.asp?ID=21378 [Accessed 23 October
2000].


61
desired ends. Organizational goals include creating goods and services, attaining
satisfactory profits and other economic, social and political benefits, and ensuring the
survival of the enterprise. Objectives include constructing production facilities, searching
for new markets, devising innovative production and marketing methods, recruiting
people to work for the organization, and reducing internal and external threats to the
integrity of the enterprise. The determination of satisfactory benefits is highly
subjective and varies among organizations. However, autonomous organizations do not
exist for long unless the value of their products exceeds the resources needed to make
them (Leavitt et al., 1973, pp. 10-11; Scott, 1998, pp. 20-21).
Tasks may be clearly defined and widely appreciated within an organization, or
they may be vaguely understood and unique to sub-parts of the collective (Leavitt et al.,
1973, pp. 23, 27). As Barnard and Simon recognized decades ago, individuals within
organizations frequently do not share a common conception of organizational goals
(Barnard, 1938; Simon, [1945] 1997). Moreover, organizations typically pursue multiple
objectives simultaneously, which can create competition and conflict among different
parts of the enterprise.
Environments
Organizations develop within environmental contexts. Environments provide
opportunities for participants to coordinate their behavior, and the material and symbolic
resources necessary to survive. In organization analysis, the environment is a residual
concept. It refers to anything outside the organization. Dills notion of the task
environment provides an element of precision to this broad concept. The task
environment refers to those parts of the environment that are relevant to the


161
of these inputs, many of which have legitimate uses unrelated to the production of
psychoactive substances (Bagley, 1999; Clawson & Lee, 1996, p. 11; Cubides, 2000;

DEA, 1993b, p. 8; Lee, 1999; Prez Valdes, 2000a; Smith, 2000a; Uribe Ramirez, 2000;
U.S. House, 1996).
In addition to finding alternative substitutes, processing labs have learned to
manufacture and recycle chemicals in short supply. Some enterprises have developed
their own filters to recycle restricted inputs, such as gasoline. Others have constructed
elaborate facilities for storing and recycling chemical inputs. In 1991, the Colombian
National Police (CNP) discovered a recycling/storage complex in the Colombian jungle
that included a hydraulic pumping system and three-story distillers to clean previously
used chemical solvents. According to the U.S. State Department, reprocessing
innovations have reduced the need for shipments of precursor chemical into certain areas
of Colombia (BINLE A, 1997).
More recently, trafficking enterprises have been producing precursor chemicals
that traditionally were imported into Colombia. Over the past two years, CNP and
Department of Administrative Security (DAS) units have discovered several different
laboratories for manufacturing potassium permanganate and caustic soda, two essential
substances for cocaine processing. Other processing innovations include using vapor
eliminating devices to conceal strong chemical odors associated with cocaine processing,
particularly in urban areas, and shifting to small, mobile processing labs that can be
packed up and moved to another location on short notice. The latter innovation
developed following the destruction of several industrial-sized processing complexes
during the 1980s.


xml version 1.0 encoding UTF-8
REPORT xmlns http:www.fcla.edudlsmddaitss xmlns:xsi http:www.w3.org2001XMLSchema-instance xsi:schemaLocation http:www.fcla.edudlsmddaitssdaitssReport.xsd
INGEST IEID EBVS9DAUI_0FZA7Q INGEST_TIME 2014-10-14T00:05:52Z PACKAGE AA00025792_00001
AGREEMENT_INFO ACCOUNT UF PROJECT UFDC
FILES


279
Cervantes, Jos. La noche de las lucierngas. Bogot: Plaza y Janes, 1980.
Chabat, Jorge. Drug Trafficking in U.S.-Mexico Relations: What You See is What
You Get. Drug Trafficking in the Americas. Eds. Bruce M. Bagley and William
O. Walker III. New Brunswick: Transaction, (1994): 373-394.
Chambliss, William J. On the Take: From Petty Crooks to Presidents. 2nd edition.
Bloomington: Indiana University Press, 1988.
Chambliss, William J. State Organized Crime The American Society of Criminology,
1988 Presidential Address. Criminology 27.2 (1989): 183-208.
Chandler, Alfred D. Jr. Strategy and Structure: Chapters in the History of the American
Industrial Enterprise. Cambridge: MIT Press, 1962.
Chepesiuk, Ron. Hard Target: The United States War Against the International Drug
Trafficking. Jefferson: McFarland & Company, 1999a.
Chepesiuk, Ron. The War on Drugs: An International Encyclopedia. Santa Barbara:
ABC-CLIO, 1999b.
Clawson, Patrick L. and Rensselaer W. Lee III. The Andean Cocaine Industry. New
York: St. Martins Press, 1996.
Clifford, Thomas. Testimony of Thomas Clifford. Organized Crime and Use of
Violence. Hearings before the Permanent Subcommittee on Investigations of the
Committee on Governmental Affairs, U.S. Senate. Part Two (2 & 5 May 1980).
Washington: U.S. Government Printing Office.
CM&. El submarino construido en Bogot contena tecnologa de punta (7 September
2000). Http://www.cmi.com.co/2000/Septiembre/Pais2597.html [Accessed 16
October 2000].
Cohen, Michael D., and Paul Bacdayan. Organizational Routines are Stored as
Procedural Memory: Evidence from a Laboratory Study. Cognition Within and
Between Organizations. Eds. James R. Meindl, Charles Stubbart, and Joseph F.
Porac. Thousand Oaks: Sage, (1996): 341-367.
Cohen, Michael D., James G. March, and Johan P. Olsen. A Garbage Can Model of
Organizational Choice. Decisions and Organizations. Ed. James G. March. New
York: Basil Blackwell (1988): 294-334.
Constantine, Thomas A. Prepared statement by Thomas A. Constantine, Administrator,
Drug Enforcement Administration. International Organized Crime Syndicates
and their Impact on the United States. Hearings before the Senate Foreign


170
profound changes can be implemented within a month or so, after which the cell can
resume its illegal activities. In cases where the damage is extreme, cells are completely
shut down, and participants transferred to other parts of the enterprise or temporarily
retired from active involvement in the organizations illegal activities. Some enterprises
transfer existing cells or create new ones in different areas where local drug enforcers
have fewer resources (Butterfield, 2002; Fuentes, 1998, pp. 234-235; Mermelstein, 1990,
pp. 122-123).
Colombian distribution cells also change their operational practices and
procedures even before law enforcers identify them. Cell workers are often required to
change communications equipment and stash house locations, sometimes as frequently as
every few months. In some operations, leaders rotate participants and even managers
among different cells on an annual or even biannual basis. Cells also develop new
routines for delivering drugs, collecting money from wholesalers, and repatriating profits
on a regular basis. Periodic adaptation is a standard risk reduction strategy designed to
shield these criminal operations from the vagaries of drug enforcers and illicit
competitors. By changing their operations on a regular basis, distribution cells stay
several steps ahead of law enforcement, allowing them to survive hostile task
environments for extended periods (Fuentes, 1998, p. 253).
Strategic Learning in Colombian Trafficking Organizations
Trafficking enterprises develop strategies in order to satisfy basic aims, such as
attaining sufficient profits and ensuring organizational survival. These immediate and
long-range action plans exist in virtually all phases of their illicit operations. While
trafficking enterprises routinely tinker with techniques and tactics, it is less clear whether


KEY TO ABBREVIATIONS
AUC
Autodefensas Unidas de Colombia (United Self-Defense Forces of
Colombia)
BINLEA
Bureau of International Narcotics and Law Enforcement Affairs
BINM
Bureau of International Narcotics Matters
CIA
Central Intelligence Agency
CNN
Cable News Network
CNNespaol
Cable News Network Espaol
CNP
Polica Nacional de Colombia (Colombian National Police)
DAS
Departamento de Seguridad Administrativo (Department of
Administrative Security)
DEA
Drug Enforcement Administration
DOJ
Department of Justice
ELN
Ejrcito de Liberacin Nacional (Army of National Liberation)
EPIC
El Paso Intelligence Center
FAA
Federal Aviation Administration
FARC
Fuerzas Armadas Revolucionarias de Colombia (Revolutionary Armed
Forces of Colombia)
Fiscala
Fiscala General de la Nacin (National Prosecutor General)
FBI
Federal Bureau of Investigation
FY
Fiscal Year
GAO
General Accounting Office
XI


150
The lack of detailed scientific knowledge involved in drug trafficking (see
Chapter 4) has significant implications for the exchange of information among
participants in the Colombian narcotics industry. A large amount of the knowledge
required to produce, transport, and distribute illegal psychoactive drugs can be acquired
rather easily, often without formal scientific training. Participants often learn by doing,
watching more experienced others carry out activities and talking with them about their
work. Mermelstein acquired much of his trafficking knowledge through practical, on-
the-job training exercises, in which he observed other participants conduct different drug
processing and distribution activities. Emil experienced a similar training process when
he traveled to Miami to learn about maritime transportation and cell management (Emil,
2000; Mermelstein, 1990, p. 59).
Over the past several decades, the informal exchange of knowledge and
experience among different participants and organizations in the Colombian drug trade
has facilitated an illegal tradition of sorts, with the relatively simple know-how involved
in drug trafficking becoming increasingly widespread throughout Colombian society
(Caldern, 2000; Krauthausen & Sarmiento, 1991, p. 139).7
Drug Smuggling U: Acquiring Trafficking Knowledge in Prison
In addition to weddings and dance halls, prisons provide an ideal setting for
participants from different enterprises to gather and share knowledge and experience
regarding drug trafficking. Social commentators and criminologists have long
recognized that prisons serve as schools for delinquent behavior, allowing career
criminals to acquire knowledge of their trade and establish useful contacts for future
71 am not suggesting that all or even most Colombian citizens are involved in drug trafficking. Rather, my
point is that due, in part, to its relatively simple technology, the knowledge contained in the drug trade has
become increasingly widespread throughout the country.


APPENDIX B
QUESTIONNAIRE
Interview conducted by Michael Kenney, University of Florida
Respondent
Job title
Date
Location
Informed Consent: You have been asked to participate in a semi-structured interview. I
am a graduate student at the University of Florida, and I am conducting research on the
Colombian drug trade and government counter-narcotics efforts. The objective of this
study is to determine if Colombian drug trafficking enterprises learn from government
drug enforcement efforts. The interview contains a series of questions relating to the
research. Your participation in this study is completely voluntary. You are free to refuse
to answer any or all questions. With your permission, I would like to audio tape the
interview.
INTRODUCTION: Please tell me a little about your professional background
CURRENT STATE OF INDUSTRY/ORGANIZATIONS
1. How would you describe the current state of the illegal drug industry in
Colombia?
2. How would you describe the enterprises that manage the production and
distribution of cocaine and heroin in Colombia?
3. Are these enterprises organizations (hierarchical chain-of-command, roles for
members, rules that guide behavior) or networks (decentralized, not hierarchic,
fluid networks of individuals loosely connected by their criminal activities)? Do
these enterprises possess a combination of these characteristics (or none of them)?
4. PROBE for organizational ontology: How many levels of hierarchy w/in
organization? Do subordinates receive and follow orders or are these just
suggestions passed between friends/colleagues? Are orders generally disregarded
or obeyed? Are roles w/in organization/entity fixed or fluid?
5. To what extent can we generalize about these enterprises? Is the organizational
structure of the Cali enterprise generalizable to the Medellin and North Valley
enterprises?
266


Emergence of the Core Organizations 38
Smuggling Methods of Core Organizations 40
Diversifying to Heroin 43
Government Drug Enforcement Efforts During Phase Four 44
Phase FiveMid-1990s to 2000: Atomization of Colombian Narcotics Industry.... 49
Government Drug Enforcement Efforts During Phase Five 52
Conclusion 54
3 TOWARD A THEORETICAL UNDERSTANDING OF ORGANIZATIONS
AND ORGANIZATIONAL LEARNING 58
Organizations 60
Organizational Tasks 60
Environments 61
Hostile environments 63
Organizational Structure 63
Routines 64
Participants in Organizations 67
Technology 68
Defining Organizational Learning 69
Acquiring Information 70
Interpreting Information 71
Applying Information 73
Research Propositions 75
Alternative Explanations for Organizational Change 75
Learning Ecologies 78
Productive Learning 80
Levels of Learning: Tactical vs. Strategic Adaptation 81
Learning under Ambiguity 85
Properties of Organizational Learning 90
Environmental Hostility vs. Benevolence 90
Organization Size and Decision-Making Hierarchies 91
Conclusion 93
4 ORGANIZATIONAL ANALYSIS OF THE ILLICIT DRUG INDUSTRY IN
COLOMBIA 95
Tasks of Colombian Trafficking Enterprises 96
Understanding Tasks in Colombian Trafficking Enterprises 100
Environments of Colombian Trafficking Enterprises 101
Trafficking Enterprises and Law Enforcement Agencies 101
Other Organizations in the Task Environment 105
Structure of Colombian Trafficking Enterprises 106
Centralized Decision-making Hierarchies 107
Flat Decision-making Hierarchies 109
Role Specialization 112
vi


63
Hostile environments
Environments are not only a source of opportunities and resources for
organizations; they are also a source of uncertainty and risk. The ongoing flows of
personnel, resources and information between organizations and their task environments
are frequently erratic and unpredictable. While a number of factors affect the degree of
risk and uncertainty within an environment, few are as salient as the degree of
environmental hostility, as opposed to benevolence. A hostile task environment contains
two or more interconnected adversaries, at least one of which seeks to dismantle other
actors within the system. While a benevolent environment is not necessarily devoid of
competition, it is friendly in the sense that the ultimate goal is to out-perform but not
undermine the other units, so that all may survive to play another day.
Organizational Structure
Structure refers to relationship patterns among participants in organizational
settings. These patterns emerge from values, norms, roles and routines that characterize
how organizations function. Values refer to criteria for selecting desired goals of
organizations, such as increasing profits and ensuring organizational survival. Norms are
general rules that specify appropriate means for pursuing goals, such as exploiting new
markets and investing in research and development. Roles are behavioral expectations
for organization participants, often dealing with task performance. Roles define the
division of labor within organizations, assigning participants to tasks that facilitate
collective aims. Structure is not arranged randomly but ordered into coherent beliefs
and prescriptions governing the behavior of participants (Scott, 1998, pp. 17-18). These
patterns provide enduring, but not necessarily static, frameworks for structuring


230
Administrator for Intelligence, who reports to the Deputy Administrator, who reports to
the Administrator. This trajectory contains seven levels of decision-making authority,
from top to bottom. Figure 6-3 (see below) diagrams the convoluted, multi-layered
structure. DEA foreign offices contain a similar bureaucratic structure. Special Agents
report to one of four group leaders, who report to one of two Assistant Attachs, who
report to the DEA Country Attach, who reports to the Ambassador and the head of the
South American Field Division (Pineda, 2000).
The contrast in size and management layers between drug enforcement agencies
and trafficking enterprises has important implications for their competitive learning
games. Recall from Chapter 4 that these implications stem from basic premises in
organization theory. First, information and decisions travel faster when they flow
through fewer processing channels. Second, at each processing channel information is
susceptible to deliberate distortion, suppression, or misinterpretation. Therefore, the
greater the channels, the greater the opportunities for the manipulation of data required
for decision-making. On both these points, trafficking enterprises enjoy a critical
advantage over their state competitors. With their small size and flat structures,
trafficking groups tend to gather and process information more quickly than law
enforcement agencies. With fewer channels for moving and processing information,
decision-making cycles within trafficking enterprises tend to be faster than within law
enforcement agencies. Information and decision flows within trafficking enterprises are
also subject to fewer manipulation points.17 With fewer opportunities for information
distortion or suppression, decisions by entrepreneurs and upper-level managers are
17 However, there is nothing to suggest that the probability of information manipulation per channel is less
in trafficking enterprises than law enforcement agencies.


157
the social process through which participants share perceptions and construct inter-
subjective understandings of reality. Sensemaking is triggered when expectations,
practices or procedures are interrupted by unforeseen events. Trafficking organizations
operate in hostile and dynamic environments, which provide frequent opportunities for
sensemaking. Interdicted drug shipments, destroyed processing laboratories, dismantled
trafficking routes, seized profits, arrested participants, and impending legal proceedings
all represent unforeseen events that disrupt the everyday activities of trafficking
enterprises. In an effort to make sense of the these problematic situations, participants
communicate, share knowledge, and construct linkages between the present situation and
prior conditions through conversations and ad hoc meetings. These meetings allow
managers and participants to share perceptions of problems and analyze potential
solutions (Daft & Weick, 1984, p. 286; Furden, 2000; Homero, 2000; Kacerosky, 1994;
Pallomari, 1997, V. 36 & 37; Restrepo, 2000; Velsquez Romero, 2000a; Weick, 1995,
pp. 45-46).
Discovery Process
The discovery process offers an interesting example of how some core trafficking
organizations engaged in sensemaking processes. Discovery refers to a set of procedures
in the U.S. legal system, including pre-trial hearings, through which parties in civil or
criminal disputes access information that is likely to be used against them in court
proceedings. Beginning in the 19th Century, these procedures were instituted to help
disputants prepare for trial, encourage pre-trial settlement by informing all parties of the
efficacy of their claims, and expose insubstantial claims that should not go to trial
(Encyclopedia Britannica Online, 2001; Fuentes, 2000, p. 208).


314
Perrow, Charles. Normal Accidents: Living with High-Risk Technologies. New York:
Basic Books, 1984.
Piaget, Jean, Judgment and Reasoning in the Child. In collaboration with E.Cartalis and
others. Translated from the French by Maijorie Warden. London: Routledge &
K.Paul, 1969.
Pineda, Ignacio, [pseud]. Country Attach, Drug Enforcement Administration, Colombia.
Bogot. 13 April 2000.
Poggi, Gianfranco. The State: Its Nature, Development and Prospects. Stanford: Stanford
University Press, 1990.
Polica Nacional de Colombia (CNP). Modus Operandi y Organizacin del
Narcoterrorista Pablo Escobar Gaviria (April 1991).
Polica Nacional de Colombia (CNP). Fuerza del cambio: Plan Estratgico Institucional
del Nuevo Siglo, 1999-2002 (1999). Bogot.
Polica Nacional de Colombia (CNP). Revista de Criminalidad (2000). Bogot: Fondo
Rotatorio de la Polica.
Polica Nacional de Colombia (CNP). Transformacin cultural y mejoramiento
institucional [No date], Power point slide show provided by Dr. Perla Pinillos to
author (4 July 2000).
Polica Nacional de Colombia (CNP). Estructura Organizacional [No date],
Http://www.policia.gov.co/indicel.htm [Accessed 26 July 2001].
Polo, Michele. Internal Cohesion and Competition among Criminal Organizations. The
Economics of Organised Crime. Eds. Gianluca Fiorentini and Sam Peltzman.
Cambridge: Cambridge University Press (1995): 87-109.
Popper, Micha and Raanan Lipshitz. Organizational Learning Mechanisms: A Structural
and Cultural Approach to Organizational Learning. Journal of Applied
Behavioral Science 34.2 (June) (1998): 161-179.
Porter, Bruce. 1993. Blow: How a Smalltown Boy Made $100 Million With the Medellin
Cocaine Cartel and Lost It All. New York: St. Martins Griffin.
Posada-Carb, Eduardo. Ed. Colombia: The Politics of Reforming the State. New York:
St. Martins Press, 1998.
Posen, Barry. Sources of Military Doctrine. Ithaca: Cornell University Press, 1984.


134
Many trafficking activities, including drug processing, transportation, and
distribution, require a fair amount of practical knowledge on the part of participants. A
worker in a distribution cell must be able to communicate with his colleagues using
simple code words, drive motor vehicles to specified locations, inspect and count
packages containing either drugs or money, and record notations in notebooks or other
forms of organizational memory. A drug chemist must be able to mix different chemicals
in varying quantities, cook them for a specified period, and filter and desiccate the
resulting solution. A cell manager must be able to supervise the criminal activities of
numerous subordinates, maintain frequent contact with the home office using a variety of
telecommunications technologies, and plan and implement distribution and collection
programs.
The amount of scientific knowledge required to carry out these activities is
minimal. Driving a car, communicating in code, writing simple notations are practical
activities that require a certain degree of general knowledge but not extensive scientific
training. The symbolic expertise involved in drug processing is fairly rudimentary and
can be acquired without any formal training in chemistry. Processing technicians need
only memorize a multi-step recipe for mixing, cooking, filtering, and drying various
substances. They do not have to understand why hydrochloric acid causes cocaine base
to crystallize into cocaine hydrochloride, only that it does. Recipes and procedures
capture this practical knowledge, and together they constitute a straightforward
performance program. While the technicians that carry out this process frequently
possess considerable knowledge about their craft, they do not require a great deal of


59
In this literature, criminal organizations and law enforcement agencies have rarely
been topics of empirical interest. This research builds on existing studies by examining
organizations that stand at the intersection of the rule of law and national security. The
lack of attention provided trafficking enterprises and drug enforcement agencies in the
organizational learning literature raises a number of questions. Most basically, what is a
drug trafficking organization that it may learn, and undertake other collective actions?
Do police agencies and trafficking enterprises learn from experience and information? If
these state and non-state collectives alter their practices and procedures, how can we be
sure that such changes are attributable to learning rather than alternative considerations?
What are the individual and organizational obstacles to learning? Assuming that
trafficking enterprises and law enforcement agencies do on occasion learn, what
conditions facilitate this process?
I do not propose to answer these demanding questions in this chapter. That
challenge unfolds later in this study. Here, my aim is more modest: to develop the
theoretical framework that guides subsequent analysis. I begin by discussing a concept
that subsumes organizational learning: organization. In order to evaluate whether
trafficking enterprises and drug enforcement agencies learn as organizations, I am
compelled to explain the latter concept. I do so by drawing on the open-systems
perspective of organization theory. Following this discussion, I proceed with a definition
of organizational learning that disaggregates the process into separate but overlapping
(1992), Haas & Haas (1995), Hall (1993), Heclo (1974), Khong (1992), Leng (1992), Le Prestre (1995),
Mann (1992), McCoy (2000), Mendelson (1993), Moltz (1993), Nagl (1999), Ndumbaro (1998), Nolan
(1994), Nye (1987), Posen (1984), Reiter (1996), Sabatier (1988), Sagan (1993), Sikkink (1997), Snyder
(1991), Stein (1994), Weir and Skocpol (1985), Welch Larson (1985), Zarkin (2000), and Zeng (1999).
For recent overviews of the political science literature on learning see Bennett & Howlett (1992) and Levy
(1994).


287
El Tiempo. Brasil confirma canje de coca por armas. (15 July 2000c).
Http://www.eltiempo.com/hoy/iud n001tn0.html [Accessed 15 July 2000].
El Tiempo. Submarino Made in Faca (8 September 2000d).
Http://www.eltiempo.com/08-09-2000/iudi O.html [Accessed 10 September
2000],
El Tiempo. Africanos ilegales, traficantes (5 March 2001a).
Http://eltiempo.terra.com.co/05-03-2001/iudi-bre O.html [Accessed 5 March
2001],
El Tiempo. Cae laboratorio qumico (26 June 2001b). Http://eltiempo.terra.com.co/26-
07-2001/judi-bre O.html [Accessed 26 July 2001].
El Tiempo. Decomiso en Nario (29 June 2001c). Http://eltiempo.terra.com.co/29-07-
2001/judi-bre O.html [Accessed 29 July 2001],
El Tiempo. Fabio Ochoa fue extraditado a E.U. (8 September 2001 d).
Http://eltiempo.terra.com.co/08-09-2001/iudi96287.html [Accessed 9 September
2001],
El Tiempo. Extraditado a E.U. alias Juvenal (30 October 2001).
Http://eltiempo.terra.com.co/30-10-2001 /iudi 123601 .html [Accessed 30 October
2001],
Emil [pseud]. Former drug trafficker. Federal Correctional Complex. Coleman. 19
September 2000.
Encyclopcedia Britannica Online. Procedural law (2001).
Http://www.eb.com:180/bol/topic?eu=l 17325&sctn=19 [Accessed 19 May 2001].
Epple, Dennis, Linda Argote, and Rukmini Devadas. Organizational Learning Curves.
Organizational Learning. Eds. Michael D. Cohen and Lee S. Sproull. Thousand
Oaks: Sage, (1996): 83-100.
Epstein, Edward Jay. Agency of Fear: Opiates and Political Power in America. New
York: G. P. Putnams Sons, 1977.
Etheridge, Lloyd S. Can Governments Learn? New York: Pergamon, 1985.
Evan, William M. The Organization Set: Toward a Theory of Interorganizational
Relations. Approaches to Organizational Design. Ed. James D. Thompson.
Pittsburgh: University of Pittsburgh Press, (1966): 173-188.
Evans, Peter B., Dietrich Rueschemeyer, and Theda Skocpol. Eds. Bringing the State
Back In. Cambridge: Cambridge University Press, 1985.


235
efforts, allowing trafficking enterprises to gather information about these activities. One
high-ranking DEA official elaborates on the last point:
it is more difficult for us [law enforcers] to learn from them [traffickers] than for
them to learn from us... Because everything we do is above board, everything we
do is done legally, we do it and it is exposed, it is discovered in court, and that is
the law and they have rights and... essentially that is the way the game is played,
thank God, that is what democracy is all about, the rule of law, due process and
everything, and it is satisfying to know that we comply with the rules and we
adhere to their rights, and we still win on occasion, and that is the stuff we seize
and the people we arrest, we do not seize and we do not arrest, and we do not
know how many tons are out there getting by us because they have this greater
insight, because we do things above water... they are very covert in their
activities, we are covert in our activities to try to penetrate them but this must be
over eventually to put them in jail and court and that is our exposure (Fernandez,
2000).
Under the U.S. legal system, the same norms that protect the rights of alleged
traffickers force drug enforcement agencies to expose the details of their criminal
investigative techniques to the information sharing process known as discovery. As
discussed in Chapter 5, a number of Colombian trafficking enterprises have manipulated
discovery to gather information about drug enforcement efforts and change their
transportation and distribution practices accordingly.
Some observers may counter that law enforcers violations of citizens rights and
the prevalence of drug-related corruption mollify the practical implications of the red tape
trap. It is indisputable that corruption and human rights violations continue to undermine
the integrity and efficacy of drug enforcement. Many U.S. and Colombian law
enforcement agencies contain at least a few corrupt participants. However, as formal
organizations their task is to uphold the rule of law rather than break it. Managers and
participants that repeatedly fail to abide by the corresponding bureaucratic constraints can
be held accountable for their actions and face a variety of disciplinary and legal


45
substantive disagreement over a number of policy issues, particularly extradition, the two
governments worked together to implement a variety of supply-reduction programs
targeting the Medellin and Cali-based trafficking networks.
During this period the Colombian government engaged in a series of
crackdowns directed at the leaders of the Medellin-based trafficking enterprises and
subsequently the Cali organizations. Several crackdowns were undertaken in response to
specific criminal acts attributed to Medellin trafficking groups, such as the assassination
of Rodrigo Lara Bonilla, the Colombian Minister of Justice, in 1984.10 Each time
Colombian authorities responded by carrying out dozens of raids against the properties of
leading Medellin traffickers, destroying processing laboratories, seizing cocaine and
money, an apprehending low and mid-level participants in these criminal enterprises.
However, the leaders remained beyond the reach of drug enforcers, developing
sophisticated security arrangements allowing them to evade government efforts. After
several weeks, the crackdowns would subside, and traffickers could return to business as
usual (Gugliotta, 1992, pp. 122-123; Matthiesen, 2000).
In 1989 Colombian president Virgilio Barco launched a fourth crackdown in
response to the assassination the leading presidential candidate for the 1990 general
elections, by sicarios (paid assassins) working for the Medellin traffickers. By this point,
the Colombian president and other political elites were convinced that the Medellin
organizations presented a legitimate threat to the countrys political system. In a
televised address, President Barco declared a nationwide state of siege against the
traffickers. At the behest of President Bush, the U.S. offered their beleagured drug war
10 Bonillas assassination was carried out in retaliation for his role in planning a police raid on the
Tranquilandia cocaine refining complex owned by several leaders of the Medellin smuggling network
(Gugliotta & Leen, 1989, pp. 132-137; Lee, 1989, p. 171; Tokatlin, 1990: 317).


14
produce thick descriptions of how organizations acquire and manipulate information and
experience.
To evaluate the explanatory power of organizational learning I have systematically
gathered primary and secondary source data from a variety of sources in Colombia and
the U.S. The theory of organizational learning and the method of structured, focused
comparison guided data collection and analysis. I identify Colombian trafficking
enterprises and U.S. and Colombian drug enforcement agencies as primary units of
analysis that can be fruitfully studied through intra-case comparison.10 From the learning
literature I developed a semi-structured instrument for conducting interviews.11
Questions address the acquisition, interpretation, and application of knowledge and
experience by Colombian trafficking enterprises and government counter-narcotics
agencies. Other questions focus on their organizational structures, routines, and practices.
Using (non-probability) snowball-sampling techniques, I interviewed seventy-six U.S.
and Colombian government officials, researchers, and former drug traffickers.
I collected additional primary and secondary source data from government
documents located at the National Archives in College Park, Maryland, the U.S.
District Court in Miami, and several government offices and collections in Bogot,
Colombia. Secondary sources of information include government documents, transcripts
10 For discussion of the method of structured focused comparison, see George (1979), George & McKeown
(1985), and King, Keohane & Verba (1994), especially p. 45. For discussion of intra-case comparison, see
Lebow (2001).
111 chose the semi-structured interview format for their two principal strengths: structure and flexibility.
The first ensures that all relevant topics will be covered during the course of interviews; the second allows
researchers to purue leads into other topics as they arise. For discussion of these issues, see Bernard
(1995).
12 The sample contained fifty-one government officials, many of whom worked in law enforcement agencies
such as the Drug Enforcement Administration and the Colombian National Police, fourteen academic


Drug trafficking enterprises are not the only organizational learners of interest in
this study. Law enforcement agencies also alter their practices and procedures in
response to knowledge and experience, thereby improving task performance and
facilitating their survival as bureaucratic organizations. To understand interactions
between these interdependent, rationally bounded actors, the study introduces the concept
of competitive learning games, and discusses numerous organizational and environmental
factors that tilt the playing field in favor of criminal enterprises. The study also includes
an historical analysis of the Colombian drug trade dating back to the 1930s. Outsmarting
the State is of broad theoretical and empirical interest to scholars and policymakers. The
study contributes to diverse bodies of literature in political science, sociology, economics,
organization theory, and criminology. The research has significant implications for
existing U.S. drug and crime control policies in Colombia, Mexico, and elsewhere.
xiv


236
sanctions. Consequently, drug enforcement agencies are largely constrained by the
legalistic and bureaucratic constraints contained in the red tape trap.
Conclusion
Drug enforcement organizations learn. Whether in response to feedback from
more powerful political and bureaucratic actors, or perceived changes in the illicit drug
industry, police agencies in Colombia and the U.S. gather, interpret, and apply
knowledge and experience to their activities. Lessons from experience are encoded in
routines that guide criminal investigations and drug enforcement operations. Over the
last twenty years, law enforcers have developed a number of undercover innovations,
including reverse stings and controlled deliveries, allowing them to better target major
trafficking organizations. Today, Colombian and U.S. nares are more proficient in
carrying out conspiracy investigations, using electronic surveillance technologies,
conducting undercover operations, and tracing illicit drug and money flows across
national borders than they were in decades past.
Like the criminal enterprises they seek to dismantle, drug enforcement agencies
learn through repeated interactions with actors that share their task environment,
including trafficking organizations. The concept of competitive learning games describes
how sets of interlocked, rationally bounded players seek to outwit their adversaries by
gathering, analyzing, and acting on intelligence to change their practices and procedures.
The concept also suggests that traffickers enjoy a number of advantages over their state
competitors. From the start of these contests law enforcers struggle to catch-up with
traffickers due to the clandestine nature of drug smuggling and the reactive nature of drug
enforcement. Trafficking enterprises fuel their initial advantage with faster decision


112
groups in consumer markets, which are represented in the three cells in the diagram.
Role Specialization
Role specialization within trafficking enterprises varies according to the size and
task complexity of the operation. Small enterprises that supply a single good or service,
such as providing precursor chemicals to processing laboratories or processing cocaine
base into cocaine hydrochloride, contain less role specialization than large organizations
involved in a variety of processing, transportation and distribution tasks. Transaction
quantity and frequency also influences functional specialization. The larger and more
frequent a transportation rings drug shipments, the more roles leaders will develop to
complete these transactions. In well-managed trafficking enterprises, leaders and
managers attempt to match participants with roles that suit their experience, training, and
skills. In large, well-established enterprises, roles tend to be stable and survive
considerable participant turnover. In small, nascent groups or ad hoc operations roles are
contingent on individual members and may not survive participant turnover (Fernandez,
2000; Thompson, 2000; Vargas, 2000).
One small-scale, multi-task enterprise active in the 1970s relied on a system of
couriers and mail packages to ship between two and four kilograms of cocaine hidden
inside specially prepared wooden handcrafts. This operation, described in detail by
Robert Sabbag, included the following roles: a leader based in New York, a coordinator
of operations in Colombia, a carpenter, several couriers, and a receiver of the
merchandise in New York. Three peoplethe leader, the Bogot-based coordinator and
the carpenterformed the nucleus of the operation, with each one performing multiple
support and line functions, such as purchasing cocaine from suppliers, packing cocaine


328
Velsquez Romero, Carlos. Actividades-Necessidades. Internal report from
Comandante, CEC Ejrcito, prepared for Fuezas Militares de Colombia,
Ejrcito Nacional (4 May 1994). Cali.
Velsquez Romero, Carlos. Former head, Cali search bloc, Colombian Armed Forces.
Bogot. 1 May 2000a.
Velsquez Romero, Carlos. Former head, Cali search bloc, Colombian Armed Forces.
Bogot. 15 July 2000b.
Viafara Salinas, Diego. Testimony of Diego Viafara Salinas, a Protected Witness.
Structure of International Trafficking Organizations. Hearings before the
Permanent Subcommittee on Investigations of the Committee on Governmental
Affairs, U.S. Senate. One Hundred First Congress, first session (12-13 September
1989). Washington: U.S. Government Printing Office.
Villega, Paula [pseud]. Assistant Director, National Directorate of Dangerous Drugs,
Republic of Colombia. Bogot. 3 March 2000.
Virany, Beverly, Michael L. Tushman, and Elaine Romanelli. Executive Succession and
Organization Outcomes in Turbulent Environments: An Organization Learning
Approach. Organizational Learning. Eds. Michael D. Cohen and Lee S. Sproull.
Thousand Oaks: Sage, (1996): 302-329.
Walker, William O. III. Drug Control in the Americas. Revised edition. Albuquerque:
University of New Mexico, 1989.
Walker, William O. III. International Collaboration in Historical Perspective. Drug
Policy in the Americas. Ed. Peter H. Smith. Boulder: Westview Press, (1992):
265-281.
Walker, William O. III. The Limits of Coercive Diplomacy: U.S. Drug Policy and
Colombian State Stability, 1978-1997. The Illicit Global Economy and State
Power. Eds. H. Richard Friman and Peter Andreas. Lanham: Towman &
Littlefield, (1999): 143-171.
Washington Office on Latin America (WOLA). The Colombian National Police, Human
Rights and U.S. Drug Policy. Washington: WOLA (May), 1993.
Westrate, David L. The Role of Law Enforcement. Drugs and Foreign Policy: A
Critical Review. Ed. Raphael F. Perl. Boulder: Westview Press, (1994): 79-99.
Weick, Karl E. The Social Psychology of Organizing. 2nd edition. New York: Random
House, 1979.
Weick, Karl E. Sensemaking in Organizations. Thousand Oaks: Sage, 1995.


249
Today the governments of Colombia and the United States persist in their quest to stamp
out the drug trade through supply-reduction policies. These efforts continue to privilege
punitive enforcement programs designed to reduce psychoactive drug consumption by
increasing the risks of transacting in these illicit commodities, as well as the costs of
consuming them. While state authorities have learned, much of it has been at the level of
tactical routines. Over the past decades, drug enforcement agencies have developed
reverse undercover stings and controlled deliveries, made better use of electronic
surveillance technologies, and improved their ability to track transnational flows of drugs
and money. In making effective use of such innovations police agents today are more
proficient at investigating and incarcerating major narcotics traffickers and disrupting
their illicit operations. However, the more drug enforcers tactics change, the more their
strategies remain the same. The basic approach to drug enforcement remains identical to
the strategy criticized by Mermelstein. The tactics may be harder to beat, but there is no
shortage of people willing to try.
The basic finding of this research is that drug enforcement is doomed to failure
because the people willing to try change their practices and procedures faster than the
state can accommodate. Quicker decision cycles, fewer impediments to action, and clear
incentive structures give narcos the edge in their repeated encounters with the state.
Even when drug enforcers demolish major trafficking organizations, other groups pick up
where their predecessors left off. This is due to the widespread diffusion of trafficking
technologies and other requisite inputs, and the continued profitability of transacting in
controlled substances.


308
Montoya, Federico. El FBI Busca Laboratorios de Drogas Heroicas en Colombia. El
Espectador (22 May 1959a): 3.
Montoya, Federico. 35 libras de marijuana incineran en Medellin. El Espectador (22
May 1959b): 3.
Moore, Molly. Latin Drugs Flow North Via Pacific: Traffickers' Ships Hard to
Intercept. Washington Post (30 January 1997).
Http://www.washingtonpost.com/wp-srv/WParch/1997-01/30/095F-013097-
idx.html [Accessed 1 April 1997].
Moore, Molly. Cocaine Seizures by U.S. Double in Pacific Ocean. Washington Post (3
September 2000). Http://www.washingtonpost.com/wp-srv/WPlate/2000-
09/03/1181-090300-idx.html [Accessed 26 May 2001],
Moore, Molly and Douglas Farah. Mexican Heroin on Rise in U.S. Washington Post (2
June 1998): A01. Http://www. washingtonpost.com/wp-srv/WParch/1998-
06/02/095F-060298-idx.html [Accessed 10 July 1998].
Morales, Pedro [pseud]. Government prosecutor, National Prosecutors Office. Bogot.
11 July 2000.
Moreau, Ron. Colombia: The Coke Trade. Newsweek (20 December 1976): 51.
Http://web.lexis-nexis.com/universe [Accessed 18 October 2000].
Moreno, Lus G. Head, Narcotics Assistance Section in Colombia, U.S. Department of
State. Bogot. 17 June 1997.
Morganthau, Tom. Hitting the Drug Lords. Newsweek (4 September 1989): 18-23.
Morganthau, Tom with Mark Miller, Richard Sandza, Joseph Contreras, Charles Lane,
and Thomas M. DeFrank. Hitting the Drug Lords. Newsweek (4 September
1989): 18-23.
Morrison, Shona. The Dynamics of Illicit Drug Transshipment and Potential Transit
Points for Australia. Transnational Organized Crime 3.1 (1997): 1-22.
Mullen, Francis M., Jr. Prepared Statement of the Acting Administrator, Francis M.
Mullen, Jr. DEA Oversight and Authorization. Hearing before the Subcommittee
on Security and Terrorism of the Committee in the Judiciary, United States
Senate, Ninety-eighth Congress, first session (23 February 1983). Washington:
U.S. Government Printing Office.
Mullen, Francis M. Testimony of Francis M. Mullen, Administrator, Drug Enforcement
Administration. DEA Oversight and Budget Authorization. Hearing before the


86
overwhelming abundance of incoming stimuli. However, these cognitive schemas are
based on prior beliefs that bias interpretations in subtle ways.
Within organizations, attention is a scarce resource; the time and effort these
collectivities direct towards information acquisition and interpretation is constrained by
other needs (March, 1988; March & Olsen, 1988). Because problems influence priorities,
attention is often forthcoming only in times of crisis, which means that information
acquisition and interpretation are often made under conditions of stress. A number of
these general difficulties manifest themselves as organizations acquire, interpret, and
apply information and experience to routines.
As they acquire and interpret feedback from their environment organizations are
vulnerable to a host of learning disabilities, including blind spots, filtering, flawed
interpretation, and secrecy. Blind spots occur when search is narrow or misdirected,
leading managers to miss or misinterpret important feedback signals. Filtering occurs
when critical information is ignored or downplayed because it is inconsistent with
decision-makers prior beliefs or powerful interests within the organization (Garvin, 2000,
pp. 28-29). Organization leaders tend to accept information that attributes success to
their own actions and failure to the actions of others or to exogenous sources (Levitt &
March, 1988, p. 324). As Scott Sagan notes, problematic situations are not simply
internal data points to be used to improve organizational performance. They are also
political events for which credit and blame must be assigned. Interpretation is easily
politicized within organizations. Sometimes it may be easier to kill the messenger (i.e.
suppress or distort information) than assign blame to powerful coalitions or individuals
(Sagan, 1993, p. 208; Huber, 1991, p. 95). While organizational learning depends on the


242
communicated informally among conspirators through conversations and meetings. The
diffusion of knowledge and experience regarding trafficking technologies is not
necessarily restricted to members of the same enterprise. In this context, organizational
learning is often inter-organizational, as knowledge spreads across criminal firms through
participants that share social networks. Venues that facilitate network learning include
social gatherings and prison.
However, not all trafficking organizations learn. Participants operate under
conditions of profound uncertainty. They struggle to make sense of ambiguous
experience. They are prone to biases when interpreting feedback, and they make
decisions based on incomplete and inaccurate information. The need to maintain secrecy
about their illegal activities further impedes learning. Practices and procedures designed
to protect core technologies from exploitation by drug enforcers and illicit competitors,
such as compartmentalized cell structures, restrict communication flows, reducing
participants ability to gather information from other parts of the enterprise. The use of
code words and aliases fosters frequent miscommunication among participants as they
seek to make sense of ambiguous situations. The lack of formal organizational memories
leads to the loss of organizational knowledge and expertise, particularly when participant
turnover is high. In light of these various impediments, it is not surprising that many
trafficking enterprises fail to adjust their tactical and strategic routines as needed. The
intensity of drug enforcement efforts in recent years has increased the costs of this
failure. Weak adapters are selected out of the Colombian psychoactive drug trade
relatively quickly, leaving behind enterprises with values, structures, and practices
conducive to tactical and strategic learning.


265
Name of
Date of
Type of
Trafficker
State learning?:
respondent
interview
respondent
learning?:
Affirmative,
negative, not
discussed
Affirmative,
negative, not
discussed
William J. Olson
7/20/99
U.S. government
Affirmative
Not discussed
Monica Passas
[pseud]
3/9/00
U.S. government
Affirmative
Not discussed
Ignacio Pineda
[pseud]
4/13/00
U.S. government
Affirmative
Affirmative
Rafael Perl
7/22/99
U.S. government
Affirmative
Not discussed
Miguel Restrepo
[pseud]
6/28/00
U.S. government
Affirmative
Not discussed
Thomas Sanchez
[pseud]
7/22/99
U.S. government
Affirmative
Not discussed
Joseph Smith
[pseud]
3/3/00
3/15/00
4/13/00
U.S. government
Affirmative
Affirmative
Jack Thompson
[pseud]
2/15/00
U.S. government
Affirmative
Not discussed
Sidney Zabludoff
7/19/99
U.S. government
(retired)
Affirmative
Not discussed
Klaus Nyholm
3/31/00
United Nations
Affirmative
Affirmative
Carlos [pseud]
4/5/00
Direct
participant
Affirmative
Not discussed
Emil [pseud]
9/19/00
Direct
participant
Affirmative
Not discussed
Homero [pseud]
9/19/00
Direct
participant
Affirmative
Affirmative
Nstor [pseud]
9/19/00
Direct
participant
Affirmative
Not discussed
Jorge Valds
8/29/00
Direct
participant
Affirmative
Not discussed


104
traffickers.7 Colombian trafficking enterprises monitor the telecommunications of police
agencies, infiltrate the same organizations through paid informants, suborn law
enforcement and military officials, prison authorities, public prosecutors, judges and
politicians, kidnap and assassinate honest officials, and detonate car bombs at
government offices.8
Table 4-1 National Government Agencies Involved in Drug Enforcement
Unites States
Drug Enforcement Administration (DEA)
Office of National Drug Control Policy
Federal Bureau of Investigation
Customs Service
Central Intelligence Agency
National Security Council
Bureau of International Narcotics and
Law Enforcement Affairs
Agency for International Development
Marshals Service
Financial Crimes Enforcement Network
Coast Guard
Bureau of Alcohol, Tobacco and Firearms
Secret Service
Internal Revenue Service
U.S. Attorney Offices
National Security Agency
U.S. Army
U.S. Navy
Colombia
Colombian National Police (CNP)
Directorate of Anti-Narcotics
CNP Directorate of Intelligence
CNP Directorate of Operations
CNP Directorate of Judicial Police
Department of Administrative Security
National Prosecutors Office
Attorneys General Office
National Directorate of Dangerous
Drugs
National Alternative Development Plan
Colombian Coast Guard
Colombian Navy
Colombian Army
Colombian Air Force
Colombian Marine
7 Numerous examples of each activity have been found in this research from primary and secondary
materials. For example, several cocaine entrepreneurs, including Pablo Escobar, Gonzalo Rodrguez
Gacha, and Jos Santacruz Londoo, have been shot and killed in Colombia by the police and armed forces
while attempting to avoid apprehension (Serrano Cadena, 1999; Can, 1994; Velsquez, 1993; Gallego
Castrilln 2000).
8 Specific examples can be found in Gugliotta & Leen (1989), Lee (1989), Rice (1989), Pallomari (1997, V.
38), Thoumi (1995), and Treaster (1989).


106
particularly intense. With few bureaucratic or legalistic constraints to action, criminal
adversaries use deception, intimidation, and violence to defeat their opponents.
Traffickers are vulnerable to theft and robbery from suppliers, customers, corrupt police
officials, even their own associates. When apprehended by police authorities, associates
may turn states evidence and share their knowledge of the enterprise with law enforcers.
In the face of such hostility, it is remarkable that some enterprises survive for months,
even years, on end. The puzzle of longevity is explained, in part, by their ability to
develop practices and procedures that minimize their exposure to hostile adversaries.
These social structures are described in detail below.
Structure of Colombian Trafficking Enterprises
Trafficking enterprises contain values, norms, roles, rules, practices, and
procedures that structure relations among participants. These relations shape how they
make decisions, perform tasks, and communicate information. Trafficking organizations
draw on routines to process narcotics, organize shipments, distribute drugs, collect and
launder proceeds, send profits back to Colombia, assess responsibility for lost shipments,
recruit new members, and perform a host of other drug-related activities.
Formalization of structure varies among enterprises. Criminal organizations that
complete multiple transactions over a number of years develop more elaborate norms,
decision-making hierarchies, and procedures than ad hoc groups that come together for
the purpose of conducting a single transaction (Caldern, 2000; Curtis & Wendel, 2000).
Routines develop through the accumulation of experience as trafficking enterprises
conduct repeated transactions over time. Organizations that transact in large quantities of
illicit drugs, such as the core cocaine enterprises, develop more elaborate structures than


74
behavior of multiple participants is interlocked (Feldman, 1989, p. 136; Hutchins,
1996, p. 50; Weick, 1979, p. 3). Routines are transmitted over time through socialization,
training and personnel movement. They are recorded in formal and informal
organizational memories. Moreover, they are independent of the individual actors who
execute them and are capable of surviving considerable turnover in individual actors
(Levitt & March, 1988, p. 320). Members may come and go and the organizations
leadership may change, but when an organization learns, knowledge is preserved in
formal and informal memories and routines that outlast individual participants. Routines
that are associated with desirable consequences, such as improved task performance, tend
to be selected and retained by organizations.5 If participants can be shown to alter a
routine or series of routines as a result of acquiring and interpreting experience, then the
organization may be said to have learned from that experience (Daft & Weick, 1984, p.
285; Fiol & Lyles, 1985, p. 804; Hedberg, 1981, p. 6; Levitt & March, 1988, p. 326; Nye,
1987, p. 381; Zeng, 1999, p. 39).
Organizations differ with respect to the extent that their behavior is determined by
formal or informal routines. Large government bureaucracies with extensive institutional
histories tend to document experience through codified rules and procedures. Small
craft-based enterprises rely more on informal understandings transmitted verbally among
participants. Organizations that face complex uncertainties or hostile task environments
rely more on informally shared understandings than organizations that function with
relatively simple, stable environments (Levitt & March, 1988, p. 327; Ouchi, 1980).
5 However, this is not always the case, as we shall see below when discussing learning under ambiguity.


275
Bureau of International Narcotics and Law Enforcement Affairs (BINLEA), U.S.
Department of State. International Narcotics Control Strategy Report, 1997.
Washington (March 1998).
Http:/Avww.state.gov/www/global/narcotics law/1997 narc report/samer97.html
[Accessed 10 July 2001].
Bureau of International Narcotics and Law Enforcement Affairs (BINLEA), U.S.
Department of State. International Narcotics Control Strategy Report, 1998.
Washington (February 1999).
Http://www.state.gov/www/global/narcotics law/1998 narc report/samer98.html
[Accessed 10 July 2001].
Bureau of International Narcotics and Law Enforcement Affairs (BINLEA), U.S.
Department of State. International Narcotics Control Strategy Report, 1999.
Washington, D.C. (March 2000).
Http://www.state.gov/www/global/narcotics law/1999 narc report/samer99 part
3.html [Accessed 5 July 2001].
Bureau of International Narcotics and Law Enforcement Affairs (BINLEA), U.S.
Department of State. International Narcotics Control Strategy Report, 2000.
Washington (March 2001).
Http://www.state.gov/g/inl/rls/nrcrpt/2000/index.cfm?docid=883 [Accessed 5
July 2001],
Bureau of International Narcotics and Law Enforcement Affairs (BINLEA), U.S.
Department of State. International Narcotics Control Strategy Report, 2001.
Washington (March 2002).
Http://www.state.gov/g/inl/rls/nrcrpt/2001/c6085.htm [Accessed 21 March 2002],
Bureau of International Narcotics Matters (BINM), U.S. Department of State.
International Narcotics Control Strategy Report, 1990. Washington (March
1991).
Bureau of International Narcotics Matters (BINM), U.S. Department of State.
International Narcotics Control Strategy Report, 1992. Washington (April 1993).
Bureau of International Narcotics Matters (BINM), U.S. Department of State.
International Narcotics Control Strategy Report, 1993. Washington (April 1994).
Bureau of Justice Statistics, U.S. Department of Justice. Felony Sentences in the United
States, 1996 (1996). Http://www.oip.usdoi.gov/bis/abstract/fsus96.htm [Accessed
15 March 2001].
Butte, Dick. El Sueo de la Tierra Propia: Un reportaje grfico sobre una familia de
colonos colombianos cultivadores de coca. Bogot: El ncora Editores, 1990.


190
Dozens of Colombian and U.S. law enforcement, military and intelligence
agencies collect, produce, and analyze relevant information about Colombian narcotics
organizations. Three separate directorates within the Colombian National Police (CNP)
produce tactical, operational, and strategic intelligence: Anti-Narcotics, Intelligence, and
the Judicial and Investigative Police. The Department of Administrative Security (DAS)
and various branches of the Colombian Armed Forces, including the Army, Navy,
Marine Infantry and Air Force, also perform intelligence functions. In recent years,
Colombian law enforcement and military organizations have combined their drug
intelligence resources through the Joint Intelligence Center. Twenty-two U.S. federal
agencies are directly involved in collecting or manufacturing domestic and foreign drug
enforcement intelligence. Table 6-2 summarizes these U.S. agencies, the type of
intelligence they produce, and the domestic and foreign scope of their intelligence
gathering activities.
The proliferation of agencies involved has led to the duplication of drug
enforcement intelligence products. One General Accounting Office report found that
four different counter-drug centers analyze air traffic data between Colombia and
Mexico. The GAO also reported that six law enforcement and military organizations
produce intelligence regarding drug trafficking along the southwest border of the U.S.
Policy makers have addressed these concerns by implementing two changes to U.S. drug
intelligence efforts. First, in 1988 lawmakers gave the Office of National Drug Control
Policy the authority to establish intelligence priorities for the drug enforcement
intelligence centers. Second, in 1993 the federal government created the National Drug
Intelligence Center (NDIC) to serve as a centralized clearinghouse for strategic drug


know who they are and I thank them. I also thank Gabriela Rovillon Acosta, Maxine
Downs, and Emilia Gioreva for their professional transcriptions of dozens of interviews.
The Center for International Security and Cooperation (CISAC) and Stanford
University provided a stimulating and supportive environment to write most of this
dissertation. For their support and friendship during my stay in Palo Alto, I thank Herb
Abrams, Chris Chyba, Nisha Fazal, Karen Guttieri, Carole Hyde, Jacques Hymans, Wu
Jun, James March, Dinshaw Mistry, Barry ONeill, Barbara Platt, Woody Powell, Scott
Sagan, Steve Stedman, and Jeremi Suri. Special thanks go to Lynn Eden, who served as
my mentor at Stanford. Lynn scrutinized all of the chapters I wrote at CISAC, always
tendering her enthusiastic support insightful criticism. Robert Axelrod, Martha Feldman,
and Richard Valcourt were kind enough to share their perceptive comments on separate
chapters.
At the University of Florida I have beneffited enormously from my association
with professors and fellow students over the past several years. My dissertation
committee chair, Philip Williams, and co-chair, Terry McCoy, guided me through
graduate school and this study. I am enormously grateful for their friendship and support.
The other members of my committee, Leann Brown, Joseph Spillane, and Larry Dodd,
provided assistance throughout the project. For additional support at the University of
Florida, I thank Leslie Anderson, Ralph DiMuccio, Errol Henderson, Lalitha Henderson,
Peter Hildebrand, Goran Hyden, Michael Martinez, Geraldine Nichols, Richard Nolan,
Hazel Phillips, Sandra Russo, Steve Sanderson, Marty Swilley, Les Thiele, and Debbie
Wallen. For their friendship during my years in Gainesville, I thank Shawn Bird, Parakh
Hoon, Ajent Shriar, Juan Carlos Valencia, Eric Cooper, Fabiano Toni, Scott Richards,
iii


194
plans and implements DEA policies regarding the collection of drug enforcement
intelligence. The El Paso Intelligence Center (EPIC) is a clearinghouse for tactical and
operational intelligence that specializes in illicit drug movement throughout the world.6
The Intelligence Division also includes a Special Field Intelligence Program that provides
funding for the collection of specialized data on drug processing methods, transportation
routes and methods, money-laundering techniques, and drug-related terrorism (GAO,
1993, 1998; Richelson, 1999, pp. 139-141).
CNP Drug Enforcement Intelligence
Traditionally, the intelligence capabilities of Colombian law enforcement
agencies have lagged far behind their U.S. counterparts. In the past U.S. officials have
been reluctant to share information with their Colombian counterparts for fear that
corruption would cause the intelligence to reach suspected traffickers. Today, bilateral
intelligence sharing has improved considerably due to a more professional Colombian
police force and a substantial effort by the CNP to upgrade its intelligence management
systems, often with U.S. assistance. In 1993, the U.S. State Department reported that
several CNP intelligence units had computerized their data collection and analysis
systems and were becoming increasingly effective in producing reliable drug
enforcement intelligence. Two years later, the CNP created a new directorate for
intelligence that it housed in a state-of-the-art intelligence center located on the outskirts
of Bogot (BINM, 1993, p. 107; Nadelmann, 1993, p. 349).
The creation of the Intelligence Directorate and its $20 million Centro de
Inteligencia reflects a broader trend within the Colombian National Police to give greater
emphasis to intelligence regarding specific threats to citizen security, such as drug
6 EPIC is described in greater detail below.


