You are on page 1of 288

Cognition and Crime

The rational choice perspective developed by Cornish and Clarke in 1986 provides
criminologists with a valuable and practical framework for the purposes of crime
control and prevention. More than 25 years later, Cognition and Crime pushes the
boundaries of this field of research by bringing together international leading (or
emerging) researchers in this area of script analysis into a single volume for the first
time. It also presents a series of original contributions on offender decision making
during the crime commission process as well as offering a critical perspective of
what could be achieved in the future to further help develop this field of research
for prevention purposes. In addition, each empirical chapter treats a specific and
important form of crime such as stalking violence, drug dealing, human trafficking
for sexual exploitation, child sexual abuse, and the transnational illegal market of
endangered species.
Academics and students from various backgrounds, and interested in investigat-
ing and preventing crime, will benefit from this book. It discusses new and future
developments in crime script analysis and the rational choice perspective.

Benoit Leclerc is a criminologist and a Senior Lecturer in the School of Criminology

and Criminal Justice at Griffith University, Brisbane, Australia. He has been
involved in research and clinical work with adolescent and adult sex offenders at
the Philippe-Pinel Institute of Montréal from 1999 to 2006 and received his PhD
from the Université of Montréal, Canada. His research interests include crime script
analysis, situational crime prevention, and sexual offending. A recently funded pro-
ject involves the study of the effectiveness of situational crime prevention to prevent
sexual offences.
Richard Wortley is Director of the Jill Dando Institute of Security and Crime
Science at University College London. He has published widely in the areas of
situational crime prevention, corrections, and sexual offending. He has recent
books entitled Situational Prison Control: Crime prevention in correctional institu-
tions (Cambridge University Press), Situational Prevention of Child Sexual Abuse
(co-edited with Stephen Smallbone—Criminal Justice Press), Environmental
Criminology and Crime Analysis (co-edited with Lorraine Mazerolle—Willan
Publishing), and Preventing Child Sexual Abuse (co-written with Stephen Smallbone
and William Marshall—Willan Publishing).
Crime Science Series
Edited by Richard Wortley, UCL
Crime science is a new way of thinking about and responding to the problem of
crime in society. The distinctive nature of crime science is captured in the name.
First, crime science is about crime. Instead of the usual focus in criminology on
the characteristics of the criminal offender, crime science is concerned with the
characteristics of the criminal event. The analysis shifts from the distant causes of
criminality—biological makeup, upbringing, social disadvantage and the like—to
the near causes of crime. Crime scientists are interested in why, where, when and
how particular crimes occur. They examine trends and patterns in crime in order
to devise immediate and practical strategies to disrupt these patterns.
Second, crime science is about science. Many traditional responses to crime
control are unsystematic, reactive and populist, too often based on untested
assumptions about what works. In contrast, crime science advocates an evidence-
based, problem-solving approach to crime control. Adopting the scientific method,
crime scientists collect data on crime, generate hypotheses about observed crime
trends, devise interventions to respond to crime problems, and test the adequacy
of those interventions.
Crime science is utilitarian in its orientation and multidisciplinary in its
foundations. Crime scientists actively engage with front-line criminal justice prac-
titioners to reduce crime by making it more difficult for individuals to offend, and
making it more likely that they will be detected if they do offend. To achieve these
objectives, crime science draws on disciplines from both the social and physical
sciences, including criminology, sociology, psychology, geography, economics,
architecture, industrial design, epidemiology, computer science, mathematics,
engineering, and biology.

1 Superhighway Robbery
Graeme R. Newman and Ronald V. Clarke

2 Crime Reduction and Problem-oriented Policing

Edited by Karen Bullock and Nick Tilley

3 Crime Science
New approaches to preventing and detecting crime
Edited by Melissa J. Smith and Nick Tilley

4 Problem-oriented Policing and Partnerships

Implementing an evidence-based approach to crime reduction
Karen Bullock, Rosie Erol and Nick Tilley

5 Preventing Child Sexual Abuse

Stephen Smallbone, William L. Marshall and Richard Wortley

6 Environmental Criminology and Crime Analysis

Edited by Richard Wortley and Lorraine Mazerolle
7 Raising the Bar
Preventing aggression in and around bars, pubs and clubs
Kathryn Graham and Ross Homel

8 Situational Prevention of Organised Crimes

Edited by Karen Bullock, Ronald V. Clarke and Nick Tilley

9 Psychological Criminology
An integrative approach
Richard Wortley

10 The Reasoning Criminologist

Essays in honour of Ronald V. Clarke
Edited by Nick Tilley and Graham Farrell

11 Patterns, Prevention and Geometry of Crime

Edited by Martin A. Andresen and J. Bryan Kinney

12 Evolution and Crime

Jason Roach and Ken Pease

13 Cognition and Crime

Offender decision making and script analyses
Edited by Benoit Leclerc and Richard Wortley
This page intentionally left blank
Cognition and Crime
Offender decision making and
script analyses

Edited by Benoit Leclerc and

Richard Wortley
First published 2014
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
Simultaneously published in the USA and Canada
by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2014 Benoit Leclerc and Richard Wortley, selection and editorial material,
individual chapters, the contributors.
The right of Benoit Leclerc and Richard Wortley to be identified as author of
this work has been asserted by them in accordance with sections 77 and 78 of the
Copyright, Designs and Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in writing from the
British Library Cataloguing in Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record has been requested for this book

ISBN: 978-0-415-68860-4 (hbk)

ISBN: 978-0-203-08348-2 (ebk)

Typeset in Times by
GreenGate Publishing Services, Tonbridge, Kent

List of figures ix
List of tables x
Notes on contributors xii
Foreword by Derek B. Cornish xvii
Acknowledgments xxii
List of abbreviations xxiii

1 The Reasoning Criminal: twenty-five years on 1


2 What are violent offenders thinking? 12


3 How house burglars decide on targets: a computer-based

scenario approach 26

4 The risks and rewards of motor vehicle theft: implications

for criminal persistence 48

5 The rational choice perspective and the phenomenon

of stalking: an examination of sex differences in behaviours,
rationales, situational precipitators and feelings 70

6 Interpersonal scripts and victim reaction in child sexual abuse:

a quantitative analysis of the offender–victim interchange 101
viii Contents
7 Drug dealing: Amsterdam’s Red Light District 120

8 Human trafficking for sexual exploitation in Italy 140


9 Script analysis of corruption in public procurement 164


10 Cigarette smuggling and terrorism financing: a script approach 186


11 Script analysis of the transnational illegal market in

endangered species: dream and reality 209

12 New developments in script analysis for situational crime

prevention: moving beyond offender scripts 221

13 Rational choice and offender decision making: lessons from

the cognitive sciences 237

Index 253

3.1 Age by estimated number of B&Es for the 96 subjects 35

3.2 Total number of cues selected for each scenario 36
3.3 Total number of cues selected over the 20 scenarios by the total
estimated number of burglaries for each participant 38
3.4 Decision tree when cue 6 (inside information) is selected first
and reveals valuable goods in the house 42
6.1 Approaches employed to investigate crime-commission processes
in adult–child sexual abuse 104
7.1 General drug dealing script and track level script for street dealing
in Amsterdam’s Red Light District 125
8.1 Crime script case study 1: from Eastern Europe to Italy 149
8.2 Crime script case study 2: from Nigeria to Italy 151
9.1 (KS1m) moves in KS1 170
9.2 (KS2m) moves in KS2 173
9.3 (KS3m) moves in KS3 174
9.4 The script of CPP in works contracts 177

2.1 Examples of different methods of, and attitudes toward harm 15

3.1 List of cues and their alternatives 32
3.2 Overall number of times a cue was selected 37
3.3 Cues selected first, number and proportion, cues arranged in
descending order 39
3.4 The ten highest mean final rating scenarios and whether they had
the attractive alternative for the five most chosen cues 40
3.5 The ten lowest mean final rating scenarios and whether they had
the deterrent alternative for the four most chosen cues 41
5.1 Stalking behaviours in a convenience sample of relational stalkers
in Queensland Australia 82
5.2 Reasons for perpetrating stalking behaviours in a convenience
sample of relational stalkers in Queensland Australia 83
5.3 Feelings leading to stalking behaviours in a convenience sample
of relational stalkers in Queensland Australia 85
5.4 Violent behaviours in a convenience sample of relational stalkers
in Queensland Australia 86
5.5 Reasons for using violence in a convenience sample of violent
relational stalkers in Queensland Australia 87
5.6 Feelings leading to violent behaviours in a convenience sample
of violent relational stalkers in Queensland Australia 89
6.1 Descriptive statistics of variables in multivariate analyses 106
6.2 Mixed-effects linear regression models of the intrusiveness of
offender sexual behaviors in child sexual abuse 110
6.3 Mixed-effects linear regression models of logged intrusiveness of
victim sexual behaviors in child sexual abuse 111
6.4 Mixed-effects logistic regression models of interpersonal script stages 113
8.1 Proposed situational crime prevention techniques 158
9.1 CPP—key stages and moves 169
9.2 The facilitators of CPP in work contracts 178
10.1 Application of script and situational controls 193
12.1 Script scenes used in crime script analysis 223
Tables xi
12.2 The application of script analysis to the main actors of crime events 225
12.3 A hypothetical victim script 226
12.4 A hypothetical guardian script 228

Benoit Leclerc is a criminologist and a Senior Lecturer in the School of
Criminology and Criminal Justice at Griffith University, Brisbane, Australia.
He has been involved in research and clinical work with adolescent and adult
sex offenders at the Philippe-Pinel Institute of Montréal from 1999 to 2006
and received his PhD from the Université of Montréal, Canada. His research
interests include crime script analysis, situational crime prevention, and sexual
offending. He is the lead investigator of a major funded research project on
the effectiveness of situational crime prevention to prevent sexual offences.
Recent publications include articles in British Journal of Criminology,
Criminal Justice and Behavior, Criminology, Journal of Interpersonal
Violence, Journal of Research in Crime and Delinquency, and Sexual Abuse:
A Journal of Research and Treatment.
Richard Wortley is Director of the Jill Dando Institute of Crime Science at University
College London. He has published widely in the areas of situational crime preven-
tion, corrections and sexual offending. He has recent books entitled Situational
Prison Control: Crime Prevention in Correctional Institutions (Cambridge
University Press), Situational Prevention of Child Sexual Abuse (co-edited with
Stephen Smallbone—Criminal Justice Press), Environmental Criminology and
Crime Analysis (co-edited with Lorraine Mazerolle—Willan Publishing), and
Preventing Child Sexual Abuse (co-written with Stephen Smallbone and William
Marshall—Willan Publishing). He has been involved in a number of major funded
research projects and consultancies in the areas of official misconduct in prison,
whistleblowing, child sexual abuse and intimate partner homicide.

Other contributors
Wim Bernasco is a Senior Researcher at the Netherlands Institute for the Study of
Crime & Law Enforcement (NSCR) and endowed professor at the Department
of Spatial Economics of the Faculty of Economics and Business Administration
(FEWEB) of VU University Amsterdam. His current research includes studies
of criminal target choice, crime displacement, geographic offender profiling,
and repeat and near-repeat victimization.
Michael Cherbonneau is a doctoral student in the Criminology Program at the
University of Texas at Dallas. His primary research interest focuses on the situ-
ational and interactional influences on participation in street crime. His current
research concerns the decision-making strategies of currently active auto
thieves. His work on the subject appears in the British Journal of Criminology,
Journal of Research in Crime and Delinquency and Justice Quarterly.
Steven Chermak is a professor at Michigan State University. His research inter-
ests include domestic terrorism, media coverage of crime and terrorism, and
intelligence-led policing. His current work focuses on understanding vio-
lent acts by political extremists, the organizational characteristics of violent
extremist groups, lone wolf terrorism, funding of terrorist organizations, and
attacks on law enforcement by political extremists. He is currently involved
in several funded projects to better understand the characteristics of terrorism
plots, social networks of financial offenders, and information sharing among
intelligence analysts.
Ronald V. Clarke is University Professor at Rutgers and Visiting Professor at
the Jill Dando Institute, UCL. He worked for 15 years in the Home Office
and was head of the Research and Planning Unit from 1982–84. While at
the Home Office he helped to develop situational crime prevention and with
Derek Cornish he formulated the rational choice perspective. He is author or
joint author of some 250 publications including Designing out Crime (HMSO,
1980), The Reasoning Criminal (Springer-Verlag, 1986), Superhighway
Robbery: Preventing e-commerce crime (Willan Publishing, 2003), Crime
Analysis for Problem Solvers (US Dept of Justice, 2005) and Outsmarting the
Terrorists (Praeger, 2006). His current research interest is wildlife crime.
Heith Copes is an Associate Professor in the Department of Justice Sciences at
the University of Alabama at Birmingham. His primary research explores the
criminal decision-making process of both property and violent offenders using
qualitative methods. He has received funding from the National Institute of
Justice to interview convicted identity thieves about their lives and crimes. His
research on criminal decision making appears in British Journal of Criminology,
European Journal of Criminology, Justice Quarterly, Criminology and Public
Policy, and Social Problems.
Richard B. Felson is Professor of Crime, Law, and Justice and Sociology at
Pennsylvania State University. He is interested in race, regional, and national
differences in violence, the role of situational factors, and differences between
violence involving women and violence involving men. His books include
Violence, Aggression, and Coercive Actions (with James Tedeschi) and
Violence and Gender Reexamined.
Joshua D. Freilich is the Director of the Criminal Justice PhD program and a
member of the Criminal Justice Department at John Jay College, the City
University of New York. He is a lead investigator for the National Consortium
xiv Contributors
for the Study of Terrorism and Responses to Terrorism (START), a Center for
Excellence of the US Department of Homeland Security (DHS), and a mem-
ber of its Executive Committee. He is also a member of the Global Terrorism
Database’s (GTD) Advisory Board. Freilich’s research focuses on causes of
and responses to terrorism as well as criminological theory. Since 2006 he has
directed the open-source Extremist Crime Database (ECDB) that tracks violent
and financial crimes committed by political extremist inside the United States.
His research has been funded by DHS directly as well as through START.
Luca Giommoni is a PhD candidate in Criminology at the Università Cattolica
del Sacro Cuore of Milan and a researcher at Transcrime (
His research interests include organized crime and drug trafficking.
Alexandra Hiropoulos is a doctoral student in Criminal Justice at the CUNY
Graduate Center, focusing her doctoral research on xenophobic violence in
South Africa. She has studied in Greece, the UK and the US and has a back-
ground in psychology and policing, having previously worked for the Serious
Crime Analysis Section of the National Policing Improvement Agency of the
UK. Her research interests include domestic violence, rape, sex work and sex
trafficking, migration, and xenophobia.
Ross Homel is Foundation Professor of Criminology and Criminal Justice at
Griffith University in Brisbane, Australia, and Director of the Griffith Institute
for Social and Behavioural Research. He has published nine books on crime and
violence, as well as more than one hundred peer-reviewed papers and numer-
ous high impact government reports. He is an Officer in the Order of Australia
(AO) “for service to education, particularly in the field of criminology, through
research into the causes of crime, early intervention and prevention methods,”
and has also been recognized as a “Queensland Great.” In 2009 he was short-
listed for Australian of the Year.
Scott Jacques is an Assistant Professor of Criminal Justice and Criminology
at Georgia State University. His work focuses on crimes against and social
control by drug dealers, and on theorizing criminological method. Currently,
he is analyzing data from a study of street dealers, coffee shops, and bars in
Amsterdam. His work has been published in top ranked criminology and crim-
inal justice journals, including Criminology, Journal of Research in Crime and
Delinquency, Justice Quarterly, and Crime & Delinquency.
Stuart Macintyre graduated with his PhD from Griffith University in Criminology
and Criminal Justice in 2002. He is the Assistant Director, Risk Mitigation,
Ethical Standards Department, Victoria Police, Australia.
Marina Mancuso is a PhD candidate in Criminology at the Università Cattolica
del Sacro Cuore of Milan and a researcher at Transcrime (
Her research interests include human trafficking with particular reference to
human trafficking for sexual exploitation.
Contributors xv
William D. Moreto is a doctoral candidate at the Rutgers University School of
Criminal Justice. His research interests include environmental criminology,
situational crime prevention, spatiotemporal analysis, wildlife crime and polic-
ing. He is currently working with the Uganda Wildlife Authority to assess law
enforcement and illegal activities within protected areas in Uganda.
Graeme R. Newman is distinguished teaching professor at the School of
Criminal Justice, University at Albany and Associate Director of the Center
for Problem-Oriented Policing. Among the recent books he has written or
edited are: Super Highway Robbery with Ronald V. Clarke (Willan, 2003),
Outsmarting the Terrorists with Ronald V. Clarke (PSI International, 2006),
Crime and Immigration with Joshua Freilich (Ashgate, 2006), Designing out
Crime from Products and Systems with Ronald V. Clarke (Willan, 2006), a
new translation of Cesare Beccaria’s On Crimes and Punishments with Pietro
Marongiu (Transaction Press, 2009) and Crime and Punishment around the
World in four volumes (ABC-CLIO, 2010).
Ernesto U. Savona is Professor of Criminology at Università Cattolica in Milan and
Director of TRANSCRIME (Joint Research Centre on Transnational Crime). He
is Editor in Chief of the European Journal on Criminal Policy and Research,
published by Springer, and member of the Global Council on Organised Crime
of the World Economic Forum. His recent contribution in the field is “Infiltration
by Italian organised crime (Mafia, N’drangheta and Camorra) in the public con-
struction industry,” in K. Bullock, R. V. Clarke, N. Tilley, Situational Prevention
of Organised Crimes, Willan Publishing (UK), 2010.
Stephen Smallbone is a psychologist and Professor in the School of Criminology
and Criminal Justice at Griffith University, and an Australian Research
Council Future Fellow. His recent publications include the books Situational
Prevention of Child Sexual Abuse (Wortley and Smallbone, 2006), Preventing
Child Sexual Abuse: Evidence, policy and practice (Smallbone, Marshall, and
Wortley, 2008), and Internet Child Pornography: Causes, investigation, and
prevention (Wortley and Smallbone, 2012).
Carleen M. Thompson is a postdoctoral research fellow and lecturer in the
School of Criminology and Criminal Justice, Griffith University (Brisbane,
Australia). Dr. Thompson’s research sits within a criminology and psychology
framework whereby her principal research interests include stalking, interper-
sonal violence, and risk assessment in the justice system. Her doctoral thesis
investigated the etiology and escalation of interpersonal violence in stalking.
She has also written scholarly articles and book chapters on stalking, juror
decision making, risk assessment, intimate partner violence, violence in pris-
ons, elderly sexual assault, and work-related self-efficacy.
Marco Zanella is an Italian lawyer, and in March 2012 he was awarded a PhD
in Criminology (Catholic University of Milan, Italy). In October 2008 he
was awarded an MA in Public and Criminal Law by the University of Trento
xvi Contributors
(Italy) with a thesis on extortion racketeering, while in September 2006 he was
awarded a BA in Law with a thesis on corruption in the European Union. In
2008 he was a research assistant at the John Jay College of Criminal Justice,
City University of New York and in 2007 a visiting student at the Faculty of
Law of the Erasmus Universiteit Rotterdam (EUR). In 2006 he served as an
intern at the United Nations Office on Drugs and Crime (UNODC) within
the Terrorism Prevention Branch (TPB)—Division for Treaty Affairs (DTA).
Between 2006 and 2009 he cooperated at TRANSCRIME—Joint Research
Centre on Transnational Crime (University of Trento and Catholic University
of Milan).

Upwards of 25 years lies between the emergence of the rational choice perspective
in 1985 and the preparation of this volume; and the theory, research, and practice
in relation to the evolution of situational crime prevention has an even longer his-
tory. This volume, compiled by two well-known researchers in the field, reviews
some of the recent progress that has been made in these two areas. Its editors and
contributors do three things: first, they provide some historical background to the
development of the rational choice perspective and of situational crime prevention.
Second, since both theory and practice are constantly evolving and adapting to meet
new challenges, they highlight a number of recent areas in which new contributions
are being made. And third, they suggest some directions for future development.
The range and interests of the international contributors to this volume give
some indication of how far theory and practice have come in this area and how
widely its ideas have spread. Situational crime prevention, in particular, is flexing
its muscles as it takes on some of the major problems of the day. It now covers
the sexual abuse of children, human trafficking, violent stalking, the financing of
terrorism, drug dealing, corruption in public procurement, and the protection of
endangered species. Nevertheless, it still maintains its vigorous and longstanding
interest in serious property crimes such as burglary and auto theft. The contents
also illustrate something of the variety of methods that are being brought to
bear on the study of offender decision making and situational prevention. These
include theoretical analyses, computer-based crime scenarios, empirical research
using police and court papers and other official documentation, and interviews
with offenders and non-participant observation of their activities.
As well as providing examples of recent activity in the field of situational pre-
vention, the contributors to this volume also offer some signposts for the future,
not only for the development of situational prevention, but also for the rational
choice perspective itself. It may be useful to highlight some of the potential direc-
tions for development that these chapters raise.

Future development of the rational choice perspective

The rational choice perspective was developed in order to draw attention to
the influence of the current environment on the occurrence of crimes. Its basic
xviii Foreword
assumption is that an overemphasis on the role of criminality in offending has
obscured the powerful role played by opportunity. As a result, criminology
has neglected the many ways in which the environment can be managed in
order to prevent or reduce offending. The perspective employs the concept of
the offender as a rational decision maker to explore the ways in which current
situational factors influence criminal decision making in relation to specific
criminal activities, both in relation to crime events themselves, and in relation
to the criminal-involvement processes of initiation, habituation, and desistance
over longer timescales.
The rational choice perspective was designed to be used as a heuristic device
rather than a theory: that is, to offer a new framework within which to think about
the implications of a more situational approach for crime prevention efforts. Such
frameworks are more like metatheories than theories: part empirically based and
part conjecture; part descriptive and part prescriptive; part explanatory and part
rhetorical. As their objective is to promote new ways of thinking and doing, they
may assume what cannot yet be proven, simplify in order to persuade, and omit in
order to clarify. In other words, they are always to some extent “useful fictions.”
There is a further qualification in the case of the rational choice perspective: above
all, it is designed for getting things done, and its continuing value depends upon
its ability to do so. This depends on how effectively it communicates its underly-
ing rationale, and how well it supports situational crime prevention in terms of
systematizing, extending, and defending its practice.
Given these priorities, specific aspects of the perspective have been devel-
oped only when considered necessary. Tinkering with, or elaborating on, the
framework has tended to be resisted unless they were thought likely to further
its mission. The intention was for the perspective to stay “good enough” until
it needed to become better. The ability to be and do better rests, however, on
the continued validity—or at least plausibility—of the basic assumptions used to
guide the development of the original framework, and on the framework’s ability
to respond to new challenges. Looking over the past 25 years or so, the perspec-
tive has been quite resilient. While the overall framework has remained relatively
unchanged, parts have been elaborated as required. Examples are:

● the development of the concept of the choice-structuring properties of crimes

in relation to debates over the likelihood of crime displacement;
● the perspective’s contribution to the growing sophistication over time in the
types of situational techniques identified and developed, and in the ways in
which they are classified and used;
● the extension of the crime-event model to lay out the crime-commission pro-
cesses of simple and more complex crimes in greater detail;
● the use of this information to guide the closer targeting of situational tech-
niques to different stages of the crime-commission process;
● the application of the presumption of rationality to argue for the relevance
of situational prevention to so-called expressive crimes, such as violent and
sexual offending; and
Foreword xix
● the flexibility of the perspective, including its ability to respond to new
ideas—such as the role of situational factors in precipitating specific crimes,
and the development of situational techniques to control such precipitators—
and make adjustments, where necessary.

As a heuristic device the rational choice perspective has the advantage of operat-
ing at a rather general level of explanation. But it is not immune to challenges
from theoretical developments or empirical research in related areas (for example,
the cognitive sciences), some of which are discussed in this volume, and others
that are as much philosophical as scientific. Such debates are vital to the future of
the perspective, but its future will also depend on the extent to which it continues
to contribute to the development of situational crime prevention.
One possible source of development may be provided by the expansion of
the perspective’s crime-event model. This not only helps to provide information
for the purposes of situational prevention, but also more fully reveals the crime
event as a process involving a series of decisions and actions extended over time
and space. In this way, appropriate emphasis can be given to the importance of
the crime commission as an integral part of the offender’s lived experience and,
over time, his or her chosen lifestyle. The vicious (or virtuous, depending on
one’s viewpoint) circle underlying the habituation model suggests that successful
commission of a specific crime may gradually induct the offender into an asso-
ciated lifestyle which in turn generates or reinforces the desires and needs that
only continued offending of that, or a very similar type, can sustain. This is well
illustrated by the chapter on car theft. Not all crimes cater to multiple motives or
support multiple aspects of a satisfactory criminal lifestyle; even for those that do,
the implications for situational prevention are not immediately clear. But it does
raise the interesting issue of whether, or to what extent, knowledge of how cer-
tain criminal lifestyles are sustained could lead to the development of strategies
directly aimed at their disruption or regulation.