9
highlighted the adaptability of the organizations that coordinate this industry. Griffith
(1997) noted that [traffickers are not only creative, they are also adaptive, changing
methods and operatives depending on the success of counter-narcotics measures (p. 84).
Clawson and Lee argued that trafficking organizations have adapted to government
efforts to reduce their profit margins by increasing production efficiency, improving and
diversifying smuggling methods, increasing their downstream penetration of U.S. drug
markets, and diversifying products and markets (Clawson & Lee 1996, pp. 10-12).
Krauthausen and Sarmiento argued that pressure from drug enforcers compels traffickers
to invent new methods and strategies for carrying out their illicit activities make the point
that in the face of continuous law enforcement pressure (Krauthausen & Sarmiento, 1991,
p. 37).
Government agencies and officials also recognize the adaptability of transnational
trafficking groups and the threat they pose to U.S. interests. The 1998 National Security
Councils International Crime Control Strategy stated that, [international criminal
organizations are adaptive and resilient, responding quickly to effective law enforcement
pressure (NSC, 1998). A recent General Accounting Office report links traffickers
adaptability to the failure of government drug enforcement efforts:
A key reason for U.S. countemarcotics programs lack of success is that
international drug trafficking organizations have become sophisticated,
multibillion-dollar industries that quickly adapt to new U.S. drug control efforts.
As success is achieved in one area, the drug-trafficking organizations change
tactics, thwarting U.S. efforts (GAO, 1997, p. 3).
Although these studies suggest that the trafficking enterprises adapt their behavior
in response to drug enforcement activities, they do not provide a conceptual framework
for understanding this process. However, there is a multidisciplinary body of literature on


263
Name of
Date of
Type of
Trafficker
State learning?:
respondent
interview
respondent
learning?:
Affirmative,
negative, not
discussed
Affirmative,
negative, not
discussed
Fabio Castillo
3/3/00,
3/8/00
Colombian
journalist
Affirmative
Not discussed
Jos Gregorio
Prez Valdes
6/5/00,
7/16/00
Colombian
journalist
Affirmative
Not discussed
Gerardo Reyes
5/25/00
Colombian
journalist
Affirmative
Not discussed
Edgr Torres
6/15/00
Colombian
journalist
Affirmative
Not discussed
Douglas Farah
8/29/99
U.S. journalist
Affirmative
Not discussed
Tim Johnson
3/7/00
U.S. journalist
Affirmative
Not discussed
Jorge Aero [pseud]
3/21/00
Colombian
government
Affirmative
Affirmative
Edgr Barrero
[pseud]
4/3/00
Colombian
government
Not discussed
Affirmative
Alma Berquist
[pseud]
3/22/00
Colombian
government
Affirmative
Not discussed
Guillermo Brandy
[pseud]
5/2/00
Colombian
government
Affirmative
Affirmative
Jonathan Cook
[pseud]
6/18/97
Colombian
government
Affirmative
Not discussed
Jos Furden
[pseud]
6/2/00
Colombian
government
Affirmative
Affirmative
Gonzalo de
Francisco
6/8/00
Colombian
government
Affirmative
Not discussed
Lanny Gale
[pseud] & Helmer
Rodriguez [pseud]
6/6/00
Colombian
government
Affirmative
Affirmative
Leonardo Gallego
Castrilln
3/21/00
Colombian
government
Affirmative
Affirmative
Homero Herrera
[pseud]
7/14/00
Colombian
government
Not discussed
Not discussed
Maria Janeway
[pseud]
6/22/00
Colombian
government
Affirmative
Not discussed
Joaquin Linville
[pseud]
4/7/00
4/18/00
Colombian
government
Affirmative
Not discussed
Antonio Lupallo
[pseud]
7/11/00
Colombian
government
Affirmative
Affirmative
Adriana Mendoza
7/22/99
Colombian
government
Affirmative
Not discussed
Gabriel Merchn
4/6/00
Colombian
government
Affirmative
Not discussed


116
Trafficking organizations develop routines to achieve a wide variety of activities.
Enterprises have developed practices, procedures and performance programs to recruit
participants, process, transport and distribute illicit drugs, collect and launder proceeds,
evade drug enforcers, communicate information among participants, record transactions,
respond to problems, plan future activities, and determine responsibility for failed
transactions. Many routines are designed to minimize traffickers exposure to law
enforcement authorities and other adversaries (see Table 4-3).
Trafficking routines emerge over time through repeated interactions among
participants. Indeed, routines are repositories of criminal experience. Trafficking
enterprises tend to accumulate more routines over time, allowing them to expand and
diversify their performance programs. An organization with ten years experience
generally has more smuggling routes and shipping methods in its transportation repertoire
than a two-month operation. Trafficking enterprises develop practices and procedures
that allow them to achieve desired ends effectively, but not necessarily efficiently. Many
transportation routes are wildly inefficient, proceeding through multiple transit points and
shipping modalities before reaching the final destination. These roundabout routes are
designed to circumvent areas with greater law enforcement activity.
In his autobiography, Max Mermelstein provides a detailed description of an
inventory routine used by his enterprise to store cocaine in stash houses prior to
wholesale distribution. The action begins as Mermelstein and his supervisor, Rafa, arrive
at one of the organizations stash houses following the completion of separate
performance programs for transporting cocaine from Colombia to the U.S. by general
aviation aircraft and delivering the load by car to a stash house in North Miami:


195
trafficking and terrorism. Today, Colombian law enforcement and military organizations
receive and provide substantial time-sensitive intelligence. Colombian drug enforcement
agencies exchange tactical, operational, and strategic intelligence with the DEA, the
Central Intelligence Agency, the Federal Bureau of Investigations, the Bureau of
International Narcotics and Law Enforcement Affairs, the Customs Service, two Joint
Interagency Task Forces (South and East), and Tactical Analysis Teams. In recent years,
much of this bilateral information exchange has occurred through the Information
Analysis/Operations Center maintained at the U.S. Embassy in Bogot, created to collect
and process counter-drug intelligence in Colombia (GAO, 1999, pp. 134-135; Marshall,
1997, p. 68).
Gathering Intelligence through Criminal Investigations
In this section, I describe the practices and procedures drug enforcers use to
develop tactical and operational intelligence targeting specific trafficking enterprises.
Law enforcement agents draw on a variety of resources to conduct their criminal
investigations, including confidential informants, surveillance, and undercover
operations. They learn the practices and procedures of criminal investigation through
training programs and practical experience.7 Drug enforcement knowledge is codified in
training manuals and textbooks, such as The Narcotic Officers Notebook (Hamey &
Cross, 1973), the Drug Enforcement Handbook (DEA, 1987), and the Aviation Drug
Investigators Guide (FAA, 1989). These documents provide detailed instructions on
specific investigative and enforcement routines, including conducting foot and vehicular
surveillance, interviewing informants, and planning raids.
7 For example, DEA agents attend a 16-week instruction course at the Justice Training Center in Quantico,
Virginia, where they receive extensive training in all facets of drug law enforcement operations (DEA,
2001b).


Eds. Juan G. Tokatlin and Bruce M. Bagley. Bogot: Ediciones Uniandes,
(1990): 57-86.
293
Gootenberg, Paul. Introduction: Cocaine: the hidden histories. Cocaine: Global
Histories. Ed. Paul Gootenberg. London: Routledge, (1999): 1-17.
Grahn, Lance. The Political Economy of Smuggling: Regional Informal Economies in
Early Bourbon New Granada. Boulder: Westview, 1997.
Griffith, Ivelaw L. Caribbean Security: Retrospect and Prospect. Latin American
Research Review 30.2 (1995): 3-32.
Griffith, Ivelaw L. Drugs and Security in the Caribbean: Sovereignty Under Siege.
University Park, PA: Penn State University Press, 1997.
Griffith, Ivelaw Lloyd. The Geography of Drug Trafficking in the Caribbean. From
Pirates to Drug Lords: The Post-Cold War Caribbean Security Environment. Eds.
Michael C. Desch, Jorge I. Dominguez, and Andrs Serbin. Albany: SUNY Press,
(1998a): 97-119.
Griffith, Ivelaw Lloyd. Security Collaboration and Confidence Building in the
Americas. International Security and Democracy: Latin America and the
Caribbean in the Post-Cold War Era. Ed. Jorge I. Dominguez. Pittsburgh:
University of Pittsburgh Press, (1998b): 169-187.
Griffith, Ivelaw Lloyd. Associate Professor of Political Science, Florida International
University. Miami. 4 August 1999.
Gros, Christian. Los campesinos de las cordilleras frente a los movimientos guerrilleros
y a la droga: Actores or vctimas? Anlisis Poltico (16) (1992): 5-22.
Gugliotta, Guy. Black Pope of Cocaine is Captured in Davie. Miami Herald (16
December 1987). Http://www.herald.com/newslibrarv/l [Accessed 5 April 2000].
Gugliotta, Guy. "The Colombian Cartels and How to Stop Them." Drug Policy in the
Americas. Ed. Peter H. Smith. Boulder: Westview Press, (1992): 111-128.
Gugliotta, Guy and Jeff Leen. Kings of Cocaine: Inside the Medellin Cartel-An
Astonishing True Story of Murder, Money, and International Corruption. New
York: Simon and Schuster, 1989.
Gutirrez Sanin, Francisco. Politicians and Criminals: Two Decades of Turbulence,
1978-1998. International Journal of Politics, Culture and Society 14.1 (2000):
71-87.


311
Ochoa, Jorge Lus. Former drug trafficker. Frontline: Drug Wars. 2000.
Http://www.pbs.org/wgbh/pages/frontline/shows/drugs/interviews/ochoaiorge.h
ml [Accessed October 11, 2000].
Ochoa, Juan David. Former drug trafficker. Frontline: Drug Wars. 2000.
Http://www.pbs.org/wgbh/pages/frontline/shows/drugs/interviews/ochoaido.html
[Accessed December 12, 2000].
Office of National Drug Control Policy (ONDCP). The National Drug Control Strategy,
1997: FY1998 Budget Summary (February 1997).
Http://www.ncjrs.org/htm/icde.htm [Accessed 26 July 2001].
Office of National Drug Control Policy (ONDCP). National Drug Control Strategy: 2000
Annual Report: FY 2001 Budget Summary (2001).
Office of National Drug Control Policy. The National Drug Control Strategy Report:
2001 Annual Report (March 2001).
Http://www.whitehousedrugpolicv.gov/policv/ndcs.html [Accessed 16 March
2001],
Office of National Drug Control Policy (ONDCP). The High Intensity Drug Trafficking
Area Program: 2001 Annual Report (2001).
Http://www.whitehousedrugpolicv.gov/publications/enforce/hidta2001/index.html
[Accessed 3 July 2001].
Office of National Drug Control Policy (ONDCP). White House Press Release. Coca
Cultivation in Colombia, 2001. Washington, D.C. (7 March 2002).
Http://www.state.gov/g/inl/rls/prsrl/ps/8865pf.htm [Accessed 30 March 2002].
Olson, William J. International Organized Crime: The Silent Threat to Sovereignty.
Fletcher Forum of World Affairs 21.2 (1997): 65-80.
Olson, William J. Staff director, Caucus International Narcotics Control, U.S. Senate.
Washington. 20 July 1999.
Olson, William J. Intelligence Problems as they Relate to International Criminal
Organizations and Drug Trafficking. Transnational Threats: Blending Law
Enforcement and Military Strategies. Ed. Carolyn W. Pumphrey. Carlisle:
Strategic Studies Institute, (2000): 57-64
Oporto, Luis [pseud]. Intelligence Analyst, Anti-Narcotics Directorate, Colombian
National Police. Bogot. 3 May 2000.
Oppenheimer, Andres. Mexican cartels taking charge in world drug trade. Miami
Herald (15 March 1997). Http://www.herald.com/americas/digdocs/052028.htm
[Accessed 22 April 1997].


214
As early as 1982, the DEA appeared to be making progress in its efforts to target
major drug traffickers and dismantle their criminal operations. A General Accounting
Office (GAO) study published in 1984 presented data indicating that DEA agents were
devoting greater resources to complex conspiracy cases and fewer resources to the
traditional buy-bust enforcement technique. The GAO study also showed that by fiscal
year 1982 agents were arresting more high-level drug violators (see Table 6-3 below),
using more telephonic wiretaps (although information from criminal informants remained
critical for initiating investigations), and seizing greater financial assets from trafficking
organizations (GAO, 1984).
Throughout the latter half of the 1980s and into the 1990s, the DEA continued to
target trafficking enterprises through greater use of electronic surveillance technologies.
Agents developed the practice of using multiple telephone wiretaps covering different
participants to gather tactical intelligence about the modus operandi of targeted
enterprises. During the 1990s, the DEAs use of court authorized wiretaps increased
further. The GAO recently reported that the number of electronic surveillance orders
conducted by the DEA increased by 183% from FY 1990 to FY 1998; the number of
telephones, pagers, and fax machines covered by these orders increased by 158% (see
Table 6-4). In 1995, the DEA created a new program to help field agents develop
conspiracy investigations by improving the use of wiretap technologies (GAO, 1999, p.
51).12
12 According to the GAO, the program is designed to helps agents focus their intercept operations on the
best available targets, choose the best telephone numbers for intercept, correctly conduct the intercepts,
make the best use of collected information, and make the most efficient use of transcribers and translators
(GAO, 1999, pp. 49-50).


92
fewer participants that process knowledge and experience, the less likely it will be subject
to these intentional machinations and unintentional processing errors. Once a minimal
number of information processing participants has been attained, fewer information
handlers increases the speed with which information can be processed, and decisions
based on the information can be made. Finally, changes in organizational routines can be
selected or discarded more rapidly because feedback travels through fewer minds.
Levels of management influence learning indirectly through their impact on
information and decision flows. Management levels structure decision-making processes
within organizations. The taller the organization, the more management layers
information and decisions must flow. Moreover, each management level represents a
distinct channel of information processing. The taller the organization, the more
information processing channels; the more processing channels, the greater the
informations exposure to manipulation, distortion, and miscalculation, whether
intentional or not. Conversely, the flatter the organization, the fewer channels through
which information and decisions must flow, and the lesser the opportunities for
machinations and processing errors. Also, flatter hierarchies increase the speed of
information processing and decision cycles. Finally, adaptations in organizational
routines can be selected and retained more quickly in flat organizations because feedback
flows through fewer processing channels (Argyris & Schon, 1996, p. 187).
To the extent that drug trafficking enterprises and drug enforcement agencies
share competitive learning ecologies (Proposition #5, p. 14), their relative size and
decision-making hierarchies should exert a significant impact on their ability to learn
from one another. In particular, it is expected that smaller, flatter collectivities will be


117
Table 4-3 Risk Reduction Routines in Colombian Trafficking Enterprises
Compartmentalize structures and Gather information about law enforcement
communication flows activities
Limit the size of trafficking operations
(both in terms of numbers of participants
and the size and frequency of drug
shipments)
Recruit family members and close friends
as conspirators
Require potential recruits and wholesale
customers to provide information regarding
their immediate family members
Pool resources among multiple
entrepreneurs and investors (la apuntada,
described in Chapter 2)
Provide insurance against drug
enforcement interdiction
Use multiple routes and shipping
modalities simultaneously
Design transportation and delivery routines
that minimize contact among lower-level
participants
Use coded language and aliases to
communicate details of impending
transactions
Corrupt strategically placed police, military
and judicial officials
Contribute to election campaigns of
Colombian politicians
Lobby government officials over
impending legislation that affects
traffickers interests
Build grass-roots social support through
philanthropy and public works
Use threats and violence against those that
threaten the security of the enterprise
Carry out bombing campaigns,
kidnappings, assassinations and other
violent acts against current and former
government officials, their families, and
innocent civilians
Even as we were walking in the front door, the two drivers were unloading their
cars in the garage. They piled up the duffel bags on the floor of the adjacent
laundry room. Each was full of football-shaped duct-taped packages containing
one kilo of coke inside one Ziploc bag inserted into a second Ziploc bag,
reinforcing it. I followed Rafa through the dining room and kitchen to the
laundry, where he began the process of unloading the duffel bags and counting
and inspecting the packages one by one. Each individual one-kilo packet had a
marking on the tape [to identify separate owners] and the bags were stacked
neatly, separated into piles according to their markings... Rafa was carrying a
spiral notebook and he noted the number and code of each package. Then George
Bergin made his count. He represented the pilots. He confirmed Rafas count.
This was an on-the-job training exercise, and Rafa handed me his notebook. On a
clean page he had me count and note the codes of the bags of coke. Finally, since
Chava was responsible for the stash house, she too counted the bags and noted the
number and codes. She signed for delivery of them. Rafa co-signed with her.
When it was determined that all of us had arrived at the same count we went on to


16
This application begins with an organizational analysis of the illicit drug industry
in Colombia. Using an open-systems perspective of organization theory, Chapter 4
examines the tasks, environments, structures, participants, and technologies that together
form Colombian trafficking enterprises. Having delineated the structural and
environmental conditions of these criminal firms, Chapter 5 examines the extent to which
they gather, interpret, and apply information to practices, procedures, and performance
programs that guide collective behavior. In recognition that trafficking enterprises are not
the only organizational learners of interest, Chapter 6 applies the process-oriented
learning model to U.S. and Colombian drug enforcement agencies, in particular the Drug
Enforcement Administration and the Colombian National Police. The chapter describes
how these bureaucratic institutions develop, manufacture, and act on tactical and strategic
intelligence regarding trafficking enterprises. The chapter also develops the notion of
competitive learning games to describe interactions between these sovereignty-free and
sovereignty-bound actors. Chapter 7 concludes the study by summarizing the theoretical
and empirical findings of the research, and discussing the implications of competitive
learning games for U.S. and Colombian drug control policies and programs.


251
Diffusion of Trafficking Technologies
Meanwhile, the spread of knowledge and experience regarding drug trafficking
continues apace throughout Colombia and the United States. This is due to a number of
factors analyzed in this research, including contraband smuggling traditions, the practical
orientation of most trafficking knowledge, and the creation of epistemic trafficking
communities by supply-reduction policies that incarcerate large numbers of participants.
Contraband smuggling traditions in Colombia and the U.S. foster the diffusion of
values, norms, and skills associated with drug trafficking. This research does not
establish a causal relationship between Colombias long-standing smuggling traditions
and the growth of the countrys psychoactive drug industry. However, over the last
several decades it has become apparent that many contraband smugglers draw on their
knowledge and skills when expanding into psychoactive drugs. The Colombian and U.S.
governments facilitate this process with policies that encourage illegal trade in legal
commodities, such as sugar, cigarettes, and emeralds, by entrepreneurs hoping to avoid
prohibitive tariffs and restrictive quotas.
The practical, as opposed to the scientific, knowledge involve in drug trafficking
means that these symbolic technologies can be acquired many participants at a relatively
low cost. One does not need to go to graduate school to learn how to smuggle
psychoactive drugs (just to learn about learning to smuggle drugs). Common sense, a
decent memory, and some on-the-job training suffice for most participants. However, if
participants do receive advanced training in narcotics trafficking, their knowledge and
skills increase correspondingly. Supply-reduction policies that emphasize incarceration
have created hundreds of institutions of higher learning or, more accurately, epistemic


254
access to state sponsored arbitration mechanisms such as the courts, reduces incentives to
regulate transactions through coercion and violence.
It is plausible that legalization would reduce prices as well as certain types of
drug-related crime in producer and consumer countries. This is because drug prices, in
part, are a function of their illegal status, due to the high risks incurred by suppliers
(Reuter et ai, 1988; Thoumi, 1995). However, this research suggests that legalization is
unlikely to reduce levels of organized criminality in Colombia, or elsewhere. Similar to
contrabandistas in the 1970s that expanded to marijuana and cocaine markets, many
contemporary trafficking enterprises are likely to adapt a legalized drug industry by
expanding into other areas of criminal behavior, including motor vehicle theft, armed
robbery, kidnapping, and other types of contraband smuggling. In Colombia, there is
sufficient demand for these activities to support hundreds of criminal organizations.
Although data are scarce, there are indications that certain cities, such as Medellin,
witnessed an expansion in non-drug-related organized criminality, including armed
robbery and murder-for-hire, following the break-up of the cocaine cartels in the 1990s.
According to a senior advisor to the Colombian president for crime and terrorism control,
the Colombian police have identified organized bands of bank robbers that previously
had been involved in drug trafficking. Confronting the necessity of maintaining their
economic livelihood and the limited applicability of their knowledge and experience,
these groups have willingly expanded into new areas of delinquent behavior.3 The
advisor explains:
3 Something similar occurred in the United States following the repeal of the Eighteenth Amendment in
1933. Criminal organizations that expanded during the period of alcohol prohibition did not collapse from
its repeal. Instead, they diversified into other activities, including gambling, prostitution, and loan
sharking.


CHAPTER 7
CONCLUSION
Over the past several decades the governments of Colombia and the United States
have adopted a series of drug control policies they would have done better to avoid.
Known collectively as supply-reduction these programs seek to increase the costs of
psychoactive drug consumption in the U.S. by reducing their supply. Applied to
Colombian trafficking enterprises, supply-reduction programs interdict drug flows,
destroy processing labs, intercept communications, confiscate profits, and apprehend and
incarcerate participants. In the U.S. the approach targets criminal enterprises in
wholesale and retail markets, along with millions of consumers that have the temerity, or
ill fortune, to procure controlled substances under the watchful eyes of the state.
The supply-reduction approach to drug control is unlikely to yield satisfactory
policy outcomes. While social commentators have leveled a number of criticisms against
supply-reduction policies, my focus here is on the criminal organizations targeted by
these efforts. Building on the findings from this comparative study of organizational
learning by trafficking enterprises and government drug enforcement agencies, I argue
that supply-reduction is likely to fail because the former adapt their tactics and strategies
faster and more thoroughly than drug enforcers can counter them. As long as these
criminal enterprises are capable of learning from experience and information, and as long
as drug revenues remain robust, they are likely to continue their illegal activities.
However, I also maintain that policy makers and drug enforcers can undertake a variety
238


271
Arrow, Kenneth J. The Economic Implications of Learning by Doing. Review of
Economic Studies 29 (1962): 155-173.
Associated Press. Mexico breaks international drug ring. San Diego Union-Tribune (1
July 1985): A-19. Http://web.lexis-nexis.com/universe [Accessed 16 February
1999].
Associated Press. Cocaine trafficking's third generation: Young, low-key and thriving.
Cable News Network (6 June 1998a).
Http://cnn.com/WORLD/americas/9806/06/AP000502.ap.html [Accessed 7 June
1998],
Associated Press. Colombian police nab alleged drug kingpin. Cable News Network (7
June 1998b).
Http://cnn.com/WORLD/americas/9806/Q7/columbia.drug.boss.ap/index.html
[Accessed 7 June 1998].
Attorney General, U.S. Department of Justice. Annual Report of the Organized Crime
Drug Enforcement Task Force Program. Washington, 1984.
Ayoob, Mohammed. The Third World Security Predicament: State Making, Regional
Conflict, and the International System. Boulder: Lynne Rienner, 1995.
Bagley, Bruce M. U.S. Foreign Policy and the War on Drugs: Analysis of a Policy
Failure. Journal of Interamerican Studies and World Affairs 30.2-3 (1988a):
189-212.
Bagley, Bruce M. Colombia and the War on Drugs. Foreign Affairs 67.1 (1988b): 70-
92.
Bagley, Bruce M. Dateline Drug Wars: Colombia: The Wrong Strategy. Foreign Policy
77(1989-1990): 154-171.
Bagley, Bruce M. "Myths of Militarization: Enlisting Armed Forces in the War on
Drugs." Drug Policy in the Americas. Ed. Peter H. Smith. Boulder: Westview
Press, (1992): 129-150.
Bagley, Bruce M. Clintons Search for a Drug Policy for Latin America. North South:
The Magazine of the Americas 4.3 (1994): 36-38.
Bagley, Bruce M. Professor of International Studies, University of Miami. Miami. 5
August 1999.
Bagley, Bruce M. and Juan G. Tokatlian. Dope and Dogma: Explaining the Failure of
U.S.-Latin American Drug Policies. The United States and Latin America in the


57
technologies, ensuring that their smaller size does not equate with lesser sophistication.
The remainder of this study is devoted to understanding how these criminal enterprises,
and their state adversaries, draw on information and experience to alter their form and
function over time.


213
seize their illicit assets (GAO, 1979, pp. 85-86; GAO, 1984, pp. 24-25; Wilson, 1978, pp.
147-148).10
Innovations in Conspiracy Investigations
By the early 1980s, the growth of the international drug trade and the presence of
Colombian distribution groups in New York and Miami convinced the DEA and other
law enforcement bureaucracies to take these criticisms to heart. Over the next decade,
the DEA undertook a sustained effort to target major trafficking organizations through
long-term conspiracy investigations, undercover operations, electronic surveillance
technologies, and financial investigations. A number of adaptations in drug enforcement
operations were made during these years, prompting one high-ranking DEA official to
refer to the 1980s as a laboratory for experimentation in international drug law
enforcement (Westrate, 1994, p. 86).
Table 6-3 DEA Domestic Arrests by Class Level,
Fiscal Years 1979 and 1982
Class of Violator"
FY 1979
FY 1982
Percent change
I
1,005
1,340
+33
II
588
782
+33
III
4,915
6,618
+35
IV
3,831
3,443
-10
Totals
10,339
12,183
+18
10 Conspiracy investigations target high-level drug violators, including organization leaders and cell
managers. These high-ranking figures are more elusive because they insulate themselves from drug
transactions by communicating through intermediaries and compartmentalizing their operations into quasi-
autonomous cells (see Chapter 4). Conspiracy investigations seek to demonstrate a relationship between
two or more individuals engaged in drug trafficking (or other criminal behavior), even if the they never
actually handle the merchandise (GAO, 1984, p. 24, fn. 1).
11 At the time of this GAO study, the DEA used a classification system called the Geographic-Drug
Enforcement Program (G-DEP) to categorize narcotics violators into four classes according to quantitative
and qualitative indicators. The DEA considers class I and II violators to be major traffickers. Common
roles performed by violators in these two classes include: head of organization, narcotics procurer,
wholesaler, money launderer, and financier. The DEA considers class III and IV violators to be
lower-level participants. Common roles for these classes include courier, retailer, and wholesaler
(for smaller quantities of narcotics). As the narcotics trade grew in the United States during the 1970s and
1980s, the DEA was forced to revise the G-DEP criteria, increasing both the amount of narcotics for
violators in classes I, II, & III and the amount of time for which these substances are trafficked (GAO,
1984).


228
enforcers to use the most effective counter-drug strategies. The dilemma fosters
complacency and organizational inertia on the part of drug enforcement agencies, and
provides an additional advantage to their illicit adversaries.16
Smaller in Size, Flatter in Structure
Compared to drug enforcement bureaus, trafficking enterprises are relatively
small and flat. Although considerable variety exists among different enterprises, in the
post-cartel era many trafficking groups contain fewer than two dozen participants.
Larger organizations that coordinate transnational networks contain fewer than a hundred
participants, many of who participate in these operations on an ad hoc basis. Trafficking
enterprises also contain few management layers. Mom and pop operations may feature
only two levels of decision-making hierarchy: the boss and numerous subordinates. Even
larger, more bureaucratic enterprises, such as the core organizations, generally contain
no more than four management levels (see Chapter 4).
Federal drug enforcement agencies are larger and more bureaucratic. The DEA
and CNP contain thousands of participants organized within numerous management
levels. In fiscal year 2000, the Drug Enforcement Agency contained more than nine
thousand employees, including over 4,500 Special Agents (see Table 6-5 below). In
1997, the Colombian National Police contained over ninety thousand uniformed officers,
approximately two thousand of which worked in the Anti-Narcotics Directorate (Cook,
1997; Llrente, 1999, p. 470).
16 The drug enforcement dilemma also presents a principal-agent problem to politicians that oversee law
enforcement bureaucracies. In this case, the interests of the agent (drug enforcement agencies) are not
always consistent with the interests of the principal (politicians with oversight authority). Moreover,
information asymmetry exists between the two actors, and it is difficult for the principal to acquire the
accurate and reliable information from the agent regarding the latters task performance. On the principal-
agent problem, see Moe (1984).


75
Research Propositions
Organizational learning refers to the process by which participants gather,
interpret, and apply information on behalf of an organization. The application of
knowledge and experience occurs through changes in informal and formal routines that
guide participants behavior. This study applies a process-oriented model of
organizational learning to two units of analysis: Colombian drug trafficking enterprises
and U.S. and Colombian drug enforcement agencies.
If participants within either organization alter their routines as a result of
acquiring and interpreting knowledge and experience, then the enterprise in question may
be said to have learned. In trafficking enterprises, changes in organizational practices
and procedures may involve the production, transportation, and distribution of illicit
drugs, or money laundering activities. For law enforcement agencies, changes in
organizational practices and procedures may involve criminal investigations or drug
enforcement operations. For both organizations, changes in routines may improve
organizational task performance, but they are not required to do so.
These ideas yield the following empirically verifiable propositions.
PI: If participants in Colombian drug trafficking enterprises gather, interpret, and
apply information to collective behavior by changing existing routines or creating
new ones, then organizational learning has occurred.
P2: If participants in U.S. and Colombian drug enforcement agencies gather,
interpret, and apply information to collective behavior by changing existing
routines or creating new ones, then organizational learning has occurred.
Alternative Explanations for Organizational Change
Learning is not the only source of change in organizations, and it is important to
clarify alternative explanations for this phenomenon. Perhaps the most common source


268
SNOWBALL RESPONDENT BASE
1. Can you suggest anyone else with whom you think it would be useful for me to
talk to about this topic in Washington D.C.? Miami? Colombia?


133
several hundred pounds of cocaine between Colombia and Puerto Rico. These vessels
do not operate completely under water, but skim along the surface of the sea. As with
low profile vessels, the semi-submersibles are virtually undetectable by radar (Farah &
Kovaleski, 1998; DEA, 1994, p. 7).24 In September 2000, Colombian authorities
discovered a partially finished submarine for transporting narcotics outside of Bogot. If
completed, the 100-foot vessel, based on Russian submarine technology, would have
included a computer navigation system and engine room, and been capable of
transporting up to ten metric tons of cocaine to the U.S. (BINLEA, 2001; CNN, 2000; El
Tiempo, 2000d; CM&, 2000).
Practical vs. Scientific Knowledge
In addition to material technology, Colombian trafficking enterprises draw on
knowledge-based expertise in conducting their illicit activities. In analyzing the symbolic
inputs used by these organizations, it is useful to distinguish between practical and
scientific knowledge. Practical knowledge refers to general information and
understanding necessary to carry out every day activities, such as talking on a telephone
or driving a car. Scientific knowledge refers to technical or symbolic erudition regarding
a particular phenomenon, such as the genetic composition of human chromosomes or
machinery production in 19th Century England. Practical expertise is obtained through
on-the-job training and the accumulation of direct experience. Scientific expertise is
acquired over many years, generally through formal educational training.
23 As noted in Chapter 2, Cervantes reports that some traffickers were using semi-submersibles to transport
marijuana in the late 1970s (Cervantes, 1980, pp. 80-81).
24 One DEA intelligence analyst interviewed for this research points out that semi-submersible vessels are
not widely among Colombian narcotics organizations (Evans, 1999).


81
curvewas subsequently found in a number of manufacturing industries and other
production tasks involving repetition, although learning rates varied considerably across
industries, products and time (Levitt & March, 1988, p. 321; Epple, Argote & Devadas,
1996; Yelle, 1979).9
One problem with the efficiency conception of learning is that it disregards
instances whereby changes in routines based on the acquisition and interpretation of
information fail to improve task performance. As numerous organization theorists have
recognized, learning does not always increase effectiveness (Cook & Yanow, 1996;
Hedberg, 1991, p. 89). Like individuals, organizations may learn dysfunctional behavior,
as when they adopt routines that hinder task performance or require more organizational
resources to achieve the same outcome. Productive learning is an important subspecies
of organizational learning. However, analysts should distinguish clearly between the two
varieties and avoid defining the latter by the former (Argyris & Schon, 1996).
Organizational learning may improve task performance, and it may not.
Levels of Learning: Tactical vs. Strategic Adaptation
A number of analysts have identified different degrees or levels of
organizational learning. While a variety of terminology has been developed to explain
this process the basic idea revolves around the distinction between the goals and
objectives of organizational behavior. Simple, low or lower, single-loop
learning and adaptation describe the process by which organizations adjust the means
9 Moreover, in some industries production costs have actually increased over time and accumulated
experience, thus demonstrating little capacity for learning. The commercial nuclear reactor industry is one
notableand frighteningexample (Perrow, 1984, pp. 34-35; Bupp & Derian, 1978).


171
they change their strategies, long-range objectives, and basic organizational structures as
a result of acquiring and interpreting information. The data gathered in this research
suggest that some Colombian enterprises adjust their strategies in response to feedback.
Prominent examples of strategic learning include diversifying into new products and
markets, and restructuring criminal operations. I examine both of these below.
Diversifying into New Products and Markets
On numerous occasions Colombian trafficking organizations have expanded into
new markets and products. In the 1960s and 1970s, contraband smuggling groups
expanded into marijuana exports, fueling the short-lived bonanza marimbera. By the late
1970s, some marijuana smugglers diversified their product lines to include methaqualone
and cocaine, in effect becoming polydrug organizations.11 Smugglers kept transportation
costs low by exporting these less bulky psychoactive substances with their marijuana
shipments. Some traffickers required their wholesale customers to purchase fixed
quantities of their new commodities along with the more popular marijuana.
By the mid-1980s, Colombian methaqualone and marijuana declined, in part due
to government interdiction. However, the Colombian drug industry experienced
considerable growth as the amount of U.S. and European bound cocaine exports rose
substantially. In the middle of the cocaine boom, numerous trafficking enterprises
diversified into the heroin trade. Similar to the marimberos, cocaine traffickers
piggybacked heroin on their cocaine shipments and required U.S. wholesalers to accept
small amounts of the new product. Using distribution systems developed by the core
11 Methaqualone is a nonbarbiturate sedative better known by the trade names Quaalude, Mandrax, and
Sopors. In the 1970s it was manufactured in Germany and Hungary and imported to Colombia in powder
form, where it was packaged and re-exported to the U.S. There are also reports that the drug was at least
partly manufactured within Colombia by the countrys legitimate pharmaceutical industry (Abadinsky,
1997, p. 387; New York Times, 1980; Reuter, 1992, p. 170; Thoumi, 1995, p. 127).


OUTSMARTING THE STATE:
A COMPARATIVE CASE STUDY OF THE LEARNING CAPACITY
OF COLOMBIAN DRUG TRAFFICKING ORGANIZATIONS AND
GOVERNMENT DRUG ENFORCEMENT AGENCIES
MICHAEL C. KENNEY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENT FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2002


I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.
Ph^V^Hiams, Chair
Professor of Political Science
I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.
Terry L. VlcCoy, Cocchair
Professor of Latin American Studies
I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.
M^Leann Brown
Associate Professor of Political
Science
I certify that I have read this study and that in my opinion it conforms to acceptable
standards of scholarly presentation and is fully adequate, in scope and quality, as a
dissertation for the degree of Doctor of Philosophy.
Larry Dodd
Professor of Political Science


272
1990s: Beyond the Cold War. Eds. Jonathan Hartlyn, Lars Schoultz, & Augusto
Varas. Chapel Hill: University of North Carolina Press (1992): 214-234.
Bajak, Frank. Drug Traffickers Change Tactics. Associated Press (4 January 1997).
Http://www.washingtonpost.com/wp-srv/WAPO/l 9970104/V000656-010497
idx.html [Accessed 7 January 1997].
Bajak, Frank. Drug Busts in Colombia, Caribbean. Washington Post (13 October
1999).
Http://www.washingtonpost.com/wpsrv/aponline/l 9991013/aponline 183651 00
htm
Baldwin, David A. Ed. Neorealism and Neoliberalism: The Contemporary Debate. New
York: Columbia University Press, 1993.
Bandera, Ernesto [pseud]. Head, Eradication of Illicit Crops Program, Anti-Narcotics
Directorate, Colombian National Police. Bogot. 3 April 2000.
Barnard, Chester I. The Functions of the Executive. Cambridge, MA: Harvard University
Press, 1938.
Barr, Stephen. Aiming to Enforce Change at Customs Shake-Up Underway.
Washington Post (17 February 1999): A15.
Http ://search. washingtonpost.com/wp-srv/WPlate/1999-02/17/0051-021799-
idx.html [Accessed 17 February 1999],
Bates, Robert H., Avner Greif, Margaret Levi, Jean-Laurent Rosenthal, and Barry R.
Weingast. Analytic Narratives. Princeton: Princeton University Press, 1998.
Battard, Robert N. Prepared Statement of Robert N. Battard. Cocaine and Marijuana
Trafficking in Southeastern United States. Hearings before the Select Committee
on Narcotics Abuse and Control, U.S. House of Representatives. Ninety-fifth
Congress, second session (9-10 June 1978). Washington: U.S. Government
Printing Office.
Bearak, Barry. Mother Delivered Drug for Trip to Disneyland, Witness Says. Los
Angeles Times (12 March 1988). Http://web.lexis-nexis.com/universe [Accessed 5
July 1999],
Bejarano, Ana Maria and Renata S. Bonnett. El fortalecimiento selectivo del estado
durante el frente nacional. Controversia (169) (1996): 9-35.
Beilis, David J. Heroin and Politicians: The Failure of Public Policy to Control
Addiction in America. Westport: Greenwood Press, 1981.


218
contests feature goal-oriented, rationally bounded players (i.e. nares and narcos) that use
different strategies to outwit their adversaries and attain organizational objectives.
Competitors engage in multiple interactions, endowing the contests with a repeated-play
quality. Learning games feature multiple equilibria. Players seek to improve their
positions by changing strategies and contests yield a variety of outcomes. Trafficking
groups may partially or completely avoid drug enforcement efforts, or they may fall
victim to surveillance and disruption. Drug enforcement agencies may partially or
completely dismantle trafficking enterprises, or they may fail to identify participants.
Players understand the basic rules of competitive learning games, along their
adversaries and their own first order preferences or goals. Traffickers understand that
their goals are to produce, transport, and/or distribute sufficient quantities of
psychoactive drugs as to yield satisfactory inducements (see Chapter 4). Furthermore,
they understand that these activities have been outlawed by nation-states, and that their
environments contain dozens of drug enforcement agencies that seek to identify and
dismantle them. Drug enforcers understand that their goals are to identify and dismantle
trafficking enterprises. They are also aware that their adversaries know they exist and
seek to evade or undermine their efforts. For both nares and narcos these first-order
preferences are stable. They do not change during the course of the game.
However, second-order preferences, which consist of the strategies players use to
achieve their first-order preferences, are not fixed and may change throughout the game.
intendedly rational, they lack access to perfect or complete information, and their second-order preferences
change. Moreover, the unit of analysis in competitive learning games is the organization, not the
individual, although I agree with Simon that a theoretical understanding of organizational behavior can be
built on individualistic foundations. I present the competitive learning game as a metaphor for thinking
about narco-narc interaction rather than a fully developed theory. In developing the framework, I have
been influenced by Andreas (2000), Farkas (1998), Geddes (1995), Levitt & March (1988), Mares (1992),
Moe (1984), Simon (1997), and Snidal (1986).


215
Table 6-4 Number of DEA Electronic Surveillance Orders
Conducted and Facilities Covered, FY 1990-1998
Fiscal Year
Surveillance
orders
Facilities (telephones, pagers
and fax machines) covered
1990
223
219
1991
256
311
1992
332
302
1993
320
320
1994
357
339
1995
330
284
1996
546
584
1997
592
544
1998
631
564
Source: GAO (1999).
Innovations in Undercover Operations
During the past two decades the DEA has developed numerous innovations in its
undercover operations and other enforcement programs.13 The traditional undercover
buy-bust routine has been amended in a variety of ways, allowing DEA agents to target
upper-echelon traffickers more effectively. In one adaptation, reverse undercover
stings, the role of the undercover agent is changed from a buyer to a supplier of illicit
drugs. By posing as large-scale drug suppliers in numerous operations, agents have been
able to arrange transactions with kingpin-level traffickers that rarely meet with new
customers. Following negotiations over price and quantity, major traffickers have been
arrested on charges of attempting to distribute narcotics and criminal conspiracy, and
their vehicles, money, and other assets confiscated (Westrate, 1994, p. 89).
The reverse undercover sting proved so successful in apprehending higher-level
figures and confiscating their assets that the routine was expanded and further modified
13 The following paragraphs draw heavily on a study by David Westrate, a high-ranking DEA official that
served separate stints as the Assistant Administrator for Operations and the Assistant Administrator for
Intelligence. In the first position, Westrate was responsible for the DEAs day-to-day activities,
domestically and internationally. In the second position, he was in charge of worldwide DEA intelligence
activities.


137
While this chapter has deepened our understanding of the organizational forms
that coordinate the Colombian drug industry, my task is far from complete. If trafficking
enterprises constitute social collectives, it remains to be seen whether and to what extent
they learn as organizations. Do participants acting on behalf of these criminal enterprises
receive, interpret, and apply information by changing the organizational routines that
structure their behavior? In the next chapter, I attempt to answer this difficult question.


79
technologies believed to be worthy of emulation. Agents of diffusion include participants
that communicate with inter-connected organizations, including competitors, and
independent consultants that contain the requisite knowledge and experience (DiMaggio
& Powell, 1983, pp. 150-153; Levitt & March, 1988, p. 329; March, 1999, p. 45; Scott,
1998, p. 213).
Ecologies of learning develop when multiple organizations learn from each
other within a shared environment. The classic type is a collection of competitors, in
which rivals are closely linked through the diffusion of experience and the effects of their
activities on one another. In this type of learning ecology, organizations compete not
only for resources, customers, and profits but for political power and social legitimacy.
Organizations learn from their competitors by analyzing their competitors reactions to
their own actions (DiMaggio & Powell 1983, p. 150; Hedberg 1981, p. 13; Levitt and
March 1988, pp. 329, 331-332).
P5: Colombian drug trafficking enterprises and U.S. and Colombian drug
enforcement agencies learn from each other within common competitive learning
ecologies.
Competitive learning ecologies complicate the task of analysis. Because
organizations in these environments draw on their competitors for knowledge and
experience, learning by the one cannot be fully understood without reference to learning
by the other. In recognition of this complication, this study analyzes interactions between
trafficking enterprises and drug enforcement agencies to determine whether they learn
from each other (see Chapter 6).


307
Migdal, Joel S., Atul Kohli, and Vivienne Shue. Eds. State Power and Social Forces:
Domination and Transformation in the Third World. Cambridge: Cambridge
University Press, 1994.
Milford, James. Prepared statement by James Milford, Deputy Administrator, Drug
Enforcement Administration. Western Hemisphere. Hearing before the U.S.
House International Relations Committee, Subcommittee on the Western
Hemisphere. (16 July 1997).
Http://www.usdoi.gov/dea/pubs/cngrtest/ct970716.htm [Accessed 29 October
2000],
Miller, Maijorie. A Coke Runner with a Conscience. Book review. Los Angeles Times
(19 December 1989). Http://web.lexis-nexis.com/universe [Accessed 10
September 2000].
Mills, James. The Underground Empire: Where Crime and Governments Embrace.
Garden City, NY: Doubleday & Company, 1986.
Minority Staff Statement. Structure of International Trafficking Organizations. Hearings
before the Permanent Subcommittee on Investigations of the Committee on
Governmental Affairs, U.S. Senate. One Hundred First Congress, first session
(12-13 September 1989). Washington: U.S. Government Printing Office.
Moe, Terry M. The New Economics of Organization. American Political Science
Review 28.4 (1984): 739-777.
Molano, Alfredo. Selva adentro: una historia oral de la colonizacin del Guaviare.
Bogot: El ncora Editores, 1987.
Molano, Alfredo. Aguas Arriba: entre la coca y el oro. Bogot: El ncora Editores,
1990.
Molano, Alfredo. Rebusque Mayor: Relatos de muas, traquetos y embarques. Bogot: El
ncora, 1997.
Molano, Juan [pseud]. Head, Narcotics Unit, Judicial Police Directorate, Colombian
National Police. Bogot. 21 July 2000.
Moltz, James Clay. 1993. Divergent Learning and the Failed Politics of Soviet
Economic Reform. World Politics 45 (January): 301-325.
Montoya, Carlos [pseud]. Head, Judicial Police Unit, Anti-Narcotics Directorate,
Colombian National Police. Bogot. 4 April 2000.


240
this body of literature, developed to understand formal organizations, is applicable to
collective forms that produce, transport, and distribute psychoactive drugs in Colombia
and the U.S. In Chapter 4,1 establish the validity of learning theory to Colombian
trafficking enterprises by examining how they function as social organizations. These
target-oriented entities provide illicit goods to achieve satisfactory profits and other
benefits. To minimize their exposure to hostile task environments, they contain flat
decision-making hierarchies, compartmentalized cells, and complex performance
programs. While participants make valuable contributions to trafficking enterprises they
also represent a source of great risk and uncertainty. Leaders attempt to reduce these
risks by monitoring and controlling participants through intricate routines and constant
communication. Participants draw on a number of material and symbolic technologies to
conduct their criminal activities. The practical, as opposed to scientific, nature of most
trafficking technologies facilitates their diffusion amongst a wide number of individuals
and organizations.
If Chapter 4 establishes that many Colombian trafficking groups contain the
constitutive elements of social organizations, Chapter 2 indicates that the psychoactive
drug industry is a bubbling, seething cauldron of motion, oxidation, and regeneration.
Change is a permanent fixture in an industry where the only certainty is uncertainty.
Over the past decades Colombian narcotics organizations have undergone many changes.
In the process, they have evolved from sporadic smugglers of European produced
narcotics to leading global producers and transporters of cocaine and heroin. The states
role in this progression cannot be denied. For many years, the governments of Colombia
and the United States downplayed, denied, and disregarded the growth of trafficking


25
traffickers located in Havana. At least a few Colombian traffickers had expanded into
drug production.
Other Phase Two Trafficking Groups
The Herran Olazaga brothers were not the only drug smugglers operating in the
Antioquia region during this period. According to Arango and Child, by the mid-1950s a
number of Colombian contrabandistas decided to enter the drug trade. In making the
shift from black market cigarettes, alcohol, and domestic appliances to drugs, these illicit
entrepreneurs drew on existing resources, including their practical knowledge of
smuggling methods and Caribbean maritime routes, investment capital, Cuban-based
mafia contacts, and Colombian-based drug chemists (Arango & Child, 1984, pp. 165-
166).
In the early 1960s, the Venezuelan press began reporting about an alarming
invasion of cocaine, heroin, and marijuana, much of it originating in Colombia. A
number of smuggling rings transported drugs from Colombia into Venezuela by means of
go-fast motor boats. At least one group of Colombian smugglers was affiliated with the
well-known Italian mafiosi, Lucky Luciano. Unlike other rings, Lucianos group
preferred to move its contraband into Venezuela through ground transportation. The
cross border traffic was facilitated by numerous well-equipped processing laboratories in
Medellin run by German technicians (El Espectador, 1961a, p. 3).
While the Venezuelan press was reporting about the invasion of Colombian drugs,
El Espectador published several articles detailing the growing traffic and consumption of
marijuana in Colombia. Although marijuana had been produced in Colombia since the
times of the Spanish conquest, consumption of the drug was generally concentrated


206
centers overwhelmed the law enforcement capacities of local police officials. To respond
to this strategic innovation by trafficking enterprises, the DEA established mobile
Regional Enforcement Teams (RETs) that target major trafficking organizations
operating in small towns and rural areas, where there is a lack of sufficient drug
enforcement resources. These highly trained, well-equipped units seek to identify and
dismantle the trafficking enterprises before they become entrenched in these
communities (Butterfield, 2002; DEA, 1999, pp. 53-54).
Reforming the Colombian National Police
Colombian law enforcement agencies have also implemented a number of
administrative reforms in recent decades with the purpose of modernizing bureaucratic
hierarchies, improving task performance, and responding to changing patterns in criminal
delinquency. Perhaps no law enforcement bureau has experienced greater structural
change over the past forty years than the Colombian National Police. Since 1966, the
CNP has experienced seven administrative reorganizations, including four since 1993.
The first two reorganizations were designed to remove the CNP from the command of
Colombias armed forces and affirm its status as an independent bureau.8 Subsequent
reforms sought to modernize the agencys administrative structures, reduce human rights
abuses and drug-related corruption, and improve service delivery. Frequently, changes in
administrative structures and routines were made in response to political pressure from
domestic and foreign sources.
8 In the first half of the 20th Century, Liberal and Conservative governments in Colombia politicized and
corrupted the National Police by filling the rank and file with members of their own party whenever they
assumed control of the national government. In 1951, the chief of the Colombias armed forces, General
Rojas Pinilla, who seized presidential power in 1953, incorporated the National Police under the direction
of the General Command of the Colombian Armed Services (Torres Velasco, 1994, pp. 181-183).


216
to exploit the tendency of many trafficking enterprises to out-source the provision of
goods and services. Enforcement units posed as drug transporters and off-loaders,
suppliers of precursor chemicals and other processing inputs, money launderers and
financial advisors, and providers of airplanes, ships, automobiles, computers, generators,
microwave ovens, and other mechanical equipment. Long-term reverse sting operations,
including Operations Green Ice, Green Ice II, Grouper, Juno, Pisces, Polarcap, and
Swordfish, dismantled numerous Colombian trafficking and money laundering
enterprises and produced substantial policy outputs, including hundreds of arrests,
thousands of kilos of illegal drugs, and millions of dollars in illicit profits (Clawson &
Lee, 1996; DEA, 1999, [no date] DEA History, 1994 1998; Westrate, 1994, pp. 88-90).
In recent years, the DEA and the Customs Service also increased their use of
controlled delivery routines in which detected drug shipments are allowed to proceed
through international shipping routes to consumer markets. Drug enforcement agents
closely monitor the shipments in order to identify and apprehend the maximum number
of participants in the conspiracy. Following a number of successful operations based on
the controlled delivery of marijuana, cocaine, and heroin shipments, the enforcement
routine was expanded to include precursor chemicals, dirty money, and mechanical
equipment. These routines have evolved into complex performance programs requiring
detailed planning and coordination among numerous law enforcement agencies from
different countries. Experimentation with controlled deliveries and undercover
operations have allowed the DEA to improve its drug enforcement operations
considerably since the 1970s (Westrate, 1994, p. 93).


103
National Security Council, the Customs Service, the Financial Crimes Enforcement
Network, the U.S. Attorney offices, and several inter-agency task forces such as the
Organized Crime Drug Enforcement Task Forces, the Joint Inter-Agency Task Force, and
the DEAs Special Operations Division, not to mention hundreds of state and local police
organizations. In addition, agencies from the departments of Defense and State
participate in drug eradication and interdiction efforts in Colombia and the Caribbean,
particularly the Bureau of International Narcotics and Law Enforcement Affairs, the
Coast Guard, the Army Special Forces, and the U.S. Southern Command (see Table 4-1).
With the combined efforts of these organizations, the U.S. government represents a
formidable foe for trafficking enterprises operating in the U.S. In South Florida alone, an
area where Colombian trafficking enterprises have traditionally maintained a strong
presence, there are over twenty federal law enforcement organizations and 120 state and
local police agencies (Hutchinson, 2001; U.S. Attorneys Office [no date]).
Outside of warfare, it is hard to imagine a more hostile relationship than that
between drug traffickers and drug enforcers. Law enforcement agencies exist to
dismantle trafficking groups. The latter survive only to the extent that they can avoid or
re-direct the intentions of the state. Indeed, relationships between these actors fit the
actual definition of warfare: drug enforcers and drug traffickers engage in armed
interactions for the purpose of undermining or destroying each other. Police agencies
monitor the telecommunications of suspected traffickers, conduct undercover sting
operations, raid stash houses, destroy drug processing laboratories, coerce business
associates to turn against their fellow conspirators, and hunt down and kill leading


321
Serrano Cadena, Rosso Jos con la colaboracin de Santiago Gamboa. Jaque mate: De
cmo la polica le gan la partida a "el ajedrista y a los carteles del
narcotrfico. Bogot: Editorial Norma, 1999.
Shannon, Elaine. Desperados: Latin Drug Lords, U.S. Lawmen, and the War America
Can't Win. New York: Viking, 1988.
Shelley, Louise. Transnational organized crime: an imminent threat to the nation-state?
Journal of International Affairs 48. 2 (1995): 463-489.
Sheridan, Brian E. Prepared Statement of Department of Defense Coordinator for Drug
Enforcement Policy and Support. The Narcotics Threat from Colombia. Hearing
before the Subcommittee on Criminal Justice, Drug Policy, and Human Resources
of the Committee on Government Reform, U.S. House of Representatives, One
Hundred Sixth Congress, first session (6 August 1999). Washington: U.S.
Government Printing Office.
Sierra, Luz Mara. Investigan 13.341 rdenes de libertad. El Tiempo (3 May 2001).
Http://eltiempo.terra.com.co/03-05-2001/iudi O.html [Accessed 3 May 2001].
Sikkink, Kathryn. Development Ideas in Latin America: Paradigm Shift and the
Economic Commission for Latin America f International Development and the
Social Sciences. Ed. Frederick Cooper. Berkeley: University of California Press,
(1997): 228-256.
Simon, Herbert A. A Behavioral Model of Rational Choice. Quarterly Journal of
Economics 69 (1955): 99-118.
Simon, Herbert A. The Architecture of Complexity. General Systems: Yearbook for the
Society of General Systems Research 10. Ed. Ludwig von Bertalanffy and Anatol
Rapoport. Ann Arbor: Society for General Systems Research (1965): 63-76.
Simon, Herbert A. Bounded Rationality and Organizational Learning, Organizational
Learning. Eds. Michael D. Cohen and Lee S. Sproull. Thousand Oaks: Sage,
(1996): 175-187.
Simon, Herbert A. Administrative Behavior: A Study of Decision-Making Processes in
Administrative Organizations. 4th edition. New York: The Free Press, 1997
[1945].
Sims, Jennifer. What is Intelligence? Information for Decision Makers, U.S.
Intelligence at the Crossroads: Agendas for Reform. Eds. Roy Godson, Ernest R.
May, and Gary Schmitt. Washington: Brasseys, (1995): 3-16.
Smitten, Richard. The Godmother. New York: Pocket Books, 1990.


261
In spite of his knowledge, he failed to fully appreciate the implications of organizational
learning on his agencys efforts to disrupt these criminal enterprises. In the concluding
chapter to this study, I have discussed a number of possible improvements that could be
made to existing drug enforcement programs. However, it could be that the best drug
enforcers can hope for is to increase their policy outputs, while ultimately failing to yield
satisfactory outcomes. Instead of destroying the Colombian drug industry, drug
enforcement initiatives will continue to create a hostile environment in which trafficking
enterprises must adapt their tactical and strategic routines in order to survive. As long as
the international traffic in controlled substances remains profitable, criminal enterprises
based in Colombia will organize to benefit from this lucrative commerce.


305
Martin, John M. and Anne T. Romano. Multinational Crime: Terrorism, Espionage,
Drug & Arms Trafficking. Newbury Park: Sage, 1992.
Martin, Sam. Testimony of Chief Sam Martin. Cocaine and Marijuana Trafficking in
Southeastern United States. Hearings before the Select Committee on Narcotics
Abuse and Control, United States House of Representatives. Ninety-fifth
Congress, second session (9-10 June 1978). Washington: U.S. Government
Printing Office.
Martz, John D. The Politics of Clientelism: Democracy & the State in Colombia. New
Brunswick: Transaction. 1997.
Massing, Michael. The Fix. Berkeley: University of California Press, 1998.
Mastrofski, Stephen and Gary Potter. Controlling Organized Crime: A Critique of Law
Enforcement Policy. Criminal Justice Policy Review 2.3 (1987): 269-301.
Matthiesen, Tatiana. El arte politico de conciliar. Bogot, Colombia: FESCOL, 2000.
May, Ernest R. Lessons of the Past. New York: Oxford University Press, 1973.
McAndrew, Duncan. The Structural Analysis of Criminal Networks. The Social
Psychology of Crime: Groups, Teams and Networks. Eds. David Canter and
Laurence Alison. Aldershot: Dartmouth, (2000): 53-94.
McCluskey, Karyn and Sarah Wardle. The Social Structure of Robbery. The Social
Psychology of Crime: Groups, Teams and Networks. Eds. David Canter and
Laurence Alison. Aldershot: Dartmouth, (2000): 249-285.
McCoy, Jennifer L. Ed. Political Learning and Redemocratization in Latin America: Do
Politicians Learn from Political Crises? Coral Gables: North-South Center Press,
2000.
McDonough, James R. Phil Williams Views on Combating International Organized
Crime: Another Perspective. Transnational Threats: Blending Law Enforcement
and Military Strategies. Ed. Carolyn W. Pumphrey. Carlisle: Strategic Studies
Institute (2000): 213-219.
McFeatters, Ann. War on Terrorism Hobbles War on Drugs. Pittsburgh Post-Gazette
(7 November 2001). Http://web.lexis-nexis.com/universe [Accessed 10 January
2002],
McFeely, Richard A. Enterprise Theory of Investigation. FBI Law Enforcement
Bulletin 70.5 (2001): 19-25. Http:/Avww.fbi.gov/publications/leb/2001/leb01.htm
(Accessed 14 July 2001).


Large trafficking organizations often contain three or four management layers,
composed of the following: (1) leader(s), (2) cell manager/chief overseas representative,
(3) assistant managers/cell section leaders, and (4) cell or field workers. Many operations
lack assistant managers, reducing the number of management layers to three. In these
operations, field workers report directly to the cell manager. In other enterprises, an
intermediary, sometimes designated as exportation manager, insulates the leader by
communicating directly with cell manager(s). This person may or may not exercise
significant decision-making authority independent of the leader. A senior-level advisor
or investor with long-standing experience in the drug trade may also assist the enterprise
leader. However, this person only serves in an advisory capacity, providing information
and contacts to the leader (Brandy, 2000; Fuentes, 1998; Natarajan, 2000; Thompson,
2000).11
Whether they contain three or four management layers, Colombian trafficking
enterprises are organizationally flat, facilitating rapid information flows and decision
cycles. Leaders decisions are communicated quickly to managers and participants, and
the knowledge contained in new practices, rules, and procedures diffuses rapidly through
the enterprise. Flat hierarchies also limit opportunities for information suppression or
manipulation. There are fewer managers to inflict their biases or deliberately distort
communication from enterprise leaders. Few information choke points and quick
decision cycles translate into rapid response rates. Flat trafficking enterprises are
structurally suited to quickly adapting their practices and procedures in response to
feedback, an essential attribute for organizations operating in hostile environments.
11 Numerous investors are former leaders of their own enterprises that have subsequently retired from
direct participation in the business.


11
States and possess formal legal standing. Organizations that operate outside the rule of
lawsuch as criminal enterprises and terrorist networkshave received scant attention
from students of organizational learning. My goal is to develop a general theoretical
explanation for how the learning capacities of these sovereignty free actors facilitate their
defiance of state prerogatives, undermining governments ability to provide stability,
order, and security for its citizens.
This research is also relevant to U.S. security interests, broadly defined.9 The
research has significant implications for existing U.S. drug and crime control policies in
Colombia and other countries where transnational criminal organizations have become
institutionalized, including Mexico, Russia, Italy, and Nigeria. The ability of criminal
enterprises to alter their organizational structures and operations in response to law
enforcement suggests that headhunting approaches to drug control, as exemplified in the
Kingpin strategy, are unlikely to achieve satisfactory outcomes. This is not to suggest
that such programs fail to impact trafficking organizations. Indeed, the Kingpin program
demonstrated that when U.S. law enforcement and intelligence agencies coordinated
effectively with their Andean counterparts cartel leaders could be apprehended and their
9 Some observers have legitimately questioned the trend by international relations scholars of labeling social
problems such as the drug trade and crime security issues. As Michael Desch notes, Broadening the
definition of security in recent years was the result of the desire of many scholars and practitioners to direct
more attention and resources to serious problems that previously had not received adequate consideration or
money. As a rhetorical device for energizing scholarly and governmental discussion of, and directing
attention to, the economic, environmental, social, and political problems affecting the post-Cold War
Caribbean, this broader definition of security seems reasonable. But whether this rhetorical strategy will
actually lead to a concerted effort to deal with these problems is unclear (Desch 1998, p. 146). This point
is well taken. However, two recent developments in Colombia and the U.S., have heightened the
importance of this study to U.S. security concerns. The first development refers to the changing nature of
the long-standing social conflict in Colombia, in which armed insurgents and paramilitaries have increased
their participation in the drag trade in order to finance the escalation of their violent confrontation with each
other and the Colombian state. This development produced alarm among Washington policymakers and led
to substantial U.S. assistance for the Pastrana administrations controversial supply-reduction strategy, Plan


327
Congress, second session (9 March 1984). Washington: U.S. Government
Printing Office.
U.S. Senate. DEA Oversight and Authorization for Fiscal Year 1986. Hearing before the
Subcommittee on Security and Terrorism of the Committee in the Judiciary.
Ninety-ninth Congress, first session (19 February 1985). Washington: U.S.
Government Printing Office.
U.S. Senate. The Transit Zone: Strategy and Balance. Hearing before the Senate Caucus
on International Narcotics Control. One Hundred and Seventh Congress, first
session (15 May 2001). Washington: U.S. Government Printing Office.
Http://drugcaucus.senate.gov/hearings/transit/ [Accessed 19 June 2001].
U. S. Sentencing Commission. Annual Report (1998).
Http://www.ussc.gov/ANNRPT/1998/ar98toc.htm [Accessed 15 March 2001],
Uribe Ramirez, Sergio. Los Cultivos Ilcitos en Colombia. Drogas Ilcitas en
Colombia: Su impacto econmico, poltico y social. Ed. Francisco Thoumi.
Bogot: Planeta Colombiana Editorial, (1997): 35-135.
Uribe Ramrez, Sergio. International representative, United Nations International
Narcotics Control Board. Bogot. 7 March 2000.
Valds, Jorge. Former drug trafficker. United States. 29 August 2000.
Valds, Jorge with Ken Abraham. Coming Clean: The True Story of a Cocaine Drug
Lord and his Unexpected Encounter with God. Colorado Springs: Waterbrook,
1999.
Valencia Tovar, Alvaro. Retired Colombian army general. Gainesville. 17 February
1999.
Valencia Tovar, Alvaro. Retired Colombian army general. Bogot. 27 April 2000.
Vargas, Nelson [pseud]. Government prosecutor, National Prosecutors Office. Bogot.
13 July 2000.
Vargas Meza, Ricardo. Drogas, Mascaras y Juegos: Narcotrfico y conflicto armado en
Colombia. Bogot: Tercer Mundo, 1999.
Vargas Meza, Ricardo. Researcher, Accin Andina. Bogot. 23 March 2000.
Velsquez, Jorge Enrique. Cmo me infiltr y enga al Cartel. Bogot: Editorial Oveja
Negra, 1993.


205
routines by a more powerful actor, this does not preclude learning occurring within the
subordinate agency following a period of initial adjustment.
Other PEA Structural Adaptations
In addition to adapting to administrative reorganizations imposed by more
powerful bureaucratic actors, the DEA creates new sub-units and programs in response to
developments in the psychoactive drug industry. In response to the re-emergence of
Caribbean trafficking routes used by Colombian smugglers, the DEA created a new
Caribbean division. During the previous decade many trafficking groups abandoned
these routes in response to intensified law enforcement efforts in the Caribbean.
However, in the early 1990s trafficking enterprises based in the Colombias Atlantic
coast area took advantage of the shift in drug enforcement resources away from the
Caribbean to re-activate their traditional routes through the Windward and Mona
Passages. By 1995, the DEA was sufficiently alarmed by the increasingly larger
psychoactive drug shipments passing through these routes that it established the 21st Field
Division in San Juan, Puerto Rico (DEA, [no date] DEA History, 1994 1998).
The DEA established the Regional Enforcement Teams initiative in 1999 in
response to another development in the illicit drug trade. In recent years, a number of
major trafficking organizations have adapted to intensified drug enforcement efforts in
major U.S. drug markets, such as New York and Miami, by shifting their distribution
cells to smaller cities and rural areas, such as Charlotte, North Carolina and Des Moines,
Iowa. Over time, several enterprises established cocaine, heroin, methamphetamine, and
marijuana production centers in these areas, complete with their own command cells,
warehouses, and transshipment points. Consistent with their design, these trafficking


54
Mexico and the U.S. The DEA and CNP have identified other traffickers they believe
now lead the Atlantic Coast and North Cauca Valley networks. Moreover, drug enforcers
fear that surviving remnants of the Medellin and Cali cartels are reorganizing, led by
former mid-level managers. In addition to these medium-sized enterprises there are now
hundreds of smaller independent trafficking groups operating in Colombia. Some of
these groups are led by notorious traffickers that continue to elude drug enforcers, others
are run by individuals unknown to Colombian and U.S. authorities. With so many
criminal enterprises determined to produce and transport illicit drugs, Colombia remains
poised to continue its leading position in the international drug trade.
Table 2-3 Colombian Drug Enforcement Indicators, 1996-1999
Year
Persons
apprehended
Cocaine
interdicted
(kilos)
Cocaine
base
interdicted
(kilos)
Heroin
Interdicted
(kilos)
Marijuana
interdicted
(kilos)
Drug labs
destroyed
1996
5,703
17,808
11,142
88
101,519
436
1997
10,711
35,792
14,906
176
117,880
228
1998
18,276
46,256
11,346
341
73,085
198
1999
21,168
21,423
9,621
541
71,369
100
Source: CNP (2000)
Conclusion
Over the past seventy years, the Colombian drug trade, and the illicit enterprises
that coordinate it, have undergone a variety of changes. While the empirical record of
this clandestine industry remains sketchy, particularly in the early decades, this chapter
draws on the available materials to craft a credible, coherent narrative. In the beginning,
drug trafficking in Colombia was sporadic. Smugglers, often working alone, transported
minute quantities of European produced cocaine and heroin through Colombia and
Panama on the way to North American markets. Sometime in the 1940s or 1950s, a
number of pioneers, perhaps of German extraction, developed cocaine and heroin


53
Alberto Orlandez Gamboa, the leader of a large Atlantic Coast trafficking enterprise
(Associated Press, 1998; CNNespaol, 1998; Chepesiuk, 1999b, pp. 92, 153, 253;
Constantine, 1998; DEA, 2000; FBI, 1993, p. 31; Guzman, 1998; Los Angeles Times,
1997; Reuters, 1998a, 1998c).15
In October 1999, CNP and DEA officials implemented Operation Millennium,
targeting a highly sophisticated smuggling network exporting between ten and thirty tons
of cocaine per month. In a series of carefully orchestrated raids in Bogot, Medellin and
Cali, CNP agents arrested thirty-three members of the network, including the alleged
leader, Bernal Madrigal, and Fabio Ochoa, one of the leaders of the old Medellin cartel
(Bajak, 1999; Johnson & Davies, 1999; El Tiempo, 1999).16 Also during this period,
CNP drug enforcement units captured a number of traffickers from smaller smuggling
groups and disrupted their cocaine and heroin operations {El Tiempo, 1997a; El Tiempo,
1997b; El Tiempo, 1997c; Caracol, 2000a; Caracol, 2000b; DAS, 1997; RCN, 2000;
Reuters, 1998b).
These drug enforcement operations indicate that CNP and DEA officials continue
to target trafficking organizations in Colombia. However, in spite of these successes, a
number of important traffickers remain at large. According to the DEA, Diego Montoya
Sanchez, once a mid-level figure in the North Cauca Valley network now leads one of the
major organizations in the area. Montoya Sanchez was the principal cocaine supplier for
the Madrigal operation and continues to coordinate multi-ton shipments of cocaine to
15 During the early 1990s, Orlandez Gamboas organization was associated with the Cali-based trafficking
network. Following the capture of the leaders of Cali core organizations in 1995 and 1996, Orlandez
Gamboa branched out on his own, exploiting maritime and air routes through Mexico and the Caribbean to
smuggle multi-ton shipments of cocaine and marijuana to the U.S.
16 Since their arrest Ochoa and Madrigal have been extradited to the U.S., where they are now awaiting trial
for numerous trafficking-related criminal charges {El Tiempo, 2001d, 2001e; Forero, 2001b; U.S. Attorneys
Office, 2001a, 2001b).


248
structures on their own, drawing on the experience of enterprise participants. Whether
the source of inspiration is internal or external, these organizations change existing
routines or create new ones by gathering, interpreting, and applying knowledge and
experience. In other words, they learn.
000
O 0
n o O O
u o

Time 3
Figure 7-2 Phased Evolution of Illicit Drug Industry
KEY: In Time 1 environmental hostility increases dramatically as drug enforcers remove the largest
trafficking enterprises from the system. In Time 2, smaller, ad hoc enterprises are the only
survivors from the government crackdown. In Time 3 a rejuvenated drug industry develops as
other enterprises form, modeling themselves after the attributes of the small survivors or drawing
on their own experience.
Implications for Counter-Drug Policy:
Or How I Learned to Stop Worrying and Love the Drug Trade
As far as interdiction, we are following the basic procedures we followed for the
last 10 years. We are following the basic procedures that I found so easy to beat.
Nobody has made the changes that need to be made in order to beat the problem
(Statement of Witness, 1989, p. 15).
In 1989, Max Mermelstein, the coordinator of a Medellin-affiliated transportation
ring, made this comment while testifying before the U.S. House Subcommittee on
Financial Institutions. Twelve years later Mermelsteins observation remains prescient.


80
Productive Learning
Among organization theorists, learning is commonly equated with improved task
performance.8 A number of analysts have drawn on this assumption to define learning as
the organizations ability to improve task performance over time. According to Fiol and
Lyles, learning is the process of improving actions through better knowledge and
understanding (Fiol & Lyles, 1985, p. 803). Dodgson describes learning as the ways
firms build, supplement and organize knowledge and routines around their activities...
and develop organizational efficiency by improving the use of the broad skills of their
workforces (Dodgson, 1993, p. 377). In their early collaborative research, Argyris and
Schon equate learning with increasing organizational effectiveness, however, in their
more recent work they distinguish productive learning, which improves performance,
from generic learning that does not. The former variety remains the focus of their
theoretical and empirical work. Other analysts have chosen different labels for learning
that improves performance, including instrumental, effective, and efficiency
learning (Argyris & Schon, 1978; Argyris & Schon, 1996, p. 4; Hedberg, 1981, p. 3;
Ndumbaro, 1998, p. 39 Tetlock, 1991, p. 35).
The classic example of efficiency learning is found in early studies of airplane
manufacturing. Beginning in the 1920 and 1930s, a number of industrial economists
discovered that the costs of airplane manufacturing fell predictably with increased
production volumes (Wright, 1936; Arrow, 1962). Increases in production were seen as
proxies for greater skill and knowledge accumulated through repetition of performance
tasks (Garvin, 2000, p. 94). This phenomenonlabeled the learning or experience
8 The assumption that learning results in improved organizational performance derives from the broader
assumption in behavioral studies of organizations that organizational learningand behavior more
generallyis fundamentally geared towards targets or goals.


319
Rohter, Larry. Colombian President Took Drug Money, Turncoat Says. New York
Times (25 July 1997). Http://www.nvtimes.com/yr/mo/dav/news/world/colombia-
drugs.html [Accessed 25 July 1997].
Rohter, Larry. Haiti Paralysis Brings a Boom in Drug Trade. New York Times (27
October 1998). Http://www.nvtimes.com/Iibrary/world/americas/102798haiti-
cocaine.html [Accessed 20 May 2001].
Rohter, Larry. Ever-Present Violence Creates Nation of Fear. New York Times (21
April 2000a). Http://www.nvtimes.com/library/world/americas/0421 OOcolombia-
drugsAhtml [Accessed 16 October 2000].
Rohter, Larry. Troubles Build as Colombians Await New Aid. New York Times (25
June 2000b). Http://www.nytimes.com/librarv/world/americas/062500colombia
us.html [Accessed 10 September 2001].
Roldn, Mary. Colombia: Cocaine and the Miracle of Modernity in Medellin.
Cocaine: Global Histories. Ed. Paul Gootenberg. London: Routledge, (1999):
165-182.
Ronfeldt, David. Tribes, Institutions, Markets, Networks: A Framework about Societal
Evolution. Santa Monica: Rand, 1996.
Rosenau, James N. Turbulence in World Politics: A Theory of Change and Continuity.
Princeton: Princeton University Press, 1990.
Ruiz Hernndez, Hernando. Implicaciones sociales y econmicas de la produccin de la
marihuana. Marihuana: Legalizacin o Represin. Bogot: Asociacin Nacional
de Instituciones Financieras, (1979): 111-228.
Sabatier, Paul A. An Advocacy Coalition Framework of Policy Change and the Role of
Policy Oriented Learning Therein. Policy Sciences 21 (1988): 129-168.
Sabbag, Robert. Snowblind: A Brief Career in the Cocaine Trade. New York: Grove,
1990.
Senz Rovner, Eduardo. La prehistoria del narcotrfico en Colombia. Innovar: Revista
de ciencias administrativas y sociales (8) (July-December 1996): 65-92.
Senz Rovner, Eduardo. La prehistoria del narcotrfico en Colombia: Temores
norteamericanos y realidades colombianas durante la primera mitad del siglo
XX. La crisis socio-poltica colombiana: Un anlisis no coyuntural de la
coyuntura. Ed. Luz Gabriela Arango. Bogot: Fundacin Social, (1997): 190-212.


202
alterations in decision-making hierarchies and communication flows. The DEA itself
was the administrative stepchild of a series of reforms that sought to reduce
fragmentation among federal drug enforcement agencies. Operational changes include
procedural and technological innovations in investigative strategies and tactics to identify
and dismantle trafficking enterprises.
The following pages analyze numerous structural and operational adaptations by
U.S. and Colombian drug enforcement agencies. While some modifications have
improved intelligence sharing, reduced bureaucratic infighting, and improved criminal
investigative techniques, others have produced more ambiguous results. Irrespective of
their implications for task performance, these changes were made as decision-makers in
police organizations and policy making bodies gathered, analyzed, and applied
knowledge and experience about drug enforcement, often in response to negative
feedback. Although some of the adaptations described below did not affect drug
enforcement policy in Colombia, they had a direct bearing on Colombian-based
trafficking enterprises that operate in the U.S.
Structural Adaptation Case #1: Toward an Administrative Merger of the PEA and FBI
In the early 1980s, dissatisfaction among policy makers within the executive
branch and Congress regarding the DEA generated substantial pressure for an
administrative reorganization of federal counter-drug efforts. Created in 1973 by the
Nixon administration, the DEA had failed to live up to its billing as the nations super
drug enforcement agency. Critics charged the agency did not effectively target major
trafficking organizations, including the emergent core enterprises in Colombia, and


234
they simply go out and buy it (Passas, 2000; Pineda, 2000). When entrepreneurs or cell
managers are dissatisfied with a workers performance, the person is simply removed
from the operation. When entrepreneurs require information about drug enforcement
efforts they assign a participant or hire an outside consultant to acquire it. This is not to
suggest that trafficking enterprises are disorganized anarchies, whose members lack rules,
roles, and procedures to guide them. Indeed, Chapter 4 contains an extended discussion
of the numerous routines trafficking enterprises use to structure participants behavior.
However, these rules and procedures are more informal and less onerous than those
regulating drug enforcers. Moreover, unlike their law enforcement adversaries,
trafficking groups do not answer to a higher bureaucratic authority. The kingpins
decision is law it does not have to be cleared with a contingent of organization lawyers to
confirm its legality.
The red tape trap confers three distinct advantages on trafficking enterprises in
their competitive learning games with police organizations. First, by slowing information
and decision flows in drug enforcement agencies, these legalistic and bureaucratic
constraints prevent them from responding as quickly to environmental change as their
illicit competitors. Second, by prohibiting government officials from violating citizens
rights, legalistic constraints prevent drug enforcers from engaging in certain practices that
may otherwise be highly effective in identifying and apprehending traffickers.20 Third, to
protect citizens rights and hold police organizations accountable for their actions,
bureaucratic and legalistic constraints increase the transparency of drug enforcement
20 This does not mean that violations of citizens rights do not occur. This study documents examples of
such violations.


227
national borders, the history of international trade suggests that criminal enterprises in
other countries are likely to step into the ensuing vacuum.
However, for territorially based law enforcement organizations, such as the
Colombian National Police, this scenario may appear more plausible. If highly
specialized CNP drug enforcement units are too successful and push the drug trade out of
Colombia they are likely to lose funding, training, and other assistance as their own and
the U.S. governments direct limited counter-drug resources elsewhere. This has already
happened in Bolivia and Peru. During the 1990s, forcible eradication programs
implemented by the Banzer and Fujimori administrations achieved considerable success
in reducing the estimated production of coca leaf in both countries. During the same
period illicit drug production increased dramatically in Colombia, compelling U.S. policy
makers to reduce drug enforcement assistance to Bolivia and Peru and increase it to
Colombia.
Drug enforcers in Colombia understand this dilemma and its implications for their
professional activities. According to the director of the Intelligence Directorate of the
Colombian National Police, the level of tactical narco-narc competition in Colombia is
fierce, as drug enforcers strive to apprehend specific trafficking enterprises. However,
the level of what he calls strategic competition for the Colombian and U.S.
governments is less intense, and this has to do with higher-level bureaucratic and political
considerations (Nicosa, 2000). Drug-related corruption further complicates the
enforcement dilemma, as compromised officials lack the willingness to vigorously
prosecute criminal enterprises they are paid to protect (Cook, 1997; Lupsha, 1996, p. 35).
The crux of the enforcement dilemma is that it is not always in the best interest of law


178
t
memories of managers and participants. While this practice reduces one source of
vulnerability for trafficking enterprises, it creates another.
Knowledge and experience that exists only in the minds of individuals is more
likely to be lost when they cease to be participants. The higher the individual in the
organizations decision-making hierarchy, or the more specialized his or her personal
knowledge, the more disruptive the loss will be to the enterprise. For this reason, drug
enforcers often seek to capture high-ranking participants, such as chief accountants, that
contain a detailed knowledge of their organizations. High rates of personnel turnover, a
common feature in some trafficking enterprises, make the accumulation of experience
difficult, and can impede organizational learning (Parra, 2000). Trafficking enterprises
that lack organizational memories experience difficulty in learning from experience,
particularly when there is little continuity among participants. Such organizations are
more likely to repeat tactical and strategic errors, increasing their exposure to drug
enforcers.
Irrespective of personnel turnover, knowledge and experience that is not accessed
for long periods can also be lost. Participants not only gather, interpret and apply
information, they also forget. Organizational forgetting is particularly problematic when
knowledge lies dormant, as during periods of low productivity (Benkard, 1999; Buell,
2000). Colombian trafficking enterprises often experience periods of little or no activity
due to lag times between shipments and the need to reduce or suspend operations to avoid
drug enforcers. During these frequent intervals, participants forget important details in
their day-to-day undertakings, and when operations start back up again, they must
relearn their roles, causing further delay before the enterprise returns to peak


172
cocaine enterprises, trafficking groups marketed Colombian heroin in major U.S.
markets. In an aggressive strategy to undercut their competitors and increase market
share, Colombian enterprises offered a low-priced product at high levels of purity
(Bagley, 1999; Berquist, 2000; Betancourt & Garcia, 1994, p. 107; DEA, 1994a, pp. 6-7;
Evans, 1999; Farah, 1999a; FBI, 1993; GAO, 1999, p. 19; Olson, 1997, p. 74; Reuter,
1992, p. 170; Ruiz Hernndez, 1979, p. 180; Sanchez, 1999; Semana, 1999; Stutman,
2000; Thoumi, 1995, p. 127; Valds, 2000; Valencia Tovar, 2000; Wrobleski, 1987;
Zabludoff, 1999).
Colombian trafficking organizations have also spread to new markets. In the
1980s and 1990s, a number of core enterprises expanded into the lucrative European and
Japanese markets, where wholesale cocaine rates were approximately two to three times
the average U.S. price. To facilitate their entry, trafficking groups formed partnerships
with criminal organizations established in these markets, including Italian Mafiosi and
Japanese Yakuza. This expansion increased in the latter half of the 1990s, particularly
following the break-up of the cartels. Post-cartel Colombian operations export
psychoactive drugs to Western and Central Europe, Russia and the Newly Independent
States, Japan, Australia, in addition to the U.S. (Abadinsky, 1997, p. 268; Bagley, 1999;
DEA, 1998, pp. 32-33, 37; Farah, 1997; Griffith, 1999; Perl, 1999; Thoumi, 1995,
endnote 18, p. 150; Torres, 2000; Zabludoff 1997, p. 24).
When shifting to new markets and products, trafficking enterprises use existing
stores of knowledge and experience. Information regarding illicit smuggling is often
fungible across product lines. Contrabandistas found that many of the practices and
procedures for smuggling black market cigarettes and liquor could be used, with few


165
companies in Central and South America that export these commodities (Cash, 1999;
Kacerosky, 1994; Reyes, 2000).
Former DEA official Cash describes how the Rodriguez Orejuela organization
switched among different routes:
... you constantly saw a changing of technique and tactic. In 88 they were
hollowing out cedar boards with a router, putting cocaine in Styrofoam
containers [and] slipping them into the routed 2 x 10. And they had cut 1" off
that 2 x 10... And then they glued it back on so you could imagine the amount
of work in that... now when that was discovered... it was not too long after that
the [concrete] fence post came up... That was after the lumber... So they bought
the concrete factory that made these fence posts... And [in] the Cornerstone case
youll see that was about the 27th time that ship came in with fence posts...
Harold Ackerman was their ambassador type representative here in Miami... He
dealt in broccoli importing and exporting... they had a machine that packaged the
cocaine just like a frozen broccoli thing that wrapped it and had it labeled. Its
perfect. They bought a factory that did that. And then it was frozen too. As if it
had something in it. And of course there were real products mixed in... it wasnt
all cocaine... Its just like you get in the frozen foods in your refrig., in the
grocery store, same size box... Then they brought it in coffee. It looked like
coffee in containers of coffee... Vacuum packed... So they bought different
facilities to do different things... but once we hit a couple of loads of that, the
coffee stopped. Then they brought it in lucite blocks. They take these big lucite
blocks and put 7 kilos of cocaine in it and put the lucite blocks in tar, in cans of
roofing tar... Very, very well thought out (Cash, 1999).
When authorities dismantled a route, the leaders of the Rodrguez Orejuela
enterprise would extrapolate lessons from the mismatch, often using their lawyers to
gather the necessary information through the discovery process (see above). In court,
Guillermo Pallomari testified that on several occasions Miguel Rodrguez Orejuela
discussed the fall of the lumber route with other entrepreneurs in order not to make
the same mistakes that led to the falling apart of that route in the United States
(Pallomari, 1997, V. 36, p. 5850).
As these examples demonstrate, Colombian trafficking enterprises respond to
government interdiction pressures by changing transportation practices and programs.


281
Crittenden, Ann. From Bakersfield to Bogot. Book review. New York Times Book
Review (15 October 1989). Http://web.lexis-nexis.com/universe [Accessed 10
September 2000].
Cubides, Fernando. Professor of Sociology, Sociology Department, National University
of Colombia. Bogot. 6 April 2000.
Curtis, Ric and Travis Wendell. Toward the Development of a Typology of Illegal Drug
Markets. Illegal Drug Markets: From Research to Prevention Policy. Eds.
Mangai Natarajan and Mike Hough. Monsey: Criminal Justice Press, (2000): 121
152.
Customs Service, U.S. Department of the Treasury. Diplomatic Correspondence (Various
years) (Havana, Cuba). Various documents contained in Carton #22, Cuba
(Reports from Customs), Bureau of Narcotics and Dangerous Drugs, Subject
Files, 1916-1970. Accession No. 170-74-12. National Archives D: College Park,
MD.
Customs Service, U.S. Department of the Treasury. Memorandum report (1961).
Document contained in Carton #30, South America, 1948-1967, Bureau of
Narcotics and Dangerous Drugs, Subject Files, 1916-1970 (1961). Accession No.
170-74-12. National Archives II: College Park, MD.
Cyert, Richard M. and James G. March. A Behavioral Theory of the Firm. Englewood
Cliffs: Prentice-Hall, 1963.
Daft, Richard L. and Karl E. Weick. Towards a Model of Organizations as Interpretation
Systems. Academy of Management Review 9.2 (1984): 284-295.
Darling, Juanita. Colombias Drug Cartels Give Way to Smaller, More Flexible
Traffickers. Los Angeles Times (29 September 1996). Http://web.lexis
nexis.com/universe [Accessed 30 June 1999].
Darling, Juanita. Colombian Cartels Find New Drug Paths to U.S. Los Angeles Times
(17 November 1997). Http://web.lexis-nexis.com/universe [Accessed 30 June
1999],
David. Black market peso broker. Frontline: Drug Wars. 2000.
Http://www.pbs.org/wgbh/pages/frontline/shows/drugs/interviews/david.html
[Accessed 5 March 2001].
Decker, Scott H., Tim Bynum, and Deborah Weisel. A Tale of Two Cities: Gangs as
Organized Crime Groups. Justice Quarterly 15.3 (1998): 395-425.
Departamento Administrativo de Seguridad (DAS). Balance de operaciones. Revista
DAS (Julio-Septiembre 1997): 19-23.


13
This case study is what Eckstein refers to as a plausibility probe. Plausibility
probes are preliminary tests of candidate-theories, or in this case candidate-hypotheses,
which are undertaken before more rigorousand costlycross-national tests are
attempted (Eckstein, 1992, pp. 147-149). This probe is undertaken to determine whether
the organizational learning proposition is worth further consideration in future studies of
the international drug trade and transnational organized crime. If the proposition does not
receive empirical support in the case of Colombian trafficking organizations, then it is
unlikely to have explanatory value in other cases. This is because many Colombian
enterprises are among the most sophisticated, highly organized criminal groups in the
world. Moreover, the learning ecology within which they operate is intensely
competitive, largely due to anti-drug pressure exerted by the U.S. and Colombian
governments. If Colombian trafficking groups are not learning in this environment and if
this process does not help explain their ability to undermine government efforts to stop
them, then it is unlikely that the explanation would be valid in other cases where criminal
organizations are less sophisticated.
To verify learning by trafficking enterprises and drug enforcement agencies, I
must demonstrate that changes in practices, procedures, and performance programs are
due to information processing, rather alternative considerations, such as power and
environmental selection. The process-oriented model of learning used in this study
necessitates an in-depth examination of how these collectivities function, rather than
superficial measurements of improved task performance. Learning cases should
their aftermath.


G-DEP
Geographic-Drug Enforcement Program
IEPRI
Institute of Political Studies and International Relations
INTERPOL
International Criminal Police Organization
IRS
Internal Revenue Service
NADDIS
Narcotics and Dangerous Drugs Information System
NAS
Narcotics Assistance Section
NDIC
National Drug Intelligence Center
NSC
National Security Council
ONDCP
Office of National Drug Control Policy
PCOC
President's Commission on Organized Crime
Pseud
Pseudonym
RCN
Radio Cadena Nacional de Colombia (National Radio Network of
Colombia)
WOLA
Washington Office on Latin America
Xll


43
In reflection of the smuggling capacity of the core enterprises, in the 1980s U.S.
and Colombian authorities began to discover containerized shipments and stash houses
containing several tons of cocaine hydrochloride. As early as March 1982 U.S. Customs
authorities discovered 3,096 pounds (or 1.38 tons) of cocaine mixed in mixed in a
shipment of hundreds of cartons of clothing apparel. The cocaine was wrapped with
yellow plastic in one-kilo packages bearing different coded markings. According to DEA
investigators the load belonged to fifteen different trafficking organizations, including the
Ochoa and Escobar groups. In 1989, DEA and Customs agents discovered approximately
12,000 pounds (5.35 tons) of cocaine hidden inside drums of powdered lye. The cocaine,
wrapped in red, yellow and blue plastic and marked with the code name Baby I, was
discovered at a warehouse in Queens, New York used by a trafficking ring associated
with the Cali-based trafficking network. Also in 1989, U.S. authorities confiscated
cocaine shipments of 11,200 pounds (5 tons) on a Panamanian freighter in the Gulf of
Mexico, 20,160 pounds (9 tons) hidden in a stash house in Texas, and 44,800 pounds (20
tons) in an unguarded warehouse near Los Angeles. The latter seizure remains the largest
ever recorded in the United States. Two years later, DEA and Customs officials
discovered 26,880 pounds (12 tons) of cocaine hidden inside concrete fence posts and
cornerstones. In 1994, federal authorities uncovered 8,000 pounds (3.57 tons) of cocaine
hidden inside empty barrels in a tractor-trailor truck in Texas (Dillow, 1994; Gugliotta &
Leen, 1989, pp. 71-73; Isikoff, 1991; McKinley, 1989).
Diversifying to Heroin
At some point during the 1980s several core enterprises expanded into the heroin
business. It is believed that this development was driven by the same factors that caused


312
Oppenheimer, Andres and Tim Johnson. Colombia threatens to probe S. Florida firms.
Miami Herald (7 March 1997).
Http://www.herald.com/americas/digdocs/043367.htm [Accessed 22 April 1997],
Oijuela, Luis. J. Narcotrfico y Poltica en la Decada de los Ochenta: Entre la Represin
y el Dilogo. Narcotrfico en Colombia: Dimesiones Polticas, Econmicas,
Jurdicas, e Internacionales. Eds. Carlos G. Arrieta, Luis J. Oijuela, Eduardo
Sarmiento P., and Juan G. Tokatlin. Bogot, Colombia: Tercer Mundo Editores,
(1990): 199-276.
Ostrow, Ronald J. Reno Rejects Proposal to Merge the DEA and FBI. Los Angeles
Times (22 October 1993). Http://web.lexis-nexis.com/universe [Accessed 24 June
2001],
Ostrow, Ronald J. FBI Chief Outlines Moves to Halt Turf War with DEA. Los Angeles
Times (15 June 1994). Http://web.lexis-nexis.com/universe [Accessed 24 June
2001],
Otis, John. The Drug Quagmire. Houston Chronicle (16 July 2000). Http://web.lexis
nexis.com/universe [Accessed 10 August 2000].
O'Toole, Thomas. Satellites Used to Round Up U.S.-Bound Marijuana Ships.
Washington Post (11 July 1978). Http://web.lexis-nexis.com/universe [Accessed
30 September 2000],
Ouchi, William G. Markets, Bureaucracies and Clans. Administrative Science
Quarterly 25 (March) (1980): 129-141.
Pallomari, Guillermo. Personal letter from Guillermo Pallomari to Mr. William M.
Pearson (7 November 1996). Court Exhibit No. 9. United States of America vs.
Michael Abbell, William Moran, Luis Grajales, Eddy Martinez, Ramon Martinez,
J.L. Pereira-Salas, et al, Defendants, Case No. 93-470-CR-WMH.
Pallomari, Guillermo. Direct Testimony of Guillermo Pallomari. Trial Transcript in the
United States of America vs. Michael Abbell, William Moran, Luis Grajales, Eddy
Martinez, Ramon Martinez, J.L. Pereira-Salas, et al, Defendants, Case No. 93
470-CR-WMH, Volumes 35 through 37, and 38 through 40 (16, 17, 21, 22,
23 & 28 July 1997).
Pan, Philip P. Five Convicted of Operating Drug Network. Washington Post (30 June
1996). Http://web.lexis-nexis.com/universe [Accessed 3 April 2001].
Pardo, Rodrigo. The Issue of Drug Traffic in Colombian-US Relations: Cooperation as
an Imperative. Journal of Interamerican Studies and World Affairs 37.1 (1995):
101-109.


253
Legalization and Harm Reduction
Dissatisfaction with drug prohibition has prompted a growing number of
detractors to support alternative drug control strategies, including harm reduction and
legalization.2 Proponents of these two approaches view the illicit drug traffic as primarily
a public health problem, rather than a law enforcement dilemma. Proponents of both also
emphasize the harms caused by drug enforcement, such as the incarceration of large
numbers of citizens for relatively minor drug-related infractions, and call for greater
support of demand-reduction programs, including prevention education and treatment for
substance abusers. Apart from these similarities, there are relevant distinctions between
the two approaches.
One difference involves the question of legalization. Unlike their fellow critics,
many, but not all, proponents of harm reduction do not favor legalizing controlled
substances. Proponents of legalization argue that removing the legal sanctions to these
commodities will significantly lower prices and thereby reduce the incidence of certain
types of drug-related criminality, such as property crime and petty theft by addicts
struggling to support costly habits. Some proponents also maintain that legalization will
assist producer countries such as Colombia by allowing these economic activities to
proceed according to the whims of the market place, rather than the violent interventions
of the state. It is also believed that the level of organized crime in producer countries,
along with drug-related corruption of public officials, will decline as trafficking
enterprises are pushed out of the industry by more efficient firms and legality, along with
2 There are a variety of harm reduction and legalization approaches. At the risk of oversimplifying, I limit
my discussion to these broad, and commonly accepted, categories. For discussion of harm reduction and
legalization, see Bertram et al. (1996), MacCoun & Reuter (2001), and Nadelmann (1998).


72
are doing.... Interpretation is the process of translating these events, of
developing models for understanding, of bringing out meaning, and of assembling
conceptual schemes among managers (Daft & Weick, 1984, p. 286).
Interpretation is a social process through which organizational participants share
their perceptions and construct inter-subjective understandings that allow them to make
sense of relevant events. When engaged in interpretation, or sensemaking, participants
communicate, share knowledge and experience, and construct linkages between the
present situation and prior experiences, often with the assistance of analogical reasoning
(Daft & Weick, 1984, p. 286; Khong, 1992; Neustadt & May, 1986; Weick, 1995, pp. 45-
46).
Organization managers and other participants are individual cognitive agents;
they draw on schemas when analyzing information.4 Schemas are knowledge
structures that individuals use to lend order to an otherwise incomprehensible confusion
of information and experience (Reiter, 1996, p. 21). Schemas help individuals make
sense of what has gone before them and what transpires around them by providing
cognitive shortcuts or frames of reference. As personalized frames, they can be an
impediment to organizational learning, preventing decision-makers from interpreting a
problem or solution in the same light.
Within organizations, participants share, record, and interpret knowledge and
experience through formal and informal organizational memories. Formal organizational
memories include files, manuals, correspondence, databases, financial accounts, and
physical objects that record and store knowledge, such as monthly accounting statements,
minutes from meetings, electronic mail, and computers (Arygris & Schon, 1996, p. 16;
4 Schemas are knowledge structures that individuals use to lend order to an otherwise incomprehensible
confusion of information and experience (Reiter, 1996, p. 21). Schemas help individuals make sense of
experience and their environment by providing cognitive shortcuts or frames of reference.


89
support established routines. If decision-makers feel that their interests will suffer by
adopting new strategies or routines they may be less likely to do so. To the extent that it
occurs, learning remains localized within the organization. Individual decision-makers
learn but changes in their belief structures fail to produce organizational change (Argyris
& Schdn, 1996, p. 17; Hannan & Freeman, 1977, p. 931; March & Olsen, 1988, p. 347).
When decision-makers overcome organizational inertia to apply knowledge to
routines, there is still no guarantee that such adaptation will lead to desired outcomes.
Organizational learning need not result in improved task performance, which is why it is
important to distinguish between productive learning and generic learning. Competency
traps occur when favorable performance becomes associated with an inferior routine or
technology, causing decision-makers to gather more experience with it; even when
superior alternatives are available (Levitt & March, 1988, p. 322; March & Olsen, 1989,
p. 63). Organizations may be more susceptible to competency traps when external
pressure causes them to choose among alternative routines within a relatively short time
period. This type of fast learning can result in maladaptive specialization within the
organization (Levitt & March, 1988, pp. 322-323).
The difficulties described above call into question the assumptions of simple
stimulus-response models of organizational learning. Analysts cannot assume that
organizations learn as they interact with their task environments. Learning is not
automatic, but problematic: It can be blocked at multiple points throughout the
acquisition-interpretation-application process. Nor does learning always lead to
improved performance or more intelligent behavior (March, 1999; Jarosz & Nye, 1993,
p. 179). In light of these ambiguities, learning by Colombian trafficking enterprises and


306
McGee, Jim and Brian Duffy. Main Justice: The Men and Women who Enforce the
Nation s Criminal Laws and Guard its Liberties. New York: Touchstone, 1997.
McKinley, James C. Jr. Seizure of Cocaine is Called Largest in New York City. New
York Times (5 November 1989): Al. Http://web.lexis-nexis.com/universe
[Accessed 18 October 2000].
McRae, Patricia B. Impact of the Illegal Narcotics Trade on Economic and Legal
Institutions in Colombia. Unpublished Ph.D. dissertation, University of South
Carolina, 1995.
McWilliams, John C. The History of Drug Control Policies in the United States,
Handbook of Drug Control in the United States. Ed. James A. Inciardi. Westport:
Greenwood Press, (1990): 29-50.
Medina Gallego, Carlos. Autodefensas, paramilitares y narcotrfico en Colombia.
Bogot: Editorial Documentos Periodsticos, 1990.
Meier, Kenneth J. The Politics of Sin: Drugs, Alcohol, and Public Policy. Armonk: M.E.
Sharpe, 1994.
Mel, Jorge Orlando. The Drug Trade Trade, Politics and the Economy: The Colombian
Experience. Translated by Pam Decho. Latin America and the Multinational
Drug Trade. Ed. Elizabeth Joyce and Carlos Malamud. New York: St. Martins
(1998): 63-96.
Mendelson, Sarah E. Internal Battles and External Wars: Politics, Learning, and the
Soviet Withdrawal from Afghanistan. World Politics 45 (April) (1993): 327-360.
Menzell, Sewall H. Cocaine Quagmire: Implementing the U.S. Anti-Drug Policy in the
North AndesColombia. Lanham: University Press of America, 1997.
Merchn, Gabriel. Director, National Directorate of Dangerous Drugs, Republic of
Colombia. Bogot. 6 April 2000.
Mermelstein, Max as told to Robin Moore and Richard Smitten. The Man Who Made it
Snow. New York: Simon and Schuster, 1990.
Meyer, John W. & Brian Rowan Institutionalized Organizations: Formal Structure as
Myth and Ceremony. American Journal of Sociology 83.2 (1977): 340-363.
Miami Herald. Colombian rebels richer than big firms. (6 July 1998).
Http://www.herald.com/americas/digdocs/030349.htm [Accessed 15 July 1998].
Migdal, Joel S. Strong Societies and Weak States: State-Society and State Capabilities in
the Third World. Princeton: Princeton University Press, 1988.


318
Reyes Posada, Alejandro. Drug Trafficking and the Guerrilla Movement in Colombia.
Drug Trafficking in the Americas. Eds. Bruce M. Bagley and William O. Walker
III. New Brunswick: Transaction, (1994): 121-130.
Reyes Posada, Alejandro. Compra de tierras por narcotraficantes. Drogas ilcitas en
Colombia: Su impacto econmico, poltico y social. Ed. Francisco E. Thoumi.
Bogot: Planeta Colombiana Editorial, (1997): 279-346.
Reyes Posada, Alejandro. Analytical Narrative. Prepared for United Nations
Transnational Organized Crime Assessment Form (Pilot Survey), Center for
International Crime Prevention (November 1999).
Rice, Berkeley. The Cocaine Express. The New York Times Business World Magazine
(27 March 1988): 20. Http://web.lexis-nexis.com/universe [Accessed 22 February
1999],
Rice, Berkeley. Trafficking: The Boom and Bust of the Air America Cocaine Ring. New
York: Charles Scribners Sons, 1989.
Richani, Nazih. The Political Economy of Violence: The War System in Colombia.
Journal of Interamerican Studies and World Affairs 39.2 (1997): 37-81.
Richelson, Jeffrey T. The U.S. Intelligence Community. 4th edition. Boulder: Westview
Press, 1999.
Rincn, Fabio. Leyenda y verdad de El Mejicano. Bogot: Aqui y Ahora, Editores, 1990.
Rincn, Fabio. El libro sellado de Pablo Escobar. Bogot, Colombia: Aqui y Ahoras
Editores, 1994.
Rindskopf-Parker, Elizabeth. Transnational Threats Vis--Vis Law Enforcement and
Military Intelligence: Lessons on the Emerging Relationship, Transnational
Threats: Blending Law Enforcement and Military Strategies. Ed. Carolyn W.
Pumphrey. Carlisle: Strategic Studies Institute, (2000): 65-83.
Restrepo, Miguel [pseud]. U.S. military official. Bogot. 28 June 2000.
Roco Vsquez, Claudia. 63 das detrs de 'Femandio'. El Tiempo (24 April 2001)
Http://eltiempo.terra.com.co/24-04-2001/prp.shtml [Accessed 24 April 2001].
Rodriguez, Mr. Testimony of Mr. Rodriguez, former member of the Medellin Drug
Cartel. Anatomy of a Colombian Drug Trafficking Operation in the United
States. Hearings before the Subcommittee on Crime of the Committee on the
Judiciary, U.S. House of Representatives. One hundred fifth Congress, first
session (16 October 1997). Washington: U.S. Government Printing Office.


96
infrastructures that support iterated transactions, Colombian trafficking enterprises
demonstrate fluidity in form and function that defies easy classification.
What is the extent of social organization within these enterprises? Do Colombian
trafficking groups pursue mutually agreed upon aims or tasks? Do they contain customs,
rules, and procedures that guide participants behavior? To what degree are these
routines formalized? What can be said of their task environment and the technologies
they use to perform their activities? To what extent does their external milieu shape their
internal structures and operations? Consistent with the structure of the concept of social
organization presented in Chapter 3, the following pages explore the illicit drug industry
in Colombia according to the tasks, environments, structures, participants, and
technologies of the criminal enterprises that coordinate it.
Tasks of Colombian Trafficking Enterprises
Criminal organizations exist to profit from the transaction of goods and services
that have been declared illegal by nation-states (Southerland & Potter, 1993, p. 251;
Reuter, 1983). In addition to attaining satisfactory or positive profits, trafficking
enterprises seek to increase market share, and minimize their exposure to risk and
uncertainty.1 To achieve these goals, trafficking enterprises provide illicit drugs for
consumer markets, launder and repatriate revenues from these activities, and supply
auxiliary goods and services. From the site of production (farm) to the point of
consumption (arm), entrepreneurs and enterprises engage in a series of transactions along
1 My usage of the terms satisfactory or positive profits rather than maximized profits is intentional and is
based on a rejection of the standard profit maximization assumption found in neoclassical economic
theories of the firm. The distinction is not superfluous: some students of organized crime base their
analyses on these widely rejected neoclassical assumptions. In a oft-cited essay on transnational criminal
organizations, Peter Lupsha claims that the goals of organized crime are typical of those of any rational
actor in market systems, the maximization of wealth, influence and power and the minimization of risk
(Lupsha, 1996, p. 34). For theoretical discussion of the unrealistic and unnecessary nature of the profit
maximization assumption, see Alchian (1950) and Moe (1984).


39
Also in the early 1970s Pablo Escobar, who would emerge a decade later as one
of the leaders of the Medellin cocaine network, worked as an enforcer for a contraband
smuggler that specialized in whiskey, cigarrettes, watches, and second-hand pianos. In
an interview with a Colombian journalist, Escobar referred to this contrabandista as his
maestro, from whom he learned the smuggling business (Castro Caycedo, 1996a, p.
284). Around 1975 the ambitious Escobar became involved in cocaine trafficking,
organizing small-scale smuggling ventures. Through contacts and intermediaries he
arranged for the purchase of kilo quantities of cocaine hydrochloride from an Ecudorian
supplier, and transported the drug to the United States through human couriers. In 1976
he was arrested by Colombian authorities near Medellin for transporting thirty-nine
kilograms of cocaine hidden inside the spare tire of a truck. Escobar was never
successfully prosecuted for this charge, and later in the decade convinced Fabio Ochoa,
an experienced contraband smuggler, to use his well-established and well-connected
smuggling routes for the more profitable drug business (Chepesiuk, 1999a, p. 142). The
old-time contrabandista apparently agreed and a business partnership gradually emerged
between the senior Ochoa, his three sons, and Escobar. This joint venture endured for a
number of years and, along with the participation of other leading traffickers, formed the
basis of what later became known as the Medellin cartel (Can, 1994, pp. 59-60;
Castillo, 1987, p. 60; Kraar, 1988, p. 34).
However, this partnership never functioned as a cartel in the strict sense that
economists use the term. Rather Escobar and the Ochoas pooled their criminal resources,
divided responsibilities, and coordinated large-scale cocaine shipments to American and


Nelson, Richard R. and Sydney G. Winter. An Evolutionary Theory of Economic
Behavior and Capabilities. Cambridge: Harvard University Press, 1981.
310
Nstor, [pseud]. Former drug trafficker. Federal Correctional Complex. Coleman. 19
September 2000.
Nolan, Jeremy [pseud]. Head, U.S. Military Group, Colombia. Bogot. 12 June 1997.
New York Times. 7 Colombian Aliens Get 15 Years in Cocaine Case. (1 August 1976):
20.
New York Times. Jersey Police Arrest 4 in Alleged Plot to Sell 555,000 Pills. (3 August
1980). Http://web.lexis-nexis.com/universe [Accessed 4 June 2001].
Nicosa, Eduardo [pseud]. Director, Intelligence Directorate, Colombian National Police.
Bogot. 14 June 2000.
Nieves, Robert J. Colombian Cocaine Cartels: Lessons From the Front. Washington:
National Strategy Information Center, 1997.
Nolan, Richard A. Learning in Foreign Policy: The Case of U.S. Development
Assistance, 1982-1992. Unpublished Ph.D. Dissertation, University of Florida,
1994.
Nordheimer, Jon. U.S. Details Workings of Vast Drug Ring. New York Times (19
November 1986). Http://web.lexis-nexis.com/universe [Accessed 12 May 1999].
Nordheimer, Jon. In a Quiet Setting, U.S. Concludes Its Biggest Drug Trafficking
Case. New York Times (11 May 1988). Http://web.lexis-nexis.com/universe
[Accessed 15 May 1999],
Ndumbaro, Laurean J. P. Learning and Policy Change: The Case of Swedish
International Development Cooperation Agency (Sida) 1980-1995. Unpublished
Ph.D. Dissertation, University of Florida, 1998.
Nye, Joseph S. Nuclear Learning and U.S.-Soviet Security Regimes, International
Organization 41.3 (1987): 371-402.
Nyholm, Klaus. Director, United Nations Drug Control Program, Colombia. Bogot. 31
March 2000.
Nystrom, Paul C. and William H. Starbuck. To Avoid Organizational Crises, Unlearn.
Organizational Dynamics (Spring) (1984): 53-65.