The future of situational crime prevention

The crime script—an elaboration of the rational choice perspective’s event
model—is another heuristic device, using ideas borrowed from the fields of cog-
nitive science and dramaturgy. Its purpose is to provide tools for analyzing the
requirements and procedural stages of specific crimes. These include casts, roles,
props, locations, decisions, actions, and goal-directed action sequences, guided
by an overall plot—the successful commission of the crime. Many of the chapters
use the concept of the crime script as a way of providing a more systematic and
comprehensive account of how specific crimes are committed and, in doing so,
offer illustrations of directions in which the field is moving.
Complex crimes, and the specific offences that contribute to them, are dis-
cussed in many of the chapters. In these cases a “master script” is often used to
provide a synoptic view of the stages of the crime as a whole, and offer a frame-
work within which contributory scripts with their own stages may be lodged.
xx Foreword
Although master scripts show how complex crimes are organized, they do not
necessarily indicate the presence of an overall planner. They may, like simpler
crime scripts, require adjusting, extending, or breaking up into further stages in
the light of the specific crimes to which they are being applied. It is important
that adequate attention be paid to these component crimes as well as to the over-
all process, since they may well indicate crucial differences in the way that the
complex crime is being carried out in specific cases—and hence the implications
for situational prevention. Master scripts characteristically operate at a relatively
high level of generality. This provides a useful tool for organizing what often
comprises a complex sequence of component crimes. However, they do not tell
one much about the content of these contributory subscripts, some of which may
vary in important ways from one instance of a specific complex crime to another.
While master scripts may enhance analyses of specific complex crimes at a stra-
tegic level, the situational prevention techniques to be employed will be applied
at the tactical level—that is, on disrupting the component crimes specific to the
particular instance of the complex crime in question.
Complex crimes constitute only one type of linkage: that among its compo-
nent crimes. There seems, however, to be a natural tendency for links to develop
among various criminal activities, and for this interconnectedness to grow over
time. Such script elaboration and concatenation takes many forms. Think, for
example, of individual crime scripts as resembling tracks in the snow—crossing,
intersecting, joining, sometimes sharing starting-points, destinations, and way-
stations—until the links they make in passing become woven into complex webs
of criminal activity, sharing resources as they become more mutually dependent
over time.
Those involved in the study of violent and sexual offending are also using the
crime script concept in their investigation of the crime-commission strategies of
child sexual abusers. As a number of chapters point out, however, crimes involv-
ing interactions between offender and victim during the course of a crime add
a new set of complications and suggest that attention needs to be given in such
scripts to the varied outcomes of such interactions and the effects of these on the
outcome of the crime as a whole. Such research also suggests the value of investi-
gating the specific scripts developed in the interest of self-protection by potential
and past victims of particular crimes, and the scripts of others, such as the police
and place managers, who have responsibility for crime prevention.
In a more general way, the growing number of scripts now being developed
provides opportunities for researchers to borrow creatively from existing ones:
what light can knowledge of one form of trafficking throw on another? Can an
understanding of how victims are recruited for child sexual abuse inform our
understanding of the processes of recruiting men and women for suicide bombing
in any way?
As the examples above indicate, the volume speaks to the continuing vigor
of both the rational choice perspective and of situational prevention. Much of
their staying power is owed to those, past and present, who have reached simi-
lar conclusions about the role played by the environment in the occurrence and
Foreword xxi
prevention of crime events. This impetus is currently best represented by the
annual conferences of Environmental Criminology and Crime Analysis (ECCA),
an international network of researchers and practitioners from many different
backgrounds and disciplines. The future of rational choice and situational preven-
tion is very much tied up with the support of networks such as ECCA, and with
the future prospects of crime science both in the United States and in the United
Kingdom, where it flourishes at University College, London and within the Jill
Dando Institute under the direction of Richard Wortley.
Books, articles, and conference papers provide glimpses of the substratum
of people and institutions that have supported the development of the rational
choice perspective and of situational crime prevention. If one were to make a
script of the stages whereby the theory and practice of situational crime preven-
tion has become a thriving academic enterprise, it would have to be a complex
narrative. It would proceed by way of a gallimaufry of casts, props, resources,
roles, stages, locations, and actions, shaped over time by the efforts of many.
Such a script would have to portray the impetus for the original ideas, their goals
and early development, the recruitment of collaborators and the forming of ad
hoc teams (of colleagues, students, and other early adopters), the communica-
tion of ideas (by lectures, presentations, journal articles, monographs, books and
edited books, and special-interest serials), the support of other interested parties
(e.g., politicians, civil servants, the police, and place managers), the develop-
ment of networks (through meetings, special and annual conferences, and lecture
tours), and the securing of funding sources (via research grants and institutional
affiliations). Such a long and complex narrative would indeed require a mas-
ter script in order to set out its many transitions and overall trajectory. But, to
complete the story, it would also have to include the master script writer whose
tireless dedication and vision continues to re-energize its plot: the primus inter
pares, Ron Clarke.

Derek B. Cornish
Wichita, Kansas, USA
January 21, 2013

We would like to thank the following list of researchers for their assistance
and suggestions made during the review process of the contributions included
in this volume: Wim Bernasco, Rick Brown, Ron Clarke, Derek Cornish, Susan
Dennison, Michael Levi, Carlo Morselli, Graeme Newman, Ken Pease, Ernesto
U. Savona, Stephen Smallbone, Martha Smith, Nick Tilley, and Richard Wright.

AO Office of the Order of Australia

ATF Bureau of Alcohol, Tobacco, Firearms and Explosives
B&E break and enter
CA Court of Auditors
CIA Central Intelligence Agency
CITES Convention on International Trade in Endangered Species
CJC Criminal Justice Courts
CPP corruption in public procurement
DEFRA Department of Environment, Food and Rural Affairs
DHS Department of Homeland Security
DTA Division for Treaty Affairs
ECCA Environmental Criminology and Crime Analysis
ECDB Extremist Crime Database
EUR Erasmus Universiteit Rotterdam
FBI Federal Bureau of Investigation
GTD Global Terrorism Database
ICCWC International Consortium on Combating Wildlife Crime
ICE Immigration and Customs Enforcement
INS Immigration and Naturalization Services
IPV intimate partner violence
PACT Prevent All Cigarette Trafficking
RAC Regional Administrative Courts
RCP rational choice perspective
RICO rackateer influenced and corrupt organizations
SCP situational crime prevention
START Study of Terrorism and Responses to Terrorism
TIMES Transnational Illegal Market in Endangered Species
TPB Terrorism Prevention Branch
TRAFFIC Trade Records Analysis of Flora and Fauna in Commerce
UNODC United Nations Office on Drugs and Crime
This page intentionally left blank
1 The Reasoning Criminal
Twenty-five years on
Benoit Leclerc and Richard Wortley

The birth of The Reasoning Criminal

In 1986 Derek Cornish and Ron Clarke published The Reasoning Criminal, an
edited collection of (mainly) empirical studies that examined the decision mak-
ing of offenders across a range of offences. The introductory chapter, written by
Cornish and Clarke – their sole chapter in the book – sets out in just 16 pages
a radical model (by the standards of conventional criminology) for viewing the
causes of crime. Rather than focus on the usual historical factors – the offender’s
genes, his/her family life, schooling, the influence of peers, and the like – assumed
to create the offender’s criminality they were concerned with proximal factors –
the offender’s perceptions of the potential crime scene – that led individuals to
commit crime. Rather than view the offender as a passive respondent to social or
dispositional forces outside of his/her control, they proposed a model in which
the offender was an active, decision-making agent, purposively pursuing criminal
goals. And rather than portray the pursuit of these goals as irrational, inexplicable
and senseless, they argued that offenders rationally sought to benefit themselves
from their criminal acts and were motivated to fulfil much the same needs that
drive us all.
The ideas that Cornish and Clarke were to fashion into the rational choice per-
spective had their genesis almost 20 years earlier. In 1967 Ron Clarke published
a research paper on absconding from an approved school (essentially a residen-
tial school for juvenile delinquents). Clarke noted that the best predictors of
absconding were ‘certain environmental influences’ – hours of daylight, features
of the school’s regime, and the distance home – and not any ‘deep psychological
aspects’ associated with the absconders. Despite appearing in the highly ranked
British Journal of Criminology, Clarke’s paper created little interest at the time
and even today according to Google Scholar has just six citations (one of which
is by Clarke himself). But it is worth noting in passing that this paper predated
Walter Mischel’s (1968) Personality and Assessment (which sets out the classic
arguments against dispositional models of behaviour), C. Ray Jeffrey’s (1971)
Crime Prevention Through Environmental Design, and Oscar Newman’s (1972)
Defensible Space: Crime Prevention Through Urban Design, making Clarke a
true pioneer of situational explanations of criminal behaviour.
2 Benoit Leclerc and Richard Wortley
At this point Clarke provided little in the way of theoretical analysis of why the
environment was so important in crime or how it interacted with characteristics of
the individual offender. The absconding data were findings looking for a theory.
Further situationally focused papers followed (Clarke and Martin, 1971; Sinclair
and Clarke, 1973) and by 1975 the various environmental factors conducive to
crime had become distilled into a single concept – opportunity (Mayhew, Clarke,
Sturman and Hough, 1976). By now Clarke and colleagues had dipped into the
psychological literature and they cited a limited range of psychological sources
(including Mischel) supporting the role of opportunity in crime. The first publica-
tion by Clarke that explicitly uses the term situational crime prevention appears
to be his 1980 paper ‘Situational’ Crime Prevention: Theory and Practice. In
this paper there is a brief section entitled ‘Preventative implications of a “choice”
model’, but there is really just one sentence that hints at the developments in the
rational choice perspective that were to come:

Since offenders’ perceptions of the risks and rewards attaching to different

forms of crime cannot be safely inferred from studies of the distribution of
offences, there might be additional preventative benefits if research of this
kind were more frequently completed by interviews with offenders.
(Clarke 1980: 139)

The classic features of the choice model were not fully fleshed out until the mid
1980s when they were first described in detail in an article co-authored with Derek
Cornish (Clarke and Cornish, 1985), who had developed a similar approach in
relation to gambling behaviour (Cornish, 1978). Setting the scene in the 1985
paper is a wide-ranging review of research from sociology, criminology, eco-
nomics and cognitive psychology supporting their contention that the offender’s
behaviour is best understood as the product of external situational forces, not
of individual pathology. From the array of perspectives reviewed, Clarke and
Cornish settled on a choice model as a way of capturing the goal-directed nature
of criminal conduct and of providing an account of the mechanisms by which
environmental forces translate into deviant behaviour. They also made the deci-
sion to label their choice model ‘rational’.
In the same year Clarke organised a symposium comprising an international
group of researchers interested in understanding crime from the offender’s per-
spective with the aim of taking forward the decision-making approach as an
explanation of crime. From this symposium, the idea of an edited collection was
proposed and The Reasoning Criminal was published in the following year with
an abridged version of the 1985 article as the introductory chapter.
The basic tenets of the rational choice perspective will be well known to most
readers of this book and require no more than a brief recapitulation here (see
Cornish and Clarke, 2008, for the most recent and comprehensive outline of the
approach). Instead of focusing on the criminal and his/her propensity to commit
crimes, the rational choice perspective emphasises crime events and the situa-
tional factors that facilitate the perpetration of crime. According to the rational
The Reasoning Criminal: twenty-five years on 3
choice perspective, the offender is not at the mercy of uncontrollable drives but
interacts with the immediate environment, which shapes his or her behaviours.
Within this framework, the offender is portrayed as a thinking individual who
is capable of making decisions and who seeks to gain benefits from committing
crimes. These decisions are crime-specific. The considerations involved in the
commission of a burglary, for example, are very different from those involved in
the commission of rape. Crime occurs when the perceived benefits of offending
outweigh the perceived costs and in this sense the decisions to commit crime are
rational. It is understood, however, that the offender’s capacity to make rational
choices is not perfect, but is bounded, limited by cognitive biases, lack of infor-
mation, time pressures, emotional arousal, drugs or alcohol, individual values,
and a range of other factors. The utility of an anticipated outcome, therefore, is
subjective – judged from the decision maker’s point of view – and an individual
may not always pursue a course of action that ultimately produces the greatest
objective benefits.
Cornish and Clarke were not the first to propose a choice model of offending.
The idea that offenders choose to commit crimes was among the first recognis-
able theories of criminology, proposed in the utilitarian deterrence models of
Bentham and Beccaria in the late eighteenth century. Nor did Cornish and Clarke
invent the term rational choice, which has a long history in economics, political
science and psychology. Their contribution was to adapt the concept of rational
choice to provide a way of analysing crime events. Unlike the classic deterrence
model, which largely locates the costs of offending in the criminal justice system,
Cornish and Clarke recognised that the cost–benefit analysis can take place at the
crime scene using information from the immediate environment. And by empha-
sising the bounded nature of rationality and the importance of offence-specific
decision making, Cornish and Clarke’s version of rational choice diverges sig-
nificantly from the strict expected-utility models originally employed in the social
sciences. Cornish and Clarke’s was first and foremost a pragmatic model. From
the start they were concerned with the practical implications of rational choice,
rather than with the precise details of offender decision making. Indeed, while the
term rational choice theory is common in the literature as a description of their
approach, they regarded rational choice as a heuristic rather than a theory and
have consistently preferred the terms rational choice model or perspective. Their
primary aim was to develop a ‘good enough theory’ to provide a framework for
developing crime control strategies (Cornish and Clarke, 2008: 38).

The impact of The Reasoning Criminal

Rational choice perspective has in equal measure been one of the most influential
and criticised criminological models to emerge in the latter quarter of the twenti-
eth century. Let us deal first with the criticisms.
Criticisms of the rational choice perspective have been mounted both on theo-
retical and ideological grounds. Theoretically, the central target for critics is the
presumption of offender rationality (Hayward, 2007; Trasler, 1986). The criticism
4 Benoit Leclerc and Richard Wortley
takes a number of forms. Some crimes may be portrayed as senseless – ‘random’
acts of violence or vandalism, for example – and seemingly devoid of any rational
benefits. It may also be pointed out that the decisions made by offenders are in
fact often self-defeating, ultimately causing him/her more pain than pleasure (for
example, by leading to imprisonment). Or the power of psychological forces to
override rationality might be emphasised. Crimes such as child sexual abuse are
viewed as the product of deep-seated psychological disorder, while temporary
emotional states – rage, sexual arousal, depression – may result in impulsive,
‘expressive’ crimes.
Cornish and Clarke (2008) and others (Farrell, 2010; Tedeschi and Felson,
1994) have responded to this criticism in detail and we won’t rehearse the rebut-
tals here. Suffice to say it seems clear that in many cases critics are responding
to what they assume the rational choice perspective is about rather than to what
Cornish and Clarke have actually said. The term ‘rational’ is interpreted literally,
with Cornish and Clarke’s qualifications concerning the bounded nature of ration-
ality (which were outlined in detail in the original 1985 article) largely ignored.
Hayward (2007: 233), for example, complains that ‘not all actors are economi-
cally self-interested’, a conclusion with which most rational choice adherents
would readily agree. Violence can be used as an instrumental means to avenge
an insult (Luckenbill, 1977) or to gain compliance of the victim (Tedeschi and
Felson, 1994), and child sexual abuse can be perpetrated for sexual gratification,
intimacy and revenge (Mann and Hollin, 2007).
Ideologically, rational choice perspective has been painted as a politically con-
servative paradigm, labelled ‘right realism’ (Matthews, 1987) and ‘administrative
criminology’ (Young, 1994), although few adherents to rational choice would
apply either of these terms to their work. It is seen to be guilty both of sins of
omission and sins of commission. Its sins of omission are to ignore the socio-
cultural (‘root’) causes of crime and in doing so to fail to support a social reform
agenda that is the staple of sociological criminology. Its sins of commission are to
instead propose a model of crime control that (so it is argued) can be used by the
state to suppress the behaviour of citizens and that will ultimately contribute to a
divided and unfair society.
Again there have been robust defences offered in the face of such criticisms
that we need not elaborate upon (see Cornish and Clarke, 2008; Wortley, 2010).
Again, too, one suspects that the criticisms relate more to an image of rational
choice than to its substance or intent. It is interesting to note that in the theoretical
review used to justify the adoption of a rational choice perspective, Clarke and
Cornish (1985) drew heavily on the sociological and criminological literature,
and in particular, the sociology of deviance. They saw the rational choice per-
spective as being consistent with a view of crime that ‘explicitly emphasized the
cultural relativity of definitions of delinquency, the relationship between social
control and the distribution of political and economic power in society, and the
need to appreciate the meaning of deviance from the actor’s perspective’ (p. 150).
Moreover, in the sociological tradition Clarke and Cornish saw offenders as prod-
ucts of their environment and ‘explicitly rejected deterministic and pathological
The Reasoning Criminal: twenty-five years on 5
explanations of crime’ (p. 150). It seems not a little perverse that among the criti-
cisms now levelled against the rational choice perspective is that it normalises
crime and does not pathologise offenders (Garland, 1999; Hayward, 2007).
We turn to the influence of the rational choice perspective. Despite the criti-
cisms from within criminology, rational choice perspective has become (almost)
mainstream. Indeed, the level of criticism it has attracted is a mark of its influence.
(Irrelevant challenges to orthodoxy can be safely ignored.) Most criminology
textbooks will now devote a chapter (or at a minimum a section within a chapter)
to it. The Reasoning Criminal has over 1,000 citations listed in Google scholar, a
figure that does not include citations to individual chapters within the edited col-
lection. And if you want to buy a second-hand copy of The Reasoning Criminal,
you will need to pay more than $200 on Amazon.
There is an accumulating catalogue of research revealing that elements of
offender decision making can be found in most forms of crime. Burglary has prob-
ably been the most studied crime in terms of decision making to date (e.g. Bennett
and Wright, 1984; Cromwell, Olson and Avary, 1991; Nee and Meenaghan, 2006;
Rengert and Walsilchick, 2000; Walsh, 1986; Wright and Decker, 1994). Other
crimes commonly examined include car theft (e.g. Cherbonneau and Copes,
2006; Webb and Laycock, 1992), shoplifting (Carroll and Weaver, 1986) and
crimes that may include some elements of violence such as armed robbery (e.g.
Feeney, 1986; Petrosino and Brensilber 2003; Wright and Decker, 1997), drug
selling (Jacobs, 1996) and crimes against passengers and employees in public
transport (Smith and Cornish, 2006). But the scope of crimes examined within
the rational choice perspective is steadily widening and increasingly includes
so-called ‘irrational’ crimes such as sexual offending (Beauregard and Leclerc,
2007) and terrorism (Clarke and Newman, 2006). If an individual who commits
an armed robbery, a sexual offence or an act of terrorism can be treated as a
reasoning criminal then any crime can be studied within a rational choice frame-
work. Perhaps more importantly, the rational choice perspective, as one of the
conceptual underpinnings for situational crime prevention, has been influential in
shaping crime prevention practice and policy. Few criminological approaches can
make this claim and it is this influence on policy that has earned it the somewhat
derisive label of administrative criminology from other criminologists. The influ-
ence has been particularly strong in the United Kingdom and it is no coincidence
that Clarke originally developed a rational choice perspective and situational
crime prevention during his tenure with the British Home Office Research and
Planning Unit between 1968 and 1984. The rational choice perspective makes
situational crime prevention principles accessible to policymakers and criminal
justice professionals such as police. Situational crime prevention has proved to
be successful in reducing opportunities to commit many forms of crime (Clarke,
1997) and more and more, the perspective has been shown to be applicable to all
forms of crimes, including terrorism (Freilich and Newman, 2009), child sexual
abuse (Wortley and Smallbone, 2006), cyber crime (McNally and Newman, 2007;
Newman and Clarke, 2003), crowd violence (Madensen and Knutsson, 2011) and
organised crimes (Bullock, Clarke and Tilley, 2010).
6 Benoit Leclerc and Richard Wortley
The emergence of crime scripts
An important development of the rational choice perspective was that of crime
scripts. The idea of crime as a process rather than a single event was raised in
Clarke and Cornish’s (1985) original paper. Later, Cornish (1994) borrowed the
concept of scripts from cognitive psychology to study crime-commission pro-
cesses from beginning to end. In cognitive and social psychology, a script is
referred to as an event schema, which is a knowledge structure that organises
the sequence of human actions to adopt in a particular context. Everyone has a
repertoire of behavioural sequences stored in their memory that is ready to be
activated in the appropriate circumstances. Applied to crime, a script represents
the complete sequence of actions adopted prior to, during, and following the com-
mission of a particular crime. At each of these stages the offender is required to
make rational choices about how to proceed to the next stage. With successful
repetition, the script becomes encoded as the default behavioural sequence, and is
enacted automatically to facilitate the flow of actions from stage to stage, without
much forethought or conscious attention, until the crime is completed.
There are two interlinked strengths associated with the performance of crime
script analysis. First, crime script analysis offers a template for examining the
crime-commission process of any crime systematically and in as much detail as
existing data allow (Cornish, 1994; 1998). This template gives new opportunities
to environmental criminologists and crime analysts for understanding the step-by-
step process that offenders have to go through in the commission of any crimes.
Indeed, it directly targets how a crime is committed in detail and provides a way
for criminologists to put themselves in the offender’s shoes to ‘get closer to the
crime scene’. Second, by providing a template to capture all of the stages of the
crime-commission process, script analysis assists criminologists to identify addi-
tional intervention points for situational prevention. In fact, each stage offers an
opportunity for intervention. Therefore, situational prevention measures can be
designed and mapped onto each stage to disrupt the crime-commission process
from being carried out by the offender.
At first, the concept of crime script was intended to assist in the development
of the rational choice and routine activity perspectives in criminology (Cornish,
1998). However, crime script analysis is more often employed as an investigative
tool. In the last ten years or so, the utility of crime script analysis for examining the
modus operandi of crimes has emerged in the literature. Crimes such as car theft
(Tremblay, Talon and Hurley, 2001) and cheque forgery (Lacoste and Tremblay,
2003) have been analysed under the lens of script analysis. More importantly, the
role of script analysis for situational crime prevention is now obvious due to a
recent increase of interest in adopting scripts for prevention purposes. Crime scripts
have been applied to various crimes for situational prevention purposes such as ter-
rorism (Clarke and Newman, 2006), child sex trafficking (Brayley, Cockbain and
Laycock, 2011) drug manufacturing (Chiu, Leclerc and Townsley, 2011), infiltra-
tion of Italian organised crime groups in the public construction industry (Savona,
2010), and child sexual abuse (Leclerc, Wortley and Smallbone, 2011).
The Reasoning Criminal: twenty-five years on 7
The aim and structure of this book
The idea for this book was conceived in 2010 and was prompted by the (then) upcom-
ing twenty-fifth anniversary of The Reasoning Criminal. Our idea was to release
the book in 2011 to coincide with the anniversary. Academics being academics we
missed our deadline. Nevertheless, the book is intended as homage to The Reasoning
Criminal, and if we cannot strictly claim to be celebrating the twenty-fifth anniver-
sary we can say that we are marking the beginning of the next 25 years.
In the 12 chapters that follow, our aim is to showcase developments in the
rational choice perspective since The Reasoning Criminal. While we have not
divided the book into sections, conceptually the chapters can be grouped into
three categories.
The first group of chapters examines offender decision making. In Chapter 2,
Richard Felson discusses the offender’s decision making during a violent crime.
Felson challenges the concept of ‘reactive’ violence whereby aggression is seen
to be a reflexive response to negative affect. Reviewing the research on violent
offending, he argues that even when offenders commit violent acts while in a state
of high emotional arousal they are nevertheless making a decision and exercis-
ing some level of rationality. Chapter 3, by Ross Homel, Stuart Macintyre and
Richard Wortley, investigates the decision-making process of experienced bur-
glars in selecting target dwellings. By using burglary scenarios to explore the cues
that were most salient to these offenders, this study shows that the single most
influential cue is ‘reliable inside information’ (e.g. there was a large amount of
cash inside the house or a good alarm was present). Other influential cues include
‘dog’, ‘alarm,’ and ‘people in the street’. In Chapter 4, Heith Copes and Michael
Cherbonneau adopt the rational choice perspective to understand how auto thieves
assess the risks and rewards of crime with an emphasis on how these assessments
contribute to criminal persistence. The ethnographic data analysed in this chapter
indicate that auto thieves perceive their crimes as an ideal way to sustain their
chosen lifestyles. The potential risks associated with auto theft are also greatly
minimised by offenders. In fact, this calculation led most of these offenders on the
path of persistence rather than desistance. In Chapter 5, Carleen Thompson and
Benoit Leclerc apply the rational choice perspective and situational precipitators
to the phenomenon of stalking. This study indicates that relational stalking, and
the use of violence in the context of relational stalking, is perpetrated for quite
different reasons according to the sex of the offender. Specifically, male rela-
tional stalking is driven by the pursuit of love whereas female relational stalking
is driven by resentment.
The second group of chapters focuses on empirical analyses of crime scripts.
In Chapter 6, Benoit Leclerc, Stephen Smallbone and Richard Wortley adopt an
interpersonal script framework to examine the offender–victim interchange (i.e.
offender action–victim reaction–offender response–sexual behaviours) when sex-
ual offenders try to gain the cooperation of children for sexual contact. The main
contribution of this study lies in the fact that the severity of child sexual abuse in
terms of sexual behaviours is determined in part by the reaction of the victim and
8 Benoit Leclerc and Richard Wortley
victim characteristics. Chapter 7, by Scott Jacques and Wim Bernasco, examines
the process of street drug dealing in Amsterdam’s Red Light District. Applying
the script perspective to street drug dealing, this chapter shows that dealers must
often search for customers, solicit them, agree on when and where to meet for
trade, and travel there in order to finalise their transaction. Jacques and Bernasco
argue that each of these steps happens for rational or opportunistic reasons.
Chapter 8, by Ernesto Savona, Luca Giommoni and Marina Mancuso, investi-
gates two Italian cases of human trafficking for sexual exploitation. The first case
deals with human trafficking for sexual exploitation carried out in night clubs
involving victims from Eastern Europe. The second case deals with trafficking
in human beings for sexual exploitation carried out on the street involving vic-
tims from Nigeria. Using script analysis, the authors identify four different script
stages: recruitment, transportation, exploitation and aftermath, and suggest situ-
ational crime prevention measures to disrupt them. In Chapter 9, Marco Zanella
applies crime script analysis to cases of corruption in public procurement in Italy.
Zanella reveals how the corruption script followed by corrupt agents unfolds and
highlights three types of facilitators (cognitive and social facilitators, legal and
administrative facilitators, and physical facilitators) needed for the completion of
the corrupt deal. In Chapter 10, Alexandra Hiropoulos, Joshua Freilich, Steven
Chermak and Graeme Newman adopt the script approach to study a cigarette
smuggling scheme. One particularity of this scheme is that it involves an extrem-
ist group of Hezbollah supporters and is committed for financing terrorism. After
breaking down the script, numerous situational techniques are proposed that, if
implemented, as argued by the authors, would prevent the script from being com-
pleted and consequently, not fund terrorism.
The third group of chapters critically looks at the future of the rational choice
perspective and crime script analysis. In Chapter 11, William Moreto and
Ronald Clarke explore how a script analysis of the Transnational Illegal Market
in Endangered Species could be undertaken. Moreto and Clarke argue that this
form of crime is too diverse, too complex and too dangerous and difficult to study
to perform script analysis. A more limited role of script analysis to tackle this
crime is suggested by the authors but this would require, as a first step, access
to detailed knowledge on offender modus operandi by agencies responsible for
enforcement. In Chapter 12, Benoit Leclerc argues that current applications of
script analysis have largely ignored the behaviours of other actors present during
the crime event which limits the potential of situational prevention. He insists
on performing script analysis from the perspective of the victim and guardian as
well as examining the interaction between the offender, the victim and guardian
scripts. According to Leclerc, these script developments will extend the reach
of situational prevention. The chapter by Richard Wortley concludes the vol-
ume. Wortley draws the distinction between the rational choice perspective as a
framework to guide policy and practice in crime prevention on the one hand, and
as descriptive model of offender decision making on the other. He argues that
the perspective succeeds admirably in the first purpose, but was never intended
to fulfil, and nor does it succeed in fulfilling, the second purpose. Wortley urges
The Reasoning Criminal: twenty-five years on 9
researchers to draw on developments in the cognitive sciences that have occurred
since the publication of The Reasoning Criminal in order to build more rigorous
and comprehensive theoretical models of offender decision making.

Beauregard, E. and Leclerc, B. (2007). An application of the rational choice approach to
the offending process of sex offenders: A closer look at the decision-making. Sexual
Abuse: A Journal of Research and Treatment, 19, 115–133.
Bennett, T. and Wright, R. (1984). Burglars on burglary: Prevention and the offender.
Aldershot: Gower.
Brayley, H., Cockbain, E. and Laycock, G. (2011). The value of crime scripting:
Deconstructing internal child sex trafficking. Policing, 5, 132–143.
Bullock, K., Clarke, R. V. and Tilley, N. (2010). Situational crime prevention of organised
crimes. Cullompton: Willan.
Carroll, J. and Weaver, F. (1986). Shoplifters’ perceptions of crime opportunities: A
process-tracing study. In D.B. Cornish and R.V. Clarke (eds), The reasoning criminal:
Rational choice approaches on offending (pp. 19–38). New York: Springer-Verlag.
Cherbonneau, M. and Copes, H. (2006). Drive it like you stole it: Auto theft and the illusion
of normalcy. British Journal of Criminology, 46, 193–211.
Chiu, Y., Leclerc, B. and Townsley, M. (2011). Crime script analysis of drug manufacturing
in clandestine laboratories. British Journal of Criminology, 51, 355–374.
Clarke, R.V. (1967). Seasonal and other environmental aspects of abscondings by approved
school boys. British Journal of Criminology, 7, 195–206.
Clarke, R.V. (1997). Situational crime prevention: Successful case studies. Guilderland,
NY: Harrow et Heston.
Clarke, R.V. (1980). ‘Situational’ crime prevention: Theory and practice. British Journal
of Criminology, 20, 136–147.
Clarke, R.V. and Cornish, D.B. (1985). Modeling offenders’ decisions: A framework for
research and policy. In M. Tonry and N. Morris (eds), Crime and justice: An annual
review of research (pp. 147–185). Vol. 6. Chicago: University of Chicago Press.
Clarke, R.V. and Martin, D.N. (1971). Absconding from approved schools. Home Office
Research Study no. 12. London: HMSO.
Clarke, R.V. and Newman, G.R. (2006). Outsmarting the terrorist. Westport, CT: Praeger
Security International.
Cornish, D.B. (1978). Gambling: A review of the literature and its implications for policy
and research. Home Office Research Studies No. 42. London: HMSO.
Cornish, D.B. (1994). The procedural analysis of offending and its relevance for situational
prevention. In R.V. Clarke (ed), Crime prevention studies, Vol. 3. (pp. 151–196).
Monsey, NY: Criminal Justice Press.
Cornish, D.B. (1998) Regulating lifestyles: A rational choice perspective. Paper presented
at the 7th International Seminar on Environmental Criminology and Crime Analysis,
Cornish, D.B. and Clarke, R.V. (1986). The reasoning criminal: Rational choice
approaches on offending. New York: Springer-Verlag.
Cornish, D.B. and Clarke, R.V. (2008). The rational choice approach. In R. Wortley
and L. Mazerolle (eds), Environmental criminology and crime analysis (pp. 21–47).
Cullompton: Willan.
10 Benoit Leclerc and Richard Wortley
Cromwell, P., Olson, J. and Avary, D’A.W. (1991). Breaking and entering: An ethnographic
analysis of burglary. Newbury Park, CA: Sage.
Farrell, G. (2010). Situational crime prevention and its discontents: Rational choice and harm
reduction versus ‘cultural criminology’. Social Policy and Administration, 44, 40–66.
Feeney, F. (1986). Robbers as decision-makers. In D.B. Cornish and R.V. Clarke (eds),
The reasoning criminal: Rational choice approaches on offending (pp. 53–71). New
York: Springer-Verlag.
Freilich, J.D. and Newman, G.R. (eds) (2009). Reducing terrorism through situational
crime prevention. Crime prevention studies. Vol. 25. Monsey, NY: Criminal Justice
Garland, D. (1999). The commonplace and the catastrophic: Interpretations of crime in late
modernity. Theoretical Criminology, 3, 353–364.
Hayward, K. (2007). Situational crime prevention and its discontents: Rational choice
theory versus the ‘culture of now’. Social Policy and Administration, 41, 232–250.
Jacobs, B. (1996). Crack dealers’ apprehension avoidance techniques: A case of restrictive
deterrence. Justice Quarterly, 13, 359–382.
Jeffery, C.R. (1971). Crime prevention through environmental design. Beverly Hills, CA:
Lacoste, J. and Tremblay, P. (2003) Crime and innovation: A script analysis of patterns in
check forgery. In M.J. Smith and D.B. Cornish (eds), Theory for practice in situational
crime prevention. Crime prevention studies. Vol. 16. (pp. 169–196). Monsey, NY:
Criminal Justice Press.
Leclerc, B., Wortley, R. and Smallbone, S. (2011). Getting into the script of adult child
sex offenders and mapping out situational prevention measures. Journal of Research in
Crime and Delinquency, 48, 209–237.
Luckenbill, D.F. (1977). Criminal homicide as a situated transaction. Social Problems, 25,
Madensen, T.D. and Knutsson, J. (eds) (2011). Preventing crowd violence. Crime
prevention studies. Vol. 26. Monsey, NY: Criminal Justice Press.
Mann, R.E and Hollin, C.R. (2007). Sexual offenders’ explanations of their offending.
Journal of Sexual Aggression, 13, 3–9.
Matthews, R. (1987) Taking realist criminology seriously. Crime, Law and Social Change,
11, 371–401.
Mayhew, P., Clarke, R.V., Sturman, A. and Hough, J. (1976). Crime as opportunity. Home
Office Research Study No. 34. London: HMSO.
McNally, M.M. and Newman, G.R. (eds) (2007). Perspectives on identity theft. Crime
prevention studies. Vol. 23. Monsey, NY: Criminal Justice Press.
Mischel, W. (1968) Personality and assessment. New York, NY: Wiley.
Nee, C. and Meenaghan, A. (2006). Expert decision making in burglars. British Journal of
Criminology, 46, 935–949.
Newman, G.R. and Clarke, R.V. (2003). Superhighway robbery: Preventing e-commerce
crime. Cullompton: Willan Publishing.
Newman, O. (1972). Defensible space: Crime prevention through urban design. New
York, NY: MacMillan.
Petrosino, A. and Brensilber, D. (2003). The motives, methods and decision-making of
convenience store robbers: Interviews with 28 incarcerated offenders in Massachusetts.
In M.J. Smith and D.B. Cornish (eds), Theory for practice in situational crime
prevention (pp. 237–263). Crime prevention studies. Vol. 16. Monsey, NY: Criminal
Justice Press.
The Reasoning Criminal: twenty-five years on 11
Rengert, G. and Wasilchick, J. (2000). Suburban burglary: A tale of two suburbs (2nd
edn). Springfield, IL: Charles C. Thomas.
Savona, E.U. (2010) Infiltration by Italian organized crime (Mafia, N’drangheta and
Camorra) of the public construction industry. In K. Bullok, R.V. Clarke and N. Tilley
(eds), Situational prevention of organized crimes (pp. 130–150). Cullompton: Willan
Sinclair, I.A.C. and Clarke, R.V.G. (1973) Acting-out behaviour and its significance for
the residential treatment of delinquents. Journal of Child Psychology and Psychiatry,
14, 283–291.
Smith, M.J. and Cornish, D.B. (2006). Secure and tranquil travel: Preventing crime and
disorder on public transport. London: University College London, Jill Dando Institute
of Crime Science.
Tedeschi, J.T. and Felson, R.B. (1994). Violence, aggression, and coercive Actions.
Washington, DC: American Psychological Association.
Trasler, G. (1986). Situational crime prevention and rational choice: A critique. In K. Heal
and G. Laycock (eds), Situational crime prevention (pp.17–24). London: HMSO.
Tremblay, P., Talon, B. and Hurley, D. (2001). Body switching and related adaptations in
the resale of stolen vehicles: Script elaborations and aggregate crime learning curves.
British Journal of Criminology, 41, 561–579.
Walsh, D. (1986). Victim selection procedures among economic criminals: The rational
choice approach. In D.B. Cornish and R.V. Clarke (eds), The reasoning criminal:
Rational choice approaches on offending (pp. 39–52). New York: Springer-Verlag.
Webb, B. and Laycock, G. (1992). Tackling car crime: The nature and extent of the
problem. Crime prevention unit paper 32. London: Home Office.
Wortley, R. (2010). Critiques of situational crime prevention. In B. Fisher and S. Lab (eds),
Encyclopedia of victimology and crime prevention (pp. 885–888). Thousand Oaks, CA:
Wortley, R. and Smallbone, S. (2006). Situational prevention of child sexual abuse. Crime
prevention studies. Vol. 19. Monsey, NY: Criminal Justice Press.
Wright, R. and Decker, S. (1994). Burglars on the job: Streetlife and residential break-ins.
Boston, MA: Northeastern University Press.
Wright, R. and Decker, S. (1997). Armed robbers in action: Stickups and street culture.
Boston, MA: Northeastern University Press.
Young, Jock. (1994). Recent paradigms in criminology. In M. Maguire, R. Morgan and
R. Reiner (eds) The Oxford Handbook of Criminology (pp. 69–124). Oxford: Oxford
University Press.
2 What are violent
offenders thinking?
Richard B. Felson

What are violent offenders thinking when they commit their crime? Are they
making decisions and, if so, on what are they basing those decisions? Do they
plan their crime ahead of time or is their reaction spontaneous? Is their goal to
harm the victim or are they indifferent to the victim’s suffering? Do they consider
tactics that would increase their chance of success or are they out of control? Do
they ignore the costs or morality of their behavior? Is the thought process different
for different types of crime or different types of individuals?
This chapter examines the offender’s decision making during a violent crime.
First, I discuss the actor’s attitude toward harming the adversary. Then I discuss
differences in attitudes toward harm for predatory offenses and offenses that stem
from disputes, and for instrumental vs. reactive aggression. I discuss the relation-
ship between low self-control and the offender’s consideration of costs and moral
issues. I also consider the role of self-control in crimes that are premeditated.
Finally, I discuss the implications of impulsivity and planning for situational
crime prevention.

Violence is a method of aggression that involves the use or threat of physical
force. Aggression is typically defined as any behavior intended to harm another
person (Berkowitz, 1962). The definition includes intended harms that are unsuc-
cessful and excludes behaviors that involve accidental harm. For example, driving
while intoxicated may be deviant behavior and irresponsible, but it cannot be
considered an act of aggression, even when someone is killed, since there was no
intent to harm. Finally, the definition includes intended harms that are considered
socially desirable, for example, legitimate punishment.
Social psychologists have struggled with the definition of aggression because
of uncertainty about the meaning of intent and because of reluctance to include
harmful behaviors they view as pro-social. It is important to have a clear defini-
tion because it affects how we think about aggression and how we explain it. We
need to know what we are trying to explain, namely, our dependent variable.
Generally we prefer definitions that classify together behaviors that have similar
motives and exclude behaviors that have completely different motives.
What are violent offenders thinking? 13
Our understanding of criminal violence is complicated by the fact that the
behavior involves both crime and violence. As a result we are attempting to explain
why people engage in deviance (or rule-breaking) and why they are engaging
in interpersonal aggression. Aggression and deviance are overlapping domains:
some acts of aggression are deviant and some are not. For example, punishment
involves intentional harm but is not necessarily deviant. And drug use and other
victimless crimes are deviant but do not involve intentional harm. Therefore, to
fully understand the behavior we should explain why people harm others as well
as why they break the law. Some of the causes are the same and some are differ-
ent. For example, individual characteristics that increase a person’s desire to harm
others (e.g., a hostility bias) or indifference (e.g., a lack of empathy) are likely to
lead to crimes involving deliberate harm but not necessarily to victimless crimes
or crimes of negligence (e.g., Dodge and Somberg, 1987; Mehrabian, 1997). On
the other hand, self-control, thrill seeking, and fearlessness may affect crime and
deviance, regardless of whether the behaviors produce harm. These characteristics
will lead to versatile offending. They may, however, be more strongly related to
offenses involving deliberate harm to others. For example, resting heart rates have
also been shown to be lower for violent offenders than for nonviolent offenders
(e.g., Farrington, 1987), suggesting that violent offenders are particularly likely
to be thrill seekers attempting to relieve their boredom. Since violent crime tends
to be more exciting than other crime it may be more likely to involve thrill seek-
ing. It may also be that fear is more likely to inhibit violent crime than nonviolent
crime because of the danger resulting from the confrontation with the victim.
Self-control may also play a greater role in crimes involving deliberate harm.
Crimes that include deliberate harm, and a victim, are generally considered
more serious. The tendency to avoid harming others may help explain what may
be the most prominent pattern of criminal behavior: serious crime is much less
frequent than minor crime. While serious offenders usually commit minor crime,
most offenders who commit minor crimes do not commit serious crime. This pat-
tern suggests that most offenders have enough self-control to limit themselves;
they might commit a theft but they will not commit homicide. This tendency
toward limited offending reduces the level of versatility shown by offenders
(Felson and Osgood, 2008).
Offenders vary in how much harm they intend. Many of them intend limited
harm and engage in, for example, pushing and shoving. If they are winning a
fist fight they know when to stop. Sometimes they have ambiguous intent and
sometimes they produce more harm than they intend. On the other hand, some
offenders have lethal intent.1 Offenders intend to bring about the death of their
victims for several reasons. Some may feel so angry and aggrieved with the vic-
tim that they desire the ultimate penalty. Others may kill the victim for practical
reasons. They may be attempting to eliminate a rival, make money, prevent iden-
tification and subsequent prosecution, or prevent the victim from retaliating. By
permanently incapacitating the victim, offenders secure their own personal safety.
They may, in other words, believe that it is “kill or be killed.”
14 Richard B. Felson
Research shows that, during an assault, offenders are more likely to kill vic-
tims whom they know in any way. Presumably they are killing someone who
can identify them. In addition, offenders are more likely to kill victims who
are armed (Felson and Steadman, 1983). They finish off the offender to avoid
retaliation. They are also more likely to kill males and blacks, controlling for the
type of weapon they use and their own gender and race. Presumably, they view
men and blacks as more dangerous. Similar patterns have been observed for
weapon choice: offenders use guns against more dangerous adversaries. These
adversary effects play an important role in the nature of violent incidents in
African American communities (Felson and Paré, 2010; Felson and Painter-
Davis, 2012).

Dispute-related and predatory offenses

In thinking about what offenders are thinking, it is useful to distinguish between
dispute-related and predatory behavior. The distinction depends on the offender’s
attitude toward harming the victim (Felson, 2005). In dispute-related aggression
the proximate goal of offenders is to harm the target. Harm is a proximate goal
because it achieves some distal goal (e.g., face-saving, justice, deterrence). These
offenders have grievances with their victims, they are angry, and they want to see
their victims suffer. Most homicides and assaults stem from disputes, but so do
some robberies, rapes, thefts, and frauds (Black, 1983). People sometimes choose
unusual methods to harm others (see, e.g., Jacques and Wright, 2008).
In contrast, harm is incidental to predatory offenders. They deliberately harm
victims but do not have a particular desire to do so. Rather, they have some other
goal in mind, and they are willing to harm the victim in order to achieve their
goal. In other words, they expect the victim to be harmed but they do not assign
a positive value to that outcome. For example, robbery and rape typically involve
predatory violence. Most robbers and rapists have no grievance with the victim
and are indifferent to the victim’s suffering. They might engage in violence for
thrills or to force the victim to comply in order to get something else they want
(e.g., money or sex). It’s nothing personal. It’s just business.
The relationship between attitudes toward harm and the use of violence vs.
other methods of harming the victim is depicted in Table 2.1. The cells on the
left refer to predatory behaviors in which actors are indifferent to the harm expe-
rienced by victims. Actors can use violence, as in most robberies and rapes, or
nonviolent methods (e.g., theft and fraud; threats of nonphysical harm). The cells
on the right refer to actors who prefer to harm their victim. They sometimes use
physical means, as in most homicide and assaults, but more often they use other
methods (e.g., insults). However, most dispute-related homicides and assaults
begin with verbal attacks and then escalate to violence (e.g., Luckenbill, 1977;
Felson and Steadman, 1983).
Most predatory offenders use nonviolent methods, namely, deception or stealth
(e.g., Jacques and Wright, 2008). Those who engage in theft or fraud deliberately
harm their victims but the harm is incidental to them. For example, thieves do
What are violent offenders thinking? 15
Table 2.1 Examples of different methods of, and attitudes
toward harm

Attitude toward harm

Indifference Preference
Violence Robbery Assault
Other methods Larceny Insults

not care whether the victim’s insurance can replace the object they steal. In some
cases nonviolent offenders may not believe that their behavior has harmed anyone
(Sykes and Matza, 1957). Shoplifters, for example, may believe that their crime is
harmless since the victim is a large corporation and not an individual. Their moral
inhibitions do not come into play because there is no clear victim.2 It is important
to take into account the offender’s subjective viewpoint in considering whether
their behavior involves aggression.

Instrumental vs. reactive aggression

Some scholars interpret predatory crime as instrumental aggression and dis-
pute-related violence as reactive or angry aggression (e.g., Berkowitz, 1989).
Instrumental aggression is behavior in which harm is used as a means to some
other end. It typically involves some degree of premeditation. Reactive aggres-
sion, on the other hand, is motivated by a desire to hurt someone; it has no
ulterior motive. Reactive aggression is typically viewed as an impulsive response
to frustration or aversive stimuli (e.g., Dollard et al., 1939; Berkowitz, 1989).
Individuals experience negative affect in response to anything unpleasant, and
that triggers an innate tendency to harm others. When the costs are too high they
may inhibit themselves and perhaps displace their aggression onto a safer target.
However, sometimes they are overwhelmed by their anger and attack the source
of aversive stimuli regardless of costs.
Bushman and Anderson (2001) dispute the idea that reactive aggression is
impulsive and instrumental aggression is planned. They argue that either type
of aggression can involve consciously controlled information processing and
either type can involve automatic or over-learned behavior based on schemas and
scripts. They conceive of reactive aggression as an overlearned behavior rather
than an innate response to negative affect. Aggression can be a habitual response
based on past rewards rather than the anticipation of future ones.
Tedeschi and Felson (1994) reject the idea that there is a special frustration–
aggression mechanism producing reactive aggression. They argue that what
some scholars describe as reactive aggression is actually instrumental behavior.
According to their rational choice perspective, offenders are motivated to achieve
social influence, retribution, thrills, and the enhancement of identities. They
acknowledged that offenders are often impulsive, but they claim that impulsive
16 Richard B. Felson
violence reflects careless decision making rather than an automatic response to
negative affect produced by aversive stimuli.
From this perspective individuals are making decisions even when they use
violence in a rage. Even in a rage, they must decide how to carry out the attack
and how to respond to counter-attack. They often limit their attack, as discussed
earlier. Decisions are sometimes made without much thought or concern for con-
sequences, but they are still decisions. Inhibitions may be overwhelmed when
individuals are angry and have a strong desire to hurt someone, but the source
of that desire is not a biological urge caused by aversive stimuli. Anger can be
a strong emotion, but it is a response to being mistreated and humiliated, not a
response to all types of aversive stimuli (Averill, 1983). An insult is likely to lead
to aggression while much more aversive stimuli, such as learning that a loved one
had died, is not. Laboratory studies show that the key determinant of aggressive
behavior is an intentional attack by a confederate, not generally aversive stimuli.
Participants will not deliver shock or cause harm to a confederate unless they
have been shocked or insulted themselves (Tedeschi and Felson, 1994). Also,
participants retaliate for intended shocks, even when they do not actually receive
the shock (e.g., Epstein and Taylor, 1967). The results suggest that “bad inten-
tions,” not bad experiences, lead to aggression, because they imply blame. Other
laboratory studies show that participants are more likely to retaliate when a person
shocks them than when a machine shocks them (Sermat, 1967). The pain is the
same, but the meaning differs. Participants only retaliate when they think that a
person has mistreated them.
The offender’s decisions are not necessarily based on valid perceptions. They are
responding to their “definition of the situation” and their rationality is “bounded.”
The burgeoning field of behavioral economics addresses various biases that affect
decision making but it has not had much influence in criminology (e.g., Ariely,
2009). The older literature on cognitive biases suggests that they tend to result in
more violence. First, the tendency for observers to weigh negative information
about an actor more heavily than positive information is likely to lead to griev-
ances and grudges (e.g., Kanouse and Hanson, 1971). Second, evidence suggests
that individuals tend to assign blame even when it is undeserved. They may fail to
understand their adversary’s perspective, or they may ignore exonerating circum-
stance, committing the fundamental attribution error (Ross, 1977). Sometimes they
blame others to avoid self-blame. People working together on a task are motivated
to blame the other person in cases of failure. Sometimes blaming others can help
explain traumatic events that are otherwise difficult to understand.
In addition, some people have a “hostility bias” and are quick to assign blame
and respond with aggression (Dodge and Coie, 1987). At the extreme are men-
tally ill people who suffer from paranoia: they think others are plotting against
them and are likely to become violent in response. Their aggressive response is
“rational,” given their definition of the situation (Felson et al. 2012).
Felson et al. (2012) found that paranoia is the strongest predictor of violent
offenses committed by prison inmates and that it accounted for a substantial
portion of the effect of psychosis (see also, Link et al., 1999). These offenses
What are violent offenders thinking? 17
included attacks on inmates and staff and weapons violations. The finding that
paranoia is a particularly strong predictor of weapons offenses is consistent with
an instrumental approach. Paranoid inmates do not just strike out at others, as
one might expect if they were responding automatically to negative affect; they
sometimes plan for violent offenses by obtaining weapons.