8
in Fort Lauderdale carrying 1,639 pounds of cocaine and 13 pounds of heroin, tarnishing
the drug-fighting image of Colombian president, Andrs Pastrana. The incident, which
came a week after Pastranas state visit to the U.S., prompted Colombias Defense
Minister to concede that the countrys air force has been seriously infiltrated by
trafficking organizations. U.S. and Colombian officials that interviewed in this research
also stress the role of drug-related corruption in preventing more effective counter
narcotics efforts on the part of the Colombian government (CNN, 1998d).
Toward an Alternative Explanation:
Organizational Learning by Trafficking Enterprises
Although these explanations help account for the persistence of drug production
and trafficking in Colombia, they do not provide a fully satisfactory explanation of the
drug dilemma. One reason for this is that they do not give adequate attention to the non
state actors that manage the production and distribution of cocaine and heroin within the
Western Hemisphere. Since the 1970s, Colombian trafficking enterprises have provided a
steady flow of narcotics to the U.S. and Europe. In the process, these criminal enterprises
have proven to be innovative and highly adaptive. These skills enabled them to maintain
the profitability of their enterprises and survive persistent government efforts to destroy
them.
The literature on the illicit drug industry in Colombia includes government
reports,4 journalistic accounts,5 and academic studies.6 Several recent studies have
4 See the Drug Enforcement Administration (1994, 1995, 1996, 1997), FBI (1993), White House (1998),
and Zabludoff (1997).
5 See Can (1994), Castillo (1987, 1991, 1996), Castro (1996), Duzn (1994), Eddy et al. (1988), Garcia
(1991), Giraldo (1992), Gugliotta and Leen (1989), Mermelstein (1990), Rincn (1990, 1994), and
Shannon (1988).
6 Betancourt and Garcia (1994), Clawson and Lee (1996), Dombois (1998), Geopolitical Drug Watch
(1998), Griffith (1997, 1998), Krauthausen and Sarmiento (1991), and Tokatlin (1995).


142
whether trafficking organizations learn, it is more useful to study the process by which
participants in these enterprises gather, interpret, and apply knowledge and experience to
their activities. This requires a detailed examination of the primary and secondary-source
data gathered in this research. The structure of the following analysis corresponds to the
process-oriented definition of organizational learning used in this study. Trafficking
enterprises are analyzed according to their ability to acquire knowledge and experience,
document and interpret this information through organizational memories, and apply
information by changing practices and procedures that guide their illicit activities.
Table 5-2 Do Colombian Trafficking Organizations Learn?
Results from Interviews6
Type of Respondent
Affirmative
Negative
Not Discussed
Colombian government officials
22
0
4
U.S. government officials
24
0
0
IGO official
1
0
0
Colombian journalists
4
0
0
U.S. journalists
2
0
0
Colombian scholars
6
3
2
U.S. scholars
3
0
0
Former participants
5
0
0
TOTAL
67
3
6
adaptation by narcotics organizations because it validates the difficulties they face in their work. However,
it is also true that the majority of respondents in all of the professional categories, including Colombian
academics, stated that narcotics traffickers adjust their behavior to feedback. It is particularly encouraging
that all five former participants, the respondents with direct experience in the Colombian drug trade,
responded affirmatively to the questions on learning and adaptation.
6 In the English language interviews (N=34), the primary learning question was as follows: Do drug
trafficking enterprises adjust their behavior in response to past experience or new information? In the
Spanish language interviews (N=42), the questions were altered slightly: Las organizaciones criminales
colombianas han aprendido de la poltica estatal? Es decir, suelan cambiar sus actividades ilcitas como
reaccin a la poltica estatal? Se cambian sus actividades ilcitas como reaccin a lo que estn haciendo
sus competidores ilcitos (es decir otros grupos de narcotraficantes, guerrillas, paras)? Estas
organizaciones son capaces de aprenderse a los acontecimientos del pasado? For a complete list of the
questions asked during the English and Spanish language interviews, see Appendix.


196
In the field, drug agents receive additional on-the-job training, supplementing
their pedagogical knowledge with practical experience. On-the-job training can be
provided formally through field training programs that include training and evaluation
procedures, and informally when new recruits are paired with senior agents that socialize
them to the norms and practices of everyday enforcement. Sources of practical
knowledge for police officials include their own experience, obtained through the steady
accumulation of day-to-day job performance, and the experiences of their colleagues,
acquired through diffusion. Diffusion of experience among law enforcers can occur
through formal mechanisms, such as police conferences and agency newsletters, and
informal practices, including conversations and storytelling among colleagues that
interact regularly.
Criminal Informants
The most valuable source of human intelligence in many drug investigations are
criminals and their associates, collectively referred to as criminal informants. While
average citizens and fellow agents provide useful tactical intelligence to law enforcement
officials, persons with detailed knowledge of specific criminal acts are often those that
participate in these transactions. Investigations of particular individuals, groups or
activities frequently commence with tips provided by criminal informants. One common
drug enforcement technique is to apprehend targets of opportunity, lower-level
participants that face the greatest exposure to drug enforcement efforts, and pressure
them to turn states evidence against higher-ranking figures that subsequently become
surveillance targets. Another practice is to recruit confidential informants by offering
handsome rewards to persons that supply information leading to the arrest of high-level


244
interactions. In these contests, these diametrically opposed, rationally bounded players
seek to outwit each other and attain organizational objectives. Players attempt to acquire
knowledge about their adversaries strategies and tactics, while preventing their
opponents from doing the same. Games proceed through a series of interlocked
behaviors as each players action evokes specific responses from its competitors. Players
that survive repeated interactions develop diverse operational repertoires as they counter
the adaptations of their adversaries. Over time, players become increasingly
sophisticated in their ability to develop and choose from a greater range of performance
programs.
A crucial factor in determining outcomes in these contests regards each players
ability to respond rapidly to feedback. The player with the faster decision cycle often
wins. Due to a variety of organizational and environmental conditions, including flatter
decision-making hierarchies, fewer legalistic and bureaucratic constraints to action, and
an unambiguous incentive structure, most trafficking enterprises have faster decision
cycles than drug enforcement agencies, suggesting that they will often win these contests.
When drug enforcers win, as they sometimes do, and specific trafficking enterprises are
removed from the game, other criminal players process the lessons of the game through
the diffusion of relevant knowledge among remaining, interconnected players.
Alternative Explanations of Routine Change
Learning is not the only source of change in trafficking enterprises and drug
enforcement agencies. Power and environmental selection also produce modifications in
organizational routines. However, as seen in several of the empirical examples cited in
this research, alternative sources of routine change are not mutually exclusive. In some


76
of non-learning change is the imposition of new or modified routines by more powerful
external actors on the unit of analysis. In the late 1960s and early 1970s, U.S. federal
drug enforcement agencies underwent a variety of organizational changes. However,
these changes were due to the imposition of new organizational structures and procedures
by policymakers in the Johnson and Nixon administrations, rather than information
processing by decision-makers within the drug enforcement agencies.6
It is possible to argue that learning takes place in such circumstances, but it is the
more powerful actor that learns and passes lessons on to the subordinate agency in the
form of routines that encode experience. While this may be true in some cases, the
explanation is largely unsatisfactory because it allows the analyst to change the unit of
analysis ex post facto. To reliably evaluate the research propositions, it is important to
maintain a consistent unit of analysis during the design and implementation of the
research. In this study, the units of analysis are Colombian drug trafficking enterprises
and U.S. and Colombian drug enforcement agencies. If change in routines is not due to
the information processing activities of these organizations, then learning, according to
the parameters of this research design, has not occurred.
P3: Organizational learning does not occur when changes in organizational
routines are due to the imposed preferences of more powerful political or
bureaucratic actors.
Another source of non-learning change is environmental selection. This occurs
when changes in organizational structures or operations appear as one set of actors is
selected out of the system and replaced by another. Because they operate in hostile task
environments it is not uncommon for trafficking enterprises to be selected out of the
illicit drug industry (i.e. identified and dismantled) by drug enforcement agencies. In this
6 For analysis of these reforms, see Rachal (1983) and Wilson (1978, 1989).


82
of behavior in response to environmental feedback.10 Complex, high or higher-
level, meta-level, double-loop learning, and just plain learning all refer to the
process by which organizations respond to goal-changing feedback by altering the
fundamental values and basic goals that drive organizational behavior.11 However, the
dividing line between simple and complex learning is inherently fuzzy. Within
organizations, goals and objectives are ambiguous and frequently change over time.
Moreover, the meaning of goals and objectives varies in different parts of the
organization. An objective for one part of an organization represents a goal for another.
Even worse, different parts of the same organization may have conflicting goals and
objectives (Levitt & March, 1988, p. 325; Tetlock, 1991, p. 46).
Distinguishing between the goals and objectives of organizational behavior is
tricky business. Yet, it is also true that not all learning is the same. There is a difference
between adjusting the day-to-day activities of an organizational subunit and a
fundamental transformation in the organizations raison-d'etat. Rather than focus on the
nebulous distinction between goals and objectives, an alternative approach is to
distinguish between tactical and strategic changes in organizational behavior.
In the discourse on warfare, tactics refer to small-scale actions that occur during
the course of battle. Strategy refers to the immediate or long-range planning of tactical
maneuvers that takes place prior to battle. According to the economic historian, Alfred
Chandler, strategy determines basic long-range goals and objectives of an enterprise,
and the adoption of courses of action and the allocation of resources necessary for
10 Deutsch (1966) and Nye (1987), Fiol and Lyles (1985) and Hedberg (1981), and Arygris and Schon
(1978, 1996), and Haas (1990) use these different terms in their respective studies.
" Again, Deutsch (1966) and Nye (1987), Fiol and Lyles (1985), Hedberg (1981), Arygris and Schon
(1978, 1996), and Haas (1990) use the different terminology.


118
the next phase of the operation. After they had been sorted and counted, all of us
pitched in and moved the bags from the laundry to the empty rear bedroom with
its one small window. There we stacked them on the floor and in the closet
according to their markings. Despite the heavy plastic wrapping, you couldnt
miss the hospital smell of ether and acetone. To combat the chemical stink
generated by 440 pounds of pure cocaine, Chava place several bottles of vinegar
in the comers of the room. Now the stash was complete and ready for delivery
(Mermelstein, 1990, pp. 58-59).
The routine described by Mermelstein contains several procedures that interlock
behaviors among multiple participants by matching their individual actions to ongoing
stimuli. This performance program is diagrammed in Figure 4-3 (see below).
Responding to orders from Rafa, two drivers (DD) deliver cocaine to the stash house and
unload the vehicle. Rafa (R) reacts to the visual cue of several cocaine-filled duffel bags
lying on the laundry room floor by opening the bags, inspecting, sorting and counting
each kilogram package of cocaine. Rafas action provides a stimulus for Bergin (B) to
re-count the packages and confirm Rafas tally. Bergins action provides a stimulus for
Mermelstein (M) to conduct his count, which in turn causes Chava (C) to make her count,
record the packages, and sign for the delivery. In this manner the action-stimulus-
reaction sequence continues among the six interlocked participants until the inventory
routine is complete and Rafa is ready to begin the delivery performance program.
Routine Failure
Rules, practices, procedures and performance programs are critical to the success
of trafficking enterprises. Operating as they do in hostile environments, trafficking
groups must carry out their activities with speed and precision. This requires coordinated
action that effective routines provide. However, Colombian trafficking organizations are
not sleek, precisely calibrated operations that rarely breakdown. In spite of the best
efforts of leaders and managers, practices and procedures do not always produce the


169
shopping malls, and pay for the calls with coins and prepaid debit cards. Other
distribution cells have been known to change routines for transporting narcotics and
money, managing stash houses, and repatriating illicit profits (Kacerosky, 1994; McGee
& Duffy, 1997, pp. 65-66, 147).
In addition to these procedural adaptations, distribution cells amend their
operation by modifying human resources. These adaptations include hiring new
personnel, rotating workers among different cells and stash houses, assigning participants
to different roles, and sending participants that have been compromised by law
enforcement efforts back to Colombia.
Structural and technological adaptations in distribution cells are often provoked
by problematic situations, including police surveillance, stash house raids, arrests,
seizures, and theft. These events cause considerable emotional arousal among managers
and participants and provide sufficient stimuli for gathering, interpreting and applying
information. The range of corresponding procedural and technological adaptations vary
according to the damage caused by the mismatch. The overriding objective of tactical
adjustments is to minimize damage while maintaining cell productivity, or, if operations
are suspended, reestablish production as soon as possible. If damage to the cell is minor,
adaptations are likely to be relatively modest, such as removing one or two cell workers,
changing telephone numbers are changed, and relocating stash houses. If damage is more
severe, adaptations are likely to be more substantial. In addition to changing
communications, transportation, and warehousing technologies, leaders and managers
may develop new practices and procedures for carrying out the cells daily activities.
Moreover, a number of participants are likely to be replaced. Yet, even these more


260
and the operational autonomy of field agents was constrained by administrative oversight
and institutional procedures. Elite units were also hampered by the vagaries of
bureaucratic politics. CENT AC, the program that most closely approximates the
recommendations outlined above, was eventually disbanded in spite of achieving
substantial policy outputs, in part, due to the resistance of DEA regional directors that
resented the programs intrusion into their jurisdictions and drain on common resources
(drug agents, buy-bust money). In Colombia, State Department and DEA officials have
been compelled to protect elite units from the countervailing interests of corrupt political
elites (GAO, 1980; Mills, 1986; Nadelmann, 1993; Wilson, 1978).
As these examples indicate, the reforms outlined above are unlikely to succeed
unless elite units are given considerable autonomy from drug enforcement bureaucracies
and compromised officials. The extent to which this is possible remains an open
question.
Conclusion
In the summer of 1997, the highest-ranking DEA official in Colombia pointed out
that his agencys goal was to drive the illicit drug trade out of Colombia into other
countries where trafficking enterprises were less sophisticated and easier to identify and
dismantle (Lyons, 1997). Several years later, as I finish this study, this goal remains
elusive. Contrary to this officials prediction, drug trafficking has not been driven out of
Colombia. Instead, drug production levels have increased as trafficking enterprises
continue to adapt to drug enforcement policies and programs. This official was aware
that trafficking organizations adapt their activities in response to government counter
narcotics policies and programs. Indeed, we discussed it at length during his interview.


18
chapters that follow. This chapter frames the evolution of the Colombian drug industry
in five distinct, but overlapping, phases of development. Over the past seventy years,
trafficking enterprises in Colombia have become increasingly sophisticated, increasing
the challenge for U.S. and Colombian drug enforcement efforts. The chapter begins with
a brief discussion of the countrys long-standing tradition in contraband smuggling, and
then proceeds through the different phases of the illicit drug industry. Within each
period, attention is given to Colombias position in the transnational drug trade and
government efforts to dismantle trafficking enterprises.
Colombian Smuggling Tradition
Transnational drug trafficking in Colombia has its roots in a long-tradition of
contraband smuggling tradition dating back to Colonialism, when Spanish authorities
sought to regulate trade within their Latin American dependencies. During the 17th and
18th centuries, contraband smuggling was common throughout Nueva Granada, the area
that encompasses contemporary Colombia. To avoid government duties and satisfy
consumer demand, enterprising smugglers transported food, licor, cigarettes, machinery,
and weapons across Riohacha, Santa Marta, and Cartagena. They also developed a
number of maritime smuggling routes through Caribbean sea lanes (Dye, 1998; Grahn,
1997; Junguito & Caballero, 1982; Lpez Restrepo & Camacho Guizado, 2001; Useche,
1997). In Nueva Granada, smugglers developed a number of practices to evade or co-opt
law enforcement authorities.
they [smugglers] utilized specific strategies and schemes that corresponded to
local conditions. Along unguarded coastlines, for example, smugglers sailed
close to shore and deposited their goods at prearranged sites where buyers were
waiting on the beach. In ports and near guarded anchorages, they used Spanish
intermediaries to arrange deals and bribe local officials while they waited
offshore... When local administrators threatened this illicit coastal trade,


257
the provision of necessary material and symbolic inputs to other government and private
organizations on an ad hoc basis.
In recognition of the decentralization of the Colombian drug trade, elite units
should focus their activities on disrupting networks, rather than individual organizations,
even large ones. Units should have the autonomy to pursue their investigations wherever
they may lead, including corrupt U.S. and Colombian officials that provide protection to
criminal organizations. Units should be free to pursue their investigations across local,
national, even regional law enforcement jurisdictions when necessary. However, units
should cooperate with local and regional law enforcement agencies. To encourage
cooperation, high-level administrators should publicly credit local and regional agencies
with all enforcement outputs that result from specific investigations. In general, elite
units will maintain low operational profiles and refrain from publicizing their activities in
the media.
Policy makers should strive to minimize the disruptive impact of personnel
turnover, and the corresponding loss of organizational knowledge, by placing field agents
in elite units for a period of five years. The annual rotation of one-fifth of unit agents
will facilitate the infusion of fresh knowledge on a regular basis. Members of the elite
units should undergo rigorous selection and vetting procedures, including computerized
background checks and periodic polygraph and urinalysis exams. Educational
requirements should include a college degree at the bachelors level, or its equivalent. If
selected for participation, members should receive formal training in all areas of drug
enforcement and additional on-the-job training once they join their units.


26
among socially marginal groups located in port cities and the sugar growing regions of
Valle del Cauca. According to Thoumi, increases in domestic demand in the late 1950s
and 1960s stimulated greater production of la mala hierba. A series of arrests by
Colombian police officials in the summer of 1961 demonstrated the existence of an
organized, city-wide network of marijuana traffickers in Bogot {El Espectador, 1961b,
p. 9; El Espectador, 1961c, p. 9; El Espectador, 1961d, p. 9; El Espectador, 1961e, p. 9;
Thoumi, 1995, p. 126).
Government Drug Enforcement Efforts During Phase Two
Due to the lack of data regarding drug enforcement outputs during this period,
such as numbers of participants arrested and kilograms of drugs seized, it is difficult to
assess the effectiveness of government counter-narcotics efforts. By the 1960s, officials
from both countries expressed concern at Colombias growing role in supplying
transnational drug networks. However, drug enforcement programs in Colombia during
this period were not particularly successful. In the aftermath of la Violencia, Colombian
police and military authorities had more pressing concerns than stopping the drug trade,
such as reducing political violence in the countryside (Walker, 1999, p. 146; Arango &
Child, 1988, pp. 169-170). When drug-related arrests were made, the perpetrators were
often low-level street dealers and drug addicts, ineffective targets for dismantling
growing trafficking networks.
The fate of the Herran Olazaga brothers following their arrest in Cuba is
indicative of the failure of drug enforcement efforts during these years. Rafael Herran
Olazaga fled Cuba immediately after posting bond from his 1956 arrest. While brother
Toms was eventually found guilty of drug trafficking, he served only one year in a


181
annual basis. Moreover, rotation is less disruptive when there is some continuity of
personnel in the original cell and the enterprise maintains active and accessible
organizational memories.13
Sometime participants receive critical feedback but fail to apply it to their
immediate activities, resulting in sub-par performance or apprehension by drug enforcers.
For example, participants in a distribution cell from the Rodrguez Orejuela enterprise
received feedback leading them to believe that an incoming drug shipment was under
surveillance by U.S. officials. In spite of this knowledge, Miguel Rodrguez Orejuela and
his cell manager ordered cell workers to receive and break down the containerized cargo
shipment, which led to the arrest of several workers. One participant immediately turned
states evidence, and testified against his former colleagues in several criminal
proceedings, causing significant damage to the enterprise (Johnson, 1993).
It is unclear why Rodrguez Orejuela allowed the shipment to continue when he
received information that it was under active surveillance. Informed speculation suggests
several possibilities. One is that the Rodriguez Orejuelas need for profits accruing from
the sale of the shipment outweighed the expected probability of interdiction. Second, it is
possible that he and his cell manager did not find the information regarding surveillance
to be credible. Third, they may have been unable to tease out this potentially useful
signal from environmental noise. Fourth, the cell manager may have failed to
communicate the information adequately to Rodriguez Orejuela, leading him to
misunderstand the gravity of the situation. Fifth, other demands on his and the cell
managers attention may have caused them to make a decision without adequately
13 Of course, the longer a participant or manager remains in any location, the greater his exposure to law
enforcement efforts. As with other aspects of organizational design in criminal enterprises, there are
tradeoffs to this practice.


309
U.S. Senate Subcommittee on Security and Terrorism. Ninety-Eighth Congress,
second session (9 March 1984). Washington: U.S. Government Printing Office.
Musto, David F. The American Disease: Origins of Narcotics Control. Expanded edition.
New York: Oxford University Press, 1987.
Musto, David F. Patterns in U.S. Drug Abuse and Response. Drug Policy in the
Americas. Ed. Peter H. Smith. Boulder: Westview Press, (1992): 29-44.
Musto, David F. The Global Drug Phenomenon: Lessons from History and Future
Challenges. Drugs and Foreign Policy: A Critical Review. Ed. Raphael F. Perl.
Boulder: Westview Press, (1994): 1-6.
Nadelmann, Ethan A. U.S. Drug Policy: A Bad Export. Foreign Policy 70 (1988): 83-
108.
Nadelmann, Ethan A. Latinoamrica: Economa Poltica del Comericio de Cocana.
Economa y Poltica del Narcotrfico. Eds. Juan G. Tokatlin and Bruce M.
Bagley. Bogot: Ediciones Uniandes, (1990): 31-53.
Nadelmann, Ethan A. Cops Across Borders: The Internationalization of U.S. Criminal
Law Enforcement. University Park: Pennsylvania State University Press, 1993.
Nadelmann, Ethan A. Commonsense Drug Policy. Foreign Affairs 77.1: 111-126.
Nagl, John A. Learning to Eat Soup with a Knife: British and American Army
Counterinsurgency Learning During the Malayan Emergency and the Vietnam
War. World Affairs 161.4 (1999): 193-199.
Natarajan, Mangai. Understanding the Structure of a Drug Trafficking Organization: A
Conversational Analysis. Illegal Drug Markets: From Research to Prevention
Policy. Eds. Mangai Natarajan and Mike Hough. Monsey: Criminal Justice Press,
(2000): 273-298.
National Drug Intelligence Center (NDIC). About the NDIC. (13 April 2001).
Http://www.usdoi.gov/ndic/tril .htm [Accessed 9 July 2001].
National Security Council (NSC), Executive Office of the President. International Crime
Control Strategy Report (June 1998).
Http://www.whitehouse.gov/WH/EOP/NSC/html/iccs-frm.html [Accessed 10
October 1998].
Naylor, R. Thomas. Mafias, Myths, and Markets: On the Theory and Practice of
Enterprise Crime. Transnational Organized Crime 3.3 (1997): 1-45.


68
limitations in mind, and much of their social structure can be understood as mechanisms
by which organizational leaders and managers seek to channel participants competing
aspirations, energies and activities into a coherent and consistent pattern of behavior that
facilitates the organizations purpose. However, people are not mindless automatons, and
their aspirations, hopes, and fears cannot be calculated and manipulated like threshing
gears and rotating gyroscopes (Feldman, 2000, p. 614; Posen, 1984, p. 43; Simon, 1997).
Technology
Organizations perform work. Some organizations manufacture materials or
machines; others provide services, information or other products. Technology refers to
the material and symbolic inputs that organizations use to complete work. Technology
includes the equipment, instruments, and information that participants use to complete
their activities, as well as the technical knowledge and skills of participants themselves
(Scott, 1998, p. 21; Leavitt, 1965, p. 1144; Leavitt et al., 1973, p. 4).
Whether the purpose of an organization is to produce legal or illegal products, it
draws on necessary technological resources. If the purpose of an organization is to
provide education, it will make use of material and symbolic resources that allow it to
accomplish this aim. Such resources include buildings for housing classrooms and
administrative offices, computers, projectors and books for developing and delivering
course materials, and the accumulated knowledge and skills of educators and their
support staff. Likewise, an organization that provides escort services will require an
office and/or other physical location(s) to coordinate and implement operations;
telephones, beepers and other equipment for communicating among participants,
customers, suppliers, and regulators; computers or other instruments for keeping track of


220
written notations. They use encryption technology to shield information stored on
computers and disseminated electronically. They limit intra-organization information
flows about impending activities on a need-to-know basis. They use the media, criminal
informants, and other sources of communication to spread disinformation about their
adversaries and sow confusion and mistrust.
Competitive learning games proceed through a series of interlocked behaviors
among players. One players action evokes specific, even routine responses from
adversaries, and vice versa (Weick, 1979). Players draw on their respective operational
repertoires to achieve organizational objectives. Trafficking enterprises use practices and
procedures for drug production, transportation, and distribution, and police agencies draw
on different routines for carrying out criminal investigations and enforcement operations.
As traffickers conduct repeated transactions in an inherently hostile task environment
(see Chapter 4) they expose themselves to greater risk and uncertainty.
Through human or stochastic error, traffickers face the probability of an eventual
mismatch. Drug enforcers may intercept their communications, interdict their drug
shipments, and apprehend their participants. A mismatch for traffickers is an opportunity
for their adversaries. Drug enforcers attempt to capitalize off these events by gathering
more intelligence about traffickers and their methods of operation. When drug enforcers
gather sufficient tactical intelligence they carry out enforcement operations against their
opponents. If traffickers discover, through their own informants or poorly executed
police work (such as a sloppy surveillance operation), that they are the subject of a
criminal investigation they will change their activities to avoid further exposure to law
enforcement. If law enforcers carry out a successful operation, some trafficking


209
Figure 6-2 below). Other routines put in place during this period sought to improve the
selection and training of police recruits, and change the incentive structure for field
agents by rewarding them for exemplary service (Bandera, 2000; Camacho Guizado,
1993; CNP, 1999; Leal Buitrago et al., 1999, p. 83; Llrente, 1999, pp. 391, 407;
Llrente, 2000; Puglisi, 2000; Serrano Cadena, 1999; Soto Velasco, 2000; Torres
Velasco, 1994; WOLA, 1993, pp. 1, 35-36).
In the process of implementing these changes, the Colombian National Police
transformed itself into one of the most respected institutions of the Colombian state. The
accolades reached a highpoint in 1995 and 1996, when CNP drug enforcement units
captured the Rodrguez Orejuela brothers and other leaders of the Cali core enterprises.
The director of the institution and the reform effort, General Rosso Serrano, became an
international celebrity, receiving extensive praise from U.S. and European law enforcers
and politicians, and adoration from the Colombian press and populace. The institution he
reformed became a model police force, hailed at international law enforcement
conferences and studied by police organizations from other countries eager to learn the
secrets of Serranos success (Bandera, 2000; Camacho Guizado, 1993; CNP, 1999; CNP,
2001; CNP, [no date]; Leal Buitrago et al., 1999, p. 83; Llrente, 1999, pp. 391, 407;
Llrente, 2000; Puglisi, 2000; Rohter, 2000; Serrano Cadena, 1999; Soto Velasco, 2000;
Torres Velasco, 1994; WOLA, 1993, pp. 1, 35-36).
To date, the results of the CNPs cultural transformation appear to be mixed. It is
clear that some degree of organizational learning took place. Formal routines were
changed, lines of authority altered, and new organizational sub-units created. These
changes were made by CNP decision-makers that gathered and analyzed information


288
Evans, Samuel [pseud]. Intelligence Analyst, Drug Enforcement Administration.
Washington. 22 July 1999.
Evangelista, Matthew. Innovation and the Arms Race. Ithaca: Cornell University Press,
1988.
Faiola, Anthony and Scott Wilson. U.S. Took Risks In Aiding Peru's Anti-Drug Patrols
Control. Washington Post (29 April 2001): A01. Http://washingtonpost.com/wp-
dyn/world/americas/A 17597-2001 Apr28.html [Accessed 28 April 2001].
Farah, Douglas. Colombian Police Claim A Major Blow to Cartel. Washington Post
(13 August 1990a): A13. Http://web.lexis-nexis.com/universe [Accessed 13
February 1999],
Farah, Douglas. Colombian: Israeli Aided Assassins. Washington Post (17 August
1990b): All. Http://web.lexis-nexis.com/universe [Accessed 13 February 1999].
Farah, Douglas. Colombian Drug Lords Ran Empire Behind Bars: Prison Raid Exposed
Thriving Business. Washington Post (26 December 1996).
Http://wpl .washingtonpost.com/wp-srv/WParch/1996-12/26/044F-122696
idx.html [Accessed 5 January 1997].
Farah, Douglas. New Threat From Colombia: Heroin Traffickers Aggressively
Marketing Nearly Pure Drug in U.S. Washington Post (27 March 1997).
Http://www.washingtonpost.com/wp-srv/WParch/1997-03/27/053F-032797
idx.html [Accessed 1 April 1997].
Farah, Douglas. Andean Coca Farming Declined in '97: U.S. Data Show Cultivation of
Cocaine's Raw Material Rose in Colombia. Washington Post (19 January 1998):
A21. Http://www.washingtonpost.com/wp-srv/WParch/1998-01 /19/108f-011998-
idx.html [Accessed 19 January 1998].
Farah, Douglas. Massacres Imperil U.S. Aid to Colombia: Paramilitary Groups Linked
to Army. Washington Post (31 January 1999a): Al.
Http://search. washingtonpost.com/wp-srv/WPlate/1999-01/31/0741-013199-
idx.html [Accessed 7 February 1999].
Farah, Douglas. Telephone interview with Washington Post correspondent, Latin
America bureau. Gainesville. 29 August 1999b.
Farah, Douglas. Colombian Drug Cartels Exploit Tech Advantage. Washington Post
(15 November 1999b). Http://web.lexis-nexis.com/universe [Accessed 17 April
2001],


302
Lpez Restrepo, Andrs. El narcotrfico, dato decisivo de la realidad colombiana.
Unpublished paper. Bogot: National University of Colombia, 2000.
Lpez Restrepo, Andrs. Professor of Economics, Institute of Political Studies and
International Relations, National University of Colombia. Bogot. 22 February
2000.
Lpez Restrepo, Andrs and Alvaro Camacho Guizado. 2001. From Smugglers to Drug-
Lords to "Traquetos": Changes in the Colombian Illicit Drugs Organizations.
Paper presented at the Democracy, Human Rights, and Peace in Colombia
Conference, Helen Kellog Institute for International Studies, University of Notre
Dame (March 26-27). Http://www.nd.edu/~kellogg/colconf.html [Accessed 18
June 2001].
Los Angeles Times. Colombian Drug King Convicted. (20 May 1988). Http://web.lexis-
nexis.com/universe [Accessed 18 January 2000].
Los Angeles Times. Top Drug Trafficker Turns Himself In. (30 September 1997): All.
Http://web.lexis-nexis.com/universe [Accessed 18 October 2000].
Lowenthal, Abraham. The Dominican Intervention. Cambridge: Harvard University
Press, 1972.
Lubasch, Arnold H. 12 Colombians Convicted in Drug Traffic. New York Times (27
January 1976): 21.
Lupsha, Peter A. El Trfico de Drogas: Mxico y Colombia una Perspectiva
Comparada. Economa y Poltica del Narcotrfico. Eds. Juan G. Tokatlin and
Bruce M. Bagley. Bogot: Ediciones Uniandes, (1990): 235-264.
Lupsha, Peter A. Transnational Organized Crime versus the Nation-State.
Transnational Organized Crime 2.1 (1996): 21-48.
Lyons, Larry. Country Attach, Drug Enforcement Administration, Colombia. Bogot. 12
June 1997.
Mabry, Donald. The US Military and the War on Drugs in Latin America. Journal of
Interamerican Studies and World Affairs 30.2-3 (1988): 53-76.
Mabry, Donald J. The Role of the Military. Drugs and Foreign Policy: A Critical
Review. Ed. Raphael F. Perl. Boulder: Westview, (1994): 101-130.
MacCoun, Robert J. and Peter Reuter. Drug War Heresies: Learning from Other Vices,
Times, and Places. Cambridge: Cambridge University Press, 2001.


40
European markets.8 As the size of their trafficking venture grew they gradually absorbed
or co-opted the smuggling operations of several of the pioneering trafficking groups
operating in the 1970s (Gugliotta, 1987; PCOC, 1986, pp. 101-103; Valds, 2000).
Similar arrangements developed between different core enterprises in the Cali-
based network, including the Rodrguez-Orejuela, Santacruz Londoo, and Herrera
Buitrago groups. Although the leaders of these separate groups ran their own smuggling
operations with their own personnel, they collaborated extensively on strategic matters
affecting the entire network, such as infiltrating the different security agencies of the
Colombian state or corrupting well-placed Colombian congressmen sympathetic to their
political and economic interests, such as outlawing the extradition of Colombian
nationals. Moreover, on occasion, low-level workers for one core organization, such as
cocaine chemists, truck drivers, or merchandise off-loaders, would be put to work for
another core group, without the workers knowledge that the ultimate locus of authority
had changed (Torres & Sarmiento, 1998; Velsquez Romero, 2000).
Smuggling Methods of Core Organizations
Core enterprises used a variety of methods for transporting cocaine from South
America to U.S. and European drug markets. A number of core organizations continued
the practice of using human couriers to smuggle cocaine. Although the quantities
smuggled by individual couriers remained small, generally less than four kilograms, core
enterprises had large numbers of participants at their disposal. In an effort to overwhelm
law enforcement authorities, now wise to this time-honored smuggling method, some
8 Medellin core organizations also absorbed the smuggling operations of several of the pioneering
trafficking groups operating in the 1970s. When Manuel Garcs was sentenced to jail in Colombia, the
Escobar organization took over his smuggling routes and operations (Valds interview). When Benjamin
Herrera was paroled from a U.S. federal prison in 1975, he returned to South America and collaborated
with the Medellin core organizations (Gugliotta, 1987).


119
intended behavior. In most organizations, rules are broken, procedures disregarded, and
performance programs poorly executed.
Inventory complete, cocaine ¡
stash ready for delivery
Chava puts
vinegar in
bedroom to
mask smell
of cocaine
delivery of
cocaine shipment
Bergin re-counts
cocaine packages
A
Chava counts, records
quantity & markings,
signs for delivery of
cocaine shipment
Figure 4-3 Stash House Inventory Routine
Trafficking routines fail for a variety of reasons. They may be poorly designed.
A practice or procedure intended to reduce risk from law enforcement may expose the
enterprise to a greater threat from illicit competitors, such as well-armed guerrillas and
paramilitaries. Some procedures lead to frequent communications breakdowns. The use


10
organizational learning that provides some clues as to how and why drug trafficking
organizations may learn. This literature examines the behavior of firms in competitive
markets and attempts to explain how they
build, supplement and organize knowledge and routines around their activities and
within their cultures, and adapt and develop organizational efficiency by
improving the use of the broad skills of their work forces (Dodgson, 1993, p.
377).
Significance of Research
This study is of broad theoretical and empirical interest to scholars and policy
makers. The research contributes to diverse bodies of literature in political science,
sociology, economics, organization theory, and criminology. Within political science, the
research transcends traditional sub-disciplinary boundaries between international
relations, comparative politics, and public policy. It focuses on actors (criminal
organizations and law enforcement agencies) and issue areas (transnational organized
crime and drug trafficking) often neglected in studies of international security, foreign
policy, globalization, and state making.7 While political scientists have long been
interested in organizational learning, the law enforcement agency and criminal enterprise
have rarely been topics of interest. Instead, they have directed their attention towards
more accessible subject matter, including government institutions, intergovernmental
policymaking bodies, and military organizations. These organizations are recognized by
7 Of course there are exceptions. Examples include Andreas (2000), Arquilla & Ronfeldt (1993, 2001),
Desch et al. (1998), Farer (1999), Friman & Andreas (1999), Griffith (1997), Nadelmann (1993), and
Williams (1994, 1998).
8 Examples of this literature include Allison (1971), Breslauer and Tetlock (1991), Brown (2000), Deutsch
(1966), Dodd (1994), Eden (Forthcoming), Etheredge (1985), Evangelista (1988), Feldman (1989), Golant
(1998), Goldgeier (1994), Haas (1990), Haas (1992), Hall (1993), Heclo (1974), Jervis (1976), Khong
(1992), Le Prestre (1995), Leng (1992), Leeuw et al. (1994), Levy (1994), McCoy (2000), Mendelson
(1993), Moltz (1993), Nagl (1999), Neustadt & May (1986), Nye (1987), Posen (1984), Reiter (1996),
Sagan (1993), Sikkink (1997), Snyder (1991), Stein (1994), and Weir and Skocpol (1985).


322
Smith, Dwight C., Jr. The Mafia Mystique. New York: Basic Books, 1975.
Smith, Dwight C., Jr. Paragons, Pariahs, and Pirates: A Spectrum-Based Theory of
Enterprise. Crime & Delinquency 26 (August) (1980): 358-386.
Smith, Dwight C., Jr. Illicit Enterprise: An Organized Crime Paradigm for the Nineties,
Handbook of Organized Crime in the United States. Eds. Robert J. Kelly, Ko-Lin
Chin, and Rufus Schatzberg. Westport: Greenwood Press, (1994): 121-150.
Smith, Joseph [pseud]. Intelligence Analyst, Drug Enforcement Administration,
Colombia. Bogot. 3 March 2000a.
Smith, Joseph [pseud]. Intelligence Analyst, Drug Enforcement Administration,
Colombia. Bogot. 15 March 2000b.
Smith, Joseph [pseud]. Intelligence Analyst, Drug Enforcement Administration,
Colombia. Bogot. 13 April 2000c.
Smith, Peter H. Semiorganized International Crime: Drug Trafficking in Mexico.
Transnational Crime in the Americas. Ed. Tom Farer. New York: Routledge,
(1999): 193-216.
Snidal, Duncan. The Game Theory of International Politics, Cooperation Under
Anarchy. Ed. by Kenneth A. Oye. Princeton: Princeton University Press, (1986):
25-57.
Snyder, Jack. Myths of Empire: Domestic Politics and International Ambition. Ithaca:
Cornell University Press, 1991.
Soto Velasco, Andrs. Official, National Planning Department, Republic of
Colombia. Bogot. 27 June 2000.
Southerland, Mittie D. and Gary W. Potter. Applying Organization Theory to Organized
Crime. Journal of Contemporary Criminal Justice 9.3 (1993): 251- 267.
Statement of Witness Appearing Anonymously Under the Federal Witness Protection
Program. Money Laundering. Hearings before the Subcommittee on Financial
Institutions Supervision, Regulation and Insurance of the Committee on Banking,
Finance and Urban Affairs, U.S. House of Representatives. One Hundred First
Congress, first session (14-15 November 1989). Washington: U.S. Government
Printing Office.
Stein, Janice Gross. Political Learning by Doing: Gorbachev as Uncommitted Thinker
and Motivated Learner ."International Organization 48.2 (1994): 155-183.


166
These adaptations allow them to continue to transship narcotics to consumer markets
even as drug enforcement efforts intensify. The expectation that all routes, no matter how
creative or ingenious, will eventually heat up or run into problems prompts smugglers
to continually refine existing transportation programs and develop new ones, even when
current arrangements are performing satisfactorily. As Homero emphasizes, the routes
always change... in order to avoid awakening suspicions, continually, so they dont get
heated (Homero, 2000). Nestor makes a similar point:
[Smuggling methods were] Constantly changing due to problems and
circumstances. Weather, surveillance, places, etc., all make you change. When a
place is heated up, you move... If you dont move after three years, then youre
dead meat. What happens is that within three years you will get some kind of
heat. You have to change your operation and location. You might be working
out of Grey Harbor, and everything is good there, but say some heat comes up, so
you have to check out the situation and make sure its safe again. For a month or
so, some place might get the whole heat (Nestor, 2000).
Table 5-3 Transportation Innovations in Drug Smuggling Operations
Use dry runs and small drug shipments to Break down multi-ton loads into smaller
test new routes or existing ones thought to consignments
be under surveillance
Modify transportation vehicles with secret
compartments for hiding drugs
Install extra fuel tanks and diesel engines to
increase the range of air and sea craft
Fly around secret radar installations
Develop alternate return routes in case of
law enforcement surveillance
Liquefy cocaine and transport it in water
tanks, bottles, and other recipients
Sources: Bagley (1999), BINLEA (2001), DEA (1993a, 2000) Derian (1999), Eddy et al.
(1988), Fuentes (1998), Henderson (1992), Johnson (2000), Kacerosky (1994),
Mermelstein (1990), Nstor (2000), Pan (1996), Prez Valdes (2000), Reyes (1999), Rice
(1989), Third Superceding Indictment (no date), Thompson (2000).
When trafficking enterprises develop these and other innovations, they expand
their operational repertoires, providing participants with a greater range of performance
Outfit vessels with state-of-the-art
surveillance and communications
technologies
Use decoy vessels to divert law enforcers
from vehicles that contain drugs
Air drop drugs over land or water to avoid
landing planes
Pack psychoactive drugs in ice tubes and
weld them to ship hulls


30
sisters, cousins, and in-laws as participants. However, the Herreras also recruited outside
the family for professionally trained participants. In all, the network contained more than
ninety members and exported approximately forty kilograms of cocaine a month to New
York and Miami. The Herreras acquired coca paste in Bolivia, Ecuador, and Peru. Using
human couriers carrying ordinary suitcases, the enterprise transported it to international
airports in Cali and Bogot, where the couriers whisked past compliant customs officials
on the organizations payroll. The coca paste was then transported to one of several
processing laboratories owned by the organization, where it would be processed into
cocaine hydrochloride. One processing lab captured by Colombian authorities outside of
Cali contained a twenty-five ton mechanical press for packing cocaine into fine sheets
and other equipment, materials and precursor chemicals worth approximately
US$800,000. Among the eight persons arrested at this lab was a chemistry professor
from Santiago University in Cali (Gage, 1975a, p. 26; Gugliotta, 1987; Moreau, 1976;
Semana, 1987).
Fully refined cocaine was distributed to one of three Herreras operating in
Barranquilla, Bogot, or Medellin. These transporters would export the drug to the U.S.,
again relying on drug-running couriers holding small quantities, generally between two
and four kilos. Couriers sometimes posed as students, carrying books that contained
cocaine secreted sheets. These students were paid between US$ 500 and US$ 1,000
per trip, plus expenses and a new suit of clothes to be used when transporting their
pedagogical materials to New York. False documents, including student visas, were
obtained by a family member that also served as a liaison with several of the networks
U.S.-based buyers. In addition to customs officials, the enterprise also received


247
enterprises survive drug enforcement pressure due their connections to influential public
officials. This, in part, explains the ability of the Cali cartels to elude drug enforcers
during the periodic crackdowns of the 1980s, which were aimed at their more prominent,
and violent, competitors from Medellin.
Alternatively, the small size of some enterprises allows them to slip through
enforcement nets focused on the largest trafficking conspiracies. Environmental
selection helps account for the decentralization of the Colombian drug trade during the
late 1990s. After Colombian and U.S. drug enforcers dismantled the Cali cartels and
other core organizations, surviving enterprises tended to be smaller, flatter, and more
circumspect in carrying out their activities. Enforcement pressure did not cause these
groups to change their practices and procedures, rather it made them more common by
selecting the largest organizations out of the system. To this extent, environmental
selection rather than adaptation accounts for change in the post-cartel drug industry.
However, according to officials interviewed in this research, it is also true that groups
that survived the crackdowns, including former remnants of the cocaine cartels, learned
from this experience that they were better off coordinating their illicit activities through
smaller, loosely coupled, ad hoc operations (see Chapter 5).
Learning and environmental selection are not mutually exclusive. Organizations
that survive intensified drug enforcement contain structures and behaviors more
conducive to environmental hostility. Over time, these attributes pass to other enterprises
that seek to establish themselves in the illicit drug industry. Newly emergent enterprises
may draw on surviving groups for inspiration, in which case they seek to pattern their
structures after the attributes of the organizational model. Or they may develop similar


300
Le Prestre, Philippe G. Environmental Learning at the World Bank, International
Organizations and Environmental Policy. Eds. Robert V. Bartlett, Priya A.
Kurian, and Madhu Malik. Westport: Greenwood Press, (1995): 83-101.
Leal Buitrago, Francisco. Ed. Tras las huellas de la crisis poltica. Bogot: Tercer
Mundo, 1996.
Leal, Francisco and Juan Gabriel Tokatlian. Eds. Orden mundial y seguridad: Nuevos
desafos para Colombia y Amrica Latina. Bogot: Tercer Mundo Editores, 1994.
Leal Buitrago, Francisco, Patricia Bulla Rodrguez, Mara Victoria Llrente, and Alfredo
Rangel Surez. Seguridad nacional y seguridad ciudadana: Una aproximacin
hacia la paz. Armar la paz es desarmar la guerra. Eds. Alvaro Camacho
Guizado and Francisco Leal Buitrago. Bogot: CEREC, (1999): 73-129.
Leavitt, Harold J. Applied Organizational Change in Industry: Structural, Technological
and Humanistic Approaches. Handbook of Organizations. Ed. James G. March.
Chicago: Rand McNally & Company, (1965): 1144-1170.
Leavitt, Harold J., William R. Dill, and Henry B. Eyring. The Organizational World.
New York: Harcourt Brace Jovanovich, 1973.
Lebow, Richard Ned. Social Science and History: Ranchers versus Farmers. Bridges
and Boundaries: Historians, Political Scientists, and the Study of International
Relations. Eds. Colin Elman and Miriam Fendius Elman. Cambridge: The MIT
Press, (2001): 111-135.
Ledwith, William E. Prepared statement by William E. Ledwith, Chief of International
Operations, Drug Enforcement Administration. The Narcotics Threat from
Colombia. Hearing before the Subcommittee on Criminal Justice, Drug Policy,
and Human Resources of the Committee on Government Reform, United States
House of Representatives, One Hundred Sixth Congress, first session (6 August
1999). Washington: U.S. Government Printing Office.
Ledwith, William E. Prepared statement by William E. Ledwith, Chief of International
Operations, Drug Enforcement Administration. The Crisis in Colombia: What
are We Facing? Hearing before the Subcommittee on Criminal Justice, Drug
Policy, and Human Resources of the Committee on Government Reform, United
States House of Representatives, One Hundred Sixth Congress, second session
(15 February 2000). Washington: U.S. Government Printing Office.
Lee, Rensselaer W. III. Dimensions of the South American Cocaine Industry. Journal
of Interamerican Studies and World Affairs 30.2-3 (1988): 87-103.
Lee, Rensselaer W. III. The White Labyrinth: Cocaine and Political Power. New
Brunswick: Transaction, 1989.


51
competitors, cooperate and pool resources, there appears to be no centralizing
coordination mechanism that provides direction and contract enforcement. Another
development is that many Colombian enterprises now avoid establishing distribution cells
in the U.S., often selling their loads to Mexican trafficking enterprises that complete the
international transportation and distribution process.
Trafficking enterprises in Colombia today are frequently led by people with
considerable criminal experience. A number of post-cartel start-ups were founded by
former mid-level managers of the Medellin and Cali core organizations. Other prominent
traffickers have formally retired from the day-to-day business of drug trafficking but
continue to invest in shipments and offer their advice when solicited. This suggests that
the knowledge and experience of the former cartels has not been lost. Traffickers from
the core organizations draw on contacts and experience in conducting their present
operations, applying previous knowledge to new ventures. Government officials
interviewed for this research suggest that one reason why existing operations remain
small is that surviving traffickers have learned from experience that they are better off
avoiding drug enforcers by purposely limiting the size of their operations.
Contemporary trafficking enterprises rely on basic smuggling routines developed
by their predecessors, while exploiting advances in communications and transportation
technologies when modifying these practices and procedures. Traffickers continue to use
a variety of maritime vessels to transport multi-ton loads and general aviation aircraft for
400-800 kilogram drug shipments. However, at least one enterprise has upgraded to the
next level of submarine technology. In September 2000 Colombian officials discovered a
partially finished submarine based on advanced Russian technology, including a


159
same errors as their predecessors. They do so by reading about the innumerable exploits
of the so-called cocaine cartels produced by Colombian and U.S. journalists, and by
drawing on the experience of entrepreneurs and participants that worked for these
organizations (Caldern, 2000; Camacho Guizado, 2000; Evans, 1999; Gale &
Rodriguez, 2000; Merchn, 2000; Morales, 2000; Sanchez, 1999; Vargas, 2000).
Applying Knowledge and Experience
Colombian trafficking enterprises change practices, procedures and performance
programs in response to information gathering and interpretation at virtually every stage
of the production, transportation, and distribution of psychoactive narcotics. Frequently,
though not inevitably, these adaptations are reactions to feedback from a hostile and
dynamic task environments. Facing challenges from illicit competitors and drug
enforcement agencies, trafficking enterprises adjust their practices and protocols as
quickly and often as circumstances warrant. Organizational survival, let alone
satisfactory profits, demands it. Most behavioral adjustment occurs at the level of tactical
routines. Trafficking enterprises routinely tinker with the practices and procedures used
in their day-to-day activities. Less common, but equally important, are changes in
strategic routines, including practices and procedures dealing with organizational goals
and long-range plans. Routines that help trafficking organizations meet their targets tend
to be retained and incorporated into existing performance programs; those that fail to do
so are often discarded. In the sections that follow I examine changes in tactical and
strategic trafficking routines over the last several decades. I begin by discussing
numerous tactical innovations in drug processing, transportation, and distribution
activities. I continue with an analysis of strategic innovations by trafficking enterprises,


APPENDIX A
RESPONDENT DATA
Name of
respondent
Date of
interview
Type of
respondent
Trafficker
learning?:
Affirmative,
negative, not
discussed
State learning?:
Affirmative,
negative, not
discussed
Bruce Bagley
8/5/99
U.S. academic
Affirmative
Not discussed
Ivelaw Griffith
8/4/99
U.S. academic
Affirmative
Not discussed
Renssalaer Lee
7/20/99
U.S. academic
Affirmative
Not discussed
Julian Arturo
10/28/99
Colombian
academic
Affirmative
Not discussed
Alvaro Camacho
3/6/00
Colombian
academic
Affirmative
Not discussed
Fernando Cubides
4/6/00
Colombian
academic
Affimative
Not discussed
Maria Victoria
Llrente
6/8/00
Colombian
academic
Not discussed
Affirmative
Andrs Lpez
Restrepo
2/28/00
Colombian
academic
Negative
Not discussed
Bernardo Parra
4/9/00
Colombian
academic
Negative
Not discussed
Eduardo Senz
3/31/00
Colombian
academic
Not discussed
Not discussed
Francisco Thoumi
7/19/99
Colombian
academic
Affirmative
Not discussed
Alvaro Valencia
Tovar
2/17/99
4/27/00
Colombian
academic,
retired general
Affirmative
Affirmative
Sergio Uribe
Ramrz
3/7/00
Colombian
researcher
Affirmative
Not discussed
Ricardo Vargas
Meza
3/24/00
Colombian
researcher
Negative
Not discussed
Carlos Velsquez
Romero
5/1/00;
7/15/00
Colombian
research, retired
government
official
Strong
affirmative
Not discussed
Ricardo Caldern
7/1/00
Colombian
journalist
Affirmative
Not discussed
262


210
with an eye towards ensuring the survival of their institution in the face of an increasingly
hostile environment (Soto Velasco, 2000). The temporal contiguity of the reforms and
the capture of several Cali capos produced a remarkable public relations success,
allowing General Serrano and other CNP leaders to claim that the drug enforcement
operations were a result of changes made to the institution.
However, the degree to which the administrative reforms impacted the behavior
of rank-and-file patrulleros (patrolmen) or improved service delivery remains unclear.
Some evidence indicates that the implementation of critical reforms was superficial, with
change restricted to top administrators. A critical intention of the cultural transformation
was to flatten the CNPs organic structure and eliminate a long-standing management
ethos characterized by few thinkers and many doers (Leal Buitrago et al., 1999).
Unfortunately, the CNP has been slow to decentralize due to resistance from high-
ranking officials. Several years after the 1997 reorganization, institutional authority
remains concentrated among upper-level administrators. Decisions and information
continue to flow vertically, from the top down, rather than horizontally. Generals,
colonels, and majors are now better trained in formulating strategic plans and measuring
performance indicators, but these skills have not changed how patrulleros carry out their
daily activities. Anecdotal data gleaned from interviews and readings suggests that CNP
patrolmen continue to lack input into institutional decision-making processes (Leal
Buitrago et al., 1999; Llrente, 1999; Llrente, 2000).
While the cultural transformation demonstrated the CNPs ability to respond to a
hostile environment, the learning process was restricted to upper echelon administrators.
Formal routines changed, reflecting the decision-making authority of top officials.