The discussion so far has focused on incentives for violent crime. Of course, costs
are also important from a rational choice perspective. Costs include the inhibiting
effects of internalized moral beliefs as well as negative external consequences
(e.g., punishment and retaliation). Scholars who take a frustration–aggression
approach and emphasize reactive aggression also recognize the role of costs and
decision making. For them, only the instigation of reactive aggression is irrational.
Both approaches acknowledge that people fail to consider costs in the heat of the
moment if the instigation is strong. The theories differ, however, in their claim
of how a person motivated to use aggression responds when the costs are high.
Frustration–aggression approaches claim that aggression is displaced onto inno-
cent third parties. In contrast, from a rational-choice perspective, displacement is
rare. Usually offenders have real or imagined grievances with their victims.
One of the central arguments in criminology is that criminal offenders do not
think about costs or moral implications during the offense because they have
low self-control. They focus on the immediate benefits of criminal behavior and
fail to consider future costs or abstract questions of morality. Gottfredson and
Hirschi (1990) go so far as to claim that offenders and law abiding citizens have
similar attitudes related to crime. Offenders are just as likely to think violence is
immoral, for example. The only difference is that offenders do not think about
moral values at the time because they lack self-control. They engage in behaviors
that are contrary to their values or beliefs. If they counted to ten, they would not
commit the offense.
The concept of self-control helps us to understand the offender’s decision-
making process and bounded rationality. Many acts of violence involve careless
decisions and an offender’s failure to consider consequences. The concept of
self-control also helps us to understand why many offenders do such stupid
things and how someone can commit an offense even when it violates their own
moral standards.
Numerous studies show that self-control is correlated with criminal behavior
(e.g., Pratt and Cullen, 2000). The measurement of self-control is problematic,
however, and causal inference is difficult with correlational data. Better evidence
comes from psychologists (e.g., Mischel, 1974; Baumeister et al., 1994) whose
work has been ignored by criminologists. For example, a number of experimental
studies show that manipulations of self-control affect deviant behavior. These
studies find that participants are better behaved when they are standing in front
of a mirror. The focus of their attention is inward, and their behavior is more
likely to conform to their values or standards (Wicklund, 1975). In other words,
18 Richard B. Felson
they are more likely to use self-control. Experimental research has shown that the
presence of a mirror inhibits male participants from delivering shocks to women,
a behavior inconsistent with their values (Carver, 1974; Scheier et al., 1974). On
the other hand, the presence of a mirror leads participants to deliver more intense
shocks when the target is a man and the delivery of shock is legitimated by the
experimenter. These results suggest that self-control may either inhibit or facili-
tate violence depending on the individual’s attitudes toward the use of violence in
a particular context.
The effect of self-control on the relationship between internal standards and
violence has also been demonstrated in a series of experiments carried out by
Froming et al. (1982). In the first experiment only college students who opposed
the use of physical punishments as a means of gaining compliance from others
were selected. A second criterion for selection was that these students believed
that most other people were favorable to the use of physical punishments. In a
teacher–learner situation the presence of a mirror inhibited the use of shocks,
but the presence of an evaluative audience facilitated aggression. In the second
experiment the criteria for selection of subjects were reversed. Only students who
were favorable to the use of physical punishments but believed most others were
unfavorable were selected. The results were exactly the opposite of those found
in the first experiment. When a mirror was present, participants gave more shocks
than control subjects, whereas the presence of an evaluative audience decreased
shock delivery. Again, the results showed that the effect of self-control depends
on individual attitudes and that individuals vary in whether they view violence
as an appropriate response in a particular circumstance. Attitudes determined the
direction of self-control’s impact on behavior.
These studies suggest that self-control varies across situations as well as across
individuals. For example, research shows that individuals use less self-control
when they are tired (Baumeister et al., 1994); it is a capacity that can be used up.
Research also suggests that alcohol intoxication diminishes people’s immediate
capacity for self-control. Alcohol apparently impairs performance and intellectual
functioning and alters perceptions of risk (e.g., Bushman, 1997). For example, it
is argued that alcohol decreases cognitive capacity so that the drinker’s attention
is focused on the most salient cues in a situation (Steele and Josephs, 1990). This
cognitive “myopia” leads intoxicated individuals to use violence because they fail
to consider the costs. Individuals who are under stress may also have lower self-
control and make careless decisions (Tedeschi and Felson, 1994). Thus, aversive
stimuli facilitate rather than instigate aggression, a much more limited role than
that suggested by frustration–aggression approaches.
Depression may be another factor that interferes with an actor’s ability or desire
to inhibit their behavior. Those who are depressed may be less likely to engage in
self-control because their focus of attention is on themselves. They may not think
about costs or legal rules because they are preoccupied with their own problems.
Felson et al. (2012) found that depressed inmates were more likely to engage
in nonviolent as well as violent offenses in prison. These effects were observed
with controls for prior violence and other potentially confounding factors, such as
What are violent offenders thinking? 19
alcohol and drug use. The versatility of depressed inmates supports the idea that
depression affects deviant behavior because it reduces inhibitions and not because
negative affect instigates aggressive behavior. Depressed inmates lack either the
ability or the willingness to control themselves.

Impulsive vs. premeditated crime

Impulsiveness is a characteristic generally use to explain “split-second” decisions.
It is more difficult to attribute premeditated offenses to careless decision making.
Offenders who plan their offenses are more attentive to the task and have more
time to think about the risks. One would therefore expect that low self-control has
a much stronger impact on the commission of crimes that are impulsive than those
that are planned.3 Gottfredson and Hirschi’s general theory of crime may not be
as general as they claim.
Unfortunately, there is no clear definition of what is meant by a “plan.” Plans
may be specific or general, vague or detailed, conscious or unconscious, devel-
oped at the scene of the crime or before. Perhaps the offender planned the crime
but did not plan such a serious outcome. Perhaps the offender planned behaviors
that created high-risk situations without conscious anticipation of committing a
crime (e.g., Ward and Hudson, 2000). There are likely to be mixed cases and
degrees of premeditation (e.g., Beauregard and Leclerc, 2007).
George Herbert Mead’s (1934) description of “the act” implies that all human
behavior involves at least a rudimentary plan of action. Before acting on impulses
(the “I”), the actor imagines consequences (the “Me”) before engaging in the
behavior. Fritz Heider (1958) also suggests that some planning is involved in all
intentional behavior, since an intention is a plan that guides action. Similarly,
Tedeschi and Felson (1994) argue that all cognitively mediated action, as opposed
to reflexes and automatic responses, involve some forethought. They view
aggression committed on impulse as a “bounded” rational choice in which an
actor engages in careless decision making and fails to consider consequences. In
sum, according to these theorists, even impulsive actions involve a rudimentary
plan immediately preceding behavior.
Dispute-related offenses may be less likely to be planned ahead of time than
predatory offenses because offenders are angry, and because the motivation
to commit violence is more likely to arise during the situation, after a verbally
aggressive exchange (e.g., Luckenbill, 1977; Felson and Steadman, 1983). In
other words, dispute-related offenses are more likely to be spontaneous and
context-driven. In contrast, the motivation for predatory offenses is more likely
to precede the criminal event.4 Such offenders are more likely to select the vic-
tim and the situation than offenders involved in a dispute (Birkbeck and Lafree,
1993). Some robbers, for example, may search for opportunities with low risk or
high yield and plan their offense beforehand (e.g., Jacobs, 2000).
Only a few studies have examined the extent to which particular violent
offenses are planned. In addition, the studies have produced widely disparate esti-
mates of planning, probably because they rely on different measures (see Erez,
20 Richard B. Felson
1980; Birkbeck and Lafree, 1993; Gill 2000). Most of the studies focus on rob-
bery. For example, Petersilia et al. (1978) found that 25 percent of incarcerated
robbers reported that they had planned their offense in detail, 50 percent planned
some aspect of their crime, and 25 percent did not plan at all. Walsh (1986) found
that 52 percent of armed robbers planned their offense while Feeney (1986) found
that only 15 percent planned their offense. Research on planning sexual assault
has also found inconsistent results. Amir (1971) reported that 71 percent of rapes
in his sample were planned while Glueck and Glueck (1956) found that only
40 percent of rapes involved planning. Monahan et al. (2005) interviewed thirty-
three incarcerated rapists and found that most used minimal planning (see also
Hazelwood and Warren, 2000). Beauregard and Leclerc (2007) found that 65 per-
cent of sex offenders engaged in at least some degree of premeditation.
Only one study has compared planning for different types of offenses. That
study used a large sample of inmates who had been convicted of a violent offense
(Felson and Massoglia, 2012). The evidence was mixed as to whether dispute-
related crimes were less likely to be planned ahead of time than predatory crimes.
As predicted, homicides and physical assaults were much less likely to be planned
than robbery, and physical assaults were less likely to be planned than sexual
assaults.5 However, sexual assaults were no more likely to be planned than homi-
cide and they were much less likely to be planned than robberies. Sexual assault
is apparently more likely to be an opportunistic crime than robbery. Its motivation
is more likely to arise out of the immediate situation. The pattern is consistent
with evidence that many sexual assaults occur during consensual sexual activity
when the man wants to go farther than the woman, and uses force when she resists
(Kanin, 1985).
The description of domestic violence as a “crime of passion” implies that these
offenses are more likely to be committed on impulse than violence involving peo-
ple outside the family. We found only mixed support in our analysis of inmate
violence. Male violence against partners and other family members were slightly
less likely to be planned than violence against strangers. The characterization of
domestic violence as “crimes of passion” appears to be overstated.
Our evidence suggests that individual differences in self-control should be
most important in homicide and physical assault—violence usually stemming
from disputes—since these offenses are most likely to be committed on impulse.
Self-control should play the least important role in robbery, since robbery is more
likely to be planned. Self-control should also play a more important role in sexual
assault than in robbery since sexual assault is a more impulsive crime.

Implications for situational crime prevention

Impulsivity is also relevant to the routine activities approach and its policy applica-
tion—situational crime prevention (Cornish and Clarke, 1986; M. Felson, 1994).
From this perspective, crime is an impulsive act, committed in response to opportu-
nity. While offenders are rational decision makers, their choices are formed in the
situation. Impulsive offenders do not go out looking for opportunities to commit
What are violent offenders thinking? 21
crime; they rarely plan their crime ahead of time. Motivated offenders stumble upon
criminal opportunities when they are engaged in their routine activities.
Offenders who plan their crime ahead of time, on the other hand, are not oppor-
tunists. For these offenders, contact produced by routine activities should be less
important in determining whether a crime takes place. In addition, to the extent
offenders plan their crimes, situational crime prevention will not be as effective.
When a crime opportunity is blocked, offenders will search for alternative targets.
In other words, displacement is more likely to the extent offenders plan their
crimes. Research suggests that, in general, crime prevention efforts in one loca-
tion do not lead crime to be displaced to other locations (e.g., Hesseling, 1993).
Still, some displacement does occur and it is much more likely to occur for crimes
that involve planning. The impulsiveness of crime is, therefore, a key assumption
in the study of routine activities and situational crime prevention.
We have found that, while many robbers commit their crime on impulse, they
are much more likely than other violent offenders to plan their offenses (Felson
and Massoglia, 2012). Robbery, therefore, should be more difficult to deter with
situational crime prevention than other violent offenses. In addition, displacement
should be more likely for robbery than for other violent offenses. Localized situ-
ational crime prevention strategies may affect robbery rates, but they should be
less effective than strategies that target other violent crimes. On the other hand,
situational crime prevention should be particularly effective in deterring violence
stemming from disputes, since these offenses tend to be more impulsive. In gen-
eral, attempts to design environments to avoid crime should take into account the
impulsiveness of the crime being targeted.

In the study of violent crime it is important to be clear whether we are trying to
explain violence or crime. Some of the causes of crimes involving intentional
harm-doing and victimless crimes are different. It is also useful to distinguish
predatory offenses and offenses stemming from disputes. Predatory offenders are
neutral to the harm they expect the victim to suffer. It is nothing personal for
them. In crimes stemming from disputes, on the other hand, harm is a proximate
goal rather than an incidental outcome. These offenders prefer to harm the victim
since they want retribution, influence, and a favorable image.
I argue that even violent crimes stemming from disputes involve instrumental
aggression. Ill-considered decisions rather than inner compulsions result in behav-
iors that seem irrational to the objective observer. Individuals who are intoxicated,
tired, stressed, depressed, or in a rage are still making decisions, although they
sometimes ignore costs and morality. From the perspective of bounded ration-
ality, a “loss of temper” involves responding to a provocation single-mindedly
without inhibition. The offender is still making tactical decisions and is only par-
tially out of control.
One thing on people’s minds during a conflict is the threat posed by their adver-
saries. Fear of retaliation can lead participants to use weapons and have lethal intent.
22 Richard B. Felson
Adversary effects are important in most violent encounters and they may help explain
the high levels of gun violence and homicide in African American communities.
According to the symbolic interactionists, an understanding of human behavior
requires an understanding of an individual’s “definition of the situation.” Violent
behavior is no exception. The decision to harm others is sometimes a careless one,
and sometimes based on illusion, but it is a decision nonetheless.

1 Of course, some lethal outcomes are not intended.
2 When offenders deny there is a victim after the offense, we call it an account or
3 Perhaps the offenders’ lack of inhibitions in premeditated crime reflects misinterpret-
ing rather than ignoring costs and morality.
4 The motivation to steal may be a response to a (self-inflicted) financial crisis of recent
origins. According to ethnographic accounts, offenders often commit financial crimes
after experiencing a shortage of funds produced by their party lifestyle (e.g., Wright
and Decker, 1997).
5 One might expect that serious crimes are more likely to be committed on impulse than
less serious crimes. However, the evidence suggested that physical assaults are less
likely to be planned than homicides. It may be that homicides are more likely to have a
tactical motive.

Amir, M. (1971). Patterns in Forcible Rape, Chicago, IL: University of Chicago Press.
Ariely, D. (2009) Predictably Irrational: The hidden forces that shape our decisions. New
York, NY: Harper Collins.
Averill, J.R. (1983). Studies on anger and aggression: Implications for theories of emotion.
American Psychologist, 38, 1145–60.
Baumeister, R.F., Heatherton, T.F. and Tice, D.M. (1994). Losing Control: How and why
people fail at self-regulation, San Diego, CA: Academic Press.
Berkowitz, L. (1962). Aggression: A social psychological analysis. New York, NY:
Berkowitz, L. (1989). The frustration–aggression hypothesis: An examination and
reformulation. Psychological Bulletin, 106, 59–73.
Birkbeck, C. and LaFree, G. (1993). The situational analysis of crime and deviance. Annual
Review of Sociology, 19, 113–37.
Black, D.J. (1983). Crime as social control. American Sociological Review, 48, 34–45.
Beauregard, E. and Leclerc, B. (2007). An application of the rational choice approach to
the offending process of sex offenders: A closer look at the decision-making. Sexual
Abuse: A Journal of Research and Treatment, 19, 115–33.
Bushman, B.J. (1997). Effects of alcohol on human aggression: Validity of proposed
mechanisms. In M. Galanter (ed), Recent Developments in Alcoholism (pp. 227–44).
New York: Plenium Press.
Bushman, B.J. and Anderson, C.A. (2001). Is it time to pull the plug on the hostile versus
instrumental aggression dichotomy? Psychological Review, 108, 273–79.
Carver, C.S. (1974). Physical aggression as a function of objective self-awareness and
attitudes toward punishment. Journal of Experimental Social Psychology, 11: 510–19.
What are violent offenders thinking? 23
Cornish, D.B. and Clarke, R.V.G. (eds) (1986). The Reasoning Criminal: Rational choice
perspectives on offending, New York, NY: Springer-Verlag.
Dodge, K.A. and Coie, J.D. (1987). Social-information-processing factors in reactive
and proactive aggression in children’s peer groups. Journal of Personality and Social
Psychology, 53, 1146–58.
Dodge, K.A. and Somberg, D.R. (1987). Hostile attributional biases among aggressive boys
are exacerbated under conditions of threats to self. Child Development, 58, 213–24.
Dollard, J., Doob, L.W., Miller, N., Mowrer, O.H. and Sears, R.R. (1939). Frustration and
Aggression, New Haven, CT: Yale University Press.
Epstein, S. and Taylor, S.P. (1967). Instigation to aggression as a function of degree of
defeat and perceived aggressive intent of the opponent. Journal of Personality, 35,
Erez, E. (1980). Planning of crime and the criminal career: Official and hidden offenses.
The Journal of Criminal Law and Criminology, 71, 73–6.
Farrington, D.P. (1987). Implications of biological findings for criminological research. In
S.A. Mednick, T.E. Moffitt, and S.A. Stack (eds), The Causes of Crime: New Biological
Approaches (pp. 42–64). New York, NY: Cambridge University Press.
Feeney, F. (1986). Robbers as decision-makers. In D. Cornish and R.V. Clarke (eds), The
Reasoning Criminal: Rational choice perspectives on offending (pp. 53–73). New
York, NY: Springer-Verlag.
Felson, M. (1994). Crime and Everyday Life: Insights and implications for society.
Thousand Oaks, CA: Pine Forge.
Felson, R.B. (2005).Violence as instrumental behaviour. In K. Kelloway, J. Barling and J.
Hurrell (eds), Handbook of Workplace Violence (pp. 7–28). Thousand Oaks, CA: Sage
Felson, R.B. and Massoglia, M. (2012). When is violence planned? Journal of Interpersonal
Violence, 27, 753–74.
Felson, R.B. and Osgood, D.W. (2008). Violent crime. In E. Goode (ed), Out of Control?:
Evaluating the general theory of crime, (pp. 160–72). Stanford, CA: Stanford University
Felson R.B. and Painter-Davis, N. (2012). Another cost of being a young black male: Race,
weaponry, and lethal outcomes in assaults. Social Science Research, 41, 1241–53.
Felson, R.B. and Paré, P. (2010). Firearms and fisticuffs: Region, race, and adversary
effects on homicide and assault. Social Science Research, 39, 272–84.
Felson, R.B. and Steadman, H. (1983). Situational factors in disputes leading to criminal
violence. Criminology, 21, 59–74.
Felson, R.B., Silver, E. and Remster, B. (2012). Mental disorder and offending in prison.
Criminal Justice and Behavior, 39, 125–43.
Froming, W.J., Walker, G.R. and Lopyan, K.J. (1982). Public and private self-awareness:
When personal attitudes conflict with societal expectations. Journal of Experimental
Social Psychology, 18, 476–87.
Gill, M. (2000). Commercial Robbery: Offender’s perspective on security and crime
prevention, London: Blackstone.
Glueck, S. and Glueck, E. (1956). Physique and Delinquency, New York, NY: Harper.
Gottfredson, M.R. and Hirschi, T. (1990) A General Theory of Crime, Stanford, CA:
Stanford University Press.
Hazelwood, R.R. and Warren, J.I. (2000). The sexually violent offender: Impulsive or
ritualistic? Aggression and Violent Behavior, 5, 267–79.
Heider, F. (1958). The Psychology of Interpersonal Relations, New York, NY: Wiley.
24 Richard B. Felson
Hesseling, R.B. (1993). Displacement: A review of the empirical literature. In R.V. Clarke
(ed), Crime Prevention Studies, vol. 3 (pp. 197–270). Monsey, NY: Criminal Justice
Jacobs, B.A. (2000). Robbing Drug Dealers, New York, NY: Aldine de Gruyter.
Jacques, S. and Wright, R. (2008). The relevance of peace to studies of drug market
violence. Criminology, 46, 221–53.
Kanin, E.J. (1985). Date rapists: Differential sexual socialization and relative deprivation.
Archives of Sexual Behavior, 14, 219–31.
Kanouse, D.E. and Hanson, L.R. (1971). Negativity in evaluations. In E.E. Jones,
D.E. Kanouse, H.H. Kelley, R.E. Nisbett, S. Valins and B. Weiner (eds), Attribution:
Perceiving the causes of behavior, (pp. 47–62). Morristown, NJ: General Learning
Link, B.G., Monahan, J., Stueve, A. and Cullen, F.T. (1999). Real in their consequences:
A sociological approach to understanding the association between psychotic symptoms
and violence. American Sociological Review, 64, 316–33.
Luckenbill, D.F. (1977). Criminal homicide as a situated transaction. Social Problems, 25,
Mead, G.H. (1934). Mind, Self, and Society, Chicago, IL: University of Chicago Press.
Mehrabian, A. (1997). Relations among personality scales of aggression, violence,
and empathy: Validational evidence bearing on the risk of eruptive violence scale.
Aggressive Behavior, 23, 433–45.
Mischel, W. (1974). Processes in delay of gratification. In L. Berkowitz (ed), Advances
in Experimental Social Psychology (Vol. 7) (pp. 249–293). New York: Academic
Monahan, B.A., Marolla, J.A. and Bromley, D.G. (2005). Constructing coercion: The
organization of sexual assault. Journal of Contemporary Ethnography, 34, 284–316.
Petersilia, J., Greenwood, P. and Lavin, M. (1978). Criminal Careers of Habitual Felons,
Washington, DC: U.S. Department of Justice.
Pratt, T.C. and Cullen, F.T. (2000). The empirical status of Gottfredson and Hirschi’s
general theory of crime: A meta-analysis. Criminology, 38, 931–64.
Ross, L. (1977). The intuitive psychologist and his shortcomings: Distortions in the
attribution process. In L. Berkowitz (ed), Advances in Experimental Social Psychology,
Vol. 10. (pp. 174–221). New York: Academic Press.
Scheier, M.F., Fenigstein, A. and Buss, A.H. (1974). Self-awareness and physical
aggression. Journal of Experimental Social Psychology, 10, 264–73.
Sermat, V. (1967). The possibility of influencing the other’s behavior and cooperation:
Chicken vs. prisoner’s dilemma. Canadian Journal of Psychology, 21, 204–19.
Steele, C.M. and Josephs, R.A. (1990). Alcohol myopia: Its prized and dangerous effects.
American Psychologist, 45, 921–33.
Sykes, G.M. and Matza, D. (1957). Techniques of neutralization: A theory of delinquency.
American Sociological Review, 22, 664–70.
Tedeschi, J.T. and Felson, R.B. (1994). Violence, Aggression, and Coercive Actions,
Washington, DC: American Psychological Association.
Walsh, D. (1986). Victim selection procedures among economic criminals: The rational
choice perspective. In D.B. Cornish and R.V.G. Clarke (eds), The Reasoning
Criminal: Rational choice perspectives on offending (pp. 39–52). New York, NY:
Ward, T. and Hudson, S. (2000). Sexual offenders’ implicit planning: A conceptual model.
Sexual Abuse: A Journal of Research and Treatment, 12, 189–202.
What are violent offenders thinking? 25
Wicklund, R.A. (1975). Objective self-awareness. In L. Berkowitz (ed), Advances in
Experimental Social Psychology (pp. 233–75). New York: Academic Press.
Wright, R.T. and Decker, S.H. (1997). Armed Robbers in Action: Stickups and street
culture, Boston, MA: Northeastern University Press.
3 How house burglars decide
on targets
A computer-based
scenario approach
Ross Homel, Stuart Macintyre and Richard Wortley

This chapter reports the results of a study of how a sample of burglars recently
released from prison in Melbourne, Victoria in the late 1990s went about their
business. Specifically, the focus was how these offenders selected and processed
information about potential houses to break into using a series of case studies or
scenarios presented on a computer screen, each scenario containing information
on a range of cues which were features of the house or street that burglars in a
previous study had reported were important in target selection (Macintyre, 2001).
The underlying rationale for the study was a desire to probe the utility of the
rational choice perspective applied to burglary (Cornish and Clarke, 1986). The
extent to which offences such as burglary are planned and executed with careful
attention to a reasonable range of potential risks and rewards was a subject of
debate in the late 1990s (Cromwell, 1994; Hirschi, 1986; Rengert and Wasilchick,
1985), and remains a live issue today (Bernasco and Nieuwbeerta, 2005; Garcia-
Retamero and Dhami, 2009; Nee and Meenaghan, 2006). In the ten years since
this study was completed (in 2001) a number of detailed studies of burglar deci-
sion processes have been published. As we show in the next section, on the whole
these studies support earlier research, highlighting the importance of factors such
as familiarity, occupancy, visibility, accessibility, vulnerability and potential
rewards as important when choosing a target. Features of the burglar, such as
their age and experience and whether or not they were taking drugs have also
been confirmed as influential. The study reported here traverses much the same
territory as recent research but it used methods that remain, in our view, innova-
tive and produced results that contribute new insights into burglars’ reasoning and
decision processes.
Wilkins and Chandler (1965, p. 22) advise that the key to good decision-
making research is a method that constructs a situation ‘as near to that met in
day-to-day work of the participants consistent with the need to obtain good infor-
mation and controls’. The study’s technique was accordingly designed to balance
the competing demands of tight experimental control with the need to simulate the
‘real world’. One real world advantage of the interactive computer-based scenario
approach is that it affords participants a great deal of flexibility in terms of how
much or how little information they select about a potential target, as well as the
order of selection. This means that participants can stop after one or two pieces of
How house burglars decide on targets 27
information have been elicited, or persist until they find out nearly all there is to
know about a dwelling. Not only does this allow for a comparison of experienced
and novice burglars, a theme of a growing body of research (e.g. Clare, 2011), it
permits the construction of decision trees that can shed light on underlying cogni-
tive processes.
A second advantage of the computer scenario technique is that, by incorporat-
ing different alternatives for each of the cues embedded in the scenarios, and by
carefully varying the combinations of cue alternatives across scenarios, a richer
picture can be painted of the decision process. For example, in one scenario a
participant may select the cue ‘dog’ and be told that there is ‘no dog’. They may
then ask about alarms, and be told that there is indeed one present in the house.
This combination and order may be important both to the decision to burgle (or
not), and to how many steps are needed to make a decision. In another scenario
they may select the same cues in the same or reverse order but be told that this
time there is a dog present but no alarm. A comparison of the rated attractiveness
of these targets could be instructive, as well as when the participant decides to
stop asking for more information, although in the present study the comparison
would be complicated – as it is in the real world – by the presence or absence of
other attractors or deterrents such as whether people are at home or the apparent
affluence of the occupants.