297
Junguito, Roberto and Carlos Caballero. Illegal Trade and the Underground Economy of
Colombia. The Underground Economy in the United States and Abroad. Ed.
Vito Tanzi. Lexington: Lexington Books, (1982): 285-313.
Kacerosky, Edward J. Search Warrant Affidavit of Special Agent Edward Kacerosky.
(1 fe 5 September 1994). Case No. 93-470-CR-WMH.
Kacerosky, Edward J. & David Tinsley. Search Warrant Affidavit of Special Agents
Edward J. Kacerosky and David Tinsley. Case No. 92-262-CR-DLG (1992).
Kamas, Linda. Dutch Disease Economics and the Colombian Export Boom. World
Development 14.9 (1986): 1177-1198.
Kash, Douglas A. and Jennifer A. Gorman. The International Sharing Program:
Fostering International Cooperation. FBI Law Enforcement Bulletin 70.2 (2001):
27-31. Http://www.fbi.gov/pub 1 ications/leb/2001 /lebO 1.htm [Accessed 14 July
2001],
Katzenstein, Peter J. Alternative Perspectives on National Security. ITEMS: Social
Science Research Council 49.4 (1995): 89-93.
Katzenstein, Peter J. Ed. The Culture of National Security: Identity and Norms in World
Politics. New York: Columbia University Press, 1996.
Kaufman, Herbert. The Natural History of Organizations. Administration and Society
7.2(1975): 131-149
Kaufman, Herbert. Time, Chance, and Organizations: Natural Selection in a Perilous
Environment. Chatham: Chatham House, 1985.
Keohane, Robert O. and Joseph S. Nye. Power and Interdependence. 2nd Edition, New
York: HarperCollins Publishers, 1989.
Kenney, Michael. Inter-American Narcotics Control Cooperation: A Case Study of the
U.S.-Colombian Anti-Drug Regime. Unpublished Masters Thesis, Center for
Latin American Studies, University of Florida, 1996.
Kenney, Michael. Boon or Bust?: A Macroeconomic and Sectoral Evaluation of the
Colombian Cocaine Industry. Latinamericanist 32.2 (1997): 3-10.
Kenney, Michael. A Synopsis of Guillermo Pallomaris Testimony in the 1997
Operation Cornerstone Trial of Michael Abbell feWilliam Moran. Transnational
Organized Crime 5.1 (1999).


90
state drug enforcement agencies organizations should be treated as propositions to be
confirmed or countered through careful analysis of empirical data. An attitude of
cautious skepticism is fully warranted at this stage of the inquiry.
Properties of Organizational Learning
While most organizations learn, if only in the tactical sense of matching routines
to problems, some leam better than others. Students of organizational learning have
identified a number of environmental and organizational properties that facilitate (or
impede) the process.
Environmental Hostility vs. Benevolence
As pointed out earlier in this chapter, organizational environments can be
distinguished according to their hostility or benevolence. Hostile environments contain
two or more interlinked organizations, at least one of which seeks to destroy the other.
By contrast, benevolent environments may contain competitors, but they do not seek to
destroy each other. A number of organization theorists argue that organizations in hostile
environments have strong incentives to respond to negative feedback because, if left
unchecked, problems quickly develop into threats to their survival (Hedberg, 1981, p. 14;
Popper & Lipshitz, 1998, p. 176). A parallel observation is that the costs of error in
hostile environments are potentially severe. Organizations that err in such environments
may face extinction at the hands of their competitors (Popper & Lipshitz, 1998, p. 176).
This survival imperative overrides other factors that may impede organizational learning,
including conflicting goals among participants and organizational inertia (Farkas, 1998,
p. 8).


211
However, the day-to-day informal practices and procedures of patrolmen apparently did
not. In this case, organizational learning was circumscribed by the failure of the reforms
to change the behavior of those that implement the bulk of the organizations service
delivery.
Adaptations in Drug Enforcement Operations
Although many of the basic tools of drug enforcement, including electronic
surveillance, undercover operations and controlled transactions, have been around for
decades, the investigative and enforcement strategies of the DEA and the CNP limited
their effectiveness until the 1980s. Policy makers and police officials did not recognize
that the disruption of Colombian trafficking enterprises required improved enforcement
methods until they developed into sophisticated, high-volume drug networks.
Banshee or Buy-Bust?
Operation Banshee offers a compelling case in point. In the early 1970s, the DEA
and New York City Police combined resources to investigate what appeared to be a series
of independent traffickers of Colombian origin operating in Manhattan and Queens. In
gathering and analyzing tactical intelligence from court authorized wiretaps and
participants that turned states evidence, criminal investigators gradually pieced together
the strands of a large transnational conspiracy. Operation Banshee eventually resulted in
the arrest and conviction of over a dozen members from the Bravo smuggling network
(see Chapter 2) (Hudson, 1974, pp. 1 & 18; Lubasch, 1976, p. 21; New York Times, 1976,
p. 20). While the operation was considered a success in law enforcement circles, many
DEA managers refrained from pursuing similar investigations. A former DEA official
that participated in the original investigation explains why:


67
organizational memories and other routines found in inter-subjective understandings
among participants.
Participants in Organizations
Organizations pursue their tasks through participants that make contributions in
return for inducements. Participants supply the human resources that allow organizations
to accomplish work. They define goals and objectives, form routines, perform tasks,
make decisions, share information, plan future actions, and address problems.
Participants join organizations for different reasons. Moreover, their assessments of
satisfactory inducements are subjective. Even wages are not a sufficient inducement for
all potential recruits. Human beings seek authority, status, power, fellowship, and a host
of other entitlements besides money. Complicating matters, participants frequently
belong to more than one organization, and their loyalty to the different collectives varies
considerably. Participants flow from one organization to another, carrying with them
values, norms, and beliefs about themselves and the roles they play (Barnard, 1938;
Cohen et al., 1988, p. 295; Feldman, 2000, p. 614; Leavitt, 1965, p. 1144; Posen, 1984, p.
43; Scott, 1998, p. 19).
In addition to their myriad contributions, participants are also a source of great
uncertainty for organizations. To accomplish their tasks, organizations require
coordination and control, and this is not easily achieved among participants with fluid
participation, wavering loyalties, and different conceptions of satisfactory inducements.
Moreover, even the most loyal and satisfied participants are limited in their ability to
pursue organizational aims. Like all human beings, they are bound by cognitive,
corporeal, and cathectic constraints. Of course, organizations are designed with these


7
Assistance Section (NAS) of the U.S. Embassy in Colombia complained of recent budget
cuts to his program, wondering aloud, How am I going to wipe it out? (Moreno, 1997).
Several officials from the Anti-Narcotics Directorate of the Colombian National Police
(CNP) underscored the difficulties in covering an area larger than the state of Texas with
only 2,500 agents (Buitrago, 1997).
A second prevalent explanation of the Colombian narcotics dilemma is that the
Colombian governments will to implement effective policy has been seriously
compromised by drug-related corruption. This charge has been leveled at the leaders of
the political system and the government agencies responsible for carrying out anti-drug
programs. In another article, Bagley argued that even if the Colombian government were
to receive sufficient aid, it would still face significant obstacles to implementing more
effective policies, principal among them being the significant level of drug-related
corruption among government agencies and institutions (Bagley, 1989-1990).
Nadelmann claims that during the 1970s and 1980s, drug traffickers succeeded in
corrupting all levels of government, including a significant share of the legislature and
cabinet-level ministers.... Even the Colombian military, which has remained a powerful
and relatively corruption-free organization, could not resist the inducements offered by
the drug traffickers (Nadelmann, 1993, p. 279).
In the 1990s, drug-related corruption continued to flourish. Ernesto Sampers
presidency (1994-1998) was severely weakened by credible allegations that his
presidential campaign had received a US$ 6 million donation from several cocaine
entrepreneurs based in Cali. In 1998, U.S. authorities seized a Colombian military plane
3 Andreas et al. (1991-1992) make a similar argument.


85
Learning under Ambiguity
To this point, the discussion has been deceptively simple. Organizations are
described as purposive agents that intentionally create and modify routines through
search and trial and error experimentation. However, in the real world of organizational
behavior, learning often occurs under conditions of profound ambiguity and uncertainty.
At each phase of the process, individuals and organizations confront a variety of
psychological and organizational factors that complicate learning: bounded rationality,
incomplete and imperfect information, biased inferences, ambiguous experience, and
organizational inertia. Recognition of these factors increases our understanding of
organizational learning and yields a more accurate and satisfying model of the process.
Figure 3-3 Complex Process Model of Organizational Learning
Some difficulties are due to basic limitations in human cognition and
coordination. All individuals and organizations operate under conditions of information
uncertainty and bounded rationality (Simon, 1997, 1955). Individuals and organizations
lack access to perfect information (March, 1999, p. 2). Much about past experience
remains unknown, and it is difficult to produce future anticipations based on present
actions. Individuals face significant limitations in their ability to analyze feedback from
complex environments. They use cognitive short cuts to make sense of the


296
Janeway, Maria [pseud]. Special Agent, Narcotics Unit, Judicial Police Directorate,
Colombian National Police. Bogot. 22 July 2000.
Jaramillo, Ana Maria. Samper cen con el cartel: Pallomari. El Tiempo (24 July 1997).
Http://www.eltiempo.com/diario/noticias/ppg01.htm [Accessed 25 July 1997].
Jaramillo, Jaime Eduardo, Leonidas Mora, and Fernando Cubides. Colonizacin, cocay
guerrilla. 3rd edition. Bogot, Colombia: Alianza Editorial Colombiana, 1989.
Jarosz, William W. with Joseph S. Nye, Jr. The Shadow of the Past: Learning from
History in National Security Decision Making, Behavior, Society, and
International Conflict 3. Eds. Philip E. Tetlock, Jo L. Husbands, Robert Jervis,
Paul C. Stem, and Charles Tilly. New York: Oxford University Press, 1993.
Jervis, Robert. Perception and Misperception in International Politics. Princeton:
Princeton University Press, 1976.
Johnson, John Thomas. Direct Testimony of John Thomas Johnson. Trial Transcript in
the United States of America vs. Harold Ackerman, Pedro Gomez Fernandez,
Carlos Giron, Gerardo Morales, Juan Jose Guel, Defendants, Case No. 92-262
CR-DLG, Volume 14(16 February 1993).
Johnson, Kevin. Panel Urges Merger of AFT, DEA. USA Today (2 Febmary 2000).
Http://web.lexis-nexis.com/universe [Accessed 24 June 2001].
Johnson, Loch K. Bombs, Bugs, Drugs, and Thugs. New York: New York University
Press, 2000.
Johnson, Tim. Samper now favors drug extraditions. Miami Herald (10 April 1997).
Http://www.herald.com/americas/digdocs/003218.htm [Accessed 12 April 1997].
Johnson, Tim. Latin American correspondent, Miami Herald. Bogot. 7 March 2000.
Johnson, Tim and Frank Davies. Several nations cooperate to break drug ring. (13
October 1999). Http://www.herald.com/americas/ [Accessed 16 December 1999].
Jonnes, Jill. Hep-Cats, Nares, and Pipe Dreams: A History of America's Romance with
Drugs. Baltimore: John Hopkins, 1996.
Jordan, Mary. 10 Plead Not Guilty in Loudoun Drug Case; Three Defendants Could
Face Life Sentences Under Tougher New Law. Washington Post (22 January
1985): D3. Http://web.lexis-nexis.com/universe [Accessed 16 Febmary 1999].
Jung, George. Former drug trafficker. Frontline: Drug Wars. 2000.
Http://www.pbs.org/wgbh/pages/frontline/shows/dmgs/interviews/iung.html
[Accessed 12 March 2001],


199
intelligence products, including link and telephone toll analyses of phone numbers and
financial transactions, and post-seizure analyses of narcotics, documents, computer
records, shipment containers, vehicles, and other equipment used by narcotics
organizations. These records and reports allow police officials to identify active criminal
enterprises and determine drug trafficking and money laundering trends. Collectively,
they form part of the organizational memories maintained by U.S. and Colombian law
enforcement agencies (Abadinsky 1997, p. 483; Furden, 2000; GAO, 1998, pp. 10-11).
The diffusion of intelligence among enforcement agencies is facilitated through
the extensive use of computerized information management systems, such as the DEAs
Narcotics and Dangerous Drugs Information System (NADDIS), and national
clearinghouses for counter-narcotics intelligence, including the El Paso Intelligence
Center (EPIC). NADDIS is a computerized file system housed at EPIC. As of 1998,
NADDIS contained the summaries of federal, state, and local law enforcement reports on
over 3,500,000 individuals, businesses, transportation vessels, and airfields identified
through the DEAs investigative reporting system. EPIC is an interagency organization
run by the DEA that centralizes the collection and dissemination of tactical-operational
drug intelligence. EPIC provides real time tactical and operational intelligence on the
movement of illicit drugs through land, sea, and air transportation routes in the Western
Hemisphere. With twenty-five plus years of experience, along with the agencys practice
of cooperating with law enforcement organizations in foreign countries, the DEAs drug
intelligence information management systems are unparalleled (BINLEA, 1999;
Constantine 1998; DEA, 2000; DEA, [no date] El Paso Intelligence Center; Frankel,
1997; GAO, 1993, pp. 11-13; GAO, 1998; GAO, 1999, p. 51; Nadelmann, 1993, pp. 207-


180
Additional Barriers to Learning
Other factors may also impede learning in Colombian trafficking enterprises,
including the obscure quality of tactical information regarding criminal activities, the
practice of rotating personnel among different cells, and the failure to apply negative
feedback to immediate circumstances.
Given the clandestine nature of trafficking in illegal psychoactive substances,
information regarding past and current activities is often hazy and opaque. To some
extent trafficking enterprises foster this situation by using code words and aliases to
communicate sensitive transactional details and maintaining strict secrecy about
impending operations. However, a lack of clarity in communication can have negative
implications for organizational adaptation. The judicious use of code words and aliases
can lead to frequent misunderstandings among participants as they struggle to make sense
of transactions and problematic situations. Interpretations based on wooly data and
miscommunication are likely to be distorted and may lead to ill-suited tactical and
strategic adaptations.
The practice of transferring personnel to different parts of an enterprise impedes
learning when it so frequent and extensive as to prevent the accumulation of institutional
knowledge. If participants and managers are shuttled to different cells every month or so,
they are likely to find it difficult to accumulate enough experience in any one role and
location to benefit the organization. However, periodic rotation can also enhance
learning by allowing participants with different areas of expertise to share their
knowledge and experience. Rotation of personnel is most likely to benefit trafficking
enterprises when it occurs regularly, but not too frequently, as on an annual or even bi-


219
Indeed, as players interact and learn from each other, it is expected that their second-
order preferences will change. At the commencement of learning games, players lack
access to complete information about their competitors second-order preferences. This
is due to the clandestine nature of drug trafficking and drug enforcement. However, as
learning games proceed players acquire information about their competitors behavior
through repeated interactions, and change their behavior accordingly.
Information is the lifeblood of competitive learning games. To be successful,
players require knowledge about the second-order preferences of their adversaries.
Moreover, they must prevent their opponents from acquiring knowledge about their own
tactics and strategies. In narco-narc learning games, players engage in a variety of
intelligence and counter-intelligence activities. Drug enforcement agencies gather
tactical and strategic intelligence regarding specific trafficking enterprises through a
variety of practices and procedures, including undercover operations and controlled
deliveries. Trafficking enterprises draw on human, technological, and documentary
sources to acquire intelligence about government counter-drug efforts. They monitor
radio frequencies used by Customs officials and drug enforcement agents. They exploit
the discovery process and the Freedom of Information Act to gather detailed information
about criminal investigations. They hire attorneys to research the statutory limitations on
specific drug enforcement practices, such as court-approved wiretaps (see Chapter 5).
To prevent the unwanted diffusion of information to competitors, nares and
narcos engage in counter-intelligence. Both sets of actors draw on a variety of practices
and procedures to protect sensitive communications. They use code words and
culturally-bound slang terminology to disguise the meaning of verbal communication and


109
activities. Entrepreneurs stay abreast of overall operations but intercede only when
significant problems develop (Thompson, 2000).
Given the large sums of money leaders invest in their drug shipments and the
numerous vulnerabilities facing all transactions, it is not surprising that leaders wish to
concentrate decision-making authority in their own hands. This is one way leaders seek
to avoid risk and uncertainty.10 Centralized decision structures also serve the interests of
mid-level managers and low-level participants. When something goes wrong, as it often
does in such a high-risk activity, centralization allows participants to defer responsibility
to their supervisors. It is more difficult to hold participants accountable (financially or
otherwise) for problems that develop from following the boss orders. If decision-makers
are competent, participants that face the greatest exposure to drug enforcement efforts
stand to benefit from closely adhering to their supervisors orders.
Flat Decision-making Hierarchies
While decision-making hierarchies in many Colombian trafficking enterprises are
centralized, they also tend to be organizationally flat. Small and large trafficking
enterprises contain few management levels. Small organizations typically contain two or
three layers of hierarchy. At the top is the leader who gives the orders. At the bottom are
the workers that carry them out. Sometimes a trusted intermediary stands between the
two to communicate information that could be used to incriminate the leader in a criminal
conspiracy.
10 However, centralization of authority may also serve to increase the risks facing entrepreneurs, as the case
of Miguel Rodrguez-Orejuela demonstrates. This apparent contradiction may be explained by a subtle but
important psychological factor. As in legitimate organizations, leaders of successful criminal organizations
may be prone to over-estimating their decision-making skills. Thus, an entrepreneur such as Rodrguez-
Orejuela may think that he is reducing the amount of uncertainty facing his enterprise by accepting greater
personal responsibility for decision-making at all levels of operations, when, in reality, he only ends up
increasing his own exposure to law enforcement.


131
p. 7; Gugliotta & Leen, 1989, pp. 121-122; Mermelstein, 1990, pp. 36-38; Valds, 1999,
p. 88).
Trafficking enterprises also use a variety of communications and transportation
technologies to conduct their illicit activities. Traffickers use standard telephones,
facsimile machines, beepers, cellular phones, and satellite phones to communicate
information. Some enterprises have recently exploited computer mediated
communication technologies for the same purpose. Participants in one enterprise
disrupted by government authorities in 1999 communicated through e-mail and online
chat rooms protected by firewalls. This organization also used cellular phone cloning
technology to steal phone numbers assigned to legitimate customers, and encryption
technology to cipher sensitive information regarding narcotics shipments and sales
(Farah, 1999b; Johnson, 2000; ONDCP, 2001, p. 90). A number of drug enforcement
agents interviewed in this research acknowledge that these developments have made it
increasingly difficult for them to gather intelligence through such traditional sources as
wiretaps placed on the telephone lines of suspected traffickers (Aero, 2000; Brandy,
2000; Crank, 2000; Smith, 2000a; Thompson, 2000).
Trafficking enterprises use numerous aviation, maritime and surveillance
technologies to perform work. The airplane has been a favorite tool of contraband
smugglers since the days of Prohibition, but contemporary transportation rings often use
more recent aviation technology. In the 1970s, a number of smuggling rings transported
marijuana from Colombia to the U.S. using twin-engine propeller planes such as the
Navajo Panther and Cessna 210. The following decade airborne numerous smugglers
turned to the more lucrative cocaine traffic, upgrading their aircraft to Gulfstream Aero


317
Reuters. Colombia captura al "ltimo capo de capos" del narcotrfico. (19 February
1998a).
Http://cnnenespanol.com/latin/COL/1998/Q2/19/narcotrafico.reut/index.html
[Accessed 20 February 1998].
Reuters. Colombia dismantles 'multinational' drug ring. (8 April 1998b).
Http://www.cnn.com/WORLD/americas/9804/08/RB0Q 1836.reut.html [Accessed
8 April 1998],
Reuters. U.S. Seizes $276 Million in Drug Cash. New York Times: 11A (19 December
1998c). Http://web.lexis-nexis.com/universe [Accessed 23 October 2000].
Reiter, Dan. Crucible of Beliefs: Learning, Alliances, and World Wars. Ithaca: Cornell
University Press, 1996.
Reuter, Peter. Disorganized Crime: The Economics of the Visible Hand. Cambridge: MIT
Press, 1983.
Revista Cambio 16. Tema de Portada: Thomas Constantine, Me preocupa que mis
nietos se vuelvan adictos. (1 March 1999).
Http://www.cambiol 6.com/l 999/mar l/portada2.htm#ThomasConstantine
[Accessed 6 March 1999].
Revista Cambio. La coca de hoy rinde el doble: Entrevista con Gabriel Merchan
Director Nacional de Estupefacientes. (21 February 2000).
Http://www.revistacambio.com/20000221/El Pais 02.asp [Accessed 3 March
2000],
Revista Cambio.com. Ingeniosos. (5-12 June 2000).
Http://216.35.70.57/Cambio/20000605/Secretos.cfm [Accessed 8 June 2000].
Reyes, Gerardo. Cartel bookkeeper to reveal secrets in Miami. Miami Herald (12 May
1997). Http://www.herald.com/americas/diudocs/054901 ,htm [Accessed 13 May
1997],
Reyes, Gerardo. Made in Miami: Vidas de narcos, santos, seductores, caudillos y
soplones. Bogot: Planeta Colombiana Editorial, 1999.
Reyes, Gerardo. Correspondent, El Nuevo Herald. Miami. 25 May 2000.
Reyes, Gerardo and Alejandro Santos. Witness: Cartel bought president. Miami Herald
(23 July 1997): Al. Http://www.herald.com/americas/digdocs/008346.htm
[Accessed 25 July 1997],
Reyes Posada, Alejandro. Paramilitares en Colombia: contexto, aliados y
consecuencias. Anlisis Poltico (12) (1991): 35-41.


33
countries. Consequently, drug enforcement resources were directed towards eradicating
Colombias booming marijuana industry. The year of the Cali cocaine bust, Colombian
authorities captured 78,000 kilograms of cannabis and destroyed almost 1.5 million
marijuana plants. In 1977, the year Colombian officials netted 32 kilograms of cocaine,
they also captured 187,077 kilograms of marijuana (see Table 2-1).
Table 2-1 Colombian Drug Enforcement Indicators, 1974-1979
Year
Persons
apprehended
Cocaine
interdicted
(kilograms)
Marijuana
interdicted
(kilograms)
Coca
plants
destroyed
Marijuana
plants
destroyed
Drug labs
destroyed
1974
1,305
164
90,000
0
37,500
6
1975
1,484
699
78,000
0
1,494,000
10
1976
769
138
27,000
0
25,000
15
1977
945
32
187,077
1,000
805,700
14
1978
555
194
158,272
4,195
431,614
34
1979
457
1,252
325,656
185,700
398,255
30
Source: CNP (2000)
Even regarding marijuana there was ambivalence among Colombian and U.S.
policy makers during much of the 1970s. In neither country did large groups of citizens
view marijuana consumption as a significant public health problem, and in Colombia at
least political elites did not see the growing drug industry as a threat to the countrys
national security interests. In the mid-1970s, when officials from the Ford and Carter
administrations pressured their Colombian counterparts about drug trafficking within
their national borders, Colombian President Alfonso Lpez Michelsen was quick to
remind them that U.S. consumers rather than Colombian suppliers represented the source
of the problem. Even within the U.S. policy community officials disagreed over the
degree to which the drug trade, in particular marijuana, threated U.S. interests. President
Nixons fight against crime agenda produced some changes in federal drug enforcement
policy (see Chapter 6), but the widespread moral fervor associated with the cocaine wars


32
operations from Colombia (Gage, 1975b, p. 24; Hudson, 1974, pp. 1 & 18; Lubasch,
1976, p. 21; New York Times, 1976, p. 20; Nieves, 1997, p. 5).
Government Drug Enforcement Efforts During Phase Three
In response to the growth of Colombian trafficking enterprises, the U.S. and
Colombian governments increased their drug enforcement activities, achieving a number
of successes in the latter half of the 1970s. Police agencies from both countries
conducted criminal investigations that disrupted several trafficking enterprises. Dozens
of participants from the Herrera and Bravo networks were captured, prosecuted, and
convicted of drug-related offenses. In 1975, in what amounted to the largest cocaine
seizure to date, Colombian police officials captured 600 kilograms of cocaine in a small
plane at the Cali airport (Frontline, 2000).
Notwithstanding these achievements, drug enforcement in Colombia was
inconsistent during this period, as reflected in the wide disparity of annual performance
indicators (see Table 2-1 below). For the entire year following the Cali airport bust,
Colombian officials managed to capture only 138 kilograms of cocaine. In 1977, the
amount of cocaine captured by Colombian authorities sunk to a dismal 32 kilograms. In
the U.S., complex conspiracy cases targeting cocaine operations, such as the Herrera and
Bravo networks, were few and far between. In the 1970s, many drug enforcement
managers were wary of these resource-heavy investigations, preferring low-cost buy-
bust operations targeting street dealers (CNP, 2000, p. 540; Nieves, 1997, p. 5).4
Drug enforcers mediocre counter-cocaine performance during the 1970s reflected
a widespread belief among U.S. and Colombian policy makers and police officials that
marijuana, rather than cocaine, was the primary drug threat facing their respective
4 Buy-bust operations are discussed in detail in Chapter 6.


Political Science Association meetings, San Francisco CA (29 August-1
September 1990).
286
Eden, Lynn. Constructing Destruction. Ithaca: Cornell University Press, forthcoming.
El Espectador. Invasin de Droga Heroica a Venezuela desde Colombia (9 August
1961a): 3.
El Espectador. Otro Eslabn de la Cadena de Traficantes Fue Descubierto (11 August
1961b): 9.
El Espectador. Caen Otros 6 Marihuaneros en Peligrosa Zona Bogotana (18 August
1961c): 9.
El Espectador. Otra traficante de Marihuana Fue Capturada Ayer en un Hotel (19
August 196Id): 9.
El Espectador. Nueve Capturas de Marihuaneros (20 August 196le): 9.
El Espectador. Caen 80 millones en drogas (4 June 1971a): 5A.
El Espectador. Siete presos por la cadena de traficantes de droga (5 June 1971b): 12A.
El Espectador. Caen 2 Jvenes Extranjeras con Cocana en Eldorado (4 August
1973a).
El Espectador. Capturados 6 extranjeros por Trfico de Drogas (6 August 1973b): 10A.
El Tiempo. Cae capo del cartel del Llano. (6 June 1997a): 9A.
El Tiempo. Cae sealado narco de Buenaventura (7 June 1997b): 20A.
El Tiempo. Capturan a dos cabecillas del cartel del Tolima (11 June 1997c): 12A.
El Tiempo. Los secretos del manual de vuelo de los Rodrguez (8 June 1997d): 12A.
El Tiempo. Polica captur a Fabio Ochoa Vsquez y a otros 29 capos: Redada
internacional de extraditables. (13 October 1999).
Http://www.eltiempo.com/hoy/iud a000tn0.html [Accessed 14 October 1999].
El Tiempo. Incautan herona en Cali. (5 May 2000a).
Http://www.eltiempo.com/hoy/iud n002tn0.html [Accessed 5 May 2000],
El Tiempo. Cay 'mua' de 15 aos. (7 June 2000b).
Http://www.eltiempo.com/hoy/bog a000tn0.html [Accessed 7 June 2000].


78
Organizational learning is a cognitive activity performed by participants that
observe and interpret events and make decisions based on their perceptions of reality.
While organizations are open systems, sensitive to exogenous pressures, the source of
learning is ultimately endogenous to the enterprise. This does not require participants
with access to complete information or unlimited computational abilities (see discussion
below). It does require participants that recognize they belong to a goal-oriented
collective, and act and manipulate information on behalf of this entity. To qualify as
organizational learning, change in routines must be the result of information acquisition
and processing activities on the part of participants. Otherwise, it is not certain that
organizational change is due to learning or alternative explanatory variables, such as
power or environmental selection.7
Learning Ecologies
Organizations interact with other organizations that share common task
environments. Organizations draw on task environments for resources, including
personnel, equipment, capital, and information. In common task environments, one
organizations knowledge and experience can be can be captured by others through the
diffusion of technology. Technology, understood broadly to include material equipment,
routines, and organizational memories, spreads through environments as organizations
copy or mimic one another. Often, targets of technology transfer are those organizations
that are seen as successful by their peers. These model organizations use superior
7 By distinguishing learning from other sources of organizational change, this study avoids a common
tautology in the literature. However, by making the distinction between learning and change so explicit, I
have increased the difficulty of demonstrating learning in subsequent chapters. My challenge is to show
that changes in routines by drug trafficking enterprises and drug enforcement agencies are due to
information acquisition, processing, and application, rather than other factors, such as power and
environmental selection.


77
manner, successful drug enforcement yields a process of natural selection that favors
some criminal enterprises over others. Trafficking organizations that survive drug
enforcement efforts may contain more effective combinations of decision-making
hierarchies, leadership styles, and operating procedures (Alchian, 1950; Kaufman, 1985;
Moe, 1984, p. 746; Nelson & Winter, 1981).
Over time, organizational forms better suited to survive hostile task environments
may populate the illicit drug industry (see Chapter 2). However, if survivors do not
change their routines as a result of information processing, then learning has not taken
place. Surviving enterprises may be more effective than their predecessors, at least in
terms of their ability to elude the state, but their ability is not due to organizational
learning. On the other hand, if participants in trafficking enterprises recognize that they
must change their organizational routines to survive hostile environments, and they do so
as a result of this recognition, then learning has transpired.
P4: Changes in the illicit drug industry may be due to natural selection rather than
organizational learning by individual enterprises.
Power
Learning
Environmental
selection
Change in
organizational
routines
Figure 3-2 Alternative Sources of Change in Organizational Routines


91
Organizations that operate in benevolent environments do not face such pressures.
They have fewer incentives to innovate and improve task performance, both of which
require attention and other scarce organizational resources. Organizations in benevolent
environments are also prone to organizational inertia, further reducing their ability to
learn from potential or actual errors (Hedberg, 1981, p. 14).
P6: Colombian drug trafficking enterprises (and drug enforcement agencies) are
more likely to leam if they operate in hostile environments.
P7: Colombian drug trafficking enterprises (and drug enforcement agencies) are
less likely to leam if they operate in benevolent environments.
Organization Size and Decision-Making Hierarchies
A number of organizational factors facilitate or impede learning, including size
and levels of management. Size refers to the number of participants that belong to a
particular organization. Management levels refer to the number of separate management
layers within an organization, top to bottom. Organizations with four or fewer levels of
management contain flat decision-making hierarchies. Organizations with five or more
discrete management layers contain tall hierarchies.
Size and management layers influence learning for reasons having to do with the
collection and analysis of information. Cognitive schemas, personal interests, and
computational limitations hinder participants information processing skills. Blind spots,
filtering, and biased or mistaken inferences impede the collection and analysis of accurate
and reliable data. While organizations require a minimal number of participants to
engage in information processing, the greater the number of participants, including
entrepreneurs, managers, and workers, that handle information, the greater the probability
that it will be manipulated, distorted, or miscalculated in some way. Conversely, the


31
protection from officials in the Colombian police and judiciary, as well as leading
members of Colombian society that invested in cocaine shipments (Gage, 1975a, p. 26).
Bravo Enterprise
The Bravo network contained several large and loosely connected drug rings that
smuggled large quantities of cocaine, heroin, and marijuana from Colombia to the U.S.
From 1972 to 1974, the Bravo enterprise distributed drug shipments throughout North
America, including Manhattan, Queens, Miami, Los Angeles, San Antonio, Texas and
Toronto. Large cocaine shipments, which in those days meant anything greater than four
kilos, were transported in containerized cargo, speed boats, and private planes using a
variety of routes, including Panama, Mexico, Puerto Rico, Canada, and even Munich,
West Germany. Smaller quantities of cocaine and heroin were hidden inside luggage or
clothes and transported by couriers. The Bravo network developed a number of creative
smuggling methods, including soaking clothes in a liquid solution of cocaine
hydrochloride, swallowing drug-filled prophylactic condoms, and hiding cocaine in false-
bottomed wine bottles, painting frames, hollow ski poles and wooden hangars, specially
modified bras and girdles, and a dog cage, complete with live canine. On reaching the
U.S., the cocaine was distributed by members of the Bravo network, which included
approximately 150 couriers, brokers, and distributors working in the New York
metropolitan area alone. Although U.S. police officials eventually captured, prosecuted
and convicted over a dozen members of what federal prosecutors described as the
biggest Colombian narcotics organization ever uncovered, several entrepreneurs,
including Alberto Bravo, Carlos Bravo and Griselda Blanco, were able to continue their


97
a transnational commodity network. The target-oriented tasks of participating enterprises
are contingent on the specific good or service they supply. The task of purchasing groups
is to buy coca paste, cocaine base or opium gum from farmers or intermediaries and
transport them to processing laboratories for further refinement. Processing groups turn
semi-refined substances, such as coca paste, cocaine base, opium gum and morphine, into
fully refined narcotics, such as cocaine hydrochloride and heroin.3 Transportation rings
send finished drugs to international transshipment points or directly to consumer markets
through a variety of smuggling methods. Importation groups receive shipments of
cocaine and heroin in consumer markets and distribute them to independent wholesalers.
Brokers arrange transactions among different participants, such as exporters and
importers, or importers and wholesalers. Money launderers receive illicit proceeds from
wholesale or retail transactions and clean them through the international banking
system. Along each node of the production-processing-transportation-distribution-
2 This description of the narcotics industry begins at the purchasing link because in Colombia independent
farmers that do not belong to trafficking groups generally grow coca and opium poppies. During the 1980s
and 1990s, there were some reports of leaders from core organizations providing seeds, fertilizers and
technical assistance to small-scale farmers, in return for the right to purchase the ensuing crop at pre
determined prices. However, even under these circumstances, farmers were not direct participants in
trafficking enterprises.
3 Cocaine processing occurs in three stages. In stage one, the raw material of coca leaf is processed into
coca paste, a rudimentary derivative that contains a concentration of desired alkaloids but many undesired
impurities as well. In stage two alkaloids are further concentrated and some impurities are removed as coca
paste is refined into cocaine base. The latter substance is still unsuitable for human consumption, at least
by the exacting standards of First World cocaine consumers, thus requiring another round of processing. In
stage three, more impurities are removed, as cocaine base is refined into the finished product, cocaine
hydrochloride or cocaine, which can be consumed by inhalation or injection. In the Colombian industry,
dmg farmers frequently process coca paste to cocaine base right on the farm. Coca paste is bought by
purchasing groups that transport the product to processing labs. There is some variation at this phase of the
Colombian industry. Some laboratories just process paste to base. The cocaine base then goes to other labs
that complete the final process stage. Other laboratories handle both the second and third processing
stages, converting paste to base and base to cocaine hydrochloride.


65
Organizations use routines to structure authority relations, allocate resources,
make plans, perform tasks, assign roles, communicate information, build culture, and
solve problems. Routines describe how organizations structure authority among different
layers of decision-making hierarchy. They describe the functional responsibilities of
different roles within the collective. Routines explain how organizations create the goods
or services that they produce. They show how organizations maintain records of at least
certain aspects of their operations. Routines determine how information is shared within
the collective and who has access to what data. They express important values and
provide a means of constructing organizational culture. Finally, routines describe how
organizations respond to existing problems and plan future endeavors (Cyert & March,
1963, pp. 103-104; Decker et al., 1998, p. 407; Levitt & March, 1988, p. 320).
Routines are nested in several levels of organizational behavior. Standard
operating procedures are simple rules of thumb for accomplishing basic tasks, as when a
participant responds to a stimulus by performing a specific action. Basic procedures are
combined to create more complicated action-sequences called performance programs and
repertoires. A performance program is simply a cluster of procedures for dealing with a
specific situation. A repertoire is the sum of the organizations performance programs for
organizing behavior in an area of activity (March & Simon, 1958, p. 141; Posen, 1984, p.
44).2
Like their building blocks, performance programs are triggered by stimuli. The
sounding of an alarm bell in a fire station provokes a fairly elaborate program of activity
carried out by numerous participants. Two or more procedures can be combined to form
2 This is similar to Swidlers notion of cultural tool kits. According to Swidler, culture influences action by
providing a repertoire or "tool kit" of habits, skills, and styles from which people construct "strategies of
action." See Swidler (1986) and Kier (1997).


124
and usefulness of their knowledge, they have the opportunity reduce their sentences by
cooperating with prosecutors and police officials. In some cases, they may even be
allowed to enter the U.S. Marshals Witness Security Program and begin life anew
somewhere in the U.S. under an assumed identity.16 Leaders attempt to reduce
participants temptation to defect by providing their legal fees, making subsistence
payments to their immediate family members, and reminding participants of the dire
consequences facing their loved ones should they cooperate with government officials.
Turnover in Trafficking Enterprises
Turnover in trafficking enterprises is high. Participants frequently enter and exit
these criminal organizations, contributing to their dynamic and fluid character. Sources
of turnover include apprehension by law enforcement authorities, voluntary retirement,
and involuntary expulsion. Every year Colombian police officials remove thousands of
participants through capture and arrest (see Table 4-4). However, many of these
perpetrators are later released due to insufficient evidence, making the disruption more
temporary than police officials would prefer (Garzn Saboy, 1997, pp. 392-394; Sierra,
2001).
Many participants retire from the business following a period of sustained
17
activity. Reasons for retirement are numerous and vary according to the personalities
in the federal prison system and 4 years and 7 months in state prison systems. However, these average
prison sentence figures include marijuana trafficking offenses, which are considerably less severe than
cocaine and heroin trafficking offenses and thereby pull the mean down. In the federal system, chug
trafficking alone accounted for 40% of all felony convictions for 1996 and 1998 (Bureau of Justice
Statistics, 1996; U.S. Sentencing Commission, 1998).
16 Both Mermelstein and Guillermo Pallomari, a former chief accountant and assistant to Miguel Rodriguez
Orejuela, benefited from this program. See Mermelstein (1990), Statement of Witness, (1989), and
Pallomari (1997).
17 In order to do so, however, they must be in good standing with the organization, which includes not
having any outstanding financial obligations from previous business dealings. A participants decision to
depart is often cleared with his or her immediate supervisor and perhaps the enterprise leader. Participants


129
of the family boss may signal the termination of the enterprise itself. However, larger,
more formally organized groups, such as core organizations and even medium-sized
enterprises can survive a limited amount of executive turnover. When leaders of these
organizations are incarcerated they will often turn over the day-to-day management of
trafficking operations to trusted subordinates, such as family members or criminal
associates.21
In some enterprises the act of turning over operations may be as simple as
supplying the new leader with the necessary contact information, and vouching for his
credibility with suppliers and wholesale customers. This service may be rendered by the
original entrepreneur in exchange for a fee or a cut of the profits of the continuing
conspiracy. Frequently, the replacement is a long-time confidant that has worked his or
her way up the organization. The head of an elite drug enforcement unit in the
detectives branch of the CNP explains:
So the ones known by their aliases as Chupeta and Cuchilla, they are in the prison
and they are not trafficking, but who is trafficking for them [are]... the people that
remain on the outside. They are the ones that rent [the routes]. [Chupeta and
Cuchilla] ceded their routes to these people that were their men of confidence and
they told them, Take your route, I rent it to you, I have nothing to do with this.
They turn over the routes and just receive some dividends for renting them. So
then, the ones that become stronger are the ones that were from the lower levels
[of the organization] and now they are owners of those routes (Molano, 2000,
authors translation).22
21 Following his imprisonment, Miguel Rodrguez Orejuela ceded control of his large criminal operations to
his son William. When Colombian authorities arrested Ivan Urdinola Grajales in 1992, he reportedly
relinquished control of his trafficking enterprise to his brother and criminal associate, Julio Fabio. Three
years later, Julio Fabio himself surrendered to Colombian officials, whereupon day-to-day management of
the organization passed to Ivans brothers in law, Orlando and Arcngel Henao Montoya, (DEA, 1994, p.
4; Farah, 1997; Farah & Moore, 1997; Lyons, 1997).
22 A Colombian government prosecutor with experience in narcotics investigations makes a similar point:
... every time that they capture a group of narcotics traffickers... naturally, they never catch all of them.
There are people that sell the information, people that know whom are the contacts outside the country and
here inside... (Morales, 2000, authors translation).


316
Rachal, Patricia. Federal Narcotics Enforcement: Reorganization and Reform. Boston:
Auburn House, 1982.
Radio Cadena Nacional de Colombia (RCN). Tras "milenio dos", General Serrano
reitera lucha antidrogas. (12 Abril 2000). Http:/Avww.rcn.com.co/actua3.htm
[Accessed 13 April 2000],
Ramos, George. 10-Year Search for Drug Ring Suspect Led Police to Irvine. Los
Angeles Times (21 February 1985): 1. Http://web.lexis-nexis.com/universe
[Accessed 15 February 2000].
Randall, Stephen J. Colombia and the United States: Hegemony and Interdependence.
Athens: University of Georgia Press, 1992.
Reiter, Dan. Crucible of Beliefs: Learning, Alliances, and World Wars. Ithaca: Cornell
University Press, 1996.
Rementera, Iban de. Economa y Drogas. Colombia Internacional (20) (1992).
Republic of Colombia. Embassy of Colombia. Colombia Bulletin: February/March 1997
(1997). Washington. Http://colombiaemb.org/bulletin/femar97.html [Accessed 15
April 1997],
Republic of Colombia. Colombia's War on Illegal Drugs: Actions and Results for 1997
(1998a). Http://www.colombiaemb.org/wpaper/wp97.html [Accessed 29 June
1998].
Republic of Colombia. Embassy of Colombia. Colombia Bulletin: January/February
1998 (1998b). Washington. Http://colombiaemb.org/bulletin/iafe98.html
[Accessed 29 June 1998].
Reuter, Peter. Disorganized Crime: The Economics of the Visible Hand. Cambridge: MIT
Press, 1983.
Reuter, Peter. Eternal Hope: Americas Quest for Narcotics Control. The Public
Interest 79 (Spring) (1985): 79-95.
Reuter, Peter. After the Borders are Sealed: Can Domestic Sources Substitute for
Imported Drugs? Drug Policy in the Americas. Ed. Peter H. Smith. Boulder:
Westview Press, (1992): 163-177.
Reuter, Peter, Gordon Crawford, and Jonathon Cave. Sealing the Borders: The Effects of
Increased Military Participation in Drug Interdiction. Santa Monica: RAND,
1988.
Reuter, Peter and John G. Haaga. The Organization of High-Level Drug Markets: An
Exploratory Study. Santa Monica: RAND, 1989.


237
cycles. With fewer information processing channels, they can change their routines in
response to feedback more quickly than drug enforcers. Traffickers also lack the myriad
of bureaucratic protocols and regulations that slow the action response rates of drug
enforcement agencies. Moreover, unlike their government competitors, traffickers have a
clear and compelling reason to adapt their routines when necessary. Their survival as
organizations depends on it. In contrast, drug enforcers are not always best served by
enacting the most effective counter-drug strategies. Without traffickers to battle, they
have no reason to exist. The drug enforcement dilemma fosters inertia on the part of drug
enforcement agencies, and provides an additional advantage to their criminal adversaries.
The cumulative effect of these advantages suggest that drug traffickers will
continue to experience success in even the most hostile law enforcement environments,
such as the United States. Even when law enforcers succeed in dismantling certain
enterprises, other players will emerge, driven by a desire to reap the substantial profits in
the transnational drug trade. This suggests important lessons for state counter-drug
policies that privilege law enforcement efforts. The analysis also suggests an alternative
course of state action, one that requires substantial double-loop learning on the part of
political elites and policy makers. These lessons and alternatives are examined in the
concluding chapter of this study.


221
enterprises will be immediately removed from the illicit drug industry, particularly those
with few participants and a single dominating figure. Trafficking enterprises that survive
a drug enforcement operation will be compelled to make additional changes or face
eventual removal. When counter-drug activities cause a particular player to be selected
out of the game, other parts of the transnational network or different criminal enterprises
may still reap the benefits of the interaction through the diffusion of relevant knowledge
among the remaining players.
In targeting additional trafficking groups, particularly those that have absorbed
the lessons of previous interactions, drug enforcers are compelled to alter their
enforcement tactics, but not necessarily their strategies. Change in tactics and mistakes
by traffickers will lead to the identification and apprehension of other players that
subsequently face the same change-removal decision point. In this manner, nares and
narcos adjust and re-adjust their activities as they interact repeatedly with each other.
Over time, traffickers that fail to alter existing routines or create new ones in response to
previous interactions are more likely to be selected out of the system. When this
happens, other players emerge to capture to capture their share of psychoactive drug
markets.
One consequence of competitive learning games is that players develop
increasingly sophisticated strategies and practices as they seek to outwit their adversaries.
Law enforcement success compels traffickers to come up with communication or
transportation innovations that are not recognized by police agents. The use of Internet-
based technologies and encryption software by trafficking groups developed in response
to law enforcers penetration of telephonic communications through wiretaps and other


122
Participants as a Source of Risk in Trafficking Enterprises
Participants threaten the integrity of trafficking organizations in a variety of ways.
One source of vulnerability stems from the cognitive, psychological and physical
limitations facing all, even honest and well-intentioned, participants. Trafficking in
illegal drugs is stressful, and some participants do not perform well under pressure. Press
reports and testimonial accounts relate numerous incidents of physically exhausted or
emotionally distraught drug couriers whose strange behavior attracts the suspicion of
police officials. In his autobiography, Jorge Valds recalls meeting a courier at the San
Francisco airport who had just transported three kilos of cocaine across the U.S. The
courier was so nervous that he refused to claim his drug-laden luggage from the baggage
claim. Valds finally grabbed the bag and escorted his mule to a hotel, whereupon the
latter admitted that he was not suited for this type of work (Molano, 1997; Otis, 2000;
Valds, 1999, pp. 62-63).
Other participants may perform satisfactorily under pressure but exercise poor
judgment in their professional or personal matters. Hedonistic lifestyles create problems
for the enterprise when participants attract unnecessary attention through conspicuous
consumption, or commit errors in judgment while under the influence of drugs. Poor
judgment is not the exclusive preserve of the intoxicated. Laziness, shortsightedness, and
inclement weather may cause participants to commit such egregious errors as conducting
business related telephone calls from their home residence or repeatedly using the same
payphone to discuss transactions (Derian, 1999; Thompson, 2000).
Beyond nerves and stupidity, participants present a potential threat to trafficking
enterprises whenever their individual interests run counter to those of their employers,


CHAPTER 1
INTRODUCTION
Over the past two decades, the governments of Colombia and the United States
identified drug trafficking as a national security threat to their respective countries; a
threat they sought to eliminate by attacking individuals and groups engaged in production,
transportation, and distribution of controlled substances. Despite devoting considerable
resources to drug enforcement policies since the early 1980s, the two governments failed
to reduce illicit drug flows between their countries. Indeed, according to the best
available estimates, in Colombia the production of cocaine and heroin has increased
substantially in recent years.
What accounts for this intractable dilemma? Why have the Colombian and U.S.
governments been unable to impede the production of psychoactive drugs? Why are
these governments unable to resolve a national security threat that emanates not from
sovereign neighbors but from presumably weaker non-state actors? This study answers
these questions through a comparative case study of Colombian trafficking enterprises
and U.S. and Colombian drug enforcement agencies. Drawing on organization theory and
extensive primary and secondary-source data, the study develops a learning-based
explanation for the persistence of drug production in Colombia.
In this chapter, my aim is to set the stage for what follows. I begin with a
description of illicit drug dilemma in Colombia and conventional explanations offered for
this conundrum. I continue with a brief discussion of the alternative explanation that
1


48
American Weekly Report, 1995a; 1995d; Mattiesen, 2000, pp. 309-310; Nieves, 1997, p.
11; Semana, 1995, p. 28; Thoumi 1995, pp. 226-228).
In February 1995, President Samper launched a comprehensive assault on the
leaders of the Cali trafficking network. Over the next several months, the Colombian
National Police, led by General Rosso Jos Serrano, captured eighteen tons of coca
leaves and two tons of cocaine hydrochloride, destroyed 195 processing laboratories,
eradicated 6,000 hectares of coca plantings, and detained 616 suspected traffickers.
While impressive, the results did not dramatically exceed the drug enforcement indicators
in preceeding years (see Table 2-2 below).
In June the Samper government received a major boost when a special joint
military-police unit captured Gilberto Rodrguez Orejuela, the alleged maximum leader
of Cali traffickers. Over the next year, the Colombian National Police, with the
assistance of the DEA and the CIA, exerted enormous pressure on the remaining Cali
leaders, all of whom were eventually either captured or surrendered to government
authorities. In combination with the Peruvian governments controversial shoot-down
policy in the coca paste air bridge between Peru and Colombia, the Cali crackdown
produced ripples throughout the cocaine industry. By June 1995, the price of coca paste
in Colombia had dropped 50%. By mid-September of the same year, wholesale and retail
cocaine prices in New York City, the Cali networks most important U.S. market,
increased 50% and 30% respectively. Similar increases were noted in other U.S. cities
that received substantial cocaine supplies through the Cali transportation and distribution
network (Caulkins & Reuter, 1998, p. 603; Geopolitical Drug Dispatch, 1995; Krauss,


6
Explaining the Colombian Drug Dilemma
What accounts for the persistence of illicit drug production in Colombia at a time
when the Colombian government has devoted more financial and human resources to its
anti-drug efforts than ever before? Why has the Colombian government been unable to
reduce the production and transit of illicit narcotics within its national territory? Over the
past decade, scholars and policy makers have produced a cottage industry of literature
that attempts to answer these questions. Several prevalent explanations have emerged
from these studies. For example, many argue that the Colombian government is unable to
resolve the narcotics dilemma because it lacks sufficient anti-drug resources. In an article
that was published in the influential journal Foreign Affairs, Bagley argued that the
Colombian government lacks the necessary resources to effectively counter the drug
trade. In order for real progress to be made he insisted that the U.S. should provide
Colombia and the other Andean governments with the $2-3 billion annually that would
be required to establish effective control over drug producing areas (Bagley, 1988, pp.
87, 89). In a more recent study of the cocaine industry, Clawson and Lee argued that, in
spite of increased international assistance over the past ten years, Colombia and the other
Andean governments still lack the necessary resources to stop drug production and
trafficking in their countries (Clawson & Lee, 1996, p. viii).
Similarly, U.S. and Colombian officials charged with implementing counter
narcotics programs in Colombia maintain that they do not receive adequate resources to
effectively carry out their mandates. The U.S. and Colombian officials interviewed in
this research emphasize that insufficient resources were a significant impediment to
achieving greater counter-narcotics success in Colombia. The Director of the Narcotics


223
environmental and organizational conditions do not determine outcomes in competitive
learning games, they do tilt the playing field in favor of criminal organizations. I analyze
these factors below.
Playing Catch Up
As competitive learning games commence, drug enforcers do not know the
identities of specific trafficking enterprises, nor the practices and procedures they use to
conduct their activities. This provides traffickers with a head start over their competitors.
As long as they remain unidentified, traffickers can conduct their illicit operations free
from penetration by drug enforcers. While traffickers may also be unaware of the
identities and methods of police agencies, unlike their state adversaries they do not
require this information to carry out their activities. This places drug enforcers at a
distinct disadvantage vis--vis their illicit competitors. Police agents cannot penetrate the
operations of unknown entities, nor can they easily manufacture intelligence about
criminal acts prior to their commission. By nature most counter-drug intelligence is
reactive, which means that nares are generally a step, or more, behind narcos. When
drug enforcers identify a new communication or transportation technique, adaptive
trafficking enterprises have already moved onto to other routines. Drug enforcers seek to
bridge the gap between their understanding of contemporary trafficking enterprises and
what these non-state actors are actually doing. Sometimes they succeed. Witness the
disruption of dozens of major Colombian trafficking enterprises in recent years.
However, once these known enterprises are removed from the industry, drug enforcers
again confront a host of criminal operations about which they know next to nothing. The
quest for real time intelligence is a perennial challenge in drug enforcement. No matter


CHAPTER 2
HISTORICAL OVERVIEW OF THE
ILLICIT DRUG INDUSTRY IN COLOMBIA
Conventional wisdom maintains that the Colombian drug industry is a recent
phenomenon. A number of government reports, Congressional hearings, and journalistic
and academic studies date the beginning of the Colombias drug trade in the 1960s, or
even the early 1970s.1 Contrary to these accounts, Colombian smugglers have been
involved in the transnational narcotics traffic since well before the 1960s. Indeed,
criminal entrepreneurs have been smuggling cocaine and heroin from Colombia almost
since national governments and international conventions formally banned these
commodities. Since the 1930s, Colombian traffickers have built on their countrys
vibrant history in import and export contraband smuggling to organize for the purpose of
participating in this illicit commerce.
This chapter presents an historical overview of the illicit drug industry in
Colombia, focusing on the social organizations that produce, process, and export
marijuana, cocaine, and heroin in Colombia. The chapter highlights the dynamic and
fluid nature of these criminal enterprises, and lays the empirical foundation for the
1 In 1980, a Drug Enforcement Administration official testified before the U.S. Senate Subcommittee on
Investigations that prior to the mid-1960s drug trafficking in Colombia was largely confined to importing
cocaine for the domestic market and exporting small quantities of marijuana to neighboring countries
(Clifford, 1980, p. 474). Several years later, the Presidents Commission on Organized Crime presented an
influential report that traced Colombian involvement in the U.S. cocaine trade to the early 1960s (PCOC,
1986, p. 77). These views are accepted by a number of researchers that have written about the development
of the Colombian drug trade, including MacDonald (1988, p. 27), Carney (2000, p. 8) and Chepesiuk
(1999a, p. 56). In a recent edited volume on cocaine based on archival research, the contributor to the
Colombia case study inexplicably dates the formal appearance of the countrys cocaine industry to 1972
(Roldn, 1999, p. 176; also see Gootenberg, 1999).
17


245
cases, power and environmental selection may provide the stimulus for participants to
engage in organizational learning.
Power
Learning

Change in
organizational
routines
Figure 7-1 Sources of Change in Organizational Routines, Amended
On several occasions, U.S. drug enforcement agencies have proven vulnerable to
the preferences of more powerful political and bureaucratic actors. In 1968 and 1973,
federal drug enforcement agencies underwent two major administrative reorganizations,
culminating in the creation of the Drug Enforcement Administration. The
implementation of these reforms was due to the intervention of policy makers in the
White House and the U.S. Congress, rather than information gathering and interpretation
by drug enforcers. Dissatisfied with federal counter-drug efforts, more powerful
bureaucratic actors sought to impose their views of drug enforcement on police officials.
Ultimately, both reforms failed to produce the desired objectives, in part, due to a lack of
learning by drug enforcers. The two major bureaucratic agencies impacted by the


320
Senz Rovner, Eduardo. Professor of Economics, National University of Colombia.
Bogot. 31 March 2000.
Sagan, Scott D. The Limits of Safety: Organizations, Accidents, and Nuclear Weapons.
Princeton: Princeton University Press, 1993.
Salazar, Alonso and Ana Maria Jaramillo. Medellin: Las subculturas del narcotrfico.
Bogot: CINEP, 1996.
Salerno, Ralph and John S. Tompkins. The Crime Confederation. Garden City:
Doubleday, 1969.
Sanchez, Thomas [pseud]. Special Agent, Drug Enforcement Administration.
Washington. 22 July 1999.
Schein, Edgar H. Organizational Culture. American Psychologist 45.2 (1990): 109-119.
Scott, Richard W. Organizations: Rational, Natural, and Open Systems. 4th edition.
Upper Saddle River: Prentice Hall, 1998.
Semana. El Destape de Lehder. (5 July 1983): 22-26.
Semana. Cay el pionero. (22 December 1987).
Semana. Un ao despus. (29 November 1994): 38-42.
Semana. En Mi Casa Mando Yo. (14 February 1995a): 28-29.
Semana. El golpe. (13 June 1995b): 38-46.
Semana. La flor maldita. (10 September 1999).
Semana. Los Nuevos Narcos. (8 May 2000a): 70-76.
Semana. El nio mua. (26 June 2000b): 40-42.
Semana. Yellow submarine (11 September 2000c).
Http://216.35.197.109/958/actualidad/ZZZZ3QRPWCC.asp [Accessed 12 May
2001],
Senge, Peter M. The Fifth Discipline: The Art and Practice of the Learning Organization.
New York: Doubleday, 1990.
Serrano Cadena, Rosso Jos. Las palabras del general. Bogot: Editorial Norma, 1998.


BIOGRAPHICAL SKETCH
Michael Kenney received a Bachelor of Arts degree in political science from the
University of Massachusetts at Boston in 1990. Upon graduation, he worked as a VISTA
Volunteer-drug prevention specialist for the Center for Drug-Free Living in Orlando,
Florida. In 1992, he joined the Peace Corps, working as an agricultural extension agent
in Ecuador for two years.
After completing his Peace Corps service and traveling in Colombia, Mr. Kenney
enrolled in the Center for Latin American Studies at the University of Florida. In May
1996 he graduated with a Masters degree in Latin American Studies. His thesis, titled
Inter-American Narcotics Control Cooperation: A Case Study of the U.S.-Colombian
Anti-Drug Regime, focused on U.S.-Colombian counter-narcotics policies since the
early 1980s.
In the fall of 1996, Mr. Kenney enrolled as a Ph.D. student in the Department of
Political Science at the University of Florida, where he continued his research on the
Colombian drug trade and U.S.-Colombian relations. With support from the National
Science Foundation and the Division of Sponsored Research, he conducted twelve
months of dissertation research in Bogot, Colombia, Washington, D.C., and Miami,
Florida. Mr. Kenney spent the 2000-2001 academic year as a Hamburg pre-doctoral
fellow at the Center for International Security and Cooperation at Stanford University.
After he returned to the University of Florida, where he completed his dissertation in
May 2002.
332


276
Butterfield, Fox. As Drug Use Drops in Big Cities, Small Towns Confront Upsurge.
New York Times (11 February 2002).
Http://www.nytimes.com/2002/02/l 1/national/l lCRIM.html [Accessed 11
February 2002].
Buzan, Barry. People, States and Fear: An Agenda for International Security Studies in
the Post-Cold War Era. Boulder: Lynne Rienner, 1991.
Cable News Network (CNN). Colombia says drug fight cost $1 bln last year. (13
February 1998a).
Http://www.cnn.com/WORLD/americas/98Q2/l 3/RB002117.reut.html [Accessed
13 February 1998].
Cable News Network (CNN). Colombia captura al "ltimo capo de capos" del
narcotrfico. (19 February 1998b).
Http://cnnenespanol.com/latin/COL/l 998/02/19/narcotrafico.reut/index.html
[Accessed 20 February 1998].
Cable News Network (CNN). U.S. lifts economic sanctions against Colombia: Improved
efforts to fight drugs spur move. (26 February 1998c).
Http://www.cnn.com/WORLD/98Q2/26/drugs.colombia/index.html [Accessed 27
February 1998].
Cable News Network (CNN). Cocaine trafficking's third generation: Young, low-key
and thriving. (6 June 1998d).
Http://cnn.com/WORLD/americas/9806/06/AP0005Q2.ap.html [Accessed 9 June
1998],
Cable News Network (CNN). U.S. pledges help for Colombian peace bid. (29 October
1998e). Http://cnn.com/WORLD/americas/9810/28/RB002264.reut.html
[Accessed October 29].
Cable News Network (CNN). Colombia air force "infiltrated" by drug mobs. (15
November 1998). Http://cnn.com/WORLD/americas/9811/15/BC-DRUGS-
COLOMBIA.reut/index.html [Accessed 16 November 1998],
Cable News Network (CNN). Feds Bust International Drug Trafficking Ring. (15 June
1999). Http://www.cnn.com/US/9908/17/fbi.drugs.02/ [Accessed 29 May 2001].
Cable News Network (CNN). More than 200 arrested as Mexican heroin ring is busted,
U.S. officials say. (15 June 2000a).
Http://www.cnn.com/2000/US/06/15/heroin.bust/index.html [Accessed 12 May
2001],
Cable News Network (CNN). Submarine found in Colombian Andes: Police link partly
built vessel to drug smugglers. (7 September 2000b).


188
2000; Nadelmann, 1993, p. 142; Reuter et ai, 1988, p. 40; Richelson, 1999, p. 2;
Rindskopf-Parker, 2000; Sims, 1995, p. 4).
The production of accurate, timely intelligence regarding clandestine trafficking
enterprises is an ongoing challenge for U.S. and Colombian law enforcement bureaus and
intelligence agencies. Trafficking enterprises do not publish annual reports, leaving this
to the government bureaucracies charged with disrupting them. Moreover, they
constantly change their practices and procedures, limiting the time frame for reliable drug
enforcement intelligence. Notwithstanding these difficulties, the U.S. and Colombian
governments have improved their counter-drug intelligence capabilities in recent decades.
These improvements have been instrumental in drug enforcers ability to identify and
dismantle a number of leading trafficking enterprises, including the Cali core
organizations. Better intelligence also helps police agencies keep abreast of the latest
developments in the illicit drug industry (Arreguin 2000; Francis 2000; Serrano Cadena,
1999, pp. 147-149).
Counter-drug intelligence comes in three varieties: tactical, operational, and
strategic. Tactical drug enforcement intelligence refers to information of immediate use
in ongoing criminal investigations targeting specific enterprises. Examples of tactical
intelligence include the names and aliases of participants, telephone and fax numbers,
makes and models of transportation vehicles, and stash house locations. Operational
intelligence is more general information about participants, organizations, technologies,
and smuggling routes that can be used in support of ongoing investigations. Strategic
intelligence refers to information about broad trends and patterns in the international drug
trade, such as coca leaf production estimates, country-specific drug threat assessments,


167
programs to draw on when transporting psychoactive drugs, money, supplies or other
materials. The more transportation programs within an organizations repertoire, the
greater its ability to adapt to changing circumstances. An enterprise with ten established
routes for shipping cocaine to the U.S. has a greater variety of practices, technologies,
and geographies from which to choose than an organization with only one route. Other
factors considered equal, the more diversified enterprise will be better positioned to
respond to the constant flow of problematic situations, including seizures, arrests and
theft, that characterize the hostile environment of the psychoactive drug industry.
Adaptations in Distribution Cells
Distribution cells dispense psychoactive drugs to independent wholesalers, collect
revenues from these transactions, and repatriate the profits to owners and investors in
Colombia and elsewhere. In handling large amounts of illicit drugs within consumer
countries they confront considerable environmental hostility, much of it emanating from
law enforcement agencies that seek to dismantle their operations. Colombian distribution
cells have developed a variety of technological and structural adaptations to reduce their
exposure to drug enforcers.
Technological adaptations refer to changes in the material and symbolic
technologies that participants use to communicate information, transport products and
profits, and store psychoactive drugs pending wholesale distribution. Participants
routinely adapt their communications technology by changing telephone and facsimile
numbers, using multiple pay phones located in different areas, switching pagers among
different participants, purchasing new cellular phones, pagers, and prepaid telephone
calling cards and discarding old ones, and using encryption technology to block drug


Compartmentalization 114
Routines in Trafficking Enterprises 115
Routine Failure 118
Participants in Colombian Trafficking Enterprises 120
Participants as a Source of Risk in Trafficking Enterprises 122
Turnover in Trafficking Enterprises 124
Promotion in Trafficking Enterprises 127
Technologies in Colombian Trafficking Enterprises 130
Practical vs. Scientific Knowledge 133
Conclusion 136
5 ORGANIZATIONAL LEARNING BY COLOMBIAN TRAFFICKING
ENTERPRISES 138
Re-Statement of Organizational Learning Proposition 139
A First Cut: Quantifying Interview Results 139
Acquiring Knowledge and Experience 143
Counter-Surveillance Activities 145
Intelligence Capabilities of Rodrguez Orejuela Organization 147
Informal Exchange of Information 148
Drug Smuggling U: Acquiring Trafficking Knowledge in Prison 150
Recording and Storing Information through Organizational Memories 152
Interpreting Knowledge and Experience 156
Discovery Process 157
Applying Knowledge and Experience 159
Adaptations in Drug Processing Routines 160
Adaptations in Transportation Routines 162
Adaptations in Distribution Cells 167
Strategic Learning in Colombian Trafficking Organizations 170
Diversifying into New Products and Markets 171
Changing Organizational Structures: Smaller is Better 173
Obstacles to Learning by Colombian Trafficking Organizations 175
Restricted Information Flows 175
Organizational Memories and Participant Turnover 177
Additional Barriers to Learning 180
Conclusion 182
6 ORGANIZATIONAL LEARNING BY DRUG ENFORCEMENT AGENCIES. ..184
A First Cut II: Quantifying Interview Results 185
Drug Enforcement Intelligence Activities 187
DEAs Role in Drug Enforcement Intelligence 192
CNP Drug Enforcement Intelligence 194
Gathering Intelligence through Criminal Investigations 195
Criminal Informants 196
Surveillance and Undercover Operations 197
vii


329
Weir, Margaret and Theda Skocpol. State Structures and the Possibilities for
Keynesian Responses to the Great Depression in Sweden, Britain and the United
States. Bringing the State Back In. Eds. Peter B. Evans, Dietrich Rueschemeyer,
and Theda Skocpol. Cambridge: Cambridge University Press, (1985): 107-168.
Westrate, David L. The Role of Law Enforcement. Drugs and Foreign Policy: A
Critical Review. Ed. Raphael F. Perl. Boulder: Westview Press, (1994): 79-99.
Whynes, David K. The Colombian Cocaine Trade and the War on Drugs. The
Colombian Economy: Issues of Trade and Development. Eds. Alvin Cohen and
Frank R. Gunter. Boulder: Westview Press, (1992): 329-352.
Wille, Richard P. Prepared Statement of Richard P. Willie. Cocaine and Marijuana
Trafficking in Southeastern United States. Hearings before the Select Committee
on Narcotics Abuse and Control, U.S. House of Representatives. Ninety-fifth
Congress, second session (9-10 June 1978). Washington: U.S. Government
Printing Office.
Williams, Phil. Transnational Criminal Organisations and International Security.
Survival 36.1 (1994): 96-113.
Williams, Phil. The International Drug Trade: An Industry Analysis. Global
Dimensions of High Intensity Crime and Low Intensity Conflict. Ed. Graham H.
Turbiville, Jr. Chicago: Office of International Criminal Justice, (1995a): 153-
183.
Williams, Phil. Transnational Criminal Organizations: Strategic Alliances. The
Washington Quarterly 18. 1 (1995b).
Http://www.altematives.com/crime/CRIMLINK.HTML [Accessed 15 March
1999],
Williams, Phil. The Nature of Drug-Trafficking Networks. Current History 97.618
(1998a): 154-159.
Williams, Phil. Organizing Transnational Crime: Networks, Markets and Hierarchies.
Transnational Organized Crime 4.3-4 (1998b): 57-87.
Williams, Terry. The Cocaine Kids: The Inside Story of a Teenage Drug Ring. Reading:
Perseus Books, 1989.
Wilson, James Q. The Investigators: Managing FBI and Narcotics Agents. New York:
Basic Books, 1978.
Wilson, James Q. Bureaucracy: What Government Agencies Do and Why They Do It.
New York: Basic Books, 1989.


212
Operation Banshee had taken over a year to put together, involved large numbers
of agents, and was costly in terms of man-hours, dollars expended, and
investigative time. To many DEA managers, Operation Banshee was just the type
of case that we should not be getting involved in. Consequently, cases of this
type were discouraged (Nieves, 1997, p. 5).
At the time, most DEA investigations relied on the time-honored buy-bust
technique in which undercover agents or criminal informants purchased minute quantities
of illicit drugs from retailers who were then immediately arrested. Unlike the conspiracy
investigations represented by Operation Banshee, buy-bust cases were resource-light and
easy to execute. All that was required was identification of a low-level street dealer by
drug enforcement agents, and a small amount of flash money to conduct the
transaction. DEA promotion procedures enhanced the techniques attractiveness by
rewarding field agents for putting powder on the table and people in jail. DEA
managers could then aggregate arrest and seizure numbers in performance review reports
that produced impressive-sounding outputs for policy makers.
However, as the federal governments super drug enforcement agency, the DEAs
primary mission was to identify and apprehend major traffickers and dismantle their
criminal operations. The buy-bust technique was ill suited for such efforts. Throughout
the 1970s, the DEA and other enforcement organizations came under increasing criticism
from the U.S. Congress and their own agencies for targeting too many low-level drug
violators through buy-bust tactics. Critics pushed the DEA to abandon its transactional
approach to drug enforcement and devote greater resources to conducting complex
conspiracy investigations against national and transnational trafficking enterprises and


162
The combined effect of these innovations is to increase the difficulty of state
efforts to disrupt drug processing operations in Colombia. Moreover, a number of
officials attribute the rise in Colombian cocaine production in recent years to adaptations
in processing technologies and practices (BINM, 1991, p. 100; Caldern, 2000; Cambio,
2000; Cook, 1997; DEA, 1993b, p. 8; El Tiempo, 2001a, 2001b; Hatch, 1997;
McNamara, 1999; Moreno, 1997; Smith, 2000a; Westrate, 1994, p. 96).
Adaptations in Transportation Routines
Changes in drug transportation routes and methods are some of the most common,
and consequential, tactical adaptations by Colombian trafficking enterprises. Trafficking
routes contain practices, procedures, people, places, and technologies for transporting
supplies, resources, and other materials from one location to another. These performance
programs are a critical component of the international drug industry. Without secure and
effective routes, rapid movements of capital, precursor chemicals, equipment, refined
narcotics, illicit profits, and participants are infeasible. Recognizing the significance of
trafficking routes, government interdiction strategies often prioritize their disruption.
Over the last two decades, trafficking enterprises have consistently responded to
drug enforcement pressure by refining existing transportation routes and developing new
ones. Indeed, changing routes have become one of the most predictable outcomes in the
competitive struggles between nares and narcos. Once drug enforcers identify and
dismantle a particular transportation route, large trafficking enterprises require only ten to
fourteen days to re-organize their operations (DEA, 1993a, p. 14). During this time,
participants gather and interpret information relating to the problematic situation, and
adjust the geographic location of their routes to areas where the heat of drug


OUTSMARTING THE STATE:
A COMPARATIVE CASE STUDY OF THE LEARNING CAPACITY
OF COLOMBIAN DRUG TRAFFICKING ORGANIZATIONS AND
GOVERNMENT DRUG ENFORCEMENT AGENCIES
MICHAEL C. KENNEY
A DISSERTATION PRESENTED TO THE GRADUATE SCHOOL OF THE
UNIVERSITY OF FLORIDA IN PARTIAL FULFILLMENT OF THE
REQUIREMENT FOR THE DEGREE OF DOCTOR OF PHILOSOPHY
UNIVERSITY OF FLORIDA
2002

ACKNOWLEDGMENTS
As with the production, transportation, and distribution of psychoactive
substances, writing a dissertation is a collective endeavor. This study was completed
with the support of many individuals and institutions. Fellowships from the National
Science Foundation, the Tinker Foundation, and the Division of Sponsored Research at
the University of Florida supported field work in Bogot and various locations in the
United States. A residential fellowship at the Center for International Security and
Cooperation at Stanford University supported data analysis and writing.
While conducting research in Bogot, I benefitted enormously from the Institute
of Political Studies and International Relations (IEPRI) at the National University of
Colombia. The director of the Institute, William Ramirez, provided a place to hang my
hat between interviews, and numerous IEPRI scholars offered constructive criticism and
friendship. For their assistance during this period, I thank Alvaro Camacho, Nathanial
Christie, Fernando Cubides, Francisco Gutirrez, Andrs Lpez, Gregorio Prez, Alvaro
Valencia, and Carlos Velsquez. During research trips to Washington, D.C. and Miami,
Fiona Wright and Juan Carlos Valencia provided greater hospitality than I deserved.
Dozens of U.S. and Colombian officials gave generously of their time during
interviews. Without their willingness to share their knowledge, this study could not have
been written. I owe a similar debt of gratitude to the former traffickers that spoke with
me. For reasons of discretion, many of my respondents must remain anonymous. They
11

know who they are and I thank them. I also thank Gabriela Rovillon Acosta, Maxine
Downs, and Emilia Gioreva for their professional transcriptions of dozens of interviews.
The Center for International Security and Cooperation (CISAC) and Stanford
University provided a stimulating and supportive environment to write most of this
dissertation. For their support and friendship during my stay in Palo Alto, I thank Herb
Abrams, Chris Chyba, Nisha Fazal, Karen Guttieri, Carole Hyde, Jacques Hymans, Wu
Jun, James March, Dinshaw Mistry, Barry ONeill, Barbara Platt, Woody Powell, Scott
Sagan, Steve Stedman, and Jeremi Suri. Special thanks go to Lynn Eden, who served as
my mentor at Stanford. Lynn scrutinized all of the chapters I wrote at CISAC, always
tendering her enthusiastic support insightful criticism. Robert Axelrod, Martha Feldman,
and Richard Valcourt were kind enough to share their perceptive comments on separate
chapters.
At the University of Florida I have beneffited enormously from my association
with professors and fellow students over the past several years. My dissertation
committee chair, Philip Williams, and co-chair, Terry McCoy, guided me through
graduate school and this study. I am enormously grateful for their friendship and support.
The other members of my committee, Leann Brown, Joseph Spillane, and Larry Dodd,
provided assistance throughout the project. For additional support at the University of
Florida, I thank Leslie Anderson, Ralph DiMuccio, Errol Henderson, Lalitha Henderson,
Peter Hildebrand, Goran Hyden, Michael Martinez, Geraldine Nichols, Richard Nolan,
Hazel Phillips, Sandra Russo, Steve Sanderson, Marty Swilley, Les Thiele, and Debbie
Wallen. For their friendship during my years in Gainesville, I thank Shawn Bird, Parakh
Hoon, Ajent Shriar, Juan Carlos Valencia, Eric Cooper, Fabiano Toni, Scott Richards,
iii

Vilma Fuentes, Josh Gordon, Ed Greaves, Brian Gridley, Tom Nisley, Liz Oldmixon, Jim
Conley, Lee Walker, and Fiona Wright.
I extend my deepest gratitude to my family for their love and support during the
years that I have been working on this study. My parents, Peter, Chris, and Margaret
Kenney, provided unwavering affection in life and graduate school. My biggest debt of
gratitude goes to my wife, Emilia, who sacrificed as much as I, if not more, to see this
project through to its completion.
IV

TABLE OF CONTENTS
page
ACKNOWLEDGMENTS ii
LIST OF TABLES ix
LIST OF FIGURES x
KEY TO ABBREVIATIONS xi
ABSTRACT xiii
CHAPTER
1 INTRODUCTION 1
Persistence of Illicit Drug Production in Colombia 2
Explaining the Colombian Drug Dilemma 6
Toward an Alternative Explanation: Organizational Learning by Trafficking
Enterprises 8
Significance of Research 10
Research Design 12
Overview of Study 15
2 HISTORICAL OVERVIEW OF THE ILLICIT DRUG INDUSTRY IN
COLOMBIA 17
Colombian Smuggling Tradition 18
Phase One1930s: Colombia as Transit Point 20
Government Drug Enforcement Efforts During Phase One 21
Phase TwoLate 1930s to early 1960s: From Transit Point to Producer 22
Herran Olazaga Brothers Enterprise 24
Other Phase Two Trafficking Groups 25
Government Drug Enforcement Efforts During Phase Two 26
Phase ThreeMid-1960s to late 1970s: Expansion of Colombian Drug Industry.... 27
Herrera Enterprise 29
Bravo Enterprise 31
Government Drug Enforcement Efforts During Phase Three 32
Phase Four1980s to mid 1990s: Rise and Fall of the Cartels 36
v

Emergence of the Core Organizations 38
Smuggling Methods of Core Organizations 40
Diversifying to Heroin 43
Government Drug Enforcement Efforts During Phase Four 44
Phase FiveMid-1990s to 2000: Atomization of Colombian Narcotics Industry.... 49
Government Drug Enforcement Efforts During Phase Five 52
Conclusion 54
3 TOWARD A THEORETICAL UNDERSTANDING OF ORGANIZATIONS
AND ORGANIZATIONAL LEARNING 58
Organizations 60
Organizational Tasks 60
Environments 61
Hostile environments 63
Organizational Structure 63
Routines 64
Participants in Organizations 67
Technology 68
Defining Organizational Learning 69
Acquiring Information 70
Interpreting Information 71
Applying Information 73
Research Propositions 75
Alternative Explanations for Organizational Change 75
Learning Ecologies 78
Productive Learning 80
Levels of Learning: Tactical vs. Strategic Adaptation 81
Learning under Ambiguity 85
Properties of Organizational Learning 90
Environmental Hostility vs. Benevolence 90
Organization Size and Decision-Making Hierarchies 91
Conclusion 93
4 ORGANIZATIONAL ANALYSIS OF THE ILLICIT DRUG INDUSTRY IN
COLOMBIA 95
Tasks of Colombian Trafficking Enterprises 96
Understanding Tasks in Colombian Trafficking Enterprises 100
Environments of Colombian Trafficking Enterprises 101
Trafficking Enterprises and Law Enforcement Agencies 101
Other Organizations in the Task Environment 105
Structure of Colombian Trafficking Enterprises 106
Centralized Decision-making Hierarchies 107
Flat Decision-making Hierarchies 109
Role Specialization 112
vi

Compartmentalization 114
Routines in Trafficking Enterprises 115
Routine Failure 118
Participants in Colombian Trafficking Enterprises 120
Participants as a Source of Risk in Trafficking Enterprises 122
Turnover in Trafficking Enterprises 124
Promotion in Trafficking Enterprises 127
Technologies in Colombian Trafficking Enterprises 130
Practical vs. Scientific Knowledge 133
Conclusion 136
5 ORGANIZATIONAL LEARNING BY COLOMBIAN TRAFFICKING
ENTERPRISES 138
Re-Statement of Organizational Learning Proposition 139
A First Cut: Quantifying Interview Results 139
Acquiring Knowledge and Experience 143
Counter-Surveillance Activities 145
Intelligence Capabilities of Rodrguez Orejuela Organization 147
Informal Exchange of Information 148
Drug Smuggling U: Acquiring Trafficking Knowledge in Prison 150
Recording and Storing Information through Organizational Memories 152
Interpreting Knowledge and Experience 156
Discovery Process 157
Applying Knowledge and Experience 159
Adaptations in Drug Processing Routines 160
Adaptations in Transportation Routines 162
Adaptations in Distribution Cells 167
Strategic Learning in Colombian Trafficking Organizations 170
Diversifying into New Products and Markets 171
Changing Organizational Structures: Smaller is Better 173
Obstacles to Learning by Colombian Trafficking Organizations 175
Restricted Information Flows 175
Organizational Memories and Participant Turnover 177
Additional Barriers to Learning 180
Conclusion 182
6 ORGANIZATIONAL LEARNING BY DRUG ENFORCEMENT AGENCIES. ..184
A First Cut II: Quantifying Interview Results 185
Drug Enforcement Intelligence Activities 187
DEAs Role in Drug Enforcement Intelligence 192
CNP Drug Enforcement Intelligence 194
Gathering Intelligence through Criminal Investigations 195
Criminal Informants 196
Surveillance and Undercover Operations 197
vii

Manufacturing Drug Enforcement Intelligence 198
From Intelligence to Operations 200
Changes in U.S. Drug Enforcement 201
Structural Adaptation Case #1: Toward an Administrative Merger of the
DEA and FBI 202
Other DEA Structural Adaptations 205
Reforming the Colombian National Police 206
Structural Adaptation Case #2: Cultural Transformation of the CNP 207
Adaptations in Drug Enforcement Operations 211
Banshee or Buy-Bust? 211
Innovations in Conspiracy Investigations 213
Innovations in Undercover Operations 215
Competitive Learning Games 217
Factors that Affect Outcomes in Narco-Narc Learning Games 222
Playing Catch Up 223
The Survival Imperative 224
The Enforcement Dilemma 226
Smaller in Size, Flatter in Structure 228
The Red Tape Trap 232
Conclusion 236
7 CONCLUSION 238
What Have We Learned? 239
Alternative Explanations of Routine Change 244
Implications for Counter-Drug Policy: Or How I Learned to Stop Worrying
and Love the Drug Trade 248
Diffusion of Trafficking Technologies 251
What is to be Done? 252
Legalization and Harm Reduction 253
A Modest Proposal 255
Conclusion 260
APPENDIX
A RESPONDENT DATA 262
B QUESTIONNAIRE 266
BIBLIOGRAPHY 269
BIOGRAPHICAL SKETCH 332
vni

LIST OF TABLES
Table Page
2-1 Colombian Drug Enforcement Indicators, 1974-1979 33
2-2 Colombian Drug Enforcement Indicators, 1980-1995 49
2-3 Colombian Drug Enforcement Indicators, 1996-1999 54
2-4 Evolution of Illicit Drug Industry in Colombia 56
4-1 National Government Agencies Involved in Drug Enforcement 104
4-2 Role Specialization in Colombian Trafficking Enterprise 113
4-3 Risk Reduction Routines in Colombian Trafficking Enterprises 117
4-4 Number of Drug-Related Arrests by Colombian National Police, 1974-
1998 (even years) 125
5-1 Professional Affiliation and Nationality of Respondents 140
5-2 Do Colombian Trafficking Organizations Learn? Results from Interviews 142
5-3 Transportation Innovations in Drug Smuggling Operations 166
6-1 Does the State Learn? Results from Interviews 186
6-2 U.S. Drug Enforcement Intelligence Agencies, Types of Intelligence
and Scope of Activities 191
6-3 DEA Domestic Arrests by Class Level, Fiscal Years 1979 and 1982 213
6-4 Number of DEA Electronic Surveillance Orders Conducted and Facilities
Covered, FY 1990-1998 215
6-5 Authorized DEA Personnel, Fiscal Year 2000 229
IX

LIST OF FIGURES
Figure P*gg
3-1 Simple Process Model of Organizational Learning 70
3-2 Alternative Sources of Change in Organizational Routines 77
3-3 Complex Process Model of Organizational Learning 85
4-1 Wheel Structure of Illicit Drug Network 99
4-2 Management Levels within Large Trafficking Enterprise Ill
4-3 Stash House Inventory Routine 119
6-1 Organization of DEA Intelligence Division, circa 1996 193
6-2 PHVA Management Cycle 208
6-3 Example of DEA Management Levels 231
7-1 Sources of Change in Organizational Routines, Amended 245
7-2 Phased Evolution of Illicit Drug Industry 248
x

KEY TO ABBREVIATIONS
AUC
Autodefensas Unidas de Colombia (United Self-Defense Forces of
Colombia)
BINLEA
Bureau of International Narcotics and Law Enforcement Affairs
BINM
Bureau of International Narcotics Matters
CIA
Central Intelligence Agency
CNN
Cable News Network
CNNespaol
Cable News Network Espaol
CNP
Polica Nacional de Colombia (Colombian National Police)
DAS
Departamento de Seguridad Administrativo (Department of
Administrative Security)
DEA
Drug Enforcement Administration
DOJ
Department of Justice
ELN
Ejrcito de Liberacin Nacional (Army of National Liberation)
EPIC
El Paso Intelligence Center
FAA
Federal Aviation Administration
FARC
Fuerzas Armadas Revolucionarias de Colombia (Revolutionary Armed
Forces of Colombia)
Fiscala
Fiscala General de la Nacin (National Prosecutor General)
FBI
Federal Bureau of Investigation
FY
Fiscal Year
GAO
General Accounting Office
XI

G-DEP
Geographic-Drug Enforcement Program
IEPRI
Institute of Political Studies and International Relations
INTERPOL
International Criminal Police Organization
IRS
Internal Revenue Service
NADDIS
Narcotics and Dangerous Drugs Information System
NAS
Narcotics Assistance Section
NDIC
National Drug Intelligence Center
NSC
National Security Council
ONDCP
Office of National Drug Control Policy
PCOC
President's Commission on Organized Crime
Pseud
Pseudonym
RCN
Radio Cadena Nacional de Colombia (National Radio Network of
Colombia)
WOLA
Washington Office on Latin America
Xll

Abstract of Dissertation Presented to the Graduate School
of the University of Florida in Partial Fulfillment of the
Requirements for the Degree of Doctor of Philosophy
OUTSMARTING THE STATE:
A COMPARATIVE CASE STUDY OF THE LEARNING CAPACITY
OF COLOMBIAN DRUG TRAFFICKING ORGANIZATIONS AND
GOVERNMENT DRUG ENFORCEMENT AGENCIES
By
Michael C. Kenney
May 2002
Chairpersons: Philip J. Williams, Terry L. McCoy
Major Department: Political Science
Over the past several decades, Colombian drug trafficking enterprises have
transported large quantities of cocaine, heroin, and other psychoactive drugs to the United
States in spite of government efforts to disrupt this illicit commerce. Contrary to much
existing scholarship, this study argues that the persistence of the Colombian drug
dilemma is, in part, due to the ability of criminal enterprises to alter their organizational
structures and behavior in response to information and experience, store this knowledge
in practices, procedures and performance programs, and select and retain innovations that
produce satisfactory outcomes. Drawing on a multidisciplinary body of literature on
organizational learning, and interviews with U.S. and Colombian officials, researchers,
and former drug traffickers, the study demonstrates that smuggling organizations learn,
and in the process become increasingly difficult for drug enforcement agencies to identify
xiii
and dismantle.

Drug trafficking enterprises are not the only organizational learners of interest in
this study. Law enforcement agencies also alter their practices and procedures in
response to knowledge and experience, thereby improving task performance and
facilitating their survival as bureaucratic organizations. To understand interactions
between these interdependent, rationally bounded actors, the study introduces the concept
of competitive learning games, and discusses numerous organizational and environmental
factors that tilt the playing field in favor of criminal enterprises. The study also includes
an historical analysis of the Colombian drug trade dating back to the 1930s. Outsmarting
the State is of broad theoretical and empirical interest to scholars and policymakers. The
study contributes to diverse bodies of literature in political science, sociology, economics,
organization theory, and criminology. The research has significant implications for
existing U.S. drug and crime control policies in Colombia, Mexico, and elsewhere.
xiv

CHAPTER 1
INTRODUCTION
Over the past two decades, the governments of Colombia and the United States
identified drug trafficking as a national security threat to their respective countries; a
threat they sought to eliminate by attacking individuals and groups engaged in production,
transportation, and distribution of controlled substances. Despite devoting considerable
resources to drug enforcement policies since the early 1980s, the two governments failed
to reduce illicit drug flows between their countries. Indeed, according to the best
available estimates, in Colombia the production of cocaine and heroin has increased
substantially in recent years.
What accounts for this intractable dilemma? Why have the Colombian and U.S.
governments been unable to impede the production of psychoactive drugs? Why are
these governments unable to resolve a national security threat that emanates not from
sovereign neighbors but from presumably weaker non-state actors? This study answers
these questions through a comparative case study of Colombian trafficking enterprises
and U.S. and Colombian drug enforcement agencies. Drawing on organization theory and
extensive primary and secondary-source data, the study develops a learning-based
explanation for the persistence of drug production in Colombia.
In this chapter, my aim is to set the stage for what follows. I begin with a
description of illicit drug dilemma in Colombia and conventional explanations offered for
this conundrum. I continue with a brief discussion of the alternative explanation that
1

2
forms the heart of this study: the organizational learning capacity of Colombian
trafficking enterprises. After emphasizing the contribution of the research to existing
bodies of literature in political science, organization theory, and criminology, I describe
the research design in detail. The chapter concludes with an overview of the remainder of
the study.
Persistence of Illicit Drug Production in Colombia
Over the past two decades, the Colombian government has sought to eliminate the
production and transit of illicit narcotics in its national territory. Working closely with
the U.S. and other members of the inter-American narcotics control regime, the
Colombian government has implemented supply-reduction programs that eradicate
drug plantings, destroy drug processing laboratories, intercept the transportation of
narcotics and the chemicals used to make them, and apprehend suspected drug traffickers
and confiscate their illicit profits.
The costs of these programs, in terms of budget allocations and human personnel,
are significant. Since the early 1980s, the Colombian government has spent several
billion US dollars to implement supply-reduction initiatives within its national territory.
While the Colombian government has received considerable anti-narcotics assistance
from the U.S. and other foreign governments over the years, it has also invested a
substantial portion of its own resources in the war on drugs. Moreover, in recent years,
the Colombian governments anti-drug expenditures have increased significantly. In the
1980s, Colombias anti-narcotics budget varied between USS 20 and 25 million per year,
with the U.S. providing half this amount. In 1995 the Colombian government devoted
US$ 900 million of its own funds to anti-drug efforts, and in 1996 this amount increased

3
to over US$ 1.3 billion. In 1997, the Colombian government allocated US$ 1.1 billion
for counter-narcotics efforts, which represented 4.8% of the governments budget for that
year (CNN, 1998a; Lee, 1989, p. 202; Republic of Colombia, 1997).
The human costs of the Colombian governments counter-narcotics efforts are
even greater. Every year thousands of Colombian civilian and military officials
participate in various phases of planning and/or implementing supply-reduction policies.
The danger inherent in this line of work in Colombia is illustrated by the deaths of
hundreds of government officials over the past 15 years. During one gruesome stretch
from 1985 to 1988, Colombian traffickers, particularly those associated with the so-called
Medellin cartel, killed over 400 police and military officers, dozens of judges and
journalists, and several high-ranking government officials. While the number of
government officials killed by drug traffickers has declined in recent years, death remains
a persistent reality for those covering the anti-drug beat in Colombia. Since 1994, 31
counter-narcotics Colombian agents have been killed and 68 have been wounded in the
line of duty (CNN 1998a; Lee, 1989, p. 202).
Notwithstanding the considerable financial and human costs of Colombian anti
drug efforts, in recent years the government has enjoyed limited successes in
implementing supply-reduction policies. One example of success has to do with recent
increases in the eradication of illicit drug plantings. Over an 8-year period, beginning in
1986 and ending in 1993, the Colombian government eradicated an estimated 5,714
hectares of coca plantings. In 1996 alone, drug enforcement units eradicated 5,600
hectares of coca plantings and 6,028 hectares of opium poppy plantings. Four years later,
they eradicated an estimated 47,000 hectares of coca plantings and 9,2542 hectares of

4
opium plantings.1 Moreover, since 1991 Colombian anti-drug units, working closely with
their U.S. counterparts, have captured the leaders of the Medellin, Cali and the North
Valley drug trafficking organizations. These arrests are part of a government strategy
aimed at dismantling the communications, transportation, and financial infrastructures of
the larger trafficking organizations by targeting their senior managers. The goal of the so-
called kingpin strategy is to disrupt the production of narcotics in Colombia by
capturing, incarcerating, and confiscating the wealth of those individuals who direct the
illicit drug industry (CNN, 1998b; Nieves, 1997).
Unfortunately, the kingpin strategy and increased eradication efforts have failed
to resolve the Colombian narcotics dilemma. Indeed, according to estimates contained in
the most recent International Narcotics Control Strategy Report, the estimated amount of
potential coca production in Colombia increased from 50,900 hectares in 1995 to 169,800
hectares in 2001.2 Corresponding with the increase in coca production has been increases
in estimated cocaine production from 230 metric tons in 1995 to more than 580 metric
tons in 2001. The estimated cultivation of opium poppies in Colombia, used to make
heroin, has also increased in recent years. In 1995 an estimated 10,300 hectares of
Colombian farmland were devoted to opium production. By 1997, the last year for which
1 These figures are contained in the International Narcotics Control Strategy Report, which is published
annually by the Bureau of International Narcotics and Law Enforcement Affairs (BINLEA and BINM,
various years, and ONDCP 2002). A caveat is warranted in presenting these and other drug-related data in
this study. Generating estimates for illicit drug production and chug performance outputs remains, at best,
an imprecise science. The clandestine nature of narcotics trafficking and the politicized nature of chug
enforcement make it difficult to produce accurate approximations for these activities. Throughout this
study, figures regarding drug production and drug enforcement indicators are meant to illustrate trends
rather than offer precise estimates.
2 Although this report contains what many specialists regard as the most accurate production estimates of
illicit chugs available, several researchers who have conducted field studies argue that the report
underestimates the amount of cocaine and heroin produced in Colombia (Clawson & Lee, 1996; Uribe
Ramirez, 1997). A hectare is equivalent to 2.47 acres.

5
figures are available, the estimate increased to 13,572 hectares. The estimated production
of opium gum, which is used to make heroin, increased slightly from 65 metric tons in
1995 to 66 tons in 1997. Over the past several years, the number of hectares in Colombia
devoted to marijuana cultivation and the estimated production of marijuana itself have
stabilized at approximately 5,000 hectares and 4,150 metric tons respectively (BINLEA,
1998, 2001).
While the trafficking of illicit narcotics is more difficult to estimate than its
production, there are indications that the supply of the Colombian-produced cocaine to
the U.S. has stabilized in recent years. For example, the U.S. Drug Enforcement
Administration (DEA) estimates that the price and availability of cocaine in the U.S. have
remained relatively constant. For the past several years, wholesale (per kilogram) prices
for cocaine have remained within a $10,500-36,000 price band, while cocaine purity at
the retail level (per gram) has stabilized at approximately 60% (DEA, various years).
Meanwhile, the flow of Colombian-produced heroin to the U.S. appears to be rising. The
DEA reports that the nationwide average purity for heroin at the retail level has increased
from 27% in 1991 to 36% in 1996. Spearheading this trend is South American heroin
most of which originates in Colombiawhich has a retail purity of 50.3% (DEA 1997).
In addition, the amount of Colombian-produced heroin seized by U.S. and Colombian
anti-drug authorities has also increased in recent years (BINLEA, 1998).
These developments have led the State Department to observe that in spite of
improved anti-drug efforts on the part of the Colombian government, the country remains
the worlds leading producer and distributor of cocaine and an important supplier of
heroin to the U.S. (BINLEA, 1998).

6
Explaining the Colombian Drug Dilemma
What accounts for the persistence of illicit drug production in Colombia at a time
when the Colombian government has devoted more financial and human resources to its
anti-drug efforts than ever before? Why has the Colombian government been unable to
reduce the production and transit of illicit narcotics within its national territory? Over the
past decade, scholars and policy makers have produced a cottage industry of literature
that attempts to answer these questions. Several prevalent explanations have emerged
from these studies. For example, many argue that the Colombian government is unable to
resolve the narcotics dilemma because it lacks sufficient anti-drug resources. In an article
that was published in the influential journal Foreign Affairs, Bagley argued that the
Colombian government lacks the necessary resources to effectively counter the drug
trade. In order for real progress to be made he insisted that the U.S. should provide
Colombia and the other Andean governments with the $2-3 billion annually that would
be required to establish effective control over drug producing areas (Bagley, 1988, pp.
87, 89). In a more recent study of the cocaine industry, Clawson and Lee argued that, in
spite of increased international assistance over the past ten years, Colombia and the other
Andean governments still lack the necessary resources to stop drug production and
trafficking in their countries (Clawson & Lee, 1996, p. viii).
Similarly, U.S. and Colombian officials charged with implementing counter
narcotics programs in Colombia maintain that they do not receive adequate resources to
effectively carry out their mandates. The U.S. and Colombian officials interviewed in
this research emphasize that insufficient resources were a significant impediment to
achieving greater counter-narcotics success in Colombia. The Director of the Narcotics

7
Assistance Section (NAS) of the U.S. Embassy in Colombia complained of recent budget
cuts to his program, wondering aloud, How am I going to wipe it out? (Moreno, 1997).
Several officials from the Anti-Narcotics Directorate of the Colombian National Police
(CNP) underscored the difficulties in covering an area larger than the state of Texas with
only 2,500 agents (Buitrago, 1997).
A second prevalent explanation of the Colombian narcotics dilemma is that the
Colombian governments will to implement effective policy has been seriously
compromised by drug-related corruption. This charge has been leveled at the leaders of
the political system and the government agencies responsible for carrying out anti-drug
programs. In another article, Bagley argued that even if the Colombian government were
to receive sufficient aid, it would still face significant obstacles to implementing more
effective policies, principal among them being the significant level of drug-related
corruption among government agencies and institutions (Bagley, 1989-1990).
Nadelmann claims that during the 1970s and 1980s, drug traffickers succeeded in
corrupting all levels of government, including a significant share of the legislature and
cabinet-level ministers.... Even the Colombian military, which has remained a powerful
and relatively corruption-free organization, could not resist the inducements offered by
the drug traffickers (Nadelmann, 1993, p. 279).
In the 1990s, drug-related corruption continued to flourish. Ernesto Sampers
presidency (1994-1998) was severely weakened by credible allegations that his
presidential campaign had received a US$ 6 million donation from several cocaine
entrepreneurs based in Cali. In 1998, U.S. authorities seized a Colombian military plane
3 Andreas et al. (1991-1992) make a similar argument.

8
in Fort Lauderdale carrying 1,639 pounds of cocaine and 13 pounds of heroin, tarnishing
the drug-fighting image of Colombian president, Andrs Pastrana. The incident, which
came a week after Pastranas state visit to the U.S., prompted Colombias Defense
Minister to concede that the countrys air force has been seriously infiltrated by
trafficking organizations. U.S. and Colombian officials that interviewed in this research
also stress the role of drug-related corruption in preventing more effective counter
narcotics efforts on the part of the Colombian government (CNN, 1998d).
Toward an Alternative Explanation:
Organizational Learning by Trafficking Enterprises
Although these explanations help account for the persistence of drug production
and trafficking in Colombia, they do not provide a fully satisfactory explanation of the
drug dilemma. One reason for this is that they do not give adequate attention to the non
state actors that manage the production and distribution of cocaine and heroin within the
Western Hemisphere. Since the 1970s, Colombian trafficking enterprises have provided a
steady flow of narcotics to the U.S. and Europe. In the process, these criminal enterprises
have proven to be innovative and highly adaptive. These skills enabled them to maintain
the profitability of their enterprises and survive persistent government efforts to destroy
them.
The literature on the illicit drug industry in Colombia includes government
reports,4 journalistic accounts,5 and academic studies.6 Several recent studies have
4 See the Drug Enforcement Administration (1994, 1995, 1996, 1997), FBI (1993), White House (1998),
and Zabludoff (1997).
5 See Can (1994), Castillo (1987, 1991, 1996), Castro (1996), Duzn (1994), Eddy et al. (1988), Garcia
(1991), Giraldo (1992), Gugliotta and Leen (1989), Mermelstein (1990), Rincn (1990, 1994), and
Shannon (1988).
6 Betancourt and Garcia (1994), Clawson and Lee (1996), Dombois (1998), Geopolitical Drug Watch
(1998), Griffith (1997, 1998), Krauthausen and Sarmiento (1991), and Tokatlin (1995).

9
highlighted the adaptability of the organizations that coordinate this industry. Griffith
(1997) noted that [traffickers are not only creative, they are also adaptive, changing
methods and operatives depending on the success of counter-narcotics measures (p. 84).
Clawson and Lee argued that trafficking organizations have adapted to government
efforts to reduce their profit margins by increasing production efficiency, improving and
diversifying smuggling methods, increasing their downstream penetration of U.S. drug
markets, and diversifying products and markets (Clawson & Lee 1996, pp. 10-12).
Krauthausen and Sarmiento argued that pressure from drug enforcers compels traffickers
to invent new methods and strategies for carrying out their illicit activities make the point
that in the face of continuous law enforcement pressure (Krauthausen & Sarmiento, 1991,
p. 37).
Government agencies and officials also recognize the adaptability of transnational
trafficking groups and the threat they pose to U.S. interests. The 1998 National Security
Councils International Crime Control Strategy stated that, [international criminal
organizations are adaptive and resilient, responding quickly to effective law enforcement
pressure (NSC, 1998). A recent General Accounting Office report links traffickers
adaptability to the failure of government drug enforcement efforts:
A key reason for U.S. countemarcotics programs lack of success is that
international drug trafficking organizations have become sophisticated,
multibillion-dollar industries that quickly adapt to new U.S. drug control efforts.
As success is achieved in one area, the drug-trafficking organizations change
tactics, thwarting U.S. efforts (GAO, 1997, p. 3).
Although these studies suggest that the trafficking enterprises adapt their behavior
in response to drug enforcement activities, they do not provide a conceptual framework
for understanding this process. However, there is a multidisciplinary body of literature on

10
organizational learning that provides some clues as to how and why drug trafficking
organizations may learn. This literature examines the behavior of firms in competitive
markets and attempts to explain how they
build, supplement and organize knowledge and routines around their activities and
within their cultures, and adapt and develop organizational efficiency by
improving the use of the broad skills of their work forces (Dodgson, 1993, p.
377).
Significance of Research
This study is of broad theoretical and empirical interest to scholars and policy
makers. The research contributes to diverse bodies of literature in political science,
sociology, economics, organization theory, and criminology. Within political science, the
research transcends traditional sub-disciplinary boundaries between international
relations, comparative politics, and public policy. It focuses on actors (criminal
organizations and law enforcement agencies) and issue areas (transnational organized
crime and drug trafficking) often neglected in studies of international security, foreign
policy, globalization, and state making.7 While political scientists have long been
interested in organizational learning, the law enforcement agency and criminal enterprise
have rarely been topics of interest. Instead, they have directed their attention towards
more accessible subject matter, including government institutions, intergovernmental
policymaking bodies, and military organizations. These organizations are recognized by
7 Of course there are exceptions. Examples include Andreas (2000), Arquilla & Ronfeldt (1993, 2001),
Desch et al. (1998), Farer (1999), Friman & Andreas (1999), Griffith (1997), Nadelmann (1993), and
Williams (1994, 1998).
8 Examples of this literature include Allison (1971), Breslauer and Tetlock (1991), Brown (2000), Deutsch
(1966), Dodd (1994), Eden (Forthcoming), Etheredge (1985), Evangelista (1988), Feldman (1989), Golant
(1998), Goldgeier (1994), Haas (1990), Haas (1992), Hall (1993), Heclo (1974), Jervis (1976), Khong
(1992), Le Prestre (1995), Leng (1992), Leeuw et al. (1994), Levy (1994), McCoy (2000), Mendelson
(1993), Moltz (1993), Nagl (1999), Neustadt & May (1986), Nye (1987), Posen (1984), Reiter (1996),
Sagan (1993), Sikkink (1997), Snyder (1991), Stein (1994), and Weir and Skocpol (1985).

11
States and possess formal legal standing. Organizations that operate outside the rule of
lawsuch as criminal enterprises and terrorist networkshave received scant attention
from students of organizational learning. My goal is to develop a general theoretical
explanation for how the learning capacities of these sovereignty free actors facilitate their
defiance of state prerogatives, undermining governments ability to provide stability,
order, and security for its citizens.
This research is also relevant to U.S. security interests, broadly defined.9 The
research has significant implications for existing U.S. drug and crime control policies in
Colombia and other countries where transnational criminal organizations have become
institutionalized, including Mexico, Russia, Italy, and Nigeria. The ability of criminal
enterprises to alter their organizational structures and operations in response to law
enforcement suggests that headhunting approaches to drug control, as exemplified in the
Kingpin strategy, are unlikely to achieve satisfactory outcomes. This is not to suggest
that such programs fail to impact trafficking organizations. Indeed, the Kingpin program
demonstrated that when U.S. law enforcement and intelligence agencies coordinated
effectively with their Andean counterparts cartel leaders could be apprehended and their
9 Some observers have legitimately questioned the trend by international relations scholars of labeling social
problems such as the drug trade and crime security issues. As Michael Desch notes, Broadening the
definition of security in recent years was the result of the desire of many scholars and practitioners to direct
more attention and resources to serious problems that previously had not received adequate consideration or
money. As a rhetorical device for energizing scholarly and governmental discussion of, and directing
attention to, the economic, environmental, social, and political problems affecting the post-Cold War
Caribbean, this broader definition of security seems reasonable. But whether this rhetorical strategy will
actually lead to a concerted effort to deal with these problems is unclear (Desch 1998, p. 146). This point
is well taken. However, two recent developments in Colombia and the U.S., have heightened the
importance of this study to U.S. security concerns. The first development refers to the changing nature of
the long-standing social conflict in Colombia, in which armed insurgents and paramilitaries have increased
their participation in the drag trade in order to finance the escalation of their violent confrontation with each
other and the Colombian state. This development produced alarm among Washington policymakers and led
to substantial U.S. assistance for the Pastrana administrations controversial supply-reduction strategy, Plan

12
smuggling operations disruptedat least temporarily. However, no sooner were these
underworld luminaries removed than replacements emerged, eager to pick up where their
predecessors left off. These post-cartel enterprises have learned from the mistakes of the
groups that went before them, downsizing their operations to reduce their vulnerability to
government anti-drug efforts. To the chagrin of state officials, existing groups have
proven more difficult to investigate and dismantle, hampering law enforcers ability to
repeat the successes of the Kingpin strategy.
Research Design
This is a comparative case study of the criminal enterprises that coordinate the
illicit drug trade and the state drug enforcement agencies charged with disrupting these
illegal activities. The basic objective of this research is to determine whether, and to
what extent, these sovereignty-free and sovereignty-bound actors learn as organizations.
Organizational learning refers to the process by which participants acquire, interpret, and
apply knowledge to practices, rules, and procedures that guide organizational behavior.
Organizational learning theory offers a framework for understanding the behavior of
trafficking enterprises and law enforcement and intelligence agencies. This process-
oriented approach illustrates how these state and non-state collectivities gather, analyze,
and apply information to change their structures and operations in order to achieve
satisfactory outcomes. Organizational learning also helps explain why trafficking
enterprises are better learners than state agencies, and how the illicit drug industry
persists in Colombia in spite of government successes in dismantling a number of leading
cartels.
Colombia. The second development refers to the tragic events in New York City of September 11th and

13
This case study is what Eckstein refers to as a plausibility probe. Plausibility
probes are preliminary tests of candidate-theories, or in this case candidate-hypotheses,
which are undertaken before more rigorousand costlycross-national tests are
attempted (Eckstein, 1992, pp. 147-149). This probe is undertaken to determine whether
the organizational learning proposition is worth further consideration in future studies of
the international drug trade and transnational organized crime. If the proposition does not
receive empirical support in the case of Colombian trafficking organizations, then it is
unlikely to have explanatory value in other cases. This is because many Colombian
enterprises are among the most sophisticated, highly organized criminal groups in the
world. Moreover, the learning ecology within which they operate is intensely
competitive, largely due to anti-drug pressure exerted by the U.S. and Colombian
governments. If Colombian trafficking groups are not learning in this environment and if
this process does not help explain their ability to undermine government efforts to stop
them, then it is unlikely that the explanation would be valid in other cases where criminal
organizations are less sophisticated.
To verify learning by trafficking enterprises and drug enforcement agencies, I
must demonstrate that changes in practices, procedures, and performance programs are
due to information processing, rather alternative considerations, such as power and
environmental selection. The process-oriented model of learning used in this study
necessitates an in-depth examination of how these collectivities function, rather than
superficial measurements of improved task performance. Learning cases should
their aftermath.