Research on burglar decision making

Motivations for burglary include the need for money, social factors (such as gang
membership, delinquent subcultures, peer approval, status), kicks, thrills, and
rebellion, and boredom (Clare, 2011; Cromwell and Olson, 2009; Hearnden and
Magill, 2004; Nee and Meenaghan, 2006). Cromwell and Olson do emphasize
however that motivations such as thrills and rebellion are characteristic mostly of
young offenders, and that on the whole burglary is a highly instrumental crime.
Other researchers have noted the close link between burglary and heroin use
(Farabee, Vandana and Anglin, 2001; Payne and Gaffney, 2012). In an Australian
study Stevenson, Forsythe and Weatherburn (2001) found that the most common
method for adult burglars to dispose of stolen goods was by directly exchanging
them for heroin.
In their formative study, Bennett and Wright (1984) found that burglars could
be grouped into three categories: 7 per cent opportunistic, 10 per cent highly
skilled, and 76 per cent in-between. Nee and Taylor (1988) rated the burglars
in their study similarly: 4 per cent were opportunistic and spontaneous with low
levels of skill, 12 per cent were highly skilled as they conducted detailed planning
and reconnaissance, and the remaining 84 per cent were in-between. Maguire
(1988) reported similar results, grouping burglars’ sophistication with regard to
decision making under three headings: planners, those who select and inform
themselves about a target well in advance; searchers, those who seek a suitable
target and burgle it; and opportunists, those who are stimulated by an opportunity
to steal.
28 R. Homel, S. Macintyre, R. Wortley
Being a motivated burglar is not enough for an offence: a suitable target must
also be selected (Cromwell and Olson, 2009). Recent research suggests that in
most cases there is at least an element of opportunism in the target selection
process, consistent with earlier research. For example, Hearnden and Magill
(2004) reported that fewer than a fifth of their burglars (14/73) had completely
pre-planned their offences, the majority forming the intention to burgle but with
details such as the precise target and methods to be used decided later. Nee and
Meenaghan (2006) came to a similar conclusion: three-quarters of their burglars
made the initial decision to commit a burglary away from the scene of the crime
and then searched around for a suitable area until they found a target. Garcia-
Retamero and Dhami (2009) explored the cues that burglars relied on and found
that compared to police officers and graduate students burglars were more likely
to rely on only one cue when choosing a target. This, however, may reflect either
opportunism or experience, or indeed both.
The most common means of travel for burglary were walking and the use of
a car (Clare and Ferrante, 2007). Burglars generally do not like to travel far to
commit an offence, and they like familiarity with a location, the ability to obtain
money quickly, and practicalities if walking (Hearnden and Magill, 2004).
However, Clare (2011) found that expert burglars compared to novices were more
likely to travel over three kilometres to their target because of the risks associated
with offending close to where they lived.
Cromwell and Olson’s (2009) study was innovative in that it combined inter-
views with burglars with re-creations of their past crimes while the researchers
observed and asked questions. Their burglars reported a higher degree of plan-
ning and analysis in the interviews than when they were taken back to their
targets, the latter appearing to have been chosen more on the basis of opportunity
than purposeful selection. Nevertheless it seemed that even if the offending was
opportunistic burglars did go through some rational choice process. If they were
presented with an obstacle such as physical security, dogs, or the presence of
residents they were likely to move onto another target rather than deal with the

Burglars use three categories of environmental cues to assess these risk fac-
tors: (a) cues that indicate visibility of the proposed target site, (b) cues that
indicate whether the target site is occupied, and (c) cues that indicate the
degree of difficulty that might be expected in actually breaking into the site.
(Cromwell and Olson, 2009, p. 54)

Since the focus of the present study is how burglars select specific houses to break
into there is value in reviewing what is known about the effects of specific cues,
including those that fall into one of the categories identified by Cromwell and
Olson (2009).
How house burglars decide on targets 29
Level of affluence and inside information
Many studies have found that cues that signify affluence, such as a well-kept gar-
den, general upkeep and décor, or an expensive car in the driveway, were highly
regarded by subjects (Bennett and Wright, 1984; Cromwell and Olson, 2009;
Hearnden and Magill, 2004; Nee and Meenaghan, 2006; Wright and Decker,
1994). Affluence can also be determined through inside information, gained
directly or indirectly. One fifth of Wright and Decker’s burglars chose to break
into the homes of persons they knew, but Budd (1999) reported that more than
half his offenders were either well known to the victim or were at least a casual

Many studies have found that an alarm, or the obvious signs that an alarm is installed,
has a deterrent effect on at least some potential offenders (Bennett and Wright,
1984; Cromwell, 1994; Netherlands Ministry of Justice, 1991; Wright, Logie and
Decker, 1995). On the other hand, the majority of Nee and Taylor’s (1988) subjects
(60 per cent) stated that alarms were not a deterrent. Cromwell and Olson (2009)
reported that the likely presence of alarms or private security was one of the main
reasons their participants chose not to burgle affluent targets, but if there were signs
that the system has not been well implemented, or could be disarmed easily, offend-
ers were not deterred (Hearnden and Magill, 2004). Garcia-Retamero and Dhami
(2009) found that burglars (as the experts) were more likely to pay attention to the
security of the property (or absence of it) than access.

Hearnden and Magill’s (2004) burglars preferred not to deal with dogs as they
could be regarded as a form of occupancy. Large dogs represented a physical
threat while small dogs were often noisy, drawing unwelcome attention to the
intruder. In general, however, the literature is divided as to whether dogs act as
a deterrent (Bennett and Wright, 1984; Krainz, 1990; Nee and Taylor, 1988),
although it seems clear that expert burglars are less likely to be put off (Clare,
2011). As a rough generalization, dogs probably deter about half of all offenders,
but a ‘beware of the dog’ sign appears to have little effect on decisions (Wright,
Logie and Decker, 1995).

Signs of occupancy
Signs of possible occupancy such as a television, radio or lights on inside a
house deter some offenders, particularly novices (Clare, 2011; Hakim and Buck,
1991; Nee and Taylor, 1988). Scarr (1973) and Maguire (1988) found that cues
that signal a lack of occupancy were highly attractive and influential. Nee and
Meenaghan’s (2006) sample of burglars would check a property for occupancy
30 R. Homel, S. Macintyre, R. Wortley
by knocking on the door, ringing the doorbell, looking for lights, or observing
whether there was a car in the driveway or milk on the doorstep. Snook, Dhami
and Kavanagh (2011) compared the effects of eight cues: presence of vehicle,
security system, windows above ground level, curtains above ground level, land-
scaping to hide behind, deadbolt and attached garage. Whether a car was present
was the most critical cue and was the most strongly related to occupancy. If a car
were not present burglars would continue to evaluate the target until another cue
would stop them. In Nee and Taylor’s study a small minority (12 per cent) actu-
ally preferred a house to be occupied since there was more chance of valuables
such as cash, jewellery, credit cards and cheque books being on site.

Locks and security

The majority of studies, including the benchmark study of Bennett and Wright
(1984), conclude that locks are ineffective (e.g. Cromwell, 1994; Cromwell,
Olson and Avary, 1991a; 1991b; 1993; Wright and Decker, 1994). Edgar and
McInerney (1987) observed that the only deterrent effect a good lock may have
is in how long it takes to defeat it, since a delay in gaining entry may increase
the chances of detection. More recently, Nee and Meenaghan (2006) reported
that security cues were not as important as whether the property was occupied.
Burglars in their study reported that the most common reason for abandoning
a recent burglary was that they had been disturbed, not that they were deterred
by insurmountable security. Cromwell and Olson (2009) do, however, make the
important point that while the majority of burglars are not deterred by locks and
security, they would prefer not to deal with them.

Visibility refers to the extent that a house can be observed by neighbours and
people passing by. It also includes the vegetation and physical structures that may
obscure the sightlines of someone approaching a property. The less visible and
more isolated a property the more vulnerable it is. Burglars like cover from shrubs
and trees as breaking in and escape are facilitated (Bennett and Wright, 1984;
Coupe and Blake, 2006; 2011; Garcia-Retamero and Dhami, 2009; Rebscher,
1990). By contrast, all the burglars in Nee and Taylor’s (1988) study expressed
ambivalence about the effect of vegetation.

House location
Research has consistently found that a house situated on a corner is more vulnera-
ble (Repetto, 1974; Taylor and Nee, 1988; Van Dijk, Mayhew and Killias, 1990).
More generally, Bennett and Wright (1984) and Nee and Meenaghan (2006) found
that burglars rated possible targets as more attractive if they had easy access and
multiple escape routes, and that they preferred targets where reconnaissance and
access could be via the rear of the dwelling. Hakim and Gaffney (1995) argued
How house burglars decide on targets 31
that properties that back on to wooded or deserted areas such as railroad tracks are
preferred targets, partly because they have easy rear access.

Cromwell and Olson (2009, p. 43) observe that ‘the concept of limited rationality
proposes that for behaviour to be rational, it does not have to be carefully pre-
conceived and planned or require hierarchical, sequential decision making’. The
literature on burglar decision making suggests that in fact offenders span the range
of decision strategies, from the simplest single cue decisions made as an impulsive
response to an immediate opportunity, to the carefully planned, deliberative and per-
haps complex decisions of the knowledgeable veteran. Heroin using offenders may
be no less capable of reasoned decisions than other offenders, especially since they
are usually single-mindedly focused on stealing money or goods and are frequently
very experienced. Evidence on the effects of a wide variety of features of potential
targets suggests that the perceptions and judgements of many burglars, especially
those with more experience, differ from those of home occupiers or police, and
that houses differ markedly in their attractiveness. Understanding better what drives
these perceptions and judgements is the primary goal of the present study.

Ninety-six participants were accessed through a private organization that assists
the reintegration of persons on release from prisons in Melbourne, Victoria, by
providing short-term dormitory-style accommodation. All of the participants
were male, experienced burglars, and heavy users of drugs, especially heroin.
The ages of the participants ranged from 16 to 29, with the mean a little under 20
years. Two participants had committed three break and enters (B&Es), which was
the lowest number, while four participants estimated they had committed around
100 B&Es, the highest number. The mean was 33. Eight participants were in full-
time employment and 88 were unemployed. Nine of the participants were married
or in a permanent de facto relationship, while 87 were single. None of the eight
who were employed were also married.
No financial incentives were offered to participants. There was some reticence
and an expected level of suspicion from the residents in the early stages of the
data collection. However, within a few weeks the presence of the researcher was
accepted, thanks largely to longer-term residents vouching for the authenticity of
the research project and allaying any fears voiced by prospective subjects.

Development of cues
This study design was restricted to an investigation of decision making in relation
to breaking and entering alone into a private dwelling (house only) in the daytime.
32 R. Homel, S. Macintyre, R. Wortley
From the interviews conducted in the earlier study in Brisbane (Macintyre, 2001)
and a review of the literature, a list was constructed of all the main cues utilized by
burglars in daytime house burglaries. Some cues (such as temperature) were not
included because they are not mentioned often in the literature. Others, such as
property marking (Wright, Logie and Decker, 1995), were not included because
often a burglar has no awareness of them until he has entered a dwelling. Louvre
windows were often mentioned as an attractor in the Brisbane interviews, but
were omitted from the cold-climate Melbourne experiment. The 17 cues utilized
for the study and their alternatives are presented in Table 3.1.
Cue alternatives could act as a deterrent (e.g. dog present), but if an alternative
did not act as a deterrent its effects could vary from neutral to being an attractor.
Some cue alternatives merely stated that something is or is not present, such as
cue 1 (dog): no dog; or dog barking loudly. Cue 17 (street type) was the only one
that had four alternatives, the remaining cues describing two extremes, such as
cue 6 (affluence): house looks expensive; or house is run down.

Table 3.1 List of cues and their alternatives

Cue Alternative Predicted effect

cue 1 (dog) No dog Attractor
Dog barking loudly Deterrent
cue 2 (lighting) Dark Attractor
Bright Deterrent
cue 3 (alarm) No alarm Attractor
Good alarm on house Deterrent
cue 4 (occupancy – lights/ No lights/TV/radio on inside house Attractor
Lights/TV/radio on inside house Deterrent
cue 5 (occupancy – car in Car in driveway Deterrent
No car in driveway Attractor
cue 6 (affluence) House looks expensive Attractor
House is run down Neutral
cue 7 (doors/windows) Doors and windows are sturdy and Deterrent
Doors are old and weak Attractor
cue 8 (locks) All doors and windows have the best Deterrent
locks available
Locks on doors and windows are old Attractor
and weak
cue 9 (garage) Garage looks new and is securely Deterrent
Garage is very old and is unlocked Attractor
How house burglars decide on targets 33

Cue Alternative Predicted effect

cue 10 (fence) House is surrounded by a six-foot wall Deterrent
and all gates are locked
House only partially surrounded by a Attractor
one-foot fence and
gate is unlocked
cue 11 (garden) House can hardly be seen as it is Attractor
surrounded by
trees and bushes
House is an open block, garden is Deterrent
bare, only lawn
cue 12 (location) House is in the middle of the street Neutral
House is on a corner block Attractor
cue 13 (people in the street) The street is very quiet no one is Attractor
The street has many neighbours out Deterrent
washing cars
and mowing lawns etc.
cue 14 (neighbourhood The street is not in a neighbourhood Attractor
watch) watch area
The street is in a sign-posted Deterrent
neighbourhood watch area
cue 15 (weather) Weather is fine Neutral
It is very windy Attractor
cue 16 (inside information) You have no inside information Neutral
From a reliable source you are told Attractor
there could be a
a large amount of cash kept in the
cue 17 (street type) The street is a cul-de-sac Deterrent
The street has a back lane Attractor
The street is a normal suburban road Attractor
The street is a major arterial road – Deterrent
very busy

Cue combinations
A design consideration was that the information generated by the experiment
about the influence of individual cues selected at different points in the decision
process should ideally be maximized. If, for example, one alternative for a cue
was always present when a specific alternative for another cue was also present it
would be impossible to know which alternative was having an effect on the attrac-
tiveness of a target. For this reason the cue combinations were constructed so that
34 R. Homel, S. Macintyre, R. Wortley
there was no significant correlation between pairs of columns in a 20 (cues) × 20
(scenarios) table where each cell contained a 1 or 0 to represent cue alternatives.1
The combinations were created using the Solver command in Microsoft Excel,
initially imposing a fully balanced arrangement with 10 ‘1s’ and 10 ‘0s’ in each col-
umn (scenario), where 0 represented the first alternative for each cue in Table 3.1.
The restriction was that no correlation between columns could exceed 0.430, which
was just under a significance level of 0.05. These constraints permitted solutions
until cue 5 (occupancy – car in driveway), at which point the algorithm could not
continue. When 12 ‘1s’ and 8 ‘0s’ in each column was allowed, the Solver was able
to complete the process, with the exception that 3/190 correlations were just sig-
nificant, a rate considered consistent with the objective of reducing to a minimum
repeat cue-alternative combinations across scenarios.

Computerized information board

Participants were presented with targets using a computerized information board
format, facilitating the monitoring of their information acquisition. The computer
program was written using HTML software and was loaded onto a notebook com-
puter.2 The 20 scenarios were presented via a series of screens on the computer.
The participants could control the presentation of the screens and select as much
or as little information as they wished in order to arrive at a decision. They were
given two practice runs before commencing the experiment proper.

All respondents were given the same instructions:

It is daytime and you are on your own. Look at what is most important to you
first, through to least important. Don’t look at things you normally would
not consider. However, you can look at as much, or as little information as
you need to. You can look at every item if you wish. Pretend you are actually
committing a B&E. Go through the process you would use as if it were a real
B&E. Look at as much or as little information you need to until you are set in
your decision.

For each scenario the computer displayed a list of the 17 cues on the first screen.
The participants were asked to select a cue from this menu by pointing the cur-
sor and then double clicking. They were then exposed to a second screen that
revealed the alternative for that cue for that scenario. For example if the partici-
pant selected cue 1 (dog) from the menu, the second screen would present them
with the alternatives for that cue (dog barking loudly or no dog). On a third screen
they were then asked to make a certainty judgement, that is a judgement about
whether or not they would break into the dwelling. The judgement was made on a
100-point scale, with 100 representing complete certainty and a 0 indicating that
a participant was extremely unlikely to break into the dwelling. After participants
How house burglars decide on targets 35
made a rating they were asked if they required more information. If they answered
‘yes’ they were returned to the first screen and were again asked to select another
cue, this time with the cues that had already been selected displaying the relevant
information. If they answered ‘no’ they were presented with the next scenario.
This process continued until all 20 scenarios had been examined and rated by
participants. It was decided that there was no advantage in randomizing the order
of scenario presentation, so they were presented in the same order for all partici-
pants.3 The program recorded a range of data: the number of the scenario; the total
number of cues selected; the sequence of cue selection; the alternative presented;
the rating made after each alternative was revealed for every selected cue; and the
final rating.

As noted earlier, the mean number of B&Es committed by the 96 participants was
33.03 (SD – 28.53). Figure 3.1 plots the reported numbers of B&Es by the ages of
participants. Four clusters of participants are apparent:

1 over 25 years of age with more than 60 B&Es;

2 between 21 and 25 years of age with fewer than 40 B&Es;
3 under 20 years of age with fewer than 20 B&Es;
4 under 20 years of age with more than 30 B&Es.





Age 22




0 20 40 60 80 100 120
Estimated total number of burglaries committed by each subject

Figure 3.1 Age by estimated number of B&Es for the 96 subjects.

36 R. Homel, S. Macintyre, R. Wortley
The older (>20 years) and younger (<20 years) participants with a high number
of B&Es were probably those who had used B&E as one of their main sources of
income. In contrast, those who had a low number of B&Es were probably only
part-time burglars. Figure 3.1 shows that the number of B&Es is not just a simple
function of age, with some participants under 20 reporting many more burglaries
than those much older.

Cues selected
Figure 3.2 presents the total number of cues selected for each scenario. One might
have expected that participants would choose a large number of cues for the first
few scenarios until they became accustomed to the process, after which fewer
cues would be necessary to reach a final decision. Figure 3.2 shows that this did
occur to some extent, but not in a simple way.
Across the 20 scenarios the mean range was 3.54 to 5.68 cues to reach a deci-
sion. Adding over the 96 participants the lowest number of cues selected was 344
for scenario 9 and the highest was 545 for scenario 11, both scenarios occurring
around the middle of the presentation. Although a slight downward linear trend
is apparent in Figure 3.2, the pattern is more like a sine wave with diminish-
ing amplitude, with a definite levelling out apparent for the last seven scenarios.
Unfortunately the decision not to randomize the order of presentation of scenarios
means that it is difficult to decide whether this levelling out reflected features of
those specific scenarios or was a result of participants settling on a standard heu-
ristic by that point in the experiment.


Number of cues


y = –3.288x + 468.77

200 R2 = 0.096


1 2 3 4 5 6 7 8 9 1 0
11 12 13 14 15 16 17 18 19 20


Figure 3.2 Total number of cues selected for each scenario.

How house burglars decide on targets 37
Table 3.2 shows the number of times a cue was selected, regardless of order. Each
cue could have been selected a total of 1,920 times (96 participants × 20 scenar-
ios). Cue 3 (alarm) was selected 1,538 times. This was the highest selection rate
(80.20 per cent). The least selected was cue 11 (garden) which was chosen 106
times (5.52 per cent). The last column shows the relative popularity of all cues.
Six cues – cue 1 (dog), cue 3 (alarm), cue 4 (occupancy – lights/TV/radio), cue
5 (occupancy – car in driveway), cue 13 (people in the street) and cue 16 (inside
information) – accounted for 67.8 per cent of all selections made.
For any scenario the most cues selected was 14, which occurred on 17 occasions.
No participant ever selected all 17 cues to make a final decision. The minimum
number of cues selected to reach a final decision was one, which occurred on 282
occasions. Single selections involved mainly cue 3 (alarm: n=141) and cue 16
(inside information: n=138). In 88 per cent of 1,920 scenario assessments partici-
pants required seven or fewer cues to reach a final decision.

Table 3.2 Overall number of times a cue was selected

Cue Number of times Percentage Percentage of

selected of possible total selections
selections (96 x made (8687)
20 = 1920)
cue 1 (dog) 1151 59.95 13.2
cue 2 (lighting) 110 5.73 1.3
cue 3 (alarm) 1538 80.2 17.7
cue 4 (occupancy – lights /TV / 773 40.26 8.9
cue 5 (occupancy – car in 640 33.33 7.3
cue 6 (affluence) 385 20.05 4.4
cue 7 (doors/windows) 268 13.96 3.1
cue 8 (locks) 380 19.79 4.4
cue 9 (garage) 164 8.54 1.9
cue 10 (fence) 300 15.63 3.5
cue 11 (garden) 106 5.52 1.2
cue 12 (location) 300 15.63 3.5
cue 13 (people in the street) 1060 55.21 12.2
cue 14 (neighbourhood watch) 193 10.05 2.2
cue 15 (weather) 196 10.21 2.3
cue 16 (inside information) 737 38.39 8.5
cue 17 (street type) 384 20.00 4.4

Total/Mean 3685 26.6 100.0

38 R. Homel, S. Macintyre, R. Wortley
The effects of age and experience
Figure 3.3 shows that there was a negative linear relationship (r = –0.43, p<0.01)
between the total number of cues a participant needed to reach a final decision and
the estimated number of B&Es they had committed. The more B&Es a person had
committed the fewer cues they used, suggesting that the more experience one has
as a burglar the more selective one becomes with information.
Inspection of the relationship between age and number of cues selected did not
throw a great deal of extra light on the role of experience. Participants younger
than 20 were divided into two groups reflecting the two clusters of young offend-
ers identified in Figure 3.1. Inexperienced young burglars required more than 120
cues to reach their decisions, whereas the more prolific offenders required fewer
than 90. The oldest and by far the most experienced group (26–29 years) required
the fewest cues, selecting less than 50 in total. Experience – with the skills it
brings – seems more potent as an influence on the decision process than youth,
despite the risk-taking proclivities and greater impulsiveness of the young.

First cue selected

An important question is what cues were most frequently selected first, since this
would imply a higher level of importance in the decision process. Table 3.3 lists
the cues in descending order of popularity for first selection. Cue 16 (inside infor-
mation) was apparently perceived to be the most important, being selected first
most often. In fact, from Table 3.2 we can see that of all the times it was selected


total number of cues selected


100 y = –0.8462x + 118.42

R2 = 0.2715


0 20 40 60 80 100 20
Estimated number of break and enters commited

Figure 3.3 Total number of cues selected over the 20 scenarios by the total estimated
number of burglaries for each participant.
How house burglars decide on targets 39
Table 3.3 Cues selected first, number and proportion, cues arranged in descending order

Cue Number of times Percentage of total Cumulative

chosen first first selections percentage
cue 16 (inside information) 694 36.15 36.15
cue 3 (alarm) 459 23.90 60.05
cue 1 (dog) 368 19.17 79.22
cue 13 (people in the street) 241 12.55 91.77
cue 5 (occupancy – car in 38 1.98 93.75
cue 17 (street type) 25 1.30 95.05
cue 6 (affluence) 16 0.83 95.88
cue 12 (location) 14 0.72 96.60
cue 10 (fence) 13 0.68 97.28
cue 4 (occupancy – lights/tv/radio) 13 0.68 97.96
cue 8 (locks) 12 0.63 98.59
cue 7 (doors/windows) 12 0.63 99.22
cue 15 (weather) 12 0.63 99.85
cue 14 (neighbourhood watch) 2 0.10 99.95
cue 2 (lighting) 1 0.05 100.00

Totals 1920 100

(737) cue 16 was only requested second or later 5.8 per cent of the time and, as
we saw earlier, in 138 assessments it was the only cue selected. Four cues – cue 16
(inside information), cue 3 (alarm), cue 1 (dog) and cue 13 (people in the street)
– accounted for 91.77 per cent of all first selections.
However, these four cues were not the four most selected in total. Cue 16, cue
4 (occupancy – lights/TV/radio) and cue 5 (occupancy – car in driveway) were
selected approximately the same number of times, but cue 4 was only selected
first 13 times and cue 5 first 38 times, compared to the 694 first selections for cue
16. A number of cues were selected many times, indicating their importance to the
burglars despite the fact that they were rarely selected first.

Deterrent and attractive cue combinations

Table 3.4 shows the properties of the five most chosen cues in the ten scenarios
that received the highest mean final ratings (that is, where the target houses were
judged most likely to be burgled). A tick symbol in a cell means for that cue–sce-
nario combination the deterrent alternative was absent. Scenarios 7 and 17 were
the only ones that had the attractive alternative for the five most chosen cues, and
significantly these scenarios had the highest mean final ratings (80.89 and 86.98).
40 R. Homel, S. Macintyre, R. Wortley
Table 3.5 parallels Table 3.4 but this time for the ten scenarios that received
the lowest mean final ratings, with a tick symbol indicating that a deterrent alter-
native was present. Cue 16 (inside information) was not included in Table 3.5
because there was no deterrent alternative for this cue. Table 3.5 confirms that
when the deterrent alternative was most often present the mean final rating was
lower, indicating that the house was judged as unlikely to be burgled. Scenario
8 was the only scenario where the four most chosen cues had the deterrent alter-
native, and not surprisingly this scenario received the lowest mean final rating
(17.50). Similarly scenario 10 and scenario 18 were the only two with three of the
four deterrent alternatives, and these received the next lowest mean final ratings
(20.47 and 19.48 respectively).
The most striking feature of Tables 3.4 and 3.5 is that the ten least attractive
scenarios all had an alarm, while none of the ten most attractive did. Overall, cue
3 (alarm) was the most frequently chosen and it was chosen first the second most
often. Clearly, alarms loomed very large in the decision making of the partici-
pants. None of the remaining 16 cues were as clear in their effects.

Number of cues selected

There was a positive linear relationship (r = .564, p< 0.01) between the attractive-
ness rating and the number of cues needed to reach a final decision. This suggests
that the participants were easily deterred by a few negative cue values, but when
the first few cue choices were not deterrent alternatives the participants selected
more cues to be sure of their decision.

Table 3.4 The ten highest mean final rating scenarios and whether they had the attractive
alternative for the five most chosen cues

Scenario No alarm No dog No people in Occupancy – no Inside information

the street lights/TV/radio – cash kept inside
on inside the house

1 ✓ ✓
2 ✓
4 ✓ ✓ ✓
5 ✓ ✓ ✓
6 ✓ ✓ ✓
7 ✓ ✓ ✓ ✓ ✓
11 ✓
12 ✓
17 ✓ ✓ ✓ ✓ ✓
20 ✓
How house burglars decide on targets 41
Table 3.5 The ten lowest mean final rating scenarios and whether they had the deterrent
alternative for the four most chosen cues

Scenario Good alarm Dog barking Many people in Occupancy – lights/

loudly the street TV/radio on inside

3 ✓
8 ✓ ✓ ✓ ✓
9 ✓
10 ✓ ✓ ✓
13 ✓
14 ✓
15 ✓ ✓
16 ✓ ✓
18 ✓ ✓ ✓
19 ✓

A decision tree
It was noted earlier that an advantage of the design employed in this study is that by
incorporating different alternatives for each of the cues, and by varying the combi-
nations of cue alternatives across scenarios, a richer picture can be painted of the
decision process. One way of educing the sequences of decisions that were used
by participants, in a fashion that is clearer than has been possible with the analyses
reported so far, is to represent them in branching diagrams or decision trees.
Figure 3.4 shows a portion of a decision tree that grew from cue 16 (inside
information) when it was the first cue selected and the alternative was ‘From
a reliable source you are told there could be a large amount of cash kept in the
house’.4 Figure 3.4 aggregates the decisions of all participants across all 20 sce-
narios, so the ‘n’ at the bottom of each box refers to the number of selections, not
the number of participants. There is also an identification number in brackets at
the top of every box and circle.
To illustrate how to read the diagram, Boxes 1–15 are described briefly. Box 1
shows that on 694 occasions participants chose cue 16 (inside information) first.
For those 694 selections, Box 2 shows that the positive alternative was displayed
419 times. The mean attractiveness rating for these 419 instances was 83.84, mak-
ing the house a highly likely B&E target at this stage. By contrast, on 275 occasions
participants received the neutral alternative (you have no inside information).5 The
mean rating after this alternative was displayed was 58.45, so clearly having inside
information greatly increased the attractiveness of the dwelling.
After Box 2, selections were spread across five different choices. Circle 7
shows that on 136 occasions no more information was requested. On 283 occa-
sions participants made a second choice after learning that valuables were present
(2) (1) 1st choice
Across all scenario(s) and all subjects
(20 × 96 = 1920)
INFORMATION YES Inside information
n = 419 N = 694
mean = 83.84

(3) (4) (5) (6) (7)

Cue 1 Cue 3 Cue 6 Cue 13 No second
Dog Alarm Affluence People in street choice
n = 25 n = 225 n = 21 n = 12 n = 136

(8) (9) (10) (11) (12) (13) (14) (15)

Dog Dog Alarm Alarm Affluence Affluence People People in street
YES NO YES NO Expensive Run down in street NO
n = 12 n = 13 n = 143 n = 82 n = 21 n=0 YES n = 12
Mean = 80.83 Mean = 97.69 Mean = 55.45 Mean = 77.2 Mean = 97.14 n=0 mean = 98.33
[1–95.00] [1–97.69] [1–74.27] [1–73.66] [1–96.67] [1–98.33]

(16) (17) (18) (19) (20) (21) (22) (23) (24) (25)
No third Cue 1 Cue 10 Cue 13 Cue 1 Cue 4 Cue 7 Cue 8 Cue 13 No third
choice Dog Fence People Dog Occupancy Doors and Locks People choice
n = 13 n = 81 n = 12 in street n = 50 lights/ TV windows n=8 in street n=8
n = 37 n = 21 n=1 n=2
(26) (27) (28) (27)
[1–…] previous means for this sub-sample only,
Dog Dog Dog Dog
after first selection.
[2–…] previous means for this sub-sample only,
n = 33 n = 48 n = 36 n = 14
after second selection.
Mean = 30.61 Mean = 47.50 Mean = 69.31 Mean = 64.29
General stopping rule – splitting stops if a subsequent
[1–56.67] [1–61.67] [1–68.06] [1–56.43]
split leaves a subsample with an ‘n’ of less than 30.
[2–46.06] [2–47.50] [2–62.14] [2–62.14]

Figure 3.4 Decision tree when cue 6 (inside information) is selected first and reveals valuable goods in the house.
How house burglars decide on targets 43
in the house, and Boxes 3, 4, 5 and 6 show the next cues requested. Boxes 8 to
15 demonstrate the influence of each alternative for each of the four cues cho-
sen. Box 3 shows that on 25 occasions cue 1 (dog) was chosen, and of those 25
instances the deterrent alternative (dog barking loudly) was displayed 12 times
(Box 8) and the attractive alternative (no dog) 13 times (Box 9). Box 8 shows that
a dog in the house reduced the mean rating from 95.00 to 80.836 while ‘no dog’
produced no change at all in the mean rating.
Figure 3.4 shows just one of the many decision trees that can be constructed
from the data. The value of decision trees is to show the order of subjects’ selec-
tions and the interdependence among cues. The trees illustrate how the effect
of a cue is relative to the presence or absence of the deterrent, attractive or neu-
tral alternatives for other cues. Selections vary depending upon prior information
received. Subjects often use selections to follow a specific line of enquiry (e.g.
how much risk is involved?), seeking to clarify the situation if initial selections
reveal conflicting information.

The 96 participants in this study demonstrated that they were capable of exercis-
ing considerable skill in their trade as burglars. Although it is possible that the
computer-based scenario approach (like Cromwell and Olson’s interviews) pro-
duces results that exaggerate somewhat the degree of rationality actually used by
burglars when on the job, it is nevertheless impressive that they had the domain
expertise to generate such detailed and logical decision sequences.
The method has revealed the critical role of five features of a dwelling that
could make it extremely attractive as a burglary target: knowing that there are
valuable goods inside, that there is an alarm, no dog, no people about in the street,
and no sign of anyone being at home (as indicated particularly by no car in the
driveway but also by lights, TV or radio being off). These were also the cues most
often selected first, but significantly they were by no means equally popular as
first selections. The decision tree in Figure 3.4 illustrates the primary influence
of knowledge about valuable goods, which in about a third of instances resulted
immediately in a move to the next scenario without any attempt to access further
information. This strategy was most characteristic of the experienced burglars,
who generally arrived at a decision on the basis of less information than the inex-
perienced participants (Figure 3.3). Participants in this study were often asked,
‘Why did you select only one cue?’ Their response was generally along the lines:
‘If I know, from a reliable source, that a house has a large amount of cash inside I
don’t care about any of the other information. I will find a way to get in and steal
the money.’ They also stated that often a large amount of cash would indicate a
drug dealer’s house with drugs inside, adding to its attractiveness.
Most decision sequences were of course longer than one step, as Figure 3.4
shows, although overall participants accessed only one-third of the available
information. Even when it was known that valuable goods were inside, in most
cases participants wanted to know about alarms (especially) but also the affluence
44 R. Homel, S. Macintyre, R. Wortley
of the occupants, the presence of dogs or whether there were people in the street.
In the international literature alarms and security seem to have a somewhat uncer-
tain influence on potential burglars, so perhaps their importance to the offenders
in this study is a little surprising. It should be remembered, however, that at the
time of the study only a minority of dwellings in Australia were equipped with
security systems, meaning that there were many other houses that could be bur-
gled without the effort involved in circumventing an alarm.
The mean final attractiveness ratings varied from 17.50 to 86.98 across the
20 scenarios. Moreover, participants arrived at their decision more quickly for
the less attractive targets. Putting this another way, low initial ratings seemed
to encourage participants to move on, especially if they were more experienced.
In terms of prevention this suggests that a home occupant should seek to deter
a potential burglar quickly with a few key cues, otherwise he may increase the
range of information he selects and be harder to deter. Noisy little dogs, alarms,
and children in the street or teenage children with unpredictable routines spring
to mind as possible deterrents. Conversely, if the first piece of information a
potential burglar elicits is that a large amount of cash could be kept in the house,
he will probably be almost impossible to deter. This is no doubt one reason for
the ‘once bitten, twice bitten’ (i.e. repeat victimization) phenomenon (Farrell
and Pease, 1993) – what better way to gain inside knowledge than from a previ-
ous visit?
This study might also assist in the development of prevention strategies
through a risk assessment instrument for houses. Such an instrument might be
able to be made a little more sophisticated than a regression model based on a
weighted sum of cues if it took account of how the house appears to a potential
burglar and the order in which key pieces of information may be elicited. Caution
needs to be exercised when attempting to roll out prevention strategies based on
research from other locations; replication studies have produced mixed results
(Laycock and Tilley, 1995). However, based on the current study, five features of
potential burglary targets (the presence of valuable goods, no alarm, no dog, no
people on the street, and no signs of occupancy) are key risk factors. Clearly these
five features are a logical starting point for any prevention initiative.
However, the primary value of this study might lie not in its implications for
prevention – important as these are – but in its potential to support theoretical
understanding of offender decision processes. As Nee and Meenaghan (2006, p.
935) observe: ‘There is now ample evidence to suggest a model of target appraisal
in the burglar, which is rational and discriminating in nature and involves the kind
of ‘bounded’ decision making described as expertise in other fields of cognitive
science.” The findings of the present study are certainly consistent with this posi-
tion, and in our view interpretation of decision trees like the one in Figure 3.4
could be greatly enriched by theoretical insights stimulated by the cognitive sci-
ence literature, especially models of expert decision making.
The primary strength of the research methods employed in this study was the
tight control over the variables under examination. Because of this, we were able
to isolate and examine the effect of specific cues, as well as various combinations
How house burglars decide on targets 45
of cues. However, it is acknowledged that the research strategy adopted comes
with potential costs in terms of external validity. No simulation can capture the
subtle and fine distinctions among cues nor the stress and tensions involved in
committing a real burglary. It is possible that the remoteness and precision of the
computer-simulated case studies increased accuracy at the expense of realism. For
example, in the study subjects were able to control the order of cues, a situation
that typically does not occur in the real world. All research methods have inherent
strengths and weaknesses. Notwithstanding these acknowledged qualifications,
the current study brings to the field a level of scientific objectivity and rigour that
is often absent in much contemporary burglar decision-making research.

The study reported in this chapter was carried out as a PhD project in the School
of Criminology and Criminal Justice at Griffith University under the supervision
of Ross Homel and Richard Wortley. It was completed in 2001, and has not been
published previously. We are very grateful to Jessica Ritchie for her assistance in
searching and summarizing the recent literature on burglars and burglary.

1 Cue 17 (street type) with four alternatives was represented by four rows, making
16+4=20 rows.
2 Netscape Communicator v4.04 (1994–1997).
3 But see the discussion of Figure 3.2 below: randomization would have permitted
clearer conclusions about the efficiency with which participants came to final deci-
sions as the experiment progressed.
4 The other half of the tree, corresponding to ‘You have no inside information’ is not
5 This half of the decision tree is not shown.
6 The ‘1-’ in the square brackets in Box 8 denotes the mean for these 12 instances after
the positive alternative for cue 16 (inside information) was displayed.

Bennett, T. and Wright, R. (1984). Burglars on burglary: Prevention and the offender.
Aldershot: Gower.
Bernasco, W. and Nieuwbeerta, P. (2005). How do residential burglars select target areas?
British Journal Criminology, 45, 295–315.
Budd, T. (1999). Burglary of Domestic Dwellings: Findings from the British Crime Survey.
London: Home Office Statistical Bulletin 4/99, Research and Statistics Directorate,
Home Office.
Clare, J. (2011). Examination of systematic variations in burglars’ domain-specific
perceptual and procedural skills. Psychology, Crime and Law, 17, 199–214.
Clare, J. and Ferrante, A. (2007). Typical burglaries in WA: Decision-making and processes.
Special Burglary Series No. 5. Crime Research Centre and Office of Crime Prevention.
Cornish, D. B. and Clarke, R. V. (1986). The reasoning criminal. New York, NY:
46 R. Homel, S. Macintyre, R. Wortley
Coupe, T. and Blake, L. (2006). Daylight and darkness targeting strategies and the risks of
being seen at residential burglaries. Criminology, 44, 431–464.
Coupe, T. and Blake, L. (2011). The effects of target characteristics on the sightings and
arrest of offenders at burglar emergencies. Security Journal, 24, 157–178.
Cromwell, P.F. (1994). The burglar’s perspective. In A.R. Roberts (ed), Critical Issues in
Crime and Justice (pp. 35–50). Thousand Oaks, CA: Sage Publications.
Cromwell, P. and Olson, J.N. (2009). The reasoning burglar: Motives and decision-making
strategies. In P. Cromwell (ed), In their own words: Criminals on crime (5th edn)
(pp. 42–56). Oxford: Oxford University Press.
Cromwell, P.F., Olson, J.N. and Avary, W.A. (1991a). Breaking and entering: An
ethnographic analysis of burglary. Newbury Park, CA: Sage Publications.
Cromwell, P.F., Olson, J.N. and Avary, W.A. (1991b). How residential burglars choose
targets: An ethnographic analysis. Security Journal, 2, 195–199.
Cromwell, P.F., Olson, J.N. and Avary, W.A. (1993). Breaking and entering: An
ethnographic analysis of burglary. Studies on Crime and Crime Prevention Annual
Review, 2, 113–121.
Deusinger, I.M. (1989). Questioning convicted burglars: A contribution to crime
prevention. In H. Wegener, F. Lšsel and J. Haisch (eds), Criminal behaviour and the
justice system: Psychological perspectives. New York, NY: Springer-Verlag.
Edgar, J.M. and McInerney, W.D. (1987). Locks: Attacks and countermeasures. Technical
Notebook: How to defeat a lock and how to keep a lock from being defeated. Security
Management, 37, 61–63.
Farabee, D., Vandana, J. and Anglin, M.D. (2001). Addiction, careers and criminal
specialisation. Crime and Delinquency, 47, 196–220.
Farrell, G. and Pease, K. (1993). Once bitten, twice bitten: Repeat victimization and its
implications for crime prevention. Crime Prevention Unit Paper, No. 46. London:
Home Office.
Garcia-Retamero, R. and Dhami, M.K. (2009). Take-the-best in expert–novice decision
strategies for residential burglary. Psychomonic Bulletin and Review, 16, 163–169.
Hakim, S. and Buck, A.J. (1991) Deterrence of suburban burglaries. Rockville, MD:
National Institute of Justice, National Criminal Justice Reference Service.
Hakim, S. and Gaffney, M.A. (1995). Residential burglary patterns. Locksmith Ledger,
Hearnden, I. and Magill, C. (2004). Decision making by house burglars: Offenders’
perspectives. (Home Office Research Findings No. 249). London: Home Office
Research, Development and Statistics Directorate.
Hirschi, T. (1986). On the capability of rational choice and social control theories of
crime. In D. Cornish and R. Clarke (eds), The reasoning criminal: Rational choice
perspectives on offending (pp. 105-118). New York, NY: Springer Verlag.
Krainz, K. (1990). Residential burglaries. Wohnhauseeinbrueche, Beichte des
Kiminalistischen Institutes/ Reports of the Institute of Criminology, Bundeskiminalamt,
Weisbaden Germany.
Laycock, G. and Tilley, N. (1995). Implementing crime prevention programs. Crime and
Justice: A Review of Research, 19, 535–584.
Maguire, M. (1988). Searchers and opportunists: Offender behaviour and burglary
prevention. Journal of Security Administration, 11, 70–76.
Macintyre, S.D. ( 2001). Burglar decision making. Unpublished PhD, Griffith University.
How house burglars decide on targets 47
Nee, C. and Meenaghan, A. (2006). Expert decision making in burglars. The British
Journal of Criminology, 46, 935–949.
Nee, O. and Taylor, M. (1988). Residential burglary in the Republic of Ireland: A situational
perspective. The Howard Journal, 27, 104–116.
Netherlands Ministry of Justice (1991). Domestic burglary: Motives and modus operandi
from the offender’s perspective. The Hague, Netherlands.
Payne, J. and Gaffney, A. (2012). How much crime is drug or alcohol related? Self-reported
attributes of police detainees. Trends and issues in crime and criminal justice, No. 439.
Australian Institute of Criminology.
Rebscher, E. (1990). Crime prevention in the field of burglary: Empirical police research
in the Federal Republic of Germany. Police Studies, 13, 10–13.
Rengert, G. and Wasilchick, J. (1985). Suburban burglary: A time and a place for
everything. Springfield, IL: Charles C. Thomas.
Reppetto, T.A. (1974). Residential crime. Cambridge, MA: Ballinger.
Scarr, H.A. (1973). Patterns of burglary. Washington, DC: U.S. Government Printing
Snook, B., Dhami, M.K. and Kavanagh, J.M. (2011). Simply criminal: Predicting burglars’
occupancy decisions with a simple heuristic. Law and Human Behavior, 35, 316–326.
Stevenson, R.J., Forsythe, L.M.V. and Weatherburn, D. (2001) .The stolen goods market
in New South Wales. British Journal of Criminology, 41, 101–118.
Taylor, M. and Nee, C. (1988). The role of cues in simulated residential burglary. The
British Journal of Criminology, 28, 396–403.
Van Dijk, J.J.M., Mayhew, P. and Killias, M. (1990). Experiences of crime across the
world: Key findings of the 1989 International Crime Survey. Deventer, MO: Kluwer.
Wilkins, L.T. and Chandler, A. (1965). Coincidence and competence in decision making.
British Journal of Criminology, 5, 22–35.
Wright, R.T. and Decker, S. (1994). Burglars on the job: Street life and residential break-
ins. Saint Louis, MO: Northwestern University Press.
Wright, R., Logie, R.H. and Decker, S.H. (1995). Criminal expertise and offender decision
making: An experimental study of the target selection process in residential burglary.
Journal of Research in Crime and Delinquency, 32, 39–53.
4 The risks and rewards of motor
vehicle theft
Implications for criminal persistence
Heith Copes and Michael Cherbonneau

The risks and rewards of auto theft: implications for

criminal persistence
The 1980s brought about an increased emphasis in understanding the ways
offenders contemplate and decide to commit crime. This renewed emphasis on
exploring the notion of strategic thinking and decision making among offend-
ers changed the way academics and policymakers thought about crime and
criminality. At the forefront of this movement was Derek Cornish and Ronald
Clarke’s (1986) edited collection, The Reasoning Criminal. Here, and elsewhere,
they argued the importance of using a “rational choice” perspective to under-
stand offenders’ involvement in specific forms of crime and ways they commit
discrete criminal acts (see also Clarke and Cornish 1985; Cornish and Clarke
2008). One reason the rational choice approach was so well received was that
its advocates found a way to incorporate the writings and ideas of scholars
from a variety of academic backgrounds and interests. Equally appealing was
the emphasis on criminal behavior as a byproduct of reasoned action on the
part of the offender, which offered those tasked with preventing crime action-
able suggestions for developing effective crime control policies. In the climate of
“nothing works” to deter crime, giving clear and effective policy was welcomed.
Those who took up the mantle of uncovering the decision-making process of
offenders now had a clear framework to guide their research and interpret their
results. Whether they were interested in elaborating on why people decide to
embark on a career in crime, the various decisions made to carry out an offense,
or the cognitive aspects of the decision-making process itself, the rational choice
perspective provided the rationale for exploring such matters and offered a clear
and simple means for making sense of the findings and uniting them under a
coherent framework. Such insight led to a growing body of literature that explores
the involvement and event decisions of those who committed a wide range of
crimes (e.g., Beauregard, Proulx, Rossmo, Leclerc and Allaire 2007; Bennett and
Wright 1984; Copes and Vieraitis 2012; Corbett and Simon 1992; Cromwell,
Olson, and Avary 1991; Feeney 1986; Gill 2000; Jacobs 1999; Paternoster and
Simpson 1993). Much of this research focused on event decisions; that is, scholars
sought to describe the various choices offenders make when searching for targets,
Risks and rewards of motor vehicle theft 49
enacting the crime, and disposing of goods and other proceeds. Armed with this
knowledge, significant improvements in crime prevention were made. For exam-
ple, information about the “crime scripts” of auto burglars has allowed managers
of car parks to reduce these crimes by better designing parking lots (Michael,
Hull, and Zahm 2001).
Criminologists working in the tradition of rational choice have also sought
to model the subjective decisions offenders make when choosing or abstaining
from crime. These involvement decisions include the factors offenders weigh
when deciding to begin, continue, and desist from crime. Interviews with street
offenders suggest that the decision to begin crime is motivated by a combination
of background (e.g., community disadvantage, structural strain, broken homes)
and foreground factors (e.g., peer approval, status, sensation seeking, financial
desperation, alcohol/drug use, participation in “streetlife”) (e.g., see Jacobs and
Wright 1999). Conversely, the decision to give up crime can be sudden or gradual
(Cusson and Pinsonneault 1986). Sudden desistance is often due to a “shocking”
event such as being injured or injuring another in the course of crime. Gradual
desistance occurs as offenders begin to reassess the long-term costs and benefits
of crime alongside changes in other important domains of their lives (e.g., gainful
employment, marriage, or parenthood). In light of these social and maturational
changes the risks of crime seem more daunting than they once were and the
rewards lose the luster they once had (Shover 1985).
Despite the call to look at all three stages of involvement (see Clarke and
Cornish 1985; Cornish and Clarke 1986, 2008), it appears that most research
has focused on those factors that lead to initiation or desistance. Undoubtedly,
knowing the reasons why people are drawn to and move away from crime are
important, but so too is knowing what keeps people on a criminal path. Here we
use a rational choice framework to understand how offenders (specifically auto
thieves) assess the risks and rewards of crime with an emphasis on how these
assessments contribute to criminal persistence. Specifically, we examine how
offenders’ lifestyles, development of skills, changing social networks, and excuse
making encourage them to continue offending because doing so is seen as reason-
able, if not rational. While this research was undertaken in part to learn about the
viewpoints of a particular type of thief (those who steal vehicles), we think that
the findings can inform discussions of criminal habituation in general.

Rational choice perspective in context

Grounded in the economic theory of expected utility, early rational choice theo-
ries depicted offenders as “pure” rational calculators who carefully considered
the risks and rewards of behavior before choosing a course of action (e.g., Becker
1968). The decision to engage in crime, like all other decisions, is often preceded
by a decision-making process where individuals assess available options and their
potential outcomes, while paying attention particularly to the odds and severity
of aversive consequences (Shover and Hochstetler 2006). Economists, cognitive
psychologists, and many others within criminology proposed models of criminal
50 Heith Copes and Michael Cherbonneau
decision making that were often stated in mathematical terms and grounded in
these assumptions (e.g., Becker 1968; Carroll 1978; Reynolds 1985).
Contemporary theorists, however, argued that this depiction of offenders’ deci-
sions was inconsistent with the suboptimal decision strategies offenders employ
and the “opportunistic, ill-considered, and even reckless nature of much crime”
(Clarke and Felson 1993:5). Therefore, to better understand the decision-making
process of offenders, criminologists began examining the ways offenders evalu-
ate their options and choose crime within a sociocultural context. This body of
research examined the components of criminal decisions, including the choice
to offend instead of pursuing legitimate activities, the target selection process,
and individuals’ perceptions of various rewards and costs of crime. Such lines
of inquiry led researchers to develop a more cautious and subjective understand-
ing of criminal choice, or what has been termed limited or bounded rationality
(Clarke and Cornish 1985; Opp 1997; Shover and Honaker 1992). The assump-
tion these scholars make is that a person’s social position influences the ways they
interpret the risks and rewards of crime. For example, crime offers different risks
and rewards to the down-and-out and desperate than it does for those from less
humble means. As such, if we wish to understand the decision-making process of
offenders “it is useful to examine the worlds in which much of their life is spent”
(Shover 1996:93; see Jacobs and Wright 1999). To do this it is necessary to situate
their decisions within the lifestyle that frames their choices.
Decades of research on persistent street offenders shows that they typically
lead lifestyles that encourage the enjoyment of good times and the dismissal
of all that is restrictive (Shover 1996; Shover and Honaker 1992; Wright and
Decker 1994; 1997). This lifestyle promotes self-indulgence and is characterized
by adoration of autonomy, action seeking, and the promotion of material excess,
especially when it comes to drug use, mating partners, and personal style (Jacobs
and Wright 1999). With such a worldview offenders frequently find themselves
strapped for cash and desperate to find ways of sustaining their lifestyle. This
worldview also erodes their ability to obtain money through legitimate means.
Few are qualified for well-paying jobs and most cannot hold on to the low paying
ones (Fleisher 1995). Some have burned family and friends to a point that they
can no longer go to them for help (Shover and Honaker 1992), which cements
their social isolation and engulfs them deeper into criminal roles. In sum, the
erosion of legitimate resources for earning or obtaining money, their social isola-
tion from conventionality, and the interpersonal constraints of significant others
contribute to them interpreting criminal behavior as a reasonable solution to their
problems. This is the case whether crime is relied on to resolve immediate crises
or chosen as an ongoing repertoire.
Those who espouse bounded models of criminal decision making assume
that people subjectively assess the costs and benefits of a line of action and
that their assessments are malleable. Under the right circumstances, “risks that
once deterred become manageable and rewards that were previously overlooked
become powerful lures” (Copes and Vieraitis 2009:243). Offenders’ changing
assessments of risks and rewards suggest that these subjectively weighed options
Risks and rewards of motor vehicle theft 51
are instrumental in determining whether they persist with crime or turn away from
it altogether. Offenders habituate to crime due to “changing conditions and per-
sonal circumstances that confirm the offender in his readiness to commit [crime]”
(Clarke and Cornish 1985:170). Previous explorations into criminal persistence
have pointed to a variety of crime facilitative factors, which include an increase
in professionalism, a change in lifestyle, and a change in peer and associate net-
works (Shover 1996).
Regardless of the type of decisions being modeled, Cornish and Clarke (1986)
argued that it is important to remember that relevant factors in the decision-mak-
ing process will vary among crime types. Those factors that lead to initiation,
habituation, and desistance for robbery are not the same as those that lead to
property crimes such as burglary, fraud, or motor vehicle theft, regardless of
the lifestyle they lead. Thus, it is important to be crime-specific when modeling
criminal involvement decisions. Such crime-specific approaches have been used
as a theoretical framework to shed light on a variety of street crimes including
burglary, low-level drug sales, and robbery (Bennett and Wright 1984; Cromwell
et al. 1991; Jacobs 1999, 2000; Shover and Honaker 1992; Wright and Decker
1994, 1997). Historically, fewer have examined the involvement decisions of auto
thieves, though there are some recent exceptions (Cherbonneau and Copes 2006;
Copes 2003a; Fleming 2003; Mullins and Cherbonneau 2011).
Here we elaborate on those factors that contribute to the habituation of auto
theft. Specifically, we discuss the ways auto thieves articulate the rewards of
stealing cars, the ways they justify their offending, the risks they associate with
their crimes, and the strategies they use to manage these risks. We then focus our
discussion on how these factors act in concert to give their decisions sustenance
and rationality. Thus, we add to our current understanding of motor vehicle theft
in particular, but more importantly we shed light on those factors that encourage
the habituation of crime, in general.