14
produce thick descriptions of how organizations acquire and manipulate information and
experience.
To evaluate the explanatory power of organizational learning I have systematically
gathered primary and secondary source data from a variety of sources in Colombia and
the U.S. The theory of organizational learning and the method of structured, focused
comparison guided data collection and analysis. I identify Colombian trafficking
enterprises and U.S. and Colombian drug enforcement agencies as primary units of
analysis that can be fruitfully studied through intra-case comparison.10 From the learning
literature I developed a semi-structured instrument for conducting interviews.11
Questions address the acquisition, interpretation, and application of knowledge and
experience by Colombian trafficking enterprises and government counter-narcotics
agencies. Other questions focus on their organizational structures, routines, and practices.
Using (non-probability) snowball-sampling techniques, I interviewed seventy-six U.S.
and Colombian government officials, researchers, and former drug traffickers.
I collected additional primary and secondary source data from government
documents located at the National Archives in College Park, Maryland, the U.S.
District Court in Miami, and several government offices and collections in Bogot,
Colombia. Secondary sources of information include government documents, transcripts
10 For discussion of the method of structured focused comparison, see George (1979), George & McKeown
(1985), and King, Keohane & Verba (1994), especially p. 45. For discussion of intra-case comparison, see
Lebow (2001).
111 chose the semi-structured interview format for their two principal strengths: structure and flexibility.
The first ensures that all relevant topics will be covered during the course of interviews; the second allows
researchers to purue leads into other topics as they arise. For discussion of these issues, see Bernard
(1995).
12 The sample contained fifty-one government officials, many of whom worked in law enforcement agencies
such as the Drug Enforcement Administration and the Colombian National Police, fourteen academic

15
and other court documents from the U.S. trials of prominent traffickers, memoirs of
former traffickers and anti-drug agents, journalistic and academic studies, and newspaper
accounts. These research methods have been applied in a number of studies on organized
crime13 and Colombian drug trafficking.14 While there are an abundance of data available
on the Colombian drug trade, these data are not always reliable. To minimize this
problem, I cross-check information against independent, alternative accounts wherever
possible and my own knowledge of the drug trade.15
Overview of Study
The remainder of this study proceeds as follows. Chapter 2 lays the empirical
foundation for additional analysis by tracing the evolution of the illicit drug industry in
Colombia over the past seven decades. This historical narrative centers on the
organizational forms that coordinate the Colombian drug trade, and the efforts by U.S.
and Colombian drug enforcers to stop them. Chapter 3 provides the theoretical
framework for the research. Following an extended treatment of this routine-based,
process-oriented approach to organizational learning, I discuss alternative explanations
for organizational change, including power and environmental selection. The chapter
also emphasizes psychological, organizational, and environmental impediments to
organizational learning, suggesting that learning cannot be presumed to exist a priori but
requires sustained application to the units of analysis.
researchers, six professional journalists, and five former participants that worked in different areas of the
Colombian drug trade.
13 See, for example, Abadinsky (1983), Adler (1993), and Reuter (1983).
14 Arango (1988), Castillo (1991, 1996), Gugliotta and Leen (1989), and Krauthausen and Sarmiento
(1991).
15 For a discussion of crosschecking procedures used in conducting this type of research, see Adler (1993).

16
This application begins with an organizational analysis of the illicit drug industry
in Colombia. Using an open-systems perspective of organization theory, Chapter 4
examines the tasks, environments, structures, participants, and technologies that together
form Colombian trafficking enterprises. Having delineated the structural and
environmental conditions of these criminal firms, Chapter 5 examines the extent to which
they gather, interpret, and apply information to practices, procedures, and performance
programs that guide collective behavior. In recognition that trafficking enterprises are not
the only organizational learners of interest, Chapter 6 applies the process-oriented
learning model to U.S. and Colombian drug enforcement agencies, in particular the Drug
Enforcement Administration and the Colombian National Police. The chapter describes
how these bureaucratic institutions develop, manufacture, and act on tactical and strategic
intelligence regarding trafficking enterprises. The chapter also develops the notion of
competitive learning games to describe interactions between these sovereignty-free and
sovereignty-bound actors. Chapter 7 concludes the study by summarizing the theoretical
and empirical findings of the research, and discussing the implications of competitive
learning games for U.S. and Colombian drug control policies and programs.

CHAPTER 2
HISTORICAL OVERVIEW OF THE
ILLICIT DRUG INDUSTRY IN COLOMBIA
Conventional wisdom maintains that the Colombian drug industry is a recent
phenomenon. A number of government reports, Congressional hearings, and journalistic
and academic studies date the beginning of the Colombias drug trade in the 1960s, or
even the early 1970s.1 Contrary to these accounts, Colombian smugglers have been
involved in the transnational narcotics traffic since well before the 1960s. Indeed,
criminal entrepreneurs have been smuggling cocaine and heroin from Colombia almost
since national governments and international conventions formally banned these
commodities. Since the 1930s, Colombian traffickers have built on their countrys
vibrant history in import and export contraband smuggling to organize for the purpose of
participating in this illicit commerce.
This chapter presents an historical overview of the illicit drug industry in
Colombia, focusing on the social organizations that produce, process, and export
marijuana, cocaine, and heroin in Colombia. The chapter highlights the dynamic and
fluid nature of these criminal enterprises, and lays the empirical foundation for the
1 In 1980, a Drug Enforcement Administration official testified before the U.S. Senate Subcommittee on
Investigations that prior to the mid-1960s drug trafficking in Colombia was largely confined to importing
cocaine for the domestic market and exporting small quantities of marijuana to neighboring countries
(Clifford, 1980, p. 474). Several years later, the Presidents Commission on Organized Crime presented an
influential report that traced Colombian involvement in the U.S. cocaine trade to the early 1960s (PCOC,
1986, p. 77). These views are accepted by a number of researchers that have written about the development
of the Colombian drug trade, including MacDonald (1988, p. 27), Carney (2000, p. 8) and Chepesiuk
(1999a, p. 56). In a recent edited volume on cocaine based on archival research, the contributor to the
Colombia case study inexplicably dates the formal appearance of the countrys cocaine industry to 1972
(Roldn, 1999, p. 176; also see Gootenberg, 1999).
17

18
chapters that follow. This chapter frames the evolution of the Colombian drug industry
in five distinct, but overlapping, phases of development. Over the past seventy years,
trafficking enterprises in Colombia have become increasingly sophisticated, increasing
the challenge for U.S. and Colombian drug enforcement efforts. The chapter begins with
a brief discussion of the countrys long-standing tradition in contraband smuggling, and
then proceeds through the different phases of the illicit drug industry. Within each
period, attention is given to Colombias position in the transnational drug trade and
government efforts to dismantle trafficking enterprises.
Colombian Smuggling Tradition
Transnational drug trafficking in Colombia has its roots in a long-tradition of
contraband smuggling tradition dating back to Colonialism, when Spanish authorities
sought to regulate trade within their Latin American dependencies. During the 17th and
18th centuries, contraband smuggling was common throughout Nueva Granada, the area
that encompasses contemporary Colombia. To avoid government duties and satisfy
consumer demand, enterprising smugglers transported food, licor, cigarettes, machinery,
and weapons across Riohacha, Santa Marta, and Cartagena. They also developed a
number of maritime smuggling routes through Caribbean sea lanes (Dye, 1998; Grahn,
1997; Junguito & Caballero, 1982; Lpez Restrepo & Camacho Guizado, 2001; Useche,
1997). In Nueva Granada, smugglers developed a number of practices to evade or co-opt
law enforcement authorities.
they [smugglers] utilized specific strategies and schemes that corresponded to
local conditions. Along unguarded coastlines, for example, smugglers sailed
close to shore and deposited their goods at prearranged sites where buyers were
waiting on the beach. In ports and near guarded anchorages, they used Spanish
intermediaries to arrange deals and bribe local officials while they waited
offshore... When local administrators threatened this illicit coastal trade,

19
t
smugglers transacted their business at sea, out of sight of roving ships and
lookouts (Grahn, 1997, pp. 28-29).
While many countries in Latin America and the Caribbean have distinguished
traditions in smuggling contraband imports, Colombia is one of the few countries in the
region to have a long history in contraband exports as well. For centuries, criminal
entrepreneurs based in Colombia have smuggled sugar, manufactured goods and
livestock to other countries in Latin America, including Brazil, Ecuador, Peru and
Venezuela. In recent decades, Colombian smugglers have established a flourishing trade
in contraband coffee with numerous countries in order to bypass quotas established by
the International Coffee Agreement.
However, the most significant drug-free contraband export has been illegal
emeralds. For years, workers in loosely-controlled government mines have stolen
emeralds, often under cover of nightfall, and sold them to emerald traders. These traders
work within close-knit, clan-based organizations that rely on secrecy, loyalty, and
coercion to run their illicit operations, attributes that would prove useful to those who
later expanded into other illicit industries. Emerald smugglers also learned the mechanics
of selling commodities on national and international black markets, laundering foreign
exchange, and the importance of hiring pistoleros to provide security to their operations.
Over time, a number of leading esmeralderos, such as Gonzalo Rodrguez Gacha, used
the knowledge and capital they acquired in the emerald trade to expand into the
production and exportation of marijuana and cocaine (Arango & Child, 1984, pp. 188-
189; Krauthausen, 1998, pp. 138-139; Lpez Restrepo 2000; Thoumi, 1995, p. 173;
Thoumi, 1995, p. 173; Thoumi, 1992, p. 51; Betancourt & Garcia, 1994, p. 106).

20
Emerald traders were not the only Colombian smugglers to expand into the illicit
drug industry. In the 1960s and 1970s, a number of black marketeers in cigarrettes,
whiskey, and domestic appliances established themselves in marijuana and cocaine
exports. Smugglers drew on contacts, customs, and capital they acquired in smuggling
these commodities to diversify into marijuana and subsequently cocaine. The subsequent
development of the marijuana and cocaine industry in Colombia was heavily influenced
by the informal practices and procedures of these contrabandistas (Arango & Child,
1987, p. 130; Betancourt & Garcia, 1994, p. 107; Lpez Restrepo, 2000; Torres, 2000;
Valds, 2000; Valencia, 2000).
The re-emergence of the Colombian heroin trade in the 1990s was stimulated by
well-established cocaine traffickers seeking to diversify their illicit activities (FBI 1993;
Farah 1997; Semana 1999; Zabludoff interview). In all three cases, criminal
entrepreneurs drew on previously acquired contacts, customs, and capital to establish
themselves in more lucrative endeavors. This knowledge and experience was critical for
smugglers operating within highly uncertain and even hazardous business environments.
# i
Phase One1930s: Colombia as Transit Point
In the early 1930s reports of captured Colombian drug smugglers began to appear
in U.S. government documents and Colombian press accounts. In 1932 a smuggler was
captured in the Panama Canal Zone carrying 25 grams of cocaine hydrochloride hidden
inside a cartridge belt. The individual claimed that he had acquired the drug from a
group of traffickers operating in Cartagena, Colombia. The following year, a different
smuggler was arrested in the Canal Zone carrying 100 grams of cocaine hydrochloride,
2 This section draws on the work of Eduardo Senz Rovner (1996, 1997), a Colombian economic historian
that conducted archival research at the National Archives in Washington, D.C.

21
also allegedly obtained in Cartagena. The same year, El Espectador, a leading
Colombian newspaper, reported that Colombian police officials confiscated a large
consignment of illicit drugs in the home of a prominent Bogotano. A confidential U.S.
State Department document from 1936 claimed that the Colombian islands of
Providencia and San Andrs were being used by smugglers for exporting illicit drugs and
alcohol to the U.S. (Senz Rovner, 1996, pp. 69-70).
Though sketchy in the details, these government documents and press reports
suggest that a number of smugglers based in Colombia were involved in illicit
transnational networks linking European drug producers with Caribbean and North
American consumers. They also indicate that Colombian ports located on the Caribbean
sea, including Barranquilla, Cartagena, and Santa Marta, were transshipment points for
opiates and cocaine produced in Europe. Given the paucity of data, I cannot say much
about the organizational forms involved in these criminal actitivities. Nor is it possible to
draw direct linkages between these early pioneers and smuggling groups in later decades.
However, it is clear that at least some trafficking was taking place in Colombia as early
as the 1930s, during which time the country served as a significant transit point for
international drug flows (Arango & Child, 1984, p. 176; Lpez Restrepo & Camacho
Guizado, 2001, pp. 3-4; Walker, 1989, pp. 75-76).
Government Drug Enforcement Efforts During Phase One
In the 1920s and 1930s the government of Colombia came to view the
consumption of coca leaves by its citizens as a social problem (Walker, 1989, p. 73).
Increased concern among national policy makers was reflected in a number of measures
during this period. According to Lpez Restrepo and Camacho Guizado, Law 11 of

22
September 15, 1920 was the first piece of national legislation outlawing the consumption,
trade, and production of certain narcotics. Eighteen years later the Directorate of
National Hygiene enacted Decree 95 to regulate the sale of coca leaves to licensed
pharmacies. In July 1938, the Colombian government established a new penal code that
increased the criminal penalties for transacting in opiates. One month later the Ministry
of Work, Hygiene, and Welfare was created with the partial purpose of applying national-
level regulations concerning the drug traffic that Colombia had accepted at international
narcotics control conferences sponsored by the League of Nations (Lpez Restrepo,
2000, p. 8; Lpez Restrepo & Camacho Guizado, 2001, p. 3; Senz Rovner, 1996, p. 72).
In the mid-1930s, drug interdiction emerged as an active component of the U.S.
governments counter-narcotics strategy. Prior to this U.S. drug enforcement largely
consisted of local officials intercepting drugs in port cities, sometimes with the help of
federal customs and counter-drug agents. The first, tentative steps towards the
internationalization of American drug enforcement were sporadic Coast Guard patrols in
Caribbean sea lanes and the Gulf of Mexico. As William Walker reports, additional
efforts were made to conduct aerial surveillance near the U.S.-Mexican border,
apparently without much success. In general, early drug enforcement activities
conducted by the Colombian and U.S. governments were haphazard, sporadic, and failed
to impede smugglers access to morphine and cocaine supplies (Walker, 1989, pp. 57,
140-141; Walker, 1992, p. 269).
Phase TwoLate 1930s to early 1960s: From Transit Point to Producer
Colombias role as a transit point in the international drug trade was not destined
to last for long. As early as 1939 the U.S. Treasury Department reported that opium was

23
being grown on large extensions of Colombian territory and that cocaine was being
processed in-country and exported to Panama. According to this confidential report,
Colombians of German extraction processed cocaine for export and transported the
product by train to Cartagena and then by automobile or pack animals to the Caribbean
ports of Tol, Cispat, or Acand, where it was hidden in banana boats and smuggled to
Panama. Once in Panama, the drugs were transferred from steamship to small launches
and transported to Puerto Piln, where they were guarded by a group of German
smugglers until it was feasible to transport the cargo by automobile to the Canal Zone
(Senz Rovner, 1996, pp. 77-78).3 During World War II, contrabandistas from the Urab
region of Colombia formed smuggling networks that transported illicit drugs, whiskey,
and cigarrettes through Panama and other Central American and Caribbean countries
(Lpez Restrepo, 2000; Walker, 1989).
Colombias shift from transit point to drug producer was facilitated by Cubas
emergence in the international drug trade following World War II. In post-war Havana
criminal organizations coordinated transnational networks composed of Latin American
cocaine suppliers and French-Corsican and Italian heroin brokers. The networks used
couriers, including seamen, stewards, passengers, and pilots, to smuggle drugs from
South America to Cuba, sometimes by way of Central America. These human mules
transported small quantities of coca paste in suitcases with specially modified
compartments, a practice that remains popular among cocaine and heroin couriers today.
In Cuba and Colombia processing groups refined the coca paste into cocaine
hydrochloride, and sent along the finished product to the United States (MacDonald,
3 However, U.S. and Colombian officials disputed this report (Senz Rovner, 1996, pp. 78-83).

24
1988, pp. 27-28; Senz Rovner, 1996, p. 89; Treasury Department, The Illicit Narcotic
Traffic in Cuba; Treasury Department, May 27, 1954).
Herran Olazaga Brothers Enterprise
In 1957 U.S. Federal Bureau of Investigation agents and Colombian police
officials discovered a cocaine and heroin processing laboratory in Medellin. The
processing lab was owned by a pair of Colombian twins that had been involved in the
narcotics trade at least since 1948, over time developing into significant suppliers for
Cuban trafficking groups. Rafael and Toms Herran Olazaga ran their clandestine
operation out of a furniture workshop that served as a front for their drug processing lab.
The Herran Olazaga brothers purchased coca leaves from the Colombian department of
Cauca and opium gum from Ecuador. They obtained the precursor chemicals used in the
refining process through a nearby commercial drug lab that also served as cover. After
processing, the brothers or their associates would smuggle the finished narcotics to
Havana, where they would sell them to independent Cuban traffickers that would
transport the drugs to the United States, Mexico, and other countries (Arango & Child,
1984, pp. 166-169; Montoya, 1959a, p. 3; Senz Rovner, 1996, p. 90).
The arrest of the Herran Olazaga brothers was important for law enforcement
officials because it confirmed long-standing rumors that Colombian traffickers were
smuggling narcotics into Cuba (Treasury Department). The case also indicates that
Colombian traffickers were increasing their role in transnational drug networks. No
longer were Colombian intermediaries merely purchasing cocaine and heroin from
French and Italian traffickers and passing it onto to Italian-American and Cuban

25
traffickers located in Havana. At least a few Colombian traffickers had expanded into
drug production.
Other Phase Two Trafficking Groups
The Herran Olazaga brothers were not the only drug smugglers operating in the
Antioquia region during this period. According to Arango and Child, by the mid-1950s a
number of Colombian contrabandistas decided to enter the drug trade. In making the
shift from black market cigarettes, alcohol, and domestic appliances to drugs, these illicit
entrepreneurs drew on existing resources, including their practical knowledge of
smuggling methods and Caribbean maritime routes, investment capital, Cuban-based
mafia contacts, and Colombian-based drug chemists (Arango & Child, 1984, pp. 165-
166).
In the early 1960s, the Venezuelan press began reporting about an alarming
invasion of cocaine, heroin, and marijuana, much of it originating in Colombia. A
number of smuggling rings transported drugs from Colombia into Venezuela by means of
go-fast motor boats. At least one group of Colombian smugglers was affiliated with the
well-known Italian mafiosi, Lucky Luciano. Unlike other rings, Lucianos group
preferred to move its contraband into Venezuela through ground transportation. The
cross border traffic was facilitated by numerous well-equipped processing laboratories in
Medellin run by German technicians (El Espectador, 1961a, p. 3).
While the Venezuelan press was reporting about the invasion of Colombian drugs,
El Espectador published several articles detailing the growing traffic and consumption of
marijuana in Colombia. Although marijuana had been produced in Colombia since the
times of the Spanish conquest, consumption of the drug was generally concentrated

26
among socially marginal groups located in port cities and the sugar growing regions of
Valle del Cauca. According to Thoumi, increases in domestic demand in the late 1950s
and 1960s stimulated greater production of la mala hierba. A series of arrests by
Colombian police officials in the summer of 1961 demonstrated the existence of an
organized, city-wide network of marijuana traffickers in Bogot {El Espectador, 1961b,
p. 9; El Espectador, 1961c, p. 9; El Espectador, 1961d, p. 9; El Espectador, 1961e, p. 9;
Thoumi, 1995, p. 126).
Government Drug Enforcement Efforts During Phase Two
Due to the lack of data regarding drug enforcement outputs during this period,
such as numbers of participants arrested and kilograms of drugs seized, it is difficult to
assess the effectiveness of government counter-narcotics efforts. By the 1960s, officials
from both countries expressed concern at Colombias growing role in supplying
transnational drug networks. However, drug enforcement programs in Colombia during
this period were not particularly successful. In the aftermath of la Violencia, Colombian
police and military authorities had more pressing concerns than stopping the drug trade,
such as reducing political violence in the countryside (Walker, 1999, p. 146; Arango &
Child, 1988, pp. 169-170). When drug-related arrests were made, the perpetrators were
often low-level street dealers and drug addicts, ineffective targets for dismantling
growing trafficking networks.
The fate of the Herran Olazaga brothers following their arrest in Cuba is
indicative of the failure of drug enforcement efforts during these years. Rafael Herran
Olazaga fled Cuba immediately after posting bond from his 1956 arrest. While brother
Toms was eventually found guilty of drug trafficking, he served only one year in a

27
Cuban penintenciary before returning to Medellin. In subsequent years, law enforcement
authorities suspected the two brothers of continuing their illicit drug activities from
Colombia. However, they apparently avoided further prosecution. According to Arango
and Child, the Antioquian trafficker died a successful businessman, owner of motels and
several food establishments (Arango & Child, 1984, p. 169; Treasury Department).
Phase ThreeMid-1960s to late 1970s: Expansion of Colombian Drug Industry
In the mid and late 1960s the involvement of Colombian smuggling groups in the
international drug trade continued to expand. After the fall of the Batista regime in Cuba
a number of Cuban and Italian-American organized crime figures based in Havana fled
the island for the United States. While some of these traffickers remained active in the
drug trade, Cubas value as a transit point was much diminshed. Colombian smugglers
with strong connections to Cuban trafficking networks found themselves well positioned
to take advantage of this development. By 1965, Colombian trafficking enterprises were
providing almost the entire cocaine supply for U.S.-based Cuban smugglers. Also during
this period, Colombian smugglers supplied cocaine and marijuana to American, Chilean,
and Mexican traffickers (Betancourt & Garcia, 1994, p. 57; Cervantes, 1980; Clifford,
1980, p. 474; Gugliotta & Leen, 1989, p. 22; Jonnes, 1996, p. 338; MacDonald, 1988, p.
28; Presidents Commission on Organized Crime [PCOC], 1986, pp. 77-78; Ruiz
Hernndez, 1979; Sabbag, 1990; Thoumi, 1995, p. 126; U.S. House, 1977, pp. 3,19;).
In the late 1960s and early 1970s, Colombian traffickers established their own
transportation routes and wholesale distribution networks in the U.S. and Europe.
Smugglers drew on their previous experiences and knowledge in trafficking illicit drugs
and other contraband goods. A number of whiskey, cigarrette and marijuana smugglers

28
from the Antioquia and Guajira regions used their skills, contacts, and capital to enter the
cocaine business in the 1970s. Enterprising marimberos and contrabandistas discovered
that the smuggling infrastructures and methods used to transport marijuana to the United
States worked well for cocaine. Marijuana distribution networks converted easily to
cocaine. A number of marimberos made the switch gradually by pigbacking small
quantities of cocaine on their marijuana loads and obligating their U.S. distributors to sell
both products. Transportation methods used by marijuana smugglers, such as flying
private planes of various sizes to clandestine airstrips along the Altantic Colombian coast,
loading them with drugs, and returning to the U.S. with the illicit cargo, also adapted
easily to cocaine trafficking. Indeed, a number of smuggling innovations pioneered by
marimberos later became popular among cocaine traffickers, such as air dropping
specially wrapped packages of marijuana in the Caribbean sea, and using large seafaring
freighters, known as mother ships, to transport huge quantities (upwards of 100 tons) of
marijuana to prearranged locations 200 miles off the U.S. coastline, where private yachts
and go-fast motor boats would converge on the mother ship, receive and transport smaller
qualities into the U.S. (Cervantes, 1980, pp. 47-48; Krauthausen & Sarmiento, 1991, p.
147; MacDonald, 1988, p. 28; Nieves, 1997, p. 3; Passic 2000; PCOC, 1986, p.
78Velsquez Romero, 2000; Valds 2000).
By the mid-1970s, Colombian trafficking enterprises were recognized as
preeminent players in the inter-American cocaine industry. In a short period of time, they
transformed themselves from mere intermediaries for trafficking groups based in other
countries to vertically integrated, production-transportation-wholesale distribution
networks. Along the way Colombian enterprises developed practices, procedures, and

29
technologies for producing and transporting illicit drugs that would be used over the next
two decades. Small, makeshift laboratories processed coca paste into cocaine
hydrochloride. Human couriers, private aviation aircraft, and maritime vessels
transported drugs across the Caribbean by sea or air. Navigation technology located drug
drop-off points in the high seas. Sofisticated communications equipment, including
Bearcat scanners, CB, VHF and single side band radios, marine radios, and telephone
paging systems, were used to communicate within smuggling groups and to monitor law
enforcement activities (Cervantes, 1980, pp. 80-81; Martin, 1978, p. 61; Battard, 1978, p.
75; U.S. House, 1977, p. 9; Wille, 1978, p. 88).
The development of Colombian trafficking enterprises did not go unnoticed by
state authorities. By the mid-1970s, U.S. drug enforcers were alarmed at the growth of
Colombian wholesale distribution rings operating in New York, Miami, and other large
urban markets. In New York City, several Colombian organizations were believed to
control the cocaine trade in Queens and Manhattan. These groups, led by such mythical
figures as Benjamin the Black Pope Herrera, Griselda the Black widow Blanco, and
Veronica the queen of cocaine Rivera, developed many of the smuggling practices and
procedures that the cocaine cartels would refine and make famous ten years later (Nieves,
1997, p. 5; Semana 1987). The following sections briefly describe the structure and
modus operandi of two of these enterprises.
Herrera Enterprise
The Herrera trafficking network was based in Cali, Colombia, and led by
Benjamin Herrera, the son of an alleged cocaine chemist who had been involved in the
drug trade for a number of years. This family enterprise featured numerous brothers,

30
sisters, cousins, and in-laws as participants. However, the Herreras also recruited outside
the family for professionally trained participants. In all, the network contained more than
ninety members and exported approximately forty kilograms of cocaine a month to New
York and Miami. The Herreras acquired coca paste in Bolivia, Ecuador, and Peru. Using
human couriers carrying ordinary suitcases, the enterprise transported it to international
airports in Cali and Bogot, where the couriers whisked past compliant customs officials
on the organizations payroll. The coca paste was then transported to one of several
processing laboratories owned by the organization, where it would be processed into
cocaine hydrochloride. One processing lab captured by Colombian authorities outside of
Cali contained a twenty-five ton mechanical press for packing cocaine into fine sheets
and other equipment, materials and precursor chemicals worth approximately
US$800,000. Among the eight persons arrested at this lab was a chemistry professor
from Santiago University in Cali (Gage, 1975a, p. 26; Gugliotta, 1987; Moreau, 1976;
Semana, 1987).
Fully refined cocaine was distributed to one of three Herreras operating in
Barranquilla, Bogot, or Medellin. These transporters would export the drug to the U.S.,
again relying on drug-running couriers holding small quantities, generally between two
and four kilos. Couriers sometimes posed as students, carrying books that contained
cocaine secreted sheets. These students were paid between US$ 500 and US$ 1,000
per trip, plus expenses and a new suit of clothes to be used when transporting their
pedagogical materials to New York. False documents, including student visas, were
obtained by a family member that also served as a liaison with several of the networks
U.S.-based buyers. In addition to customs officials, the enterprise also received

31
protection from officials in the Colombian police and judiciary, as well as leading
members of Colombian society that invested in cocaine shipments (Gage, 1975a, p. 26).
Bravo Enterprise
The Bravo network contained several large and loosely connected drug rings that
smuggled large quantities of cocaine, heroin, and marijuana from Colombia to the U.S.
From 1972 to 1974, the Bravo enterprise distributed drug shipments throughout North
America, including Manhattan, Queens, Miami, Los Angeles, San Antonio, Texas and
Toronto. Large cocaine shipments, which in those days meant anything greater than four
kilos, were transported in containerized cargo, speed boats, and private planes using a
variety of routes, including Panama, Mexico, Puerto Rico, Canada, and even Munich,
West Germany. Smaller quantities of cocaine and heroin were hidden inside luggage or
clothes and transported by couriers. The Bravo network developed a number of creative
smuggling methods, including soaking clothes in a liquid solution of cocaine
hydrochloride, swallowing drug-filled prophylactic condoms, and hiding cocaine in false-
bottomed wine bottles, painting frames, hollow ski poles and wooden hangars, specially
modified bras and girdles, and a dog cage, complete with live canine. On reaching the
U.S., the cocaine was distributed by members of the Bravo network, which included
approximately 150 couriers, brokers, and distributors working in the New York
metropolitan area alone. Although U.S. police officials eventually captured, prosecuted
and convicted over a dozen members of what federal prosecutors described as the
biggest Colombian narcotics organization ever uncovered, several entrepreneurs,
including Alberto Bravo, Carlos Bravo and Griselda Blanco, were able to continue their

32
operations from Colombia (Gage, 1975b, p. 24; Hudson, 1974, pp. 1 & 18; Lubasch,
1976, p. 21; New York Times, 1976, p. 20; Nieves, 1997, p. 5).
Government Drug Enforcement Efforts During Phase Three
In response to the growth of Colombian trafficking enterprises, the U.S. and
Colombian governments increased their drug enforcement activities, achieving a number
of successes in the latter half of the 1970s. Police agencies from both countries
conducted criminal investigations that disrupted several trafficking enterprises. Dozens
of participants from the Herrera and Bravo networks were captured, prosecuted, and
convicted of drug-related offenses. In 1975, in what amounted to the largest cocaine
seizure to date, Colombian police officials captured 600 kilograms of cocaine in a small
plane at the Cali airport (Frontline, 2000).
Notwithstanding these achievements, drug enforcement in Colombia was
inconsistent during this period, as reflected in the wide disparity of annual performance
indicators (see Table 2-1 below). For the entire year following the Cali airport bust,
Colombian officials managed to capture only 138 kilograms of cocaine. In 1977, the
amount of cocaine captured by Colombian authorities sunk to a dismal 32 kilograms. In
the U.S., complex conspiracy cases targeting cocaine operations, such as the Herrera and
Bravo networks, were few and far between. In the 1970s, many drug enforcement
managers were wary of these resource-heavy investigations, preferring low-cost buy-
bust operations targeting street dealers (CNP, 2000, p. 540; Nieves, 1997, p. 5).4
Drug enforcers mediocre counter-cocaine performance during the 1970s reflected
a widespread belief among U.S. and Colombian policy makers and police officials that
marijuana, rather than cocaine, was the primary drug threat facing their respective
4 Buy-bust operations are discussed in detail in Chapter 6.

33
countries. Consequently, drug enforcement resources were directed towards eradicating
Colombias booming marijuana industry. The year of the Cali cocaine bust, Colombian
authorities captured 78,000 kilograms of cannabis and destroyed almost 1.5 million
marijuana plants. In 1977, the year Colombian officials netted 32 kilograms of cocaine,
they also captured 187,077 kilograms of marijuana (see Table 2-1).
Table 2-1 Colombian Drug Enforcement Indicators, 1974-1979
Year
Persons
apprehended
Cocaine
interdicted
(kilograms)
Marijuana
interdicted
(kilograms)
Coca
plants
destroyed
Marijuana
plants
destroyed
Drug labs
destroyed
1974
1,305
164
90,000
0
37,500
6
1975
1,484
699
78,000
0
1,494,000
10
1976
769
138
27,000
0
25,000
15
1977
945
32
187,077
1,000
805,700
14
1978
555
194
158,272
4,195
431,614
34
1979
457
1,252
325,656
185,700
398,255
30
Source: CNP (2000)
Even regarding marijuana there was ambivalence among Colombian and U.S.
policy makers during much of the 1970s. In neither country did large groups of citizens
view marijuana consumption as a significant public health problem, and in Colombia at
least political elites did not see the growing drug industry as a threat to the countrys
national security interests. In the mid-1970s, when officials from the Ford and Carter
administrations pressured their Colombian counterparts about drug trafficking within
their national borders, Colombian President Alfonso Lpez Michelsen was quick to
remind them that U.S. consumers rather than Colombian suppliers represented the source
of the problem. Even within the U.S. policy community officials disagreed over the
degree to which the drug trade, in particular marijuana, threated U.S. interests. President
Nixons fight against crime agenda produced some changes in federal drug enforcement
policy (see Chapter 6), but the widespread moral fervor associated with the cocaine wars

34
remained several years away (Massing, 1998; Randall, 1992, p. 246; Tokatlian, 1988, p.
139; Tokatlian, 1990, pp. 293-294; Walker, 1999, pp. 148-149).
Notwithstanding this ambivalence, the 1970s witnessed an increase in the U.S.
governments drug enforcement presence in Colombia. Prior to the Nixon
Administration, U.S. drug enforcers, including the FBI and the Bureau of Narcotics and
Dangerous Drugs, operated sporadically, on a case-by-case basis, in South America.
With the creation of the Drug Enforcement Administration in 1973, the Nixon
Administration ushered in a new era of the internationalization of U.S. drug enforcement
policy. By the late 1970s, the new super drug enforcement bureau was active in dozens
of countries, and established a permanent presence in Colombia, working closely with the
Colombian National Police (CNP) (Nadelmann, 1993).
With the change of Colombian presidential administrations in August 1978, the
Colombian governments attitude towards drug trafficking shifted. President Julio
Turbay implemented a more aggressive counter-narcotics policy than his predecessors
following allegations in the U.S. media that two of his ministers and his own family were
connected to the illicit drug industry. Following his inauguration, President Turbay
placed the Guajira Peninsula, which had long served as a major production center for
marijuana and transit point for marijuana and cocaine, under the jurisdiction of
Colombia's armed forces. By the end of 1978, 10,000 army troops were in the region
engaged in the manual eradication of marijuana. Operation Fulminant, as the anti-drug
mission was called, produced immediate results: 3,500 tons of marijuana were seized,
which represented an estimated loss of US$ 70 million to Colombian smugglers,
approximately seventy drug running airplanes and seventy seafaring vessels were

captured, and 1,000 Colombians and Americans were arrested (Krause, 1979; Thoumi,
1995, p. 210; Walker, 1999, p. 150).
35
Following on the heels of the DEAs Operation Stopgap, which targeted
Colombian and American marijuana smugglers in the Caribbean, Fulminant temporarily
disrupted the marijuana industry in Colombia. However, the operation was short lived, in
part due to a lack of support from the Colombian military. The generals were concerned
that too many resources were being shifted away from more pressing anti-guerrilla
operations. They also worried about the debilitating impact of drug-related corruption on
field officers involved in Fulminant (Oijuela, 1990, p. 218).
However, the most significant impact of the marijuana eradication and
interdiction campaigns undertaken by U.S. and Colombian drug enforcers in the late
1970s was an unintended one. In successfully targeting a number of leading marimberos,
Operations Fulminant and Stopgap substantially increased the risks and costs associated
with marijuana trafficking. Given limited state resources, cocaine did not receive as
much attention from drug enforcers as marijuana, providing an additional incentive for
smugglers to diversify into cocaine.5 Many of those with sufficient capital, contacts, and
knowledge did, transferring their marijuana smuggling expertise to the more lucrative
cocaine trade. In this respect, short-sighted government drug enforcement policies and
programs that focused on marijuana trafficking helped set the stage for the dramatic
increase in cocaine trafficking over the next decade (Betancourt & Garcia, 1994; Cannon,
1979; Krauthausen & Sarmiento, 1991; Oijuela, 1990, p. 218; OToole, 1978).
5 Apart from increased marijuana drug enforcement, there were other incentives to switch to cocaine
trafficking. For one thing, cocaine was less aromatic than marijuana, making it easier to hide and transport.
Also, per unit profits for cocaine were substantially greater than for marijuana, reducing the need to
transport such large quantities to ensure adequate profits.

36
Phase Four1980s-mid 1990s: Rise and Fall of the Cartels
By the early 1980s a number of Colombian trafficking organizations were
organizing large-scale cocaine shipments from South America to the United States and
Europe. These criminal enterprises provided a measure of coordination to what had been
a highly fragmented industry. At the zenith of this transnational production-
transportation-marketing structure stood several vertically integrated core organizations
that, according to one estimate, supplied over sixty percent of the cocaine reaching U.S.
and European drug markets. Size varied considerably among different core
organizations. Small core enterprises were composed of anywhere from two dozen to a
hundred members, while large organizations contained several hundred members.
Participants were compartmentalized into discrete units, organized along functional lines,
such as drug processing, transportation, and wholesale distribution. Core organizations
generally contained three or more levels of management, with at least two layers
insulating leaders from the activities of rank-and-file workers (PCOC, 1986, pp. 100-101;
Thompson, 2000; Vargas, 2000; Zabludoff, 1997, p. 24).
The core organizations transformed the Colombian drug industry by coordinating
international cocaine shipments that measured in metric tons rather than kilograms. To
do so, they relied on networks of hundreds of individuals and groups that provided
specialized goods and services. Independent suppliers from Bolivia and Peru provided
coca paste, a semi-processed form of cocaine, which was often transported to Colombia
for further processing. In completing the cocaine refining process, some Colombian core
organizations owned their own processing laboratories, while others subcontracted
processing services to independent groups of chemists that operated hundreds of

37
kitchen labs outside of Medellin, Cali, or Bogot. On occasion different organizations
pooled resources to develop large-scale cocaine refining operations, such as the infamous
Tranquilandia complex that was discovered and destroyed by Colombian authorities in
1984. In certain regions of Colombia left-wing guerrilla fronts and paramilitary groups
provided protection for cocaine processing labs. Small trafficking groups and individuals
were recruited to help finance large-scale shipments and share the risk associated with
government interdiction efforts. American pilots were often hired to transport the
finished product to consumer markets in the U.S. and Europe. Independent money
launderers were contracted to repatriate illicit profits. In this manner, core organizations
outsourced numerous activities to independent groups linked in ad hoc support
networks. Participation in these networks was rather fluid: Depending on the needs and
circumstances involved in each shipment, core organizations could choose from among
different suppliers, processors, transporters, and financiers (Castillo, 1996; Diehl, 1982;
Morganthau, 1989, p. 23; Juan David Ochoa, 2000; Thoumi, 1999; Zabludoff, 1997, p.
25).
Core enterprises focused on enforcing agreements among network participants,
providing security for trafficking operations, gathering intelligence on government drug
enforcement efforts, and protecting leaders political and economic interests.6 Several
core organizations also maintained distribution cells in the U.S. responsible for
receiving drug shipments, storing merchandise in warehouses or stash houses,
distributing cocaine at the wholesale level, and shipping revenues back to Colombia.
However, the core organizations functioned more as voluntary export associations and
6 To perform these functions core enterprises relied on the use of intimidation and violence.

38
interest groups than monolithic firms (Clawson & Lee, 1996, pp. 39-40; Krauthausen &
Sarmiento, 1991; Lee, 1991, p. 14; Zabludoff, 1997, pp. 23-26; Zabludoff, 2000).
Emergence of the Core Organizations
Core trafficking organizations and their support networks did not spring up
overnight. Rather, they developed over time as their leaders acquired experience,
contacts, and knowledge in a variety of criminal endeavors, including smuggling
contraband goods, robbing cars, even kidnapping. The largest, and most well-known,
core organizations were led by entrepreneurs based in Medellin and Cali. During the
1980s and 1990s, such figures as Pablo Escobar Gaviria, Carlos Lehder Rivas, Gonzalo
Rodrguez Gacha, Jos Santacruz Londoo, and the Ochoa and Rodrguez Orejuela
brothers became household names as Colombian and American journalists assiduously
chronicled their criminal exploits.7
Many of the leaders of the Medellin and Cali core enterprises began their delictive
careers working as low-level members for established criminal gangs. In the early 1970s,
three of the future leaders of the Cali-based cocaine network, Gilberto and Miguel
Rodrguez Orejuela and Jos Santacruz Londoo, worked as foot soldiers in a Cali
criminal gang known as Los Chemas. This gang was primarily involved in kidnapping
and counterfeiting but gradually expanded into smuggling cocaine base from Bolivia and
Peru and converting it into cocaine hydrochloride in Colombia (Castillo, 1987, pp. 41-42;
DEA, 1994b, p. 1).
7 Due to the fascination of the Colombian and U.S. media and public with the Medellin and Cali core
organizations an enormous body of literature emerged in the 1980s and 1990s, much of it written by
journalists more interested in dramatization and hyperbole than scientific analysis. Some of the more well-
known journalistic accounts, in both English and Spanish, include: Can (1994), Castillo (1987, 1991,
1996), Castro-Caycedo (1996b), Cervantes (1980), Corts (1993), Duzn (1994), Eddy et al. (1988),
Freemantle (1985), Giraldo (1992), Gugliotta and Leen (1989), Mermelstein (1990), Mills (1986), Reyes
(1999), Rice (1989), Rincn (1990, 1994), Shannon (1988), Torres (1995), Torres & Sarmiento (1998).

39
Also in the early 1970s Pablo Escobar, who would emerge a decade later as one
of the leaders of the Medellin cocaine network, worked as an enforcer for a contraband
smuggler that specialized in whiskey, cigarrettes, watches, and second-hand pianos. In
an interview with a Colombian journalist, Escobar referred to this contrabandista as his
maestro, from whom he learned the smuggling business (Castro Caycedo, 1996a, p.
284). Around 1975 the ambitious Escobar became involved in cocaine trafficking,
organizing small-scale smuggling ventures. Through contacts and intermediaries he
arranged for the purchase of kilo quantities of cocaine hydrochloride from an Ecudorian
supplier, and transported the drug to the United States through human couriers. In 1976
he was arrested by Colombian authorities near Medellin for transporting thirty-nine
kilograms of cocaine hidden inside the spare tire of a truck. Escobar was never
successfully prosecuted for this charge, and later in the decade convinced Fabio Ochoa,
an experienced contraband smuggler, to use his well-established and well-connected
smuggling routes for the more profitable drug business (Chepesiuk, 1999a, p. 142). The
old-time contrabandista apparently agreed and a business partnership gradually emerged
between the senior Ochoa, his three sons, and Escobar. This joint venture endured for a
number of years and, along with the participation of other leading traffickers, formed the
basis of what later became known as the Medellin cartel (Can, 1994, pp. 59-60;
Castillo, 1987, p. 60; Kraar, 1988, p. 34).
However, this partnership never functioned as a cartel in the strict sense that
economists use the term. Rather Escobar and the Ochoas pooled their criminal resources,
divided responsibilities, and coordinated large-scale cocaine shipments to American and

40
European markets.8 As the size of their trafficking venture grew they gradually absorbed
or co-opted the smuggling operations of several of the pioneering trafficking groups
operating in the 1970s (Gugliotta, 1987; PCOC, 1986, pp. 101-103; Valds, 2000).
Similar arrangements developed between different core enterprises in the Cali-
based network, including the Rodrguez-Orejuela, Santacruz Londoo, and Herrera
Buitrago groups. Although the leaders of these separate groups ran their own smuggling
operations with their own personnel, they collaborated extensively on strategic matters
affecting the entire network, such as infiltrating the different security agencies of the
Colombian state or corrupting well-placed Colombian congressmen sympathetic to their
political and economic interests, such as outlawing the extradition of Colombian
nationals. Moreover, on occasion, low-level workers for one core organization, such as
cocaine chemists, truck drivers, or merchandise off-loaders, would be put to work for
another core group, without the workers knowledge that the ultimate locus of authority
had changed (Torres & Sarmiento, 1998; Velsquez Romero, 2000).
Smuggling Methods of Core Organizations
Core enterprises used a variety of methods for transporting cocaine from South
America to U.S. and European drug markets. A number of core organizations continued
the practice of using human couriers to smuggle cocaine. Although the quantities
smuggled by individual couriers remained small, generally less than four kilograms, core
enterprises had large numbers of participants at their disposal. In an effort to overwhelm
law enforcement authorities, now wise to this time-honored smuggling method, some
8 Medellin core organizations also absorbed the smuggling operations of several of the pioneering
trafficking groups operating in the 1970s. When Manuel Garcs was sentenced to jail in Colombia, the
Escobar organization took over his smuggling routes and operations (Valds interview). When Benjamin
Herrera was paroled from a U.S. federal prison in 1975, he returned to South America and collaborated
with the Medellin core organizations (Gugliotta, 1987).

41
commercial flights contained as many as many as seventeen drug-running mules.
Organizers overloaded the planes with mules with the expectation that drug enforcers
would identify and apprehend several, which was considered acceptable as long as the
majority cleared Customs. Couriers continued to use false-bottomed suitcases to smuggle
drugs, as well as dog cages, hollow coat hangars, and specially designed bras and girdles.
In addition to these external body carries, large numbers of Colombian couriers
ingested balloons or condoms containing small amounts of cocaine (DEA 1993, p. 18;
Raab, 1984; Ramos, 1985).
However, the bulk of cocaine transported by the core organizations was through
general aviation aircraft and maritime vessels. A number of core enterprises outsourced
air transporation services drug-running American pilots and support personnel. These
transportation rings used light twin engine aircraft, such as Gulfstream Aero
Commanders, Beechcraft King Air 300s, Piper Navajos, Aztecs and Cheyennes, and
Cessna 400s and Conquest IIs, to transport several hundred kilos of cocaine over a range
of approximately 1,800 miles. Many planes were equipped with rubber fuel bladders
for traveling long distances without the need to refuel, along with radar detectors and
radio scanners for tracking government planes and seafaring vessels on drug patrol. In
the 1990s, some Cali core organizations were using large, long-range, multi-engine
aircraft, such as Convair 580s and Boeing 727s, to smuggle several tons of cocaine into
Mexico and Canada. For smaller planes that required refueling, smuggling networks
relied on a number of transhipment points in Mexico, Central America, and the
Caribbean where drug-running pilots could land and, for a hefty commission, refuel.9
9 One of the leaders of the Medellin-based smuggling network, Carlos Lehder, maintained a transshipment
point on Normans Cay, a small island in the Bahamas where he owned several properties. In exchange for

42
While some transshipment points were merely refueling stations, others contained small
crews of workers that off-loaded and repackaged cocaine shipments. In Mexico a
number of trafficking groups with marijuana and heroin smuggling experience provided
this service, breaking down the cocaine into smaller shipments and transporting it
through Mexico in autmobiles and trucks (DEA, 1993, p. 17; DEA, 1994b, p. 7;
Mermelstein, 1990; Pallomari, 1997; PCOC, 1986, pp. 88-89, 98; Raab, 1984; Rice,
1989, pp. 145-146).
Maritime smuggling was conducted by seafaring vessels of various types and
transportation capacities, including high-speed cigarette or go-fast boats, sport
fishermans boats, yachts, medium-sized commercial fishing boats, large, dilapidated
freighters, and even what the DEA identified to as "submarine-like semi-submersible
vessels capable of tranporting hundred kilogram quantities of cocaine through the
Carribean (DEA, 1994b, p. 7). Maritime vessels often departed from commercial ports in
Colombia, such as Buenaventura, Barranquilla and Cartagena, as well as hundreds of
clandestine debarkation points located along the countrys northern coast. On smaller
boats, cocaine was hidden in customized storage compartments, with the vessels
frequently modified for this purpose. On large freight ships, the drug was hidden within
containerized cargo shipments, including blue jeans, ceramic floor tiles, fresh cut flowers,
fruits and vegetables, fish, coffee, lumber, concrete posts, blue jeans, and dozens of other
exports from South America (Maitland, 1981; Griffith, 1997, p. 81; Castillo, 1996, p. 51;
Mermelstein, 1990, p. 146; Nstor, 2000).
a commission, Lehder allowed independent cocaine and marijuana smugglers as well as those associated
with the Medellin network to land and refuel before continuing to the United States. For a description of
Lehders operation, see Gugliotta & Leen (1989) and Shannon (1988). Also, see the interview granted by
Jorge Luis Ochoa (2000) to Frontline.

43
In reflection of the smuggling capacity of the core enterprises, in the 1980s U.S.
and Colombian authorities began to discover containerized shipments and stash houses
containing several tons of cocaine hydrochloride. As early as March 1982 U.S. Customs
authorities discovered 3,096 pounds (or 1.38 tons) of cocaine mixed in mixed in a
shipment of hundreds of cartons of clothing apparel. The cocaine was wrapped with
yellow plastic in one-kilo packages bearing different coded markings. According to DEA
investigators the load belonged to fifteen different trafficking organizations, including the
Ochoa and Escobar groups. In 1989, DEA and Customs agents discovered approximately
12,000 pounds (5.35 tons) of cocaine hidden inside drums of powdered lye. The cocaine,
wrapped in red, yellow and blue plastic and marked with the code name Baby I, was
discovered at a warehouse in Queens, New York used by a trafficking ring associated
with the Cali-based trafficking network. Also in 1989, U.S. authorities confiscated
cocaine shipments of 11,200 pounds (5 tons) on a Panamanian freighter in the Gulf of
Mexico, 20,160 pounds (9 tons) hidden in a stash house in Texas, and 44,800 pounds (20
tons) in an unguarded warehouse near Los Angeles. The latter seizure remains the largest
ever recorded in the United States. Two years later, DEA and Customs officials
discovered 26,880 pounds (12 tons) of cocaine hidden inside concrete fence posts and
cornerstones. In 1994, federal authorities uncovered 8,000 pounds (3.57 tons) of cocaine
hidden inside empty barrels in a tractor-trailor truck in Texas (Dillow, 1994; Gugliotta &
Leen, 1989, pp. 71-73; Isikoff, 1991; McKinley, 1989).
Diversifying to Heroin
At some point during the 1980s several core enterprises expanded into the heroin
business. It is believed that this development was driven by the same factors that caused

44
numerous marimberos to diversify into cocaine in the preceding decade, including the
growing risks associated with cocaine trafficking due to successful drug enforcement
efforts, and the greater per unit profitability of heroin. In addition, traffickers may have
been attracted by profitable European drug markets, where heroin distribution networks
were well-established, and law enforcement efforts posed fewer risks to drug smugglers.
In 1994, the DEA reported that core organizations led by the Ochoa brothers and
Leonidas Vargas Vargas were using their cocaine networks to smuggle heroin. The same
report claimed that the Rodrguez Orejuela brothers and Helmer Herrera Buitrago had
also invested in the heroin industry. Two other core organizations associated with the
Cali network, the Santacruz Londoo and the Urdinola Grajales groups, began
experimenting with heroin production in the early 1980s, and supplied Colombian
farmers with the necessary inputs necessary to cultivate opium. According to a former
director of the Colombian National Police (CNP), Santacruz Londoo brought Afghani
and Pakistani opium cultivators to Colombia in order to learn how to cultivate opium
poppies. Since then Colombia has become a major source of heroin to the U.S.,
supplying a higher grade product that many U.S. consumers favor over Southeast Asian
heroin (DEA, 1994a, p. 6; Farah, 1997; Farah, 1999; FBI, 1993, p. 21; Moore & Farah,
1998; Revista Cambio 16, 1999; Semana, 1999; Serrano Cadena, 1999, pp. 50-51;
Statement of Witness, 1989, p. 11; Zabludoff, 1999).
Government Drug Enforcement Efforts During Phase Four
Throughout the 1980s and the 1990s the U.S. and Colombian governments
directed considerable resources towards dismantling the core cocaine organizations.
While bilateral cooperation between the two countries was occasionally marred by

45
substantive disagreement over a number of policy issues, particularly extradition, the two
governments worked together to implement a variety of supply-reduction programs
targeting the Medellin and Cali-based trafficking networks.
During this period the Colombian government engaged in a series of
crackdowns directed at the leaders of the Medellin-based trafficking enterprises and
subsequently the Cali organizations. Several crackdowns were undertaken in response to
specific criminal acts attributed to Medellin trafficking groups, such as the assassination
of Rodrigo Lara Bonilla, the Colombian Minister of Justice, in 1984.10 Each time
Colombian authorities responded by carrying out dozens of raids against the properties of
leading Medellin traffickers, destroying processing laboratories, seizing cocaine and
money, an apprehending low and mid-level participants in these criminal enterprises.
However, the leaders remained beyond the reach of drug enforcers, developing
sophisticated security arrangements allowing them to evade government efforts. After
several weeks, the crackdowns would subside, and traffickers could return to business as
usual (Gugliotta, 1992, pp. 122-123; Matthiesen, 2000).
In 1989 Colombian president Virgilio Barco launched a fourth crackdown in
response to the assassination the leading presidential candidate for the 1990 general
elections, by sicarios (paid assassins) working for the Medellin traffickers. By this point,
the Colombian president and other political elites were convinced that the Medellin
organizations presented a legitimate threat to the countrys political system. In a
televised address, President Barco declared a nationwide state of siege against the
traffickers. At the behest of President Bush, the U.S. offered their beleagured drug war
10 Bonillas assassination was carried out in retaliation for his role in planning a police raid on the
Tranquilandia cocaine refining complex owned by several leaders of the Medellin smuggling network
(Gugliotta & Leen, 1989, pp. 132-137; Lee, 1989, p. 171; Tokatlin, 1990: 317).