The current findings are based on insights and analysis of ethnographic data
collected from auto thieves over the course of three projects (see Cherbonneau
and Copes 2006; Copes 2003a; Copes and Cherbonneau 2006; Jacobs and
Cherbonneau, 2012; Mullins and Cherbonneau 2011). In total we interviewed
89 individuals who had admitted to stealing cars. The first study consisted of
interviews with 42 individuals on community supervision in one metropoli-
tan area in Tennessee. All of whom were under probation or parole following
convictions for various property crimes and had committed at least one motor
vehicle theft. The second study involved interviews with 12 auto thieves incar-
cerated in two medium security prisons in Louisiana. These individuals were
part of a larger study where the authors were interviewing auto thieves and car-
jackers about various aspects of the criminal decision-making process. The final
study consisted of interviews with 35 active auto thieves in a large metropolitan
city in Missouri.
52 Heith Copes and Michael Cherbonneau
While each of the projects varied in their primary intent, they all shared a com-
mon line of questioning. Namely, we asked respondents about specific aspects of
auto theft, including their motivations for committing motor vehicle theft, their
process for selecting targets, their perceptions of the risks and rewards of steal-
ing cars, the ways they managed the risks and fears associated with their crimes,
and the techniques and skills they used to accomplish their crimes. To provide
further details of the crimes we asked them to describe specific auto thefts. The
interviews were conversational by design, but questions were structured in ways
that allowed substantial probing into important topics and of details to occur.
Interviews varied in length, but none lasted more than two hours, with the major-
ity of interviews lasting approximately one hour. All interviews were recorded,
with the permission of participants, and later transcribed.
Across the samples, respondents ranged in age from 17 to 58 (mean age was
29). Fourteen participants were White, 72 were African American, two were
Hispanic, and one was Asian. All but eight were male. Some reported no previ-
ous arrests for auto theft, while 34 respondents had one or more convictions for
auto theft either as a juvenile, adult, or both. The number of auto thefts committed
varied significantly with some respondents being relatively inexperienced (com-
mitting only a handful of thefts) and others being quite prolific (stealing in excess
of 50 vehicles). In inquiring about their other criminal behaviors and looking
at their official records (when available) it appears that these auto thieves are
not specialists, at least not in the long term. Like other persistent offenders they
admitted to engaging in a variety of crimes (Shover 1996); though it was clear that
many would meet criteria for short-term specialization of auto theft.
We recognize that having three separate sampling strategies is not the norm.
However, we believe that such a strategy is advantageous. First, the samples
include auto thieves from three states and disparate backgrounds. As such, the
results are not unique to a single group of thieves or geographic region, which
we believe increases the generalizability of the findings. Second, we included
offenders who were under correctional and community supervision as well as
those who were not. While scholars do not agree on which type of offender is best
suited to shed light on criminal decision making (Copes, Hochstetler, and Brown,
2013; Jacques and Wright 2008), by including offenders in prison, on parole, and
not under supervision we are able to overcome the limitations of any one type of
sample. Overall, the inclusion of auto thieves from different regions and from
different levels of formal supervision increases the validity and generalizability
of our findings. That is, we find similar decision-making strategies regardless of
how they were recruited or where they hailed from.

Perceptions of the rewards of motor vehicle theft

Much research that addresses the motivations for stealing cars provides simple dis-
tinctions between professional auto theft (profit-oriented) and amateur auto theft
(recreation-oriented) (Challinger 1987; Clarke and Harris 1992). Undoubtedly,
the primary motives for stealing cars are money and joyriding. Like others, we
Risks and rewards of motor vehicle theft 53
argue that the motivations and benefits of stealing cars are best understood in the
context of the hedonistic culture of the street in which persistent auto thieves find
themselves. Examining the accounts of auto thieves in the context of this lifestyle
provides greater insights into why they perceived stealing cars as rewarding.
This lifestyle of “ostentatious consumption” entails major expenses, and par-
ticipants often found themselves in a self-defined desperate need of replenishing
their cash (Shover 1996:94). The auto thieves spent the spoils of their criminal
ventures with unconcern. As one auto thief said, “I don’t treasure the money that
I make. I don’t even try to save it. All down in the end I know I won’t be able to
save it anyway.” For these auto thieves, money was not something to be guarded
and squirreled away; it was to be spent immediately and frivolously.
A significant portion of those we spoke with said that money was the primary
motive for their crimes. Profits from stolen cars went a long way in bankrolling their
chosen lifestyle. Those motivated by financial rewards claimed that they could earn
anywhere from $500 to $5,000 per car, depending on the quality of the vehicle sto-
len or parts being sold, the buyers, their role in the theft (if they were part of group),
and their position in the chop shop hierarchy. In fact, the perceived ease at making
fast money persuaded some to quit the “drug game” and focus on stealing cars.

I was selling drugs and got tired of selling drugs. A friend of mine, he told
me I could make more money and it would be easier to take the cars than to
sell the drugs. … [Selling drugs] was just so hectic so I stopped to steal cars
because it was easier. The money come quicker.

Echoing this another said:

’cause when I’m really down I can always steal something nice and get a
nice little profit off of it. … With these rims, you can sell these [for] a quick
$1,000, $1,200. [In-car] TVs, you can sell them [for] $300, $400.

If auto thieves have the necessary skills and proper connections, the offense can
easily finance their lifestyle by providing a viable source of income (Mullins and
Cherbonneau 2011). The importance of having proper connections was evidenced
by the fact that many of those who did not sell stolen cars said that they would
have if they had the right connections. When asked whether he had ever sold any
of the cars one offender responded, “No, I wouldn’t mind doing that though. I
didn’t know of any chop shops.” Similarly, another replied, “I don’t know any
chop shops … they [the police] caught up with them and they’re closed down,
that was seven or eight years ago. But man, I don’t know none today. … I wish I
did.” With no connections, these individuals were limited in the way they could
dispose of the stolen car.
Consistent with Sutherland’s (1947) ideas of differential association, the
thieves were not only able to learn the techniques and rationalizations of the crime
but also provided the proper connections to sell the stolen cars. These individuals
spent much of their time hanging out in bars and on street corners, which helped
54 Heith Copes and Michael Cherbonneau
establish their connections in the underworld. When asked how he knew where to
sell stolen cars one offender said:

Basically, when you do drugs or if you sell drugs, in any large amount those
kinds of people run in circles. Drug dealers, chop shops, you know everybody
knows everybody in the underworld. Even if that person that you dealing
with don’t know, they know somebody [who does]. The underworld is pretty
well connected.

While knowledge of chop shops secured higher payoffs, offenders without these
connections could still profit from auto theft by stripping cars and selling the parts
in a loosely structured network of friends and acquaintances (Fleming 2003). As
one offender said:

Sometimes [we] sold the parts, sometimes [we] just put it on our car. But
most time we’ll strip the car all the way down to the engine and sell the
engine. When we didn’t know about going to sell the cars [to chop shops],
that’s what we were doing. We were selling body parts.

Some took orders for parts or stole cars with items that they could dispose of eas-
ily. “I can make more money off a car that I get ’cause I can sell the rims and TVs
and speakers and all that shit.” Others put the word out that they had merchandise
to sell.

They know that I got some parts and they’ll tell who[ever], they’ll just go
around. If they see somebody they know that like rims and stuff like that then
they ask them do they wanna buy a radio, rims or whatever, and if they do
they’ll come and tell me. They’ll bring the person who want to buy it to me and
I give them a little money for, you know what I’m saying, bringing them to me.

Auto theft offered them the ability to continue their party pursuits by providing the
necessary financial resources and by allowing them to travel when the desire arose.
In addition to providing money, stealing cars also allowed participants to sus-
tain their lifestyles in other ways. In a world where appearance is everything, auto
theft provided the ability to keep up displays of conspicuous consumption. If an
offender wants drugs they can rob drug dealers or pharmacies (Jacobs 2000). If
they want cash they can engage in armed robbery (Wright and Decker 1997). If
they want to travel in stylish vehicles they can steal cars. Auto thieves fulfilled
these material desires by either driving these stolen cars in their neighborhoods
and other public places where they will “be seen” (i.e., joyriding) or by stripping
the cars for parts to install on their own vehicles (Copes 2003a).
Owing to detection risks, those who stole to joyride only kept the car for a short
time, usually under three days, but sought to make the most of their time. They
used the stolen vehicle to continue the good times by visiting friends, picking up
girls, or to just be seen. For a significant number of offenders who were intent on
Risks and rewards of motor vehicle theft 55
cruising around in a stolen car, stealing a suitable one was a prerequisite. They
searched for vehicles that fit the style and image they wished to project. Some
avoided stealing cars that were wrecked or too old.

Oh yeah, I got to ride in a 2000 or better. Like for my birthday, I had a

Magnum for my birthday. You know what I’m saying, I had to ride nice for
my birthday, you know. I mean, I just like to ride good and enjoy myself

In other instances, joyriders did not always search for vehicles fitting the image
they wanted to portray. Rather, they stole cars for purposes other than cruising
(e.g., immediate transportation or financial gain) but during those searches found
vehicles that were considered too nice or flashy to not show off and flaunt for a
time. In these cases, offenders modified their immediate intents when good for-
tune shined favorably on them.
Although there were several instances where joyriding continued well into
adulthood, it was most clearly an age-graded activity that became less frequent as
many in the sample aged and progressed in their careers. Interestingly, this shift
often began around the time when offenders were able to buy cars of their own.
For example, a twenty-five year old male offered the following:

[We would joyride] when we were younger [but not anymore]. … Because I
guess now we old enough to have cars, and we’ve got cars, and I mean, it’s
not really a big thing. Why joyride now? We’re not juveniles anymore nei-
ther. So why drive around in a hot box like that and get jammed up for grand
theft auto and do some time for real?

Those who now viewed joyriding as too risky or too foolish did not necessarily
cease their auto theft participation altogether. Rather, they often redirected their
abilities toward stealing vehicles to finance their lifestyle and/or to cannibalize for
parts and desired accessories to enhance their own cars (e.g., rims, stereos, televi-
sions). In regard to the latter adaptation, one offender explained, “I took a car and
demolished it to build up my car, you know what I’m saying. I never did steal a
car to sell or anything.” Another said:

I got a Pontiac and I see you got a Pontiac, and my fenders are bent up [then]
I’ll take your car and take the fenders off and put it on my car. Do the paint
up real quick and go ditch yours. Nothing might be wrong with yours. Yours
might be brand new. Mine is second hand. But I’d jack your car to take the
body parts to put on mine. Make my shit look good and just ditch yours.

While offenders were interested in various vehicle parts and accessories such as
rims, televisions, custom seats, and interior dressings, it was stereo systems that
ranked highest on the list of sought after accessories. “Every now and then I might
take a little radio or some music out of them. Keep a little music or whatever.”
56 Heith Copes and Michael Cherbonneau
While impulsiveness is often portrayed as a psychological shortcoming in the
criminology literature, a measure of it and other indicators of a “devil may care”
attitude are respectable in street-offenders’ surroundings. It is spontaneity and
action, not reserve, that brings about “good times” for those immersed in this life-
style. This is likely the reason many described auto theft as “fun,” “thrilling,” or a
“high.” As one offender recalled, “I mean it’s like just about as good of a rush as
snorting a foot long line of cocaine … especially when you have people with you.
They are all hyper and everything, and you enjoy it.” Another said, “I liked the
shit—the speed, the flashers, going and doing something, getting into mischief or
trouble. I like it, it’s cool you know what I’m saying? It’s a thrill if you ask me—a
big thrill.”
Auto thieves’ thirst for excitement is evidenced by their desire to steal cars and
drive them recklessly (Copes 2003a; Fleming 2003; Halsey 2008). Cars were sto-
len to race, test drive, tear up, or engage in dangerous car stunts. In other words,
to “just raise hell.” As one offender said, “When I was younger, [we stole them]
for joyrides, demolition derbies. Steal a car and tear it up.” The inherent thrill
that came with successfully stealing cars was said to be a major motivation for
doing so. Being successful at crime provided the auto thieves with a sense of
accomplishment. By getting away with auto theft they were able to do things most
people could not. This is illustrated by the following offender:

But, man really though, it was like, it was more of like, a thrilling thing to me.
To be able to get away with it. I mean, it would just give me goose pimples.
Man I mean, it was like I pulled this off and I made this, you know. How I
mean I would manage to keep from getting caught.

Although the financial, instrumental, and psychic benefits of their crimes are
important, it is common for offenders to interpret their actions in a way that they
can maintain a positive self-image when violating the law. Consequently, par-
ticipants claimed that they would not engage in just any crime. Most agreed that
hurting someone for money was beyond their capacity. This mindset is consistent
with Sykes and Matza’s (1957) contention that when offenders contemplate com-
mitting criminal acts they find ways to neutralize the guilt associated with their
actions. While there are a number of ways that offenders can justify or excuse
their crimes, auto thieves relied primarily on: denial of victim, appeal to higher
loyalties, denial of responsibility, and denial of injury (see Copes 2003b).
One way that these auto thieves justified their crimes was by denying the
victim. That is, auto thieves claimed that some car owners were foolish, stupid,
or careless in the protection of their property and were, thus, deserving targets.
These offenders were alert to emergent opportunities and quick to take advantage
of any mistake such as vehicles that were parked and unlocked, vehicles with
the keys inside, or vehicles that were unattended and left running (Copes and
Cherbonneau 2006). One offender summed up the dynamics of such opportuni-
ties nicely when he said, “I would capitalize on their carelessness.” Whereas some
offenders saw their victims as deserving because they were foolish or careless in
Risks and rewards of motor vehicle theft 57
ways of handling and protecting property, others were seeking revenge against
people who treated the offender poorly. They insisted that the victims do not truly
deserve victim status. As a passenger in a vehicle he would eventually steal, one
thief said the theft stemmed from the disposition of the vehicle owner and also
having to wait outside in the car while the owner went into a house:

I’m thinking the guy gay. So now, I don’t like gay guys you know what I’m
saying, so now, I’m gonna take his goddamn car. “Fuck him. He got us sitting
out here in this hot ass sun and he’s in there drinking Kool-Aid or whatever.
Fuck him. Let’s go.”

Auto thieves also stole cars as a way to get back at the people who insulted or
belittled them. For instance one offender stated:

I was at this bar. I was with this dude, he was drinking and kept on putting me
down—talking about what I could have had. He had questioned me about my
education and stuff. What I could have had if I went to college and all that. He
had a nice car and stuff. I’m a good con artist. I done talked him into letting
me drive. I got the keys and he’s in the bar drinking and stuff. I sneaked off
with the car.

Such offenders claimed that victims deserved what happened because of their
own actions.
Another common neutralization technique used by respondents was the appeal
to higher loyalties. They were cognizant that what they were doing was wrong;
however, they argued that auto theft was something they had to do to support their
families. “Like I say, I’m a family man when you get past all that there. I bring my
money home.” Another offender echoed this sentiment:

I mean look, I need to do this here. I mean my pocket is all flat. I got to go
[steal]. I got a son. I got to provide for him. … No, I ain’t had no guilt, no
shame because every time I felt like a little guilt mood was coming I put my
son in front [of me] and I’m doing that for him. I’m doing that for him. [Guilt]
just disappear.

Denial of responsibility was a third neutralization technique used by respondents.

This technique allowed offenders to escape guilt by blaming their behaviors on
extenuating circumstances. Many reported that they only committed crime when
they were under the influence of drugs or alcohol. As one drug-addicted offender
said, “Crack have you do some crazy shit, man.” Others report that they reluc-
tantly went along with or were coerced into stealing by friends. One joyrider
said that he “just got wrapped up with a stolen car” and that he was unaware of
his friend’s intention to steal a car until it was “too late to back out.” In these
accounts, car theft “wasn’t something planned—it just happened.” It was beyond
their control. These offenders claimed they had no intention of stealing a car, but
58 Heith Copes and Michael Cherbonneau
the next thing they knew, they were driving or riding in a stolen car. Victims of
circumstance, they shouldered less blame as a result.
A final technique respondents used to excuse their thefts was the denial of
injury. By convincing themselves that no one was physically harmed, offenders
minimized the seriousness of their illegal behavior. Some relied on the fact that
car owners had insurance to validate their belief that no real harm was caused.
One offender, who set some of the stolen cars on fire, said, “I guess burning it was
actually like doing the person a favor in a way. If you burn the car at least they
get all their insurance.” Some car thieves did their best to minimize damage to the
car. Others tried to leave the car in a spot where the police can find it or bring it
back to the spot from which they stole it. “I just figured I’d steal a car and bring it
back. I didn’t realize it was going to be more to it than it was.” These respondents
claimed that no real harm came from their thefts because they were merely “bor-
rowing” the car. They believed that because they did not financially profit from
the thefts and the owners were able to get their vehicles back no real harm was
caused. The words of one car thief perhaps best illustrate this belief:

The police are going to find [the car]. One time, one of my friends wanted to
just ditch [the car] in the river. I said, “No we not. We are not ditching that
car.” And I meant that. They [were] going to find that car. They goin’ to be
happy. We happy, let them be happy. They goin’ to find it. [My friend said]
“They got insurance on it. Let’s dump it in the river.” [I said], no, we not
goin’ that way and you not either. They goin’ to get their car back.

Perceptions of the risks of motor vehicle theft

Regardless of what motivated them, those we spoke with indicated that they sel-
dom dwelled on the potential risks of their crimes (Copes and Tewksbury 2011;
see also Shover 1996; Wright and Decker 1994, 1997). In fact, few stated that
the possibility of being apprehended by police was the most pressing risk they
associated with stealing cars. Even those who did actively consider the risks of
apprehension thought it was unlikely to occur in the short term (Shover 1996).
Such a belief, as well as a lack of concern over apprehension, was supported
by both their own experiences with sanctioning practices for auto theft (or lack
thereof) and knowledge of the experiences of others. That is, the perceived inepti-
tude of the criminal justice system to detect and sanction auto theft at rates suitable
for deterrence combined with their personal success at evading arrest reinforced
beliefs about the unlikeliness that such eventualities would befall them. For some,
this realization came early in their careers and sparked inspiration to explore the
offense further:

I had a few older friends that was doing it. You know, I rode with ’em before
[but] I didn’t know it was stolen. … [O]nce I found out at the end of the day
that it was stolen … it was like the police weren’t tripping off of stuff like
this, so it made me think, this is easy. I won’t get caught.
Risks and rewards of motor vehicle theft 59
Unsanctioned experiences such as these, coupled with their continued success
helped to shake inexperienced offenders of their perception that engaging in crim-
inal acts (in this case auto theft) threatens a high certainty of arrest.
The apparent monotonic relationship between evasion and perceived risk of
arrest was especially pronounced in offenders’ calculations as they progressed in
their careers. Some offenders reportedly maintained prolific careers at auto theft
that were largely uncomplicated by the criminal justice system. Such was the case
of one offender who claimed to have stolen over 200 motorcycles and at least 50
cars without ever being arrested. As he continued his thieving with impunity, any
concerns about arrest became remote. Even those who had been arrested were
able to minimize the deterrent effect of formal sanctions. As one offender said:

Shoot, I’m not going to lie, I can’t count [the cars I stole]. Just that many. …
If I got caught for all the crimes that I did I’d never see daylight. I was only
caught five times. I been doing it since 1974.

Being “only caught five times” was seen as a mild exception to the rule, and not
something for which the individual was concerned. Thus, it appears that because
of their success at evading law enforcement many thought they had little or no
need to fear being arrested. Even when apprehension did occur, it was usually
buttressed between substantial periods of unsanctioned offending; thus, offenders
perceived it as the exception rather than the rule and not necessarily a reason to
discontinue their crimes.
When asked what they thought would happen if they were caught, many
assumed that the consequences for their actions would be minor, allowing
them to dismiss easily the thought and continue on with their plans (Copes and
Tewksbury 2011; Fleming 2003; Light, Nee and Ingham 1993). Others believed
that the gravity of the punishment imposed would depend on their role at the
time of capture; namely, with more leniency granted to those in group-based
offenses. For example, if they were apprehended while someone else was driv-
ing, many believed that they would receive a lesser charge: “The person driving,
he gonna get a little bit more because he’s behind the wheel. The passengers,
they gonna get lesser time ’cause they just passengers.” Apprehended on foot
following a vehicle chase, one offender told arresting officers he was in the
back seat (even though he was the driver during the pursuit). He was able to
escape the charges completely whereas the remaining passengers apprehended
with him, did not fare as well:

All of us got the same charge. But my … lawyer got mine threw out because
I said I was in the backseat. If you in a stolen car and you in the backseat, you
getting off. It’s the person who driving and the person in passenger side and
the oldest.

Regardless of where they were sitting (or claimed to be sitting) upon detec-
tion, others went so far as to say that as a passenger, they could simply declare
60 Heith Copes and Michael Cherbonneau
ignorance of the vehicle’s illicit origins. According to a female offender who was
pulled over in a car she helped to steal with two men and another female:

They arrested them [male occupants] and just kinda like talked to us [female
occupants] and let us go so we weren’t the ones driving the stolen car. I
played it off like I didn’t know it was a stolen car and they just came and
picked me up.

While the anticipation that even if consequences did arise they would be minor was
expressed by most, this was an especially common sentiment among the younger
thieves we interviewed who viewed their juvenile status as an insulator against seri-
ous consequences. This finding is not only supported in the literature (e.g., Fleming
2003; Light et al. 1993) but articulated in the narratives of older offenders as they
made sense of the way they approached crimes committed as adults:

When were young and juveniles it was just a slap on your wrist. Go over
there on [street where local juvenile detention facility is located] and wait for
a weekend until your momma come get you, go to your court date, get you a
probation officer or something for it. You good, you know. But now you’re
older now. I mean you’ve got to think smarter now. That’s way the little
homies drive like that [i.e., joyride] ’cause if they get jammed up in it, they’ll
do a little slap on the wrist [and] go home.

While acknowledging the possibility of arrest and criminal justice processing,

many also assumed that they would simply be given probation, a sentence that
did not instill fear, trepidation, or even minor concern. Probation was almost
universally perceived as an easily managed and adjusted-to sentence, and was
not something that led to changes in their decisions. In short, probation, or any
diversionary sentence, was perceived as “a joke” by most. One young offender
expressed this sentiment when he was asked what he thought would happen if he
got caught: “I would just go to the police station and get signed out. I was just a
juvenile so I couldn’t go to jail.” Another offender’s response to being caught by
police was, “Fuck the police. They ain’t going to do anything but bring me to jail.
I ain’t worried about 5-0 [police], shit.”
When it came to less convenient and more restrictive forms of punishment,
quite a few claimed that they were unconcerned with being incarcerated. This
indifference was likely built from previous incarcerations, which reinforced in
them that there was no real reason to fear confinement—it too was an easily
managed and adjusted-to sentence (Shover 1996). In the words of one offender,
“Yeah [I feared getting caught], when I was young and shit. Now, I been to
jail already so now when I do it, it ain’t shit.” Undoubtedly, auto thieves did
not want to be apprehended, but such formal sanctions did not appear to factor
heavily in their decisions.
While the potential for jail time raised the pulse of some, many said that an
important, and the more likely risk in auto theft, was being confronted by the
Risks and rewards of motor vehicle theft 61
owner. They were aware that attempts to take the property of others could quickly
turn into serious violence from victims trying to prevent the loss and/or from their
own attempts to protect themselves. It is noteworthy that a number had suffered
serious injuries at the hands of victims while many more reported instances where
they came close to suffering a similar fate.

I know the chance that I’m taking with what I’m doing. I could go to jail. I could
get killed. I could get killed very easily. I’ve been shot at several times. I’ve had
people come up sticking guns in my face when I was in their car, several times.

Of course, most avoided confronting the owners of the vehicles they were steal-
ing, as they neither want to be seen and reported nor killed or injured by defensive
owners. One car thief succinctly explained, “The police take you to jail, but the
owner’ll kill you.” When asked whom he feared more, the police or the owner, a
young thief who ventured into rural areas to steal responded, “The owners! Shoot,
this is country folk. These people don’t believe in calling the cops. You in their
yard stealing something; they coming out with shotguns.”
For all their brazen talk about being able to avoid arrest, downplaying the likely
penalties, and claims about being able to handle what punishments they would receive,
it should be said that auto thieves did not relish the prospect of apprehension and many
expressed concern about being arrested. When these concerns did arise, offenders were
keen to not let them curtail their behavior. For some, immediate action was the stand-
ard remedy for such misgivings: “Like I say when I get scared or something, for me,
the best way to overcome it is to just do it. Just go ahead and do it.” For others, peace of
mind was restored by pushing thoughts of arrest from their minds. “I just blank it out.
I never tried to worry about it.” Still others detailed a more complex set of cognitive
gymnastics that helped to alter how they viewed the realities of such prospects:

You just know in your heart and in your mind that what you’re doing nobody
will ever know about it and you’ll never get caught for it. And even if things
get hot, you’ll get out of it. It’s always worked. I guess it’s a Karma thing.
… I thought about [being caught] a few times but it was the positive think-
ing thing, you know. If I don’t think about it, they won’t catch me. If they
[police] get behind me I will get away.