46
ally an emergency assistance package of US$ 90 million, and the Drug Enforcement
Administration undertook Operation Bolivar, specifically targeting the Medellin
organizations.
The fourth crackdown proved more effective than previous efforts because lasted
over a year, as opposed to six weeks, and possessed a more coherent strategy that
included extradition. The relentless pressure exerted by Colombian and U.S. authorities
forced the leaders of the Medellin organizations into hiding, reducing their ability to
coordinate large-scale cocaine shipments. Several leading traffickers were captured or
killed during police raids. Moreover, the crackdown produced ripple effects throughout
the Colombian drug industry. The inability of processing groups associated with the
Medellin network to buy the precursor chemicals needed for cocaine processing on a
regular basis reduced the price for Peruvian and Bolivian coca leaf to less than half of
production costs. The scarcity of cocaine supply affected consumer markets, causing
wholesale and retail cocaine prices in the United States to rise between fifty and one
hundred percent. However, this price spike proved short-lived, lasting approximately
eighteen months (Bagley, 1989-1990, p. 155; Caulkins & Reuter, 1998, p. 603; Farah
1990a, 1990b; Gugliotta 1992, p. 124; Krauss, 1995, p. A12; Nieves, 1997, p. 11;
Velsquez, 1993; Whynes, 1992, p. 333).
Nevertheless, by the early 1990s many of the Medellin core organizations had
been severely damaged by Colombian and U.S. anti-drug efforts. Prompted by effective
police pressure and a decree by a new Colombian president, several leading traffickers,
most notably the three Ochoa brothers, turned themselves in to the authorities. In June
1991, when it became clear that the new Colombian constitution would ban extradition of

47
Colombian nationals, Pablo Escobar also turned himself in to Colombian authorities.
However, Escobars surrender turned out to be largely meaningless as he continued his
criminal activities in a luxurious correctional facility of his own construction, from which
he escaped after a year of imprisonment. Over the next two and a half years, Escobar
managed to elude Colombian and U.S. authorities and former business associates that
turned against him. But the effort to stay alive required all of his resources, eliminating
his role as a major trafficker.11 Meanwhile, competing trafficking enterprises based in
Cali took advantage of drug enforcers single-minded focus on capturing Escobar by
expanding their own transnational operations (Can, 1994; Nieves, 1997, p. 12).
With the death of Escobar at the hands of an elite Colombian police unit in
December 1993, the Colombian government could now turn its attention to the Cali
enterprises. However, it would be another fourteen months before Colombian authorities
systematically cracked down against the leading Cali traffickers. The impetus for the
1995 offensive was not an act of violence by the Caleos, who relied more on bribery
than intimidation to protect their interests, but political pressure from Washington.
Disgusted with Colombian president Ernesto Sampers ability to stay in power despite a
campaign finance scandal in which he was believed to have knowlingly accepted
contributions from the Rodrguez Orejuela brothers and other traffickers, the U.S. Senate
threatened to impose economic sanctions on Colombia if the government failed to
achieve certain drug enforcement benchmarks in 1995. These included capturing the
leaders of the Cali core organizations, confiscating their illicit assets, and dismantling
their criminal operations (Can 1994; Economist Intelligence Unit, 1995, p. 4; Latin
11 For further discussion of these events see Luis Can El Patrn: Vida y muerte de Pablo Escobar
(Bogot, Colombia: Planeta, 1994).

48
American Weekly Report, 1995a; 1995d; Mattiesen, 2000, pp. 309-310; Nieves, 1997, p.
11; Semana, 1995, p. 28; Thoumi 1995, pp. 226-228).
In February 1995, President Samper launched a comprehensive assault on the
leaders of the Cali trafficking network. Over the next several months, the Colombian
National Police, led by General Rosso Jos Serrano, captured eighteen tons of coca
leaves and two tons of cocaine hydrochloride, destroyed 195 processing laboratories,
eradicated 6,000 hectares of coca plantings, and detained 616 suspected traffickers.
While impressive, the results did not dramatically exceed the drug enforcement indicators
in preceeding years (see Table 2-2 below).
In June the Samper government received a major boost when a special joint
military-police unit captured Gilberto Rodrguez Orejuela, the alleged maximum leader
of Cali traffickers. Over the next year, the Colombian National Police, with the
assistance of the DEA and the CIA, exerted enormous pressure on the remaining Cali
leaders, all of whom were eventually either captured or surrendered to government
authorities. In combination with the Peruvian governments controversial shoot-down
policy in the coca paste air bridge between Peru and Colombia, the Cali crackdown
produced ripples throughout the cocaine industry. By June 1995, the price of coca paste
in Colombia had dropped 50%. By mid-September of the same year, wholesale and retail
cocaine prices in New York City, the Cali networks most important U.S. market,
increased 50% and 30% respectively. Similar increases were noted in other U.S. cities
that received substantial cocaine supplies through the Cali transportation and distribution
network (Caulkins & Reuter, 1998, p. 603; Geopolitical Drug Dispatch, 1995; Krauss,

49
1995; Latin American Weekly Report, 1995b, 1995c, 1995e, 1995f, 1995g, 1995h, 1995i;
Serrano Cadena, 1999).
However, like earlier crackdowns against the Medellin enterprises, these price
increases disappeared completely within a year, reflecting the drug industrys ability to
absorb even the most intense drug enforcement efforts in the long term. Indeed,
following a brief period of regeneration, new and revitalized trafficking groups emerged
in Colombia to produce greater amounts of cocaine and heroin than before.
Table 2-2 Colombian Drug Enforcement Indicators, 1980-1995
Year
Persons
apprehended
Cocaine
interdicted
(kilos)
Cocaine
base
interdicted
(kilos)
Heroin
Interdicted
(kilos)
Marijuana
interdicted
(kilos)
Drug labs
destroyed
1980
358
748
0
0
192,422
29
1981
798
339
0
0
3,302,242
31
1982
1,139
651
0
0
3,283,878
165
1983
1,073
2,083
0
0
3,537,387
113
1984
5,251
19,582
9,448
0
4,301,263
137
1985
1,951
4,239
3,674
0
1,021,046
696
1986
3,699
3,039
4,070
2
846,000
572
1987
4,732
8,326
6,712
2
1,287,272
1,359
1988
4,929
12,047
2,554
0
842,994
655
1989
5,217
24,668
9,601
0
617,925
389
1990
6,253
16,000
3,429
850
659,047
268
1991
6,349
59,347
8,223
0
381,157
235
1992
6,770
28,016
5,289
36
206,934
95
1993
8,136
19,137
6,945
42
505,274
241
1994
7,221
28,145
22,580
94
161,322
334
1995
8,053
34,577
15,375
184
171,347
331
Source: CNP (2000)
Phase FiveMid-1990s to 2000: Decentralization of Colombian Drug Trade
With the fall of the Medellin and Cali cartels the Colombian drug trade has
decentralized. Instead of a handful of vertically integrated core organizations that
dominate the industry, hundreds of independent trafficking groups operate throughout

50
Colombia, conducting their activities on a smaller, more discrete scale. Some
organizations, particularly several based in the Northern Cauca Valley and Atlantic Coast
areas, are long-standing enterprises that escaped the drug enforcement net as the CNP and
DEA pursued more prominent core organizations.13 In addition to these medium-sized
survivors, hundreds of small trafficking operations have emerged in recent years. There
is variation in size among existing enterprises, with medium-sized organizations ranging
from twenty to one hundred participants, and smaller groups containing ten to twenty
associates. Contemporary enterprises tend to be flatter than the core organizations, and
more circumspect in carrying out their activities. Small groups are frequently led by a
single figure that exerts substantial decision-making authority, while co-equal
subordinates perform the work. Medium-sized organizations often contain different
leaders that occupy discrete, functionally-specific managerial positions, similar to the
core enterprises (Aero, 2000; DEA, 2000; FBI, 1993; Gallego Castrilln, 2000; Nicosa,
2000; Semana, 2000a).
Instead of coordinating several phases of trafficking activities like the core
enterprises, many trafficking operations today specialize in single-phases of drug
production, processing or transportation. While different groups, including former
12 It is not clear how many trafficking enterprises actually exist in Colombia, which is reflected in the wide
range of expert estimates. General Serrano, the former head of the Colombian National Police, estimates
that are between 350 and 400 minor trafficking organizations operating in Colombia. Alejandro Reyes
Posada, a leading Colombian drug researcher, is even more precise. According to Reyes Posada there are
438 narcotics organizations in Colombia, 256 of which are re-structured off-shoots of the former cartels,
while 182 represent new groups. A recent report in Semana magazine estimates there are between 80 and
100 groups in Colombia that specialize solely in the exportation of cocaine or heroin (Reyes Posada, 1999,
p. 2; Semana, 2000a, p. 71; Serrano Cadena, 1999, p. 237). My own research suggests that the clandestine
nature of drug trafficker merits greater discretion in formulating estimates. While DEA and CNP
intelligence officials confirm that there are hundreds of trafficking enterprises in Colombia, they also
stress the difficulty of providing exact estimates. The highest-ranking DEA official in Colombia told me
that while he believes there are several hundred organizations operating in Colombia he does not know the
exact number (Arreguin, 2000).
13 Indeed, a number of these groups provided transportation services to the Medellin and Cali core
organizations during the 1980s and 1990s.

51
competitors, cooperate and pool resources, there appears to be no centralizing
coordination mechanism that provides direction and contract enforcement. Another
development is that many Colombian enterprises now avoid establishing distribution cells
in the U.S., often selling their loads to Mexican trafficking enterprises that complete the
international transportation and distribution process.
Trafficking enterprises in Colombia today are frequently led by people with
considerable criminal experience. A number of post-cartel start-ups were founded by
former mid-level managers of the Medellin and Cali core organizations. Other prominent
traffickers have formally retired from the day-to-day business of drug trafficking but
continue to invest in shipments and offer their advice when solicited. This suggests that
the knowledge and experience of the former cartels has not been lost. Traffickers from
the core organizations draw on contacts and experience in conducting their present
operations, applying previous knowledge to new ventures. Government officials
interviewed for this research suggest that one reason why existing operations remain
small is that surviving traffickers have learned from experience that they are better off
avoiding drug enforcers by purposely limiting the size of their operations.
Contemporary trafficking enterprises rely on basic smuggling routines developed
by their predecessors, while exploiting advances in communications and transportation
technologies when modifying these practices and procedures. Traffickers continue to use
a variety of maritime vessels to transport multi-ton loads and general aviation aircraft for
400-800 kilogram drug shipments. However, at least one enterprise has upgraded to the
next level of submarine technology. In September 2000 Colombian officials discovered a
partially finished submarine based on advanced Russian technology, including a

52
computer navigation system and engine room. When completed the submarine would
have measured over 100 meters in length and had a transportation capacity of 200 tons,
far in excess of the one to three tons of cocaine transported by the mini-subs in the
Caribbean during the 1990s. Other recent innovations include using calling cards,
satellite phones, electronic mail, and encryption technology to communicate information,
mixing cocaine hydrochloride with charcoal and iron dust, welding cocaine into cargo
ship hulls, and using double-lined latex pellets for internal body carries (Arreguin, 2000;
Associated Press, 1998a; Branigan, 1999; Bohning, 2000; Brooks & Farah, 1998; CM&,
2000; CNN, 2000; DEA, 2000; El Tiempo, 2000d; Francis, 2000; Johnson, 2000;
Markowitz, 2000; Smith, 2000a).
Government Drug Enforcement Efforts During Phase Five
Colombian and U.S. drug enforcers continued to target the largest, most well-
known trafficking organizations in the late 1990s. In the post-cartel era, this meant the
medium-sized enterprises operating in the Northern Cauca Valley and along the Atlantic
Coast. Between 1997 and 2000, the leaders of several enterprises surrendered or were
captured by government authorities. In August 1997, CNP officials arrested Julio Cesar
Nasser David, one of the architects behind the Atlantic Coast trafficking network.14 The
following month, Orlando Henao Montoya, a leader in the North Valley network,
surrendered to Colombian officials in Bogot. In February 1998, CNP forces
apprehended Jos Nelson Urrego, another leader of the North Valley network, at his
country estate outside of Medellin. Four months later, Colombian authorities captured
14 Nasser Davids participation in the Colombian narcotics industry dates back to the 1970s, when his
organization smuggled several tons of marijuana to Southern Florida using mother ships and speedboats.
In the mid-1980s, Nasser Davids organization moved into the cocaine trade, transporting and distributing
multi-hundred kilogram shipments of cocaine to the U.S. through containerized cargo and general aviation
aircraft (Constantine, 1998; FBI, 1993, p. 31; Chepesiuk, 1999b, p. 153; Reuters, 1998c).

53
Alberto Orlandez Gamboa, the leader of a large Atlantic Coast trafficking enterprise
(Associated Press, 1998; CNNespaol, 1998; Chepesiuk, 1999b, pp. 92, 153, 253;
Constantine, 1998; DEA, 2000; FBI, 1993, p. 31; Guzman, 1998; Los Angeles Times,
1997; Reuters, 1998a, 1998c).15
In October 1999, CNP and DEA officials implemented Operation Millennium,
targeting a highly sophisticated smuggling network exporting between ten and thirty tons
of cocaine per month. In a series of carefully orchestrated raids in Bogot, Medellin and
Cali, CNP agents arrested thirty-three members of the network, including the alleged
leader, Bernal Madrigal, and Fabio Ochoa, one of the leaders of the old Medellin cartel
(Bajak, 1999; Johnson & Davies, 1999; El Tiempo, 1999).16 Also during this period,
CNP drug enforcement units captured a number of traffickers from smaller smuggling
groups and disrupted their cocaine and heroin operations {El Tiempo, 1997a; El Tiempo,
1997b; El Tiempo, 1997c; Caracol, 2000a; Caracol, 2000b; DAS, 1997; RCN, 2000;
Reuters, 1998b).
These drug enforcement operations indicate that CNP and DEA officials continue
to target trafficking organizations in Colombia. However, in spite of these successes, a
number of important traffickers remain at large. According to the DEA, Diego Montoya
Sanchez, once a mid-level figure in the North Cauca Valley network now leads one of the
major organizations in the area. Montoya Sanchez was the principal cocaine supplier for
the Madrigal operation and continues to coordinate multi-ton shipments of cocaine to
15 During the early 1990s, Orlandez Gamboas organization was associated with the Cali-based trafficking
network. Following the capture of the leaders of Cali core organizations in 1995 and 1996, Orlandez
Gamboa branched out on his own, exploiting maritime and air routes through Mexico and the Caribbean to
smuggle multi-ton shipments of cocaine and marijuana to the U.S.
16 Since their arrest Ochoa and Madrigal have been extradited to the U.S., where they are now awaiting trial
for numerous trafficking-related criminal charges {El Tiempo, 2001d, 2001e; Forero, 2001b; U.S. Attorneys
Office, 2001a, 2001b).

54
Mexico and the U.S. The DEA and CNP have identified other traffickers they believe
now lead the Atlantic Coast and North Cauca Valley networks. Moreover, drug enforcers
fear that surviving remnants of the Medellin and Cali cartels are reorganizing, led by
former mid-level managers. In addition to these medium-sized enterprises there are now
hundreds of smaller independent trafficking groups operating in Colombia. Some of
these groups are led by notorious traffickers that continue to elude drug enforcers, others
are run by individuals unknown to Colombian and U.S. authorities. With so many
criminal enterprises determined to produce and transport illicit drugs, Colombia remains
poised to continue its leading position in the international drug trade.
Table 2-3 Colombian Drug Enforcement Indicators, 1996-1999
Year
Persons
apprehended
Cocaine
interdicted
(kilos)
Cocaine
base
interdicted
(kilos)
Heroin
Interdicted
(kilos)
Marijuana
interdicted
(kilos)
Drug labs
destroyed
1996
5,703
17,808
11,142
88
101,519
436
1997
10,711
35,792
14,906
176
117,880
228
1998
18,276
46,256
11,346
341
73,085
198
1999
21,168
21,423
9,621
541
71,369
100
Source: CNP (2000)
Conclusion
Over the past seventy years, the Colombian drug trade, and the illicit enterprises
that coordinate it, have undergone a variety of changes. While the empirical record of
this clandestine industry remains sketchy, particularly in the early decades, this chapter
draws on the available materials to craft a credible, coherent narrative. In the beginning,
drug trafficking in Colombia was sporadic. Smugglers, often working alone, transported
minute quantities of European produced cocaine and heroin through Colombia and
Panama on the way to North American markets. Sometime in the 1940s or 1950s, a
number of pioneers, perhaps of German extraction, developed cocaine and heroin

55
processing laboratories on Colombian soil, transforming the country from mere
transhipment point to drug producer. Early producers sold their illegal wares to
international smugglers based in Cuba who transported the drugs to the U.S. During the
1950s and 1960s, many smugglers based in Colombia were content to serve as suppliers
and transhippers for trafficking enterprises based in other countries, including Cuba,
Chile, Mexico, and the United States. The primary psychoactive drugs produced in or
passing through Colombia at this time were marijuana and cocaine.
In the late 1960s and early 1970s a fundamental change occurred in the
Colombian drug trade. Several enterprises expanded into U.S. wholesale distribution
markets. These networks developed infrastructures and technologies that laid the
foundation for core enterprises in the 1980s. With the rise of the Medellin and Cali
cartels, and their multi-tonnage trafficking capacities, Colombias dominant position in
the international cocaine trade solidified. Driven by growing U.S. appetites for marijuana
and cocaine, the amount of psychactive drugs coming out of Colombian in the 1970s and
1980s increased substantially, while the organizations coordinating this traffic also grew
and became increasingly bureaucratic.
By the mid-1980s, a number of Colombian trafficking enterprises had become too
large, their leaders too ambitious. Over the next decade, the Colombian and U.S.
governments waged an on-again, off-again campaign to dismantle the leading core
organizations. While drug enforcers achieved remarkable victories, dismantling several
Medellin and Cali cartels, their ability to disrupt the industry was limited by the fluid
nature of trafficking enterprises. Within months of even the most effective supply-
reduction programs, drug smugglers would bounce back, revitalizing existing operations

or creating new ones, driven by the strong demand for their illicit commodities in U.S.
and European drug markets.
56
Table 2-4
Evolution of Illicit Drug Industry in Colombia
Phase
Time Period
Characteristics of Colombian Drug Trade
Organizational
Exemplar
One
1930s
Individual smugglers transport cocaine and
heroin to Panama
N/A
Two
Late 193 Os-
early 1960s
Smugglers transport cocaine and heroin in
Caribbean, supply criminal organizations
based in Havana, development of cocaine
and heroin processing labs in Antioquia
Herr an
Olazaga
brothers
Three
Late 1960s-
1970s
Marijuana and cocaine trafficking groups
supply Cuban, American, Chilean &
Mexican networks; establishment of
Colombian distribution networks in U.S.
Herrera
network,
Bravo
network,
Four
Late 1970s-
early 1990s
Rise of core organizations and support
networks; spread of cocaine production
throughout Colombia; participation of
guerrillas and paramilitary groups;
dismantling of Medellin and Cali core
organizations
Core
organizations
in Medellin
and Cali
Five
Mid 19905-
present
Decentralization of Colombian drug trade;
small and medium-sized groups; growth of
cocaine & heroin production in Colombia
North Valley
network,
Atlantic Coast
network,
Madrigal
network
Today, many trafficking enterprises in Colombia bear a greater resemblance to the
pre-cartel groups of the 1960s and 1970s than to the core organizations that followed.
They are small and contain flat decision-making hierarchies. They specialize in drug
production or transportation, but not both. Their leaders avoid the limelight, managing
their operations with the utmost discretion. Yet, in one significant respect existing
groups remain similar to their larger predecessors. They continue to modify their
trafficking practices by exploiting advances in communications and transportation

57
technologies, ensuring that their smaller size does not equate with lesser sophistication.
The remainder of this study is devoted to understanding how these criminal enterprises,
and their state adversaries, draw on information and experience to alter their form and
function over time.

CHAPTER 3
TOWARD A THEORETICAL UNDERSTANDING OF
ORGANIZATIONS AND ORGANIZATIONAL LEARNING
Organizational learning is a slippery concept. It contains multiple meanings that
are open to a variety of interpretations. Learning is difficult to define, let alone measure
and apply to empirical settings. Reflecting such difficulties, the academic literature lacks
a generally agreed upon theory of organizational learning (Levy, 1994, p. 280; Fiol &
Lyles, 1985, p. 803; Keohane & Nye, 1989, p. 264; Stein, 1994, p. 224). In spite of the
theoretical and methodological challenges associated with organizational learning, over
the past several decades social scientists have drawn on a number of basic concepts to
develop a cumulative body of research that demonstrating learning to be a valid
explanation for certain types of organizational change. Dating back to Herbert Simons
seminal formulation, first published in 1945, that organizations intelligently adjust
routines in response to external stimuli and experience, political scientists have applied
notions of learning to a variety of subject matter, including government institutions,
intergovernmental bodies, military organizations, transnational epistemic communities,
and elite decision-makers.1
1 Two decades after the appearance of the first edition Administrative Behavior (Simon, [1945] 1997), Karl
Deutsch applied the concept of cybernetics to communications systems within governments and other
social organizations (Deutsch, 1966). In the early 1970s several political scientists and diplomatic
historians analyzed how policy makers draw on lessons of the past when making decisions (Lowenthal,
1972; May, 1973). In 1976, Robert Jervis drew on social and cognitive psychology to explain how
decision makers reason and leam from previous experience (Jervis, 1976). In recent years the research
agenda on learning in political science has gathered momentum as a growing number of scholars have
applied learning concepts to individual decision-making and organizational behavior. Examples of this
literature includebut are not limited to: Breslauer and Tetlock (1991), Dodd (1994), Eden (forthcoming),
Etheredge (1985), Evangelista (1988), Goldgeier (1994), Goldstein & Keohane (1993), Haas (1990), Haas
58

59
In this literature, criminal organizations and law enforcement agencies have rarely
been topics of empirical interest. This research builds on existing studies by examining
organizations that stand at the intersection of the rule of law and national security. The
lack of attention provided trafficking enterprises and drug enforcement agencies in the
organizational learning literature raises a number of questions. Most basically, what is a
drug trafficking organization that it may learn, and undertake other collective actions?
Do police agencies and trafficking enterprises learn from experience and information? If
these state and non-state collectives alter their practices and procedures, how can we be
sure that such changes are attributable to learning rather than alternative considerations?
What are the individual and organizational obstacles to learning? Assuming that
trafficking enterprises and law enforcement agencies do on occasion learn, what
conditions facilitate this process?
I do not propose to answer these demanding questions in this chapter. That
challenge unfolds later in this study. Here, my aim is more modest: to develop the
theoretical framework that guides subsequent analysis. I begin by discussing a concept
that subsumes organizational learning: organization. In order to evaluate whether
trafficking enterprises and drug enforcement agencies learn as organizations, I am
compelled to explain the latter concept. I do so by drawing on the open-systems
perspective of organization theory. Following this discussion, I proceed with a definition
of organizational learning that disaggregates the process into separate but overlapping
(1992), Haas & Haas (1995), Hall (1993), Heclo (1974), Khong (1992), Leng (1992), Le Prestre (1995),
Mann (1992), McCoy (2000), Mendelson (1993), Moltz (1993), Nagl (1999), Ndumbaro (1998), Nolan
(1994), Nye (1987), Posen (1984), Reiter (1996), Sabatier (1988), Sagan (1993), Sikkink (1997), Snyder
(1991), Stein (1994), Weir and Skocpol (1985), Welch Larson (1985), Zarkin (2000), and Zeng (1999).
For recent overviews of the political science literature on learning see Bennett & Howlett (1992) and Levy
(1994).

60
phases of information acquisition, interpretation, and application. In recognition that
learning is not the only source of change in organizational routines, I address alternative
sources of change in organizational behavior, including power and environmental
selection. I follow this with brief analyses of several different learning concepts that are
pertinent to this research, including competitive learning ecologies, productive
learning, tactical adaptation, and strategic learning. The chapter continues with a detailed
discussion of the numerous psychological, organizational, and environmental obstacles to
organizational learning, including bounded rationality, tall hierarchies, and benevolent
environments.
As fascinating as the concept of organizational learning may be, my purpose in
these pages is not to theorize for its own sake, but to develop a series of propositions that
can be applied to the units of analysis in this research. Throughout this chapter I
intersperse the discussion with falsifiable propositions that I apply to trafficking
enterprises and drug enforcement agencies in subsequent chapters.
Organizations
At the most basic level, organizations are simply collections of participants that
coordinate their activities through patterned behaviors, and draw on resources and
technologies to achieve specific goals. Organizations are social systems that contain five
inter-related components: tasks, environments, structures, participants, and technologies
(Scott, 1998).
Organizational Tasks
Organizations exist to achieve aims shared amongst participants. Tasks are
embodied in goals that drive organization behavior, and objectives to achieve these

61
desired ends. Organizational goals include creating goods and services, attaining
satisfactory profits and other economic, social and political benefits, and ensuring the
survival of the enterprise. Objectives include constructing production facilities, searching
for new markets, devising innovative production and marketing methods, recruiting
people to work for the organization, and reducing internal and external threats to the
integrity of the enterprise. The determination of satisfactory benefits is highly
subjective and varies among organizations. However, autonomous organizations do not
exist for long unless the value of their products exceeds the resources needed to make
them (Leavitt et al., 1973, pp. 10-11; Scott, 1998, pp. 20-21).
Tasks may be clearly defined and widely appreciated within an organization, or
they may be vaguely understood and unique to sub-parts of the collective (Leavitt et al.,
1973, pp. 23, 27). As Barnard and Simon recognized decades ago, individuals within
organizations frequently do not share a common conception of organizational goals
(Barnard, 1938; Simon, [1945] 1997). Moreover, organizations typically pursue multiple
objectives simultaneously, which can create competition and conflict among different
parts of the enterprise.
Environments
Organizations develop within environmental contexts. Environments provide
opportunities for participants to coordinate their behavior, and the material and symbolic
resources necessary to survive. In organization analysis, the environment is a residual
concept. It refers to anything outside the organization. Dills notion of the task
environment provides an element of precision to this broad concept. The task
environment refers to those parts of the environment that are relevant to the

62
organizations aim or purpose. Dill (1958), and subsequently Thompson (1967),
disaggregate the task environment into four areas: (1) customers or clients of the
organization (including distributors and consumers); (2) suppliers of labor, capital and
technologies; (3) competitors for market share and resources; and (4) regulatory groups,
which include government agencies. The organizations that fulfill these roles comprise
an organization set. An organization set is composed of the customers, suppliers,
competitors and regulators with which an enterprise interacts in order to achieve its
purpose (Blau & Scott, 1962; Dill, 1958; Evan, 1966; Kaufman, 1985, p. 41; Posen,
1984, p. 43; Scott, 1998, p. 147; Scott, 1998, pp. 124, 147; Smith, 1980, p. 376;
Thompson, 1967, pp. 27-28).
Organizations draw on task environments for a wide variety of physical,
technological, and symbolic inputs. Organizations obtain necessary inflows of capital,
participants, and technology from external suppliers. Capital provides the enterprise with
the financial resources necessary to acquire other material and informational resources.
Participants bring knowledge, skills and experience that organizations use to perform
tasks. Technologies include the equipment, machines, instruments, and information
necessary for the organization to accomplish work. Of course, interactions between
enterprises and task environments are not unidirectional: organizations trade outputs
with customers or clients in return for more inputs. In order to receive the necessary
inputs, organizations must produce goods or services valued by their customers or clients.
This is true for both legal and illegal organizations (Posen, 1984, p. 43; Scott, 1998, pp.
21-22, 229; Smith, 1975, pp. 340-341; Thompson, 1967, p. 28).

63
Hostile environments
Environments are not only a source of opportunities and resources for
organizations; they are also a source of uncertainty and risk. The ongoing flows of
personnel, resources and information between organizations and their task environments
are frequently erratic and unpredictable. While a number of factors affect the degree of
risk and uncertainty within an environment, few are as salient as the degree of
environmental hostility, as opposed to benevolence. A hostile task environment contains
two or more interconnected adversaries, at least one of which seeks to dismantle other
actors within the system. While a benevolent environment is not necessarily devoid of
competition, it is friendly in the sense that the ultimate goal is to out-perform but not
undermine the other units, so that all may survive to play another day.
Organizational Structure
Structure refers to relationship patterns among participants in organizational
settings. These patterns emerge from values, norms, roles and routines that characterize
how organizations function. Values refer to criteria for selecting desired goals of
organizations, such as increasing profits and ensuring organizational survival. Norms are
general rules that specify appropriate means for pursuing goals, such as exploiting new
markets and investing in research and development. Roles are behavioral expectations
for organization participants, often dealing with task performance. Roles define the
division of labor within organizations, assigning participants to tasks that facilitate
collective aims. Structure is not arranged randomly but ordered into coherent beliefs
and prescriptions governing the behavior of participants (Scott, 1998, pp. 17-18). These
patterns provide enduring, but not necessarily static, frameworks for structuring

64
authority, work, and communication within organizations (Argyris & Schon, 1996, pp. 8-
9; Donald & Wilson, 2000, p. 194; Leavitt, 1965, p. 1144; Leavitt et al., 1973, p. 4;
McCluskey & Wardle, 2000, p. 252).
Routines
Routines are an essential part of organization structure. They include rules,
procedures, conventions, and practices that guide individual behavior within
organizations. Routines interlock behavior among multiple participants by matching
action to informational cues. Routines provide rules of thumb for participants to
respond to incoming stimuli in ways that make sense. In this manner, routines coordinate
the behavior of different individuals into coherent and predictable patterns of behavior.
This coordination allows organizations to accomplish work reliably and consistently
(Cohen & Bacdayan, 1996, p. 348; Cyert & March, 1963; Feldman, 2000, p. 611; March
& Simon, 1958; Weick, 1979, pp. 112-115).
In addition to organizing individual action into collective behavior, routines
influence participants by providing rules for conduct, along with an incentive system that
matches rewards and punishments for specific action. Through training and socialization,
participants learn the rules and procedures of their organization and receive rewards (or
punishments) according to their ability to follow them. Successful participants generally
move up the ranks of the organization and tend to apply the same criteria to their
subordinates. In this manner, routines frequently become institutionalized within
organizations, outlasting individual participants and sometimes even their original
usefulness (Levitt & March, 1988, p. 320; March & Simon, 1958, p. 145; Posen, 1984, p.
44).

65
Organizations use routines to structure authority relations, allocate resources,
make plans, perform tasks, assign roles, communicate information, build culture, and
solve problems. Routines describe how organizations structure authority among different
layers of decision-making hierarchy. They describe the functional responsibilities of
different roles within the collective. Routines explain how organizations create the goods
or services that they produce. They show how organizations maintain records of at least
certain aspects of their operations. Routines determine how information is shared within
the collective and who has access to what data. They express important values and
provide a means of constructing organizational culture. Finally, routines describe how
organizations respond to existing problems and plan future endeavors (Cyert & March,
1963, pp. 103-104; Decker et al., 1998, p. 407; Levitt & March, 1988, p. 320).
Routines are nested in several levels of organizational behavior. Standard
operating procedures are simple rules of thumb for accomplishing basic tasks, as when a
participant responds to a stimulus by performing a specific action. Basic procedures are
combined to create more complicated action-sequences called performance programs and
repertoires. A performance program is simply a cluster of procedures for dealing with a
specific situation. A repertoire is the sum of the organizations performance programs for
organizing behavior in an area of activity (March & Simon, 1958, p. 141; Posen, 1984, p.
44).2
Like their building blocks, performance programs are triggered by stimuli. The
sounding of an alarm bell in a fire station provokes a fairly elaborate program of activity
carried out by numerous participants. Two or more procedures can be combined to form
2 This is similar to Swidlers notion of cultural tool kits. According to Swidler, culture influences action by
providing a repertoire or "tool kit" of habits, skills, and styles from which people construct "strategies of
action." See Swidler (1986) and Kier (1997).

66
a performance program. Posen provides a simple but illustrative example in his
organizational analysis of military doctrine. In most Western armies, it is standard
operating procedure for tanks to fire from the hull-down position as this provides
greater protection from enemy fire. Another standard procedure calls for tanks to
advance by exploiting topographical irregularities in the surrounding terrain. Military
organizations often combine these two procedures into an effective performance
program: while some tanks advance on the enemy by exploiting the terrain, others remain
in defensive firing position and provide cover for the advancing units. An organizational
repertoire refers to the combination of performance programs available to an organization
when undertaking an activity. The performance program described by Posen is one of
many available in a military organizations repertoire for engaging in land-based warfare
(March & Simon, 1958, p. 141; Posen, 1984, p. 44).
Organizations vary in the degree to which their routines are formalized. Rules
and procedures may be considered more or less formal to the extent that they are codified
in documents. Formal routines are explicitly documented in manuals, files, data bases,
financial accounts, and other forms of organizational memory. Informal routines are
not recorded in documents but exist in the inter-subjective mind of organization
participants. These customs, norms, rules and procedures are shared verbally and
reinforced amongst participants through socialization processes. Certain understandings
of how things are done around here are reinforced over others as participants engage in
their daily activities. While some organizations may contain mostly formal or informal
routines, many contain both types, with some rules and procedures formally recorded in

67
organizational memories and other routines found in inter-subjective understandings
among participants.
Participants in Organizations
Organizations pursue their tasks through participants that make contributions in
return for inducements. Participants supply the human resources that allow organizations
to accomplish work. They define goals and objectives, form routines, perform tasks,
make decisions, share information, plan future actions, and address problems.
Participants join organizations for different reasons. Moreover, their assessments of
satisfactory inducements are subjective. Even wages are not a sufficient inducement for
all potential recruits. Human beings seek authority, status, power, fellowship, and a host
of other entitlements besides money. Complicating matters, participants frequently
belong to more than one organization, and their loyalty to the different collectives varies
considerably. Participants flow from one organization to another, carrying with them
values, norms, and beliefs about themselves and the roles they play (Barnard, 1938;
Cohen et al., 1988, p. 295; Feldman, 2000, p. 614; Leavitt, 1965, p. 1144; Posen, 1984, p.
43; Scott, 1998, p. 19).
In addition to their myriad contributions, participants are also a source of great
uncertainty for organizations. To accomplish their tasks, organizations require
coordination and control, and this is not easily achieved among participants with fluid
participation, wavering loyalties, and different conceptions of satisfactory inducements.
Moreover, even the most loyal and satisfied participants are limited in their ability to
pursue organizational aims. Like all human beings, they are bound by cognitive,
corporeal, and cathectic constraints. Of course, organizations are designed with these

68
limitations in mind, and much of their social structure can be understood as mechanisms
by which organizational leaders and managers seek to channel participants competing
aspirations, energies and activities into a coherent and consistent pattern of behavior that
facilitates the organizations purpose. However, people are not mindless automatons, and
their aspirations, hopes, and fears cannot be calculated and manipulated like threshing
gears and rotating gyroscopes (Feldman, 2000, p. 614; Posen, 1984, p. 43; Simon, 1997).
Technology
Organizations perform work. Some organizations manufacture materials or
machines; others provide services, information or other products. Technology refers to
the material and symbolic inputs that organizations use to complete work. Technology
includes the equipment, instruments, and information that participants use to complete
their activities, as well as the technical knowledge and skills of participants themselves
(Scott, 1998, p. 21; Leavitt, 1965, p. 1144; Leavitt et al., 1973, p. 4).
Whether the purpose of an organization is to produce legal or illegal products, it
draws on necessary technological resources. If the purpose of an organization is to
provide education, it will make use of material and symbolic resources that allow it to
accomplish this aim. Such resources include buildings for housing classrooms and
administrative offices, computers, projectors and books for developing and delivering
course materials, and the accumulated knowledge and skills of educators and their
support staff. Likewise, an organization that provides escort services will require an
office and/or other physical location(s) to coordinate and implement operations;
telephones, beepers and other equipment for communicating among participants,
customers, suppliers, and regulators; computers or other instruments for keeping track of

69
organizational inputs and outputs; support materials and services that allow participants
to carry out their activities; and the knowledge, skills and experience of sex workers
themselves.
Depending on the degree of hostility in the task environment, criminal
organizations may be required to devote a considerable portion of their technological
resources to conceal and otherwise protect the physical integrity of their operations.
However, some illicit activities, including illegal gambling and prostitution, are more
accepted among law enforcement and government officials than others, such as narcotics
and weapons trafficking (Reuter, 1983; Smith, 1975). For this reason, it is not surprising
that many contemporary narcotics organizations devote a substantial portion of their
technological resources towards reducing uncertainty and risk emanating from the task
environment. Trafficking enterprises use these material and symbolic inputs to hide
narcotics shipments, corrupt government officials, encipher communications, gather
counter-intelligence, stockpile inputs, and maintain warehouse inventories.
Defining Organizational Learning
Organizations learn when participants receive, interpret, and apply information to
behavior through organizational routines. Organizational learning is a form of
organizational action. When an organization acts, such that participants fulfill roles
and perform functions on behalf of the collective, then it may be said to learn when
participants gather, analyze, and apply information (Argyris, 1993, p. 8; Argyris &
Schon, 1996, p. 10; Farkas, 1998; Hedberg, 1981, p. 3; Simon, 1996, p. 176).
Throughout this study, organizational learning refers to a certain type of behavior
performed by individuals acting on behalf of the collective.

70
The process of organizational learning can be conceptually disaggregated into
three steps: acquiring, interpreting, and acting on information (see Figure 3-1 below).
Organizational learning unfolds as information is discovered, shared, and applied by
participants that coordinate their activities through formal and informal routines. The
three phases of organizational learning are disaggregated for heuristic purposes; in
reality, information can be acquired, interpreted, and applied simultaneously. The
different steps are connected through feedback loops that reinforce certain interpretations
and behaviors while disregarding others. There are a number of impediments to
organizational learning, and the process may be blocked at any phase. I discuss obstacles
to learning in detail later in the chapter.
Figure 3-1 Simple Process Model of Organizational Learning
Source: Adapted from Daft & Weick (1984), Garvin (2000).
Acquiring Information
The first step in the organizational learning process is gathering information.
Information is the lifeline of organizational learning. Information includes know-how,
understanding, techniques or practices and is useful to the extent that it helps
organizations achieve targets. Organizations acquire information through their own trial
3 Although the language varies, a number of analysts conceptualize organizational learning as a process
composed of three or four discrete stages. Daft and Weick describe learning according to three analogous
stages, including scanning (data collection), interpretation (data given meaning), and learning (action taken)
(Daft & Weick, 1984, p. 286). Kim describes learning as composed of four stages: observing, assessing,
designing, and implementing (Kim, 1993, p. 44). See Garvin for discussion on this point (Garvin, 2000,
endnote 4, p. 223).

71
and error experience and exogenous search. Trial and error experience refers to the
process by which organizations interact with their environment, using a variety of tactics
and strategies designed to achieve targets. The experience that emerges from this activity
provides a valuable guide for future behavior. Search refers to the process by which
organizations gather information from sources exogenous to the organization but
endogenous to the task environment. Search provides the means to draw on the
knowledge and experience of others, including customers, consultants and competitors, in
making decisions (Argyris & Schon, 1996, p. 3; Weick, 1979).
Gathering information is costly. It involves an expenditure of organizational
resources, including time, attention, personnel, and money. Because resources are
limited, organizations are more likely to acquire knowledge and experience when they
have a reason to do so, such as the need to find a solution to a problem. As target-
oriented entities, the failure to meet targets or objectives is a common source of problems
in organizations. The failure to match expected outcomes with actual outcomes produces
emotional arousal on the part of participants that triggers a search for information,
including alternative courses of action to resolve the problem (Allison, 1971, p. 84;
Argyris & Schon, 1996, p. xxiii; Garvin, 2000, p. 23; March, 1988, p. 3; March & Olsen,
1989, p. 60; Weick, 1995, pp. 45-46).
Interpreting Information
For the learning process to proceed, participants must interpret information. Daft
and Weick provide an illuminating description of this complex process:
Organizations must make interpretations. Managers literally must wade into the
ocean of events that surround the organization and actively try to make sense of
them. Organizational participants physically act on these events, attending to
some of them, ignoring most of them, and talking to other people to see what they

72
are doing.... Interpretation is the process of translating these events, of
developing models for understanding, of bringing out meaning, and of assembling
conceptual schemes among managers (Daft & Weick, 1984, p. 286).
Interpretation is a social process through which organizational participants share
their perceptions and construct inter-subjective understandings that allow them to make
sense of relevant events. When engaged in interpretation, or sensemaking, participants
communicate, share knowledge and experience, and construct linkages between the
present situation and prior experiences, often with the assistance of analogical reasoning
(Daft & Weick, 1984, p. 286; Khong, 1992; Neustadt & May, 1986; Weick, 1995, pp. 45-
46).
Organization managers and other participants are individual cognitive agents;
they draw on schemas when analyzing information.4 Schemas are knowledge
structures that individuals use to lend order to an otherwise incomprehensible confusion
of information and experience (Reiter, 1996, p. 21). Schemas help individuals make
sense of what has gone before them and what transpires around them by providing
cognitive shortcuts or frames of reference. As personalized frames, they can be an
impediment to organizational learning, preventing decision-makers from interpreting a
problem or solution in the same light.
Within organizations, participants share, record, and interpret knowledge and
experience through formal and informal organizational memories. Formal organizational
memories include files, manuals, correspondence, databases, financial accounts, and
physical objects that record and store knowledge, such as monthly accounting statements,
minutes from meetings, electronic mail, and computers (Arygris & Schon, 1996, p. 16;
4 Schemas are knowledge structures that individuals use to lend order to an otherwise incomprehensible
confusion of information and experience (Reiter, 1996, p. 21). Schemas help individuals make sense of
experience and their environment by providing cognitive shortcuts or frames of reference.

73
Simon 1997, p. 218). Although recent advances in information technology have reduced
the costs of recording and retrieving information, much organizational experience
remains undocumented. Undocumented knowledge is often transmitted through informal
organizational memories. These include the conversations, stories, and myths through
which participants construct and conserve inter-subjective understandings regarding
appropriate practices and procedures (Levitt & March, 1988, p. 327; Argyris & Schon,
1996, p. 16).
Organizations interpret only a small portion of their experience at a time.
Organizational memories are evoked at certain times by particular parts of the collective.
Depending on how they are organized, some aspects of organizational memories may be
more accessible than others. In particular, the availability of memories is associated with
their recency and frequency of use, as well as their location within the organization.
Recently and frequently used memories are easier to retrieve than those that are used only
sparingly or have not been accessed over a considerable period of time. Information
stored in electronic documents that form part of computerized data bases, or paper
documents located in well-organized filing systems are more likely to be used than
isolated documents stored in obscure locations (Levitt & March, 1988, pp. 328-329).
Applying Information
The process of organizational learning is not complete until knowledge that has
been acquired and interpreted by participants is applied to collective behavior (Garvin,
2000, p. 26). Organizations enact knowledge and experience through routines. Routines
include rules, procedures, conventions, and practices that guide individual behavior
within organizational settings. Routines provide the mechanism through which the

74
behavior of multiple participants is interlocked (Feldman, 1989, p. 136; Hutchins,
1996, p. 50; Weick, 1979, p. 3). Routines are transmitted over time through socialization,
training and personnel movement. They are recorded in formal and informal
organizational memories. Moreover, they are independent of the individual actors who
execute them and are capable of surviving considerable turnover in individual actors
(Levitt & March, 1988, p. 320). Members may come and go and the organizations
leadership may change, but when an organization learns, knowledge is preserved in
formal and informal memories and routines that outlast individual participants. Routines
that are associated with desirable consequences, such as improved task performance, tend
to be selected and retained by organizations.5 If participants can be shown to alter a
routine or series of routines as a result of acquiring and interpreting experience, then the
organization may be said to have learned from that experience (Daft & Weick, 1984, p.
285; Fiol & Lyles, 1985, p. 804; Hedberg, 1981, p. 6; Levitt & March, 1988, p. 326; Nye,
1987, p. 381; Zeng, 1999, p. 39).
Organizations differ with respect to the extent that their behavior is determined by
formal or informal routines. Large government bureaucracies with extensive institutional
histories tend to document experience through codified rules and procedures. Small
craft-based enterprises rely more on informal understandings transmitted verbally among
participants. Organizations that face complex uncertainties or hostile task environments
rely more on informally shared understandings than organizations that function with
relatively simple, stable environments (Levitt & March, 1988, p. 327; Ouchi, 1980).
5 However, this is not always the case, as we shall see below when discussing learning under ambiguity.

75
Research Propositions
Organizational learning refers to the process by which participants gather,
interpret, and apply information on behalf of an organization. The application of
knowledge and experience occurs through changes in informal and formal routines that
guide participants behavior. This study applies a process-oriented model of
organizational learning to two units of analysis: Colombian drug trafficking enterprises
and U.S. and Colombian drug enforcement agencies.
If participants within either organization alter their routines as a result of
acquiring and interpreting knowledge and experience, then the enterprise in question may
be said to have learned. In trafficking enterprises, changes in organizational practices
and procedures may involve the production, transportation, and distribution of illicit
drugs, or money laundering activities. For law enforcement agencies, changes in
organizational practices and procedures may involve criminal investigations or drug
enforcement operations. For both organizations, changes in routines may improve
organizational task performance, but they are not required to do so.
These ideas yield the following empirically verifiable propositions.
PI: If participants in Colombian drug trafficking enterprises gather, interpret, and
apply information to collective behavior by changing existing routines or creating
new ones, then organizational learning has occurred.
P2: If participants in U.S. and Colombian drug enforcement agencies gather,
interpret, and apply information to collective behavior by changing existing
routines or creating new ones, then organizational learning has occurred.
Alternative Explanations for Organizational Change
Learning is not the only source of change in organizations, and it is important to
clarify alternative explanations for this phenomenon. Perhaps the most common source

76
of non-learning change is the imposition of new or modified routines by more powerful
external actors on the unit of analysis. In the late 1960s and early 1970s, U.S. federal
drug enforcement agencies underwent a variety of organizational changes. However,
these changes were due to the imposition of new organizational structures and procedures
by policymakers in the Johnson and Nixon administrations, rather than information
processing by decision-makers within the drug enforcement agencies.6
It is possible to argue that learning takes place in such circumstances, but it is the
more powerful actor that learns and passes lessons on to the subordinate agency in the
form of routines that encode experience. While this may be true in some cases, the
explanation is largely unsatisfactory because it allows the analyst to change the unit of
analysis ex post facto. To reliably evaluate the research propositions, it is important to
maintain a consistent unit of analysis during the design and implementation of the
research. In this study, the units of analysis are Colombian drug trafficking enterprises
and U.S. and Colombian drug enforcement agencies. If change in routines is not due to
the information processing activities of these organizations, then learning, according to
the parameters of this research design, has not occurred.
P3: Organizational learning does not occur when changes in organizational
routines are due to the imposed preferences of more powerful political or
bureaucratic actors.
Another source of non-learning change is environmental selection. This occurs
when changes in organizational structures or operations appear as one set of actors is
selected out of the system and replaced by another. Because they operate in hostile task
environments it is not uncommon for trafficking enterprises to be selected out of the
illicit drug industry (i.e. identified and dismantled) by drug enforcement agencies. In this
6 For analysis of these reforms, see Rachal (1983) and Wilson (1978, 1989).

77
manner, successful drug enforcement yields a process of natural selection that favors
some criminal enterprises over others. Trafficking organizations that survive drug
enforcement efforts may contain more effective combinations of decision-making
hierarchies, leadership styles, and operating procedures (Alchian, 1950; Kaufman, 1985;
Moe, 1984, p. 746; Nelson & Winter, 1981).
Over time, organizational forms better suited to survive hostile task environments
may populate the illicit drug industry (see Chapter 2). However, if survivors do not
change their routines as a result of information processing, then learning has not taken
place. Surviving enterprises may be more effective than their predecessors, at least in
terms of their ability to elude the state, but their ability is not due to organizational
learning. On the other hand, if participants in trafficking enterprises recognize that they
must change their organizational routines to survive hostile environments, and they do so
as a result of this recognition, then learning has transpired.
P4: Changes in the illicit drug industry may be due to natural selection rather than
organizational learning by individual enterprises.
Power
Learning
Environmental
selection
Change in
organizational
routines
Figure 3-2 Alternative Sources of Change in Organizational Routines

78
Organizational learning is a cognitive activity performed by participants that
observe and interpret events and make decisions based on their perceptions of reality.
While organizations are open systems, sensitive to exogenous pressures, the source of
learning is ultimately endogenous to the enterprise. This does not require participants
with access to complete information or unlimited computational abilities (see discussion
below). It does require participants that recognize they belong to a goal-oriented
collective, and act and manipulate information on behalf of this entity. To qualify as
organizational learning, change in routines must be the result of information acquisition
and processing activities on the part of participants. Otherwise, it is not certain that
organizational change is due to learning or alternative explanatory variables, such as
power or environmental selection.7
Learning Ecologies
Organizations interact with other organizations that share common task
environments. Organizations draw on task environments for resources, including
personnel, equipment, capital, and information. In common task environments, one
organizations knowledge and experience can be can be captured by others through the
diffusion of technology. Technology, understood broadly to include material equipment,
routines, and organizational memories, spreads through environments as organizations
copy or mimic one another. Often, targets of technology transfer are those organizations
that are seen as successful by their peers. These model organizations use superior
7 By distinguishing learning from other sources of organizational change, this study avoids a common
tautology in the literature. However, by making the distinction between learning and change so explicit, I
have increased the difficulty of demonstrating learning in subsequent chapters. My challenge is to show
that changes in routines by drug trafficking enterprises and drug enforcement agencies are due to
information acquisition, processing, and application, rather than other factors, such as power and
environmental selection.

79
technologies believed to be worthy of emulation. Agents of diffusion include participants
that communicate with inter-connected organizations, including competitors, and
independent consultants that contain the requisite knowledge and experience (DiMaggio
& Powell, 1983, pp. 150-153; Levitt & March, 1988, p. 329; March, 1999, p. 45; Scott,
1998, p. 213).
Ecologies of learning develop when multiple organizations learn from each
other within a shared environment. The classic type is a collection of competitors, in
which rivals are closely linked through the diffusion of experience and the effects of their
activities on one another. In this type of learning ecology, organizations compete not
only for resources, customers, and profits but for political power and social legitimacy.
Organizations learn from their competitors by analyzing their competitors reactions to
their own actions (DiMaggio & Powell 1983, p. 150; Hedberg 1981, p. 13; Levitt and
March 1988, pp. 329, 331-332).
P5: Colombian drug trafficking enterprises and U.S. and Colombian drug
enforcement agencies learn from each other within common competitive learning
ecologies.
Competitive learning ecologies complicate the task of analysis. Because
organizations in these environments draw on their competitors for knowledge and
experience, learning by the one cannot be fully understood without reference to learning
by the other. In recognition of this complication, this study analyzes interactions between
trafficking enterprises and drug enforcement agencies to determine whether they learn
from each other (see Chapter 6).

80
Productive Learning
Among organization theorists, learning is commonly equated with improved task
performance.8 A number of analysts have drawn on this assumption to define learning as
the organizations ability to improve task performance over time. According to Fiol and
Lyles, learning is the process of improving actions through better knowledge and
understanding (Fiol & Lyles, 1985, p. 803). Dodgson describes learning as the ways
firms build, supplement and organize knowledge and routines around their activities...
and develop organizational efficiency by improving the use of the broad skills of their
workforces (Dodgson, 1993, p. 377). In their early collaborative research, Argyris and
Schon equate learning with increasing organizational effectiveness, however, in their
more recent work they distinguish productive learning, which improves performance,
from generic learning that does not. The former variety remains the focus of their
theoretical and empirical work. Other analysts have chosen different labels for learning
that improves performance, including instrumental, effective, and efficiency
learning (Argyris & Schon, 1978; Argyris & Schon, 1996, p. 4; Hedberg, 1981, p. 3;
Ndumbaro, 1998, p. 39 Tetlock, 1991, p. 35).
The classic example of efficiency learning is found in early studies of airplane
manufacturing. Beginning in the 1920 and 1930s, a number of industrial economists
discovered that the costs of airplane manufacturing fell predictably with increased
production volumes (Wright, 1936; Arrow, 1962). Increases in production were seen as
proxies for greater skill and knowledge accumulated through repetition of performance
tasks (Garvin, 2000, p. 94). This phenomenonlabeled the learning or experience
8 The assumption that learning results in improved organizational performance derives from the broader
assumption in behavioral studies of organizations that organizational learningand behavior more
generallyis fundamentally geared towards targets or goals.

81
curvewas subsequently found in a number of manufacturing industries and other
production tasks involving repetition, although learning rates varied considerably across
industries, products and time (Levitt & March, 1988, p. 321; Epple, Argote & Devadas,
1996; Yelle, 1979).9
One problem with the efficiency conception of learning is that it disregards
instances whereby changes in routines based on the acquisition and interpretation of
information fail to improve task performance. As numerous organization theorists have
recognized, learning does not always increase effectiveness (Cook & Yanow, 1996;
Hedberg, 1991, p. 89). Like individuals, organizations may learn dysfunctional behavior,
as when they adopt routines that hinder task performance or require more organizational
resources to achieve the same outcome. Productive learning is an important subspecies
of organizational learning. However, analysts should distinguish clearly between the two
varieties and avoid defining the latter by the former (Argyris & Schon, 1996).
Organizational learning may improve task performance, and it may not.
Levels of Learning: Tactical vs. Strategic Adaptation
A number of analysts have identified different degrees or levels of
organizational learning. While a variety of terminology has been developed to explain
this process the basic idea revolves around the distinction between the goals and
objectives of organizational behavior. Simple, low or lower, single-loop
learning and adaptation describe the process by which organizations adjust the means
9 Moreover, in some industries production costs have actually increased over time and accumulated
experience, thus demonstrating little capacity for learning. The commercial nuclear reactor industry is one
notableand frighteningexample (Perrow, 1984, pp. 34-35; Bupp & Derian, 1978).

82
of behavior in response to environmental feedback.10 Complex, high or higher-
level, meta-level, double-loop learning, and just plain learning all refer to the
process by which organizations respond to goal-changing feedback by altering the
fundamental values and basic goals that drive organizational behavior.11 However, the
dividing line between simple and complex learning is inherently fuzzy. Within
organizations, goals and objectives are ambiguous and frequently change over time.
Moreover, the meaning of goals and objectives varies in different parts of the
organization. An objective for one part of an organization represents a goal for another.
Even worse, different parts of the same organization may have conflicting goals and
objectives (Levitt & March, 1988, p. 325; Tetlock, 1991, p. 46).
Distinguishing between the goals and objectives of organizational behavior is
tricky business. Yet, it is also true that not all learning is the same. There is a difference
between adjusting t