The offenders displayed a remarkable capacity to neutralize their fears and the
potential for arrest by displacing such thoughts from their minds. Be that as it may,
readers should make no mistake that offenders’ overwhelming indifference toward
the potential legal consequences of auto theft merely reflect the broad-based weigh-
ing of formal sanction threats in the abstract. When examining the ways that auto
thieves consider and manage risks in the foreground or immediacy of crime com-
mission, both informal and formal threats swayed their decision making. While the
thieves described numerous strategies to manage such threats, the remainder of this
section focuses on a limited range of tactics that offenders used to moderate the
opportunity structure of suspicion from vehicle owners, passersby, and police.
62 Heith Copes and Michael Cherbonneau
Efforts to conceal their activities and to keep outsiders from becoming wise to
their intentions began at the very early stages of the crime; namely, while search-
ing for targets and effecting entry into them (Cherbonneau and Copes 2006). A
primary way that offenders reduced their chances of having owners and all other
onlookers alerted to their presence was by blending into the locations where they
searched for targets. The easiest way for offenders to do this was by dressing like
other people in the area. The importance of blending in was best described by the
following accomplished car thief:

At first you think the whole world is watching you. But after you get used
to it you feel like you blend in. It depends, you can be sloppy like and some-
times it is the way you look that gives you away. Sometimes the way you
look just blends in and they will never notice you. So you dress like a little
punk and you step out of the car they going to keep an eye on you. You dress
like in nice slacks like you straight out of college, with Polo’s and a pair of
slacks. When you step out of the car nobody will pay attention to you.

While some attempted to blend in by dressing sharp, others did so by wearing

mechanic or tow truck driver uniforms. They believed that by wearing oily over-
alls they could create the illusion that they were there for legitimate reasons.
Namely, that they were working on the car and not stealing it. “I keep on a work
uniform. That’s what I steal in. I do my dirt in. You got on work clothes people
don’t give you much attention.”
There were other more subtle things offenders did to normalize their appear-
ance beyond manipulating how one dressed. For example, one of the males we
interviewed would explore the more affluent outlying suburban counties of the
city where he lived looking for more desirable targets to sell to a chop shop—
targets that simply could not be found in the part of the city where he lived.
Concerned that police in these communities were weary of vehicles occupied by
black males (such as himself), he chose to enlist the company of a trustworthy
female who could help allay suspicions of people and police from the area better
than his “regular” partners would. In his words:

I went and got one of my little chicks. … I called her because I was going
to the county. Make us at least look like we a couple or [we] husband and
wife handling some business. … Just throw it off [rather] than two men just
riding. … Now in the city I’ll do it … with my [male] partner. … In the city
I’m more comfortable. They use to that. That’s 24/7—motherfuckers riding
four-deep all cats [black males]. … But, she a better throw off in the county.
… We could be out having errands or something.

Another way to minimize onlookers’ suspicions was by selecting specific vehicle

models. The offenders perceived some vehicles to attract more police attention
than others. For one offender, this translated into stealing “something real decent
[but] undescriptive. Not something that’s gonna be looked at as being stolen but
Risks and rewards of motor vehicle theft 63
… just a plain Jane model. … [N]othing flashy but nice enough to move.” The fol-
lowing offender also noted the importance of stealing something that was not too
noticeable, which meant that it was especially wise to avoid the most commonly
stolen vehicle types in his area:

[I’m looking for] something we can get, something looking nice, something
that won’t be noticeable for the police. See what I’m saying nothing notice-
able. They ain’t gonna trip off this [Chrysler] 300—it’s stolen but they ain’t
up on people stealing 300s. They up on the Grand Prix, the Bonneville, the
Grand Am, you know they up on all that. The Chrysler Dodge Intrepid, they
on all that you see what I’m saying? So if they see a little young guy like me
riding around they gonna automatically want to flag me.

An ongoing concern when driving a stolen car was that the vehicle’s license plate
would come back “hot” if run by police. While offenders believed that it gener-
ally takes 24 hours until a stolen vehicle shows up on “hot sheets” used by police
to identify stolen vehicles, affixing a fresh license plate to the stolen vehicle as
quickly as possible after stealing it was desirable. By doing so, offenders gained
a worry-free driving experience for roughly three days—the typical amount of
time most were willing to drive a stolen car before it was considered “too hot.”
Though this strategy was not fool-proof, as authorities could run the plate through
dispatch and discover that it did not belong to that particular car (i.e., same make,
model, color), offenders reasoned that as long as the new plate was not associated
with a vehicle that was also stolen, the police would not create much of a fuss over
the apparent discrepancy. According to one offender, “

If they run that license plate and they know it’s stolen they gonna flag you.
But there’s some cops that will run the license plate and then they know that
[it doesn’t belong to] that car … but won’t flag though.

Others said it was better than keeping the original stolen plates on the car as that
would be a dead giveaway: “Why keep some stolen plates on the car? They’ll pull
you over even quicker. See they might run the [new] plates and be like, ‘Oh well,
she just been and got her plates registered.’”
Beyond directly replacing the plates on the stolen car with those associated
with a vehicle that was not stolen, similar methods included using plates obtained
from a car dealership or rental agency or removing plates from the vehicle alto-
gether and affixing a temporary plate tag in the rear window. One offender who
spoke about the use of car dealership plates instead of official plates claimed this
type of modification was in some ways better because the police would be more
inclined to associate “dealer” vehicles with legitimate activities.

Put a dealer plate on there and chances are police won’t flag you with a dealer
plate [because they’ll think], “Nah, he’s probably just test driving this car” …
I done pulled up on the side of police in stolen cars with dealer plates on them
64 Heith Copes and Michael Cherbonneau
[and they thought], “Man, he probably just bought the car from the auction”
or something you know?

The most universally agreed way to influence the perceptions of police and
thus, keep from having one’s license plate checked by authorities was to avoid
attention-grabbing conduct when operating the stolen car in traffic (Cherbonneau
and Copes 2006). This included things such as avoiding speeding, stopping com-
pletely at stop signs, and staying in one lane. In other words, they believed that
driving sensibly could aid in their desire to avoid attention:

I don’t drive fast and break rules while I’m driving. I don’t get the law to stop me
for anything … I drive normal, like it’s my car. When you steal a car you don’t
want to be driving it all fast. That’s going to draw attention. You might be at the
red light and the police might be right next to you, he don’t know that car is stolen
unless they done reported it. You make him think you up to something, he might
run the license plate and find out it’s stolen. You just want to drive it normal.

There were some offenders who were so adamant about not attracting police
attention that they adhered at all times to an extreme form of the sensible driving
script with little deviation.

When I’m driving I always make sure I have my seat belt on, use my blinkers.
I drive careful. … I got to make sure I stay in my lane, drive right, look in
mirrors, just like it is my car. Using blinkers to get over, turning right. Make
sure I do everything cautious.

When asked how he drove the car immediately following the theft, one young
offender replied:

Obey every law, [drive] by the book. Wear your seat belt. Real close to the
steering wheel. Not thugged out with a gangster lean and all this stuff. Do the
10 o’clock and the 2 o’clock [on the steering wheel]. Stop and check your
brakes 500 feet before the stop sign and everything.

Not all were hyper-sensitive to how they operated the vehicle. Some recognized
that the average driver speeds, does not use blinkers in every situation, and does
not drive with perfect posture, and attempts at doing so could be suspicious in and
of themself. Thus, the best way to look normal was to drive in a manner consistent
with other drivers. By taking a very casual approach to the application of basic
traffic rules by synchronizing their driving to that of fellow motorists instead cre-
ated an aura of “normalcy” that was all the more believable to onlookers.
In sum, our sample relied on a number of strategies to reduce the chances of
being detected, and consequently the fears associated with crime. They believed
that effective strategies for avoiding capture included being careful in the places
they chose to steal from, being sensitive to the types of cars they targeted, being
restrained in the way they drove after the theft, and keeping a low profile when
driving. They recognized that to avoid arrest, suspicions of onlookers, and
Risks and rewards of motor vehicle theft 65
confrontations with owners, they must present an image of belonging—whether
they were in the car driving or in the area looking for cars to steal.

The ethnographic data analyzed here has revealed much about the way auto
thieves evaluate the benefits and risks of stealing cars. In terms of benefits, auto
thieves saw their crimes as an ideal way to sustain their chosen lifestyles. By
breaking into and absconding with someone else’s car they could easily obtain
money and other symbols of success with relatively little effort. In addition, they
were adept at excusing and justifying their crimes. Their crimes were cast as being
the fault of car owners or as not involving any real financial harm so that they suf-
fered little pangs of conscience for their larcenous ways. In addition, they easily
minimized the potential risks of their endeavors. They pointed to the incompe-
tence of law enforcement or their own skills and professionalism as reasons they
need not concern themselves with being arrested or punished seriously for their
crimes. In short, auto thieves saw their crimes as richly rewarding and remarkably
easy. Few saw any reason to desist.
The auto thieves committed numerous offenses over an extended period.
Nearly all of them claimed that they either planned to continue offending (among
the active sample) or would have continued offending if they had not been appre-
hended by law enforcement (among the incarcerated sample). Their accounts
suggest that once offenders begin their auto theft career they either maintained
or increased their offending frequency. Their current life circumstances coupled
with their experiences with stealing cars led to the subjective cost–benefits scale
being weighted in favor of continued offending, which consequently led to persis-
tence in auto theft. We argue that this perception of their crimes, and subsequent
habituation, was due largely to their: increased skills to accomplish the crimes;
change in their street connections; adherence to a lifestyle conducive to crime in
general; and acceptance of various linguistic devices that allowed them to make
sense of their misdeeds.
First, the development of skills allowed auto thieves to increase their chances
of success. Specifically, their skills allowed them to steal more desired vehicles
and evade the formal sanctions associated with doing so, which may be why
so few feared detection or legal consequences. This optimistic view was likely
born from the relative success at avoiding arrest. The auto thieves believed that
they could continue offending because they had developed the necessary skills
to evade capture, thereby nullifying the deterrent effects of criminal sanctions.
Each successful theft reinforced their belief that consequences were far removed
and that auto theft was worth the risk. Instances of actual detection following
long periods of unsanctioned offending did not help to sway their thoughts from
crime either. This is especially true from what we know about “resetting effects”
(Pogarsky and Piquero 2003), whereby the notion that one’s number could be up
if they choose to reoffend becomes less relevant if one’s number just came up
(Jacobs and Cherbonneau 2012).
66 Heith Copes and Michael Cherbonneau
In addition, improved professionalism and skills make crime commission more
palatable to offenders as they minimized their fears and anxieties. For most property
thieves, auto thieves included, the overriding emotions experienced during crime
commission are fear and anxiety (Cusson 1993; Lejeune 1977). To be success-
ful, thieves must learn to manage these emotions to maintain a level of composure
and a suitable degree of aplomb while in the midst of crime. This is not an easy
task. For many people, the anxiety, fear, and paranoia that accompany crime are
enough to deter. The auto thieves learned to push fears out of their mind when they
stole cars in much the same way other persistent offenders do (Bennett and Wright
1984; Gill 2000). But entering another person’s vehicle can cause even the most
seasoned offenders to become riddled with fear and panic. If offenders cannot man-
age these emotions then their careers in crime will likely be short-lived. Auto theft
would entail far more emotional strain for those who suspected that they would be
unable to manage encounters with would-be victims, unruly bystanders, or police.
By relying on their skills, motor vehicle thieves can control the negative emotions
associated with crime, which makes their crimes easier emotionally and ultimately
facilitates criminal persistence (Cherbonneau and Copes 2006).
Second, because of offenders’ lack of desire or inability to come by funds
legitimately, their withdrawal from and support by law-abiding others, and their
desire to maintain a criminal lifestyle (including chronic drug use), auto theft
increasingly was perceived to be a viable solution to prevent uncomfortable situ-
ations (see Shover 1996). The cash intensive lifestyle of most auto thieves could
not be sustained by legitimate employment and it may have undermined both the
inclination and ability to keep employment. Even if offenders were willing to
obtain employment at the kinds of jobs they could secure, the physical and tem-
poral demands of work and their lifestyles conflicted. Consequently, for many of
those who pursued auto theft, legitimate employment was sacrificed. The absence
of income from non-criminal sources thus reinforced the need to find other ave-
nues of money and auto theft was well suited to fit these demands. It provided the
money to finance their lives and the accoutrements they desired to display—both
of which are material symbols of their lifestyle.
In addition, as auto thieves severed ties with conventional citizens, they devel-
oped associations with others living a criminal lifestyle. It is certainly possible to
commit crime without the assistance of criminal companions, but knowing other
offenders facilitates the commission of crime and the disposal of stolen goods. By
interacting with other car thieves, they learned the skills to break into and start
vehicles, how to break down and dispose of the vehicle (in parts or as a whole),
and the beliefs, values, and justifications for doing so. They learned which cars
brought the most money, who would buy stolen parts, and where they could go to
get rid of vehicles. Without such knowledge their profits would have been limited.
Thus, developing underworld connections allowed them to increase the financial
rewards of auto theft and reduce the risks associated with disposing of stolen cars
and parts (Mullins and Cherbonneau 2011).
Finally, the use of accounts, justifications, and other linguistic devices
allowed auto thieves to overcome the guilt and potential negative self-image
Risks and rewards of motor vehicle theft 67
associated with stealing cars. Recent developments in neutralization theory sug-
gest that neutralizing beliefs may not be enough to cause people to begin a
career in crime, but they may be useful in explaining criminal persistence or
desistance (Maruna and Copes 2005). In this interpretation of the theory, neu-
tralizations start as after-the-fact rationalizations, but eventually become the
rationale or moral release mechanisms facilitating future offending. By holding
on to these justifications and bringing them to the foreground when needed,
the auto thieves could continue stealing and easily make sense of their crimes.
Whether neutralizations caused initiation into auto theft is unclear, but it seems
that relying on these techniques helped them continue with their crimes. In
this light, the decision to engage in auto theft did not weigh heavily on their
Our analysis has focused on the rewards and risks that auto thieves associate
with their crimes and the implication of these assessments for sustaining their
criminal careers. We believe doing so has implications for understanding how
auto thieves (and other types of offenders more generally) perceive crime as
attractive and risk-free, which ultimately encourages them to continue offend-
ing. The mental calculation of costs and benefits of crime is a subjective process
that is always “in motion” as the relevant variables that inform such prospects
are constantly being assessed and reassessed. With experience and cognitive
reinterpretations, risks that seemed insurmountable one day can be overlooked
the next day. Correspondingly, rewards that once were thought negligible can
turn offenders’ heads under new circumstances. Interpreting the motives of their
crimes within the context of their lives, committing crime with impunity, learning
arrest avoidance skills, eroding resources, and reliance on accounts act in concert
to foster images of auto theft as being a rational and reasonable solution to fulfill-
ing their desires and needs. Consequently, these factors contributed greatly to the
habituation of auto theft.

Beauregard, E., Proulx, J. Rossmo, K. Leclerc, B. and Allaire, J.-F. (2007). Script Analysis
of the Hunting Process of Serial Sex Offenders. Criminal Justice and Behavior 39,
Becker, G. S. (1968). Crime and Punishment: An Economic Approach. Journal of Political
Economy 76, 169–217.
Bennett, T., and Wright, R. (1984). Burglars on Burglary: Prevention and the Offender.
Aldershot: Gower.
Carroll, J. (1978). A Psychological Approach to Deterrence: The Evaluation of Crime
Opportunities. Journal of Personality and Social Psychology 36, 1512–1520.
Challinger, D. (1987). Car Security Hardware: How Good Is It? In Car Theft:
Putting on the Breaks, Proceedings of Seminar on Car Theft, May 21 (pp. 42–48).
Sydney: National Roads and Motorists’ Association and the Australian Institute of
Cherbonneau, M., and Copes, H. (2006). Drive It Like You Stole It’: Auto Theft and the
Illusion of Normalcy. British Journal of Criminology 46, 193–211.
68 Heith Copes and Michael Cherbonneau
Clarke, R.V., and Cornish, D.B. (1985). Modeling Offenders’ Decisions: A Framework for
Research and Policy. In M. Tonry and N. Morris (eds), Crime and Justice: An Annual
Review of Research (Vol. 6, pp. 147–185). Chicago, IL: University of Chicago Press.
Clarke, R.V., and Felson, M. (1993). Routine Activity and Rational Choice: Advances in
Criminological Theory (Vol. 5). New Brunswick, NJ: Transaction Publishers.
Clarke, R.V., and Harris, P.M. (1992). Auto Theft and Its Prevention. In M. Tonry (ed),
Crime and Justice: A review of research (Vol. 16, pp. 1–54). Chicago, IL: University
of Chicago Press.
Copes, H. (2003a). Streetlife and the Rewards of Auto Theft. Deviant Behavior 24,
Copes, H. (2003b). Societal Attachments, Offending Frequency and Techniques of
Neutralization. Deviant Behavior 24: 101–127.
Copes, H., and Cherbonneau, M. (2006). The Key to Auto Theft: Emerging Methods
of Auto Theft from the Offenders’ Perspectives. British Journal of Criminology 46,
Copes, H., and Tewksbury, R. (2011). Criminal Experience and Perceptions of Risk: What
Auto Thieves Fear When Stealing Cars. Journal of Crime and Justice 34, 62–79.
Copes, H., and Vieraitis, L. (2009). Bounded Rationality of Identity Thieves: Using
Offender-Based Research to Inform Policy. Criminology and Public Policy 8, 237–262.
Copes, H., and Vieraitis, L. (2012). Identity Thieves: Motives and Methods. Boston, MA:
Northeastern University Press.
Copes, H., Hochstetler, A., and Brown, A. (2013). Inmates’ Retrospective Experiences of
Prison Interviews. Field Methods 25, 182–196.
Corbett, C., and Simon, F. (1992). Decisions to Break or Adhere to the Rules of the Road,
Viewed from the Rational Choice Perspective. British Journal of Criminology 32,
Cornish, D.B., and Clarke, R.V. (1986). The Reasoning Criminal. New York:
Cornish, D.B., and Clarke, R.V. (2008). The Rational Choice Perspective. In R. Wortley
and L. Mazzerolle (eds), Environmental Criminology and Crime Analysis (pp. 21–47).
Portland, OR: Willan Publishing.
Cromwell, P., Olson, J., and Avary, D.W. (1991). Breaking and Entering: An Ethnographic
Analysis of Burglary. Newbury Park, CA: Sage.
Cusson, M. (1993). Situational Deterrence: Fear During the Criminal Event. In R. Clarke
(ed), Crime Prevention Studies (Vol. 1, pp. 55–68). Monsey, NY: Willow Tree Press.
Cusson, M., and Pinsonneault. P. (1986). The decision to give up crime. In D. Cornish and
R. Clarke (eds), The Reasoning Criminal (pp. 72–82). New York: Springer-Verlag.
Feeney, F. (1986). Robbers as Decision Makers. In D. Cornish and R. Clarke (Eds.), The
Reasoning Criminal (pp. 53–71). New York: Springer-Verlag.
Fleisher, M. (1995). Beggars and Thieves: Lives of Urban Street Criminals. Madison, WI:
University of Wisconsin Press.
Fleming, Z. (2003). The Thrill of It All: Youthful Offenders and Auto Theft. In P. Cromwell
(Ed.) In Their Own Words: Criminals on Crime (3rd edn) (pp. 99–107). Los Angeles,
CA: Roxbury Publishing.
Gill, M. (2000). Commercial Robbery. London: Blackstone Press Limited.
Halsey, M. (2008). Narrating the Chase: Edgework and Young Peoples’ Experiences of
Crime. In T. Anthony and C. Cunneen (eds), The Critical Criminology Companion
(pp. 105–117). Sydney: Hawkins Press.
Risks and rewards of motor vehicle theft 69
Jacobs, B. (1999). Dealing Crack: The Social World of Streetcorner Selling. Boston, MA:
Northeastern University Press.
Jacobs, B. (2000). Robbing Drug Dealers: Violence Beyond the Law. New York, NY:
Aldine de Gruyter.
Jacobs, B., and Cherbonneau. M. (2012). Auto Theft and Restrictive Deterrence. Justice
Quarterly, doi:10.1080/07418825.2012.660977.
Jacobs, B., and Wright, R. (1999). Stick-up, Street Culture, and Offender Motivation.
Criminology 37, 149–173.
Jacques, S., and Wright, R. (2008). Intimacy with Outlaws: The Role of Relational Distance
in Recruiting, Paying, and Interviewing Underworld Research Participants. Journal of
Research in Crime and Delinquency 46, 22–38.
Lejeune, R. (1977). The Management of a Mugging. Urban Life 6, 123–147.
Light, Roy, Claire Nee, and Helen Ingham. (1993). Car Theft: The Offender’s Perspective.
Home Office Research Study no. 130. London: Home Office.
Maruna, S., and Copes, H. (2005). What Have We Learned from Fifty Years of
Neutralization Research? Crime and Justice: A Review of Research 32, 221–320.
Michael, S., Hull, R.B., and Zahm, D.L. (2000). Environmental Factors Influencing Auto
Burglary: A Case Study. Environment and Behavior 33, 368–388.
Mullins, C., and Cherbonneau, M. (2011). Establishing Connections: Gender, Motor
Vehicle Theft, and Disposal Networks. Justice Quarterly 28, 278–302.
Opp, K-D. (1997). Limited Rationality and Crime. In Graeme Newman, Ronald V.
Clarke, and S. Giora Shoham (eds), Rational Choice and Situational Crime Prevention
(pp. 47–63). Brookfield, VT: Dartmouth.
Paternoster, R., and Simpson, S. (1993). A Rational Choice Theory of Corporate Crime.
In R. V. Clarke and M. Felson (eds), Routine Activity and Rational Choice. Advances
in Criminological Theory (Vol. 5, pp. 37–58). New Brunswick, NJ: Transaction
Pogarsky, G., and Piquero, A.R. (2003). Can punishment encourage offending?
Investigating the ‘resetting’ effect. Journal of Research in Crime and Delinquency 40,
Reynolds, M. O. (1985). Crime by Choice: An Economic Analysis. Dallas, TX: Fisher
Shover, N. (1985). Aging Criminals. Beverly Hills, CA: Sage.
Shover, N. (1996). Great Pretenders: Pursuits and Careers of Persistent Thieves. Boulder,
CO: Westview.
Shover, N., and Hochstetler, A. (2006). Choosing White-Collar Crime. Cambridge, MA:
Cambridge University Press.
Shover, N., and Honaker, D. (1992). The Socially Bounded Decision Making of Persistent
Property Offenders. Howard Journal of Criminal Justice 31, 276–293.
Sutherland, Edwin. 1947. Principles of Criminology. Philadelphia, PA: Lippincott.
Sykes, G., and Matza, D. (1957). Techniques of Neutralization: A Theory of Delinquency.
American Sociological Review, 22, 664–670.
Wright, R., and Decker, S. (1994). Burglars on the Job: Streetlife and Residential Break-
Ins. Boston, MA: Northeastern University Press.
Wright, R., and Decker, S. (1997). Armed Robbers in Action: Stickups and Street Culture.
Boston, MA: Northeastern University Press.
5 The rational choice perspective
and the phenomenon of stalking
An examination of sex differences
in behaviours, rationales, situational
precipitators and feelings
Carleen M. Thompson and Benoit Leclerc
Current understanding of the theoretical underpinnings of stalking is still in its
infancy (Lyndon et al., 2012). The theoretical research that has been conducted
primarily focuses on characteristics of the offender that predispose him/her to
commit this form of crime (e.g. attachment theory and feminist theories of power
and control; see Davis, Swan and Gambone, 2012; Lyndon et al., 2012) or aspects
of the victim’s lifestyle that increase the likelihood that he/she will be selected as
a target (i.e. using routine activities theory; e.g. Fisher, Cullen and Turner, 2002).
Few researchers have applied theories that focus on the specific crime event/s
and the situations or contexts in which it occurs, such as interactions between
the parties involved. However, a number of contextual factors associated with
the crime event have been demonstrated to be pertinent for understanding stalk-
ing perpetration and the escalation to violence in stalking cases (e.g. motivation,
feelings, substance use, triggering events and victim responses; Davis, Ace and
Andra, 2000; James et al., 2009; McEwan, Mullen, MacKenzie and Ogloff, 2009;
Rosenfeld, 2004; Thompson, 2009; Thompson, Dennison and Stewart, in press).
Additionally, although there is evidence that some of these factors may vary
across male and female stalkers (e.g. triggered by different precipitating events
and different motives/reasons for stalking; Meloy, Mohandie and Green, 2011;
Purcell, Pathe and Mullen, 2001), current theoretical frameworks typically do not
capture these differences.
A theoretical framework capable of addressing these limitations is the com-
bination of the rational choice perspective employed in the field of criminology
(Cornish and Clarke, 2008) and the concept of situational precipitators devel-
oped by Wortley for situational crime prevention purposes (1998; 2001). The
innovative feature of this framework is the emphasis on how the crime event/s
occur/s as a result of the interaction between the immediate environment and the
offender’s needs, motives and decisions. Examining how these factors interplay
when stalking escalates to violence may also foster a better understanding of why
such violence occurs. Furthermore, as this framework focuses on factors associ-
ated with individuals’ decisions to perpetrate stalking behaviours, it can be used
to explore differences in males’ and females’ reasons for stalking and related situ-
ational precipitators. In this chapter, we use this framework to further understand
the phenomenon of stalking and associated sex differences. The chapter focuses
The phenomenon of stalking 71
on the behaviours selected by stalkers, the rationales for engaging in stalking
behaviours and associated feelings and situational precipitators. These dimen-
sions of offending are compared across male and female relational stalkers to
determine if and how stalking differs across sex.

Theoretical background on stalking

Stalking behaviours. Although legislation varies across jurisdictions, stalk-
ing typically encompasses the perpetration of (1) persistent or repetitive (2)
unwanted acts (3) intended to, or that should have been foreseen to, or could rea-
sonably (4) induce apprehension, fear or harm (see Beatty, 2003; Dennison and
Thomson, 2005; Kapley and Cooke, 2007; Miller and Nugent, 2002). Stalking
may encompass a range of different behaviours of varying levels of severity and
intrusiveness. Spitzberg (2002) grouped nearly 450 stalking behaviours identi-
fied from 43 different studies into seven categories: (a) hyperintimacy tactics that
include inappropriate or overzealous ‘romantic’ gestures ranging from phone calls
and gifts to sexual coercion; (b) pursuit, proximity and surveillance that encom-
passes attempts to be physically close to the victim as well as tactics to gather
information on the victim; (c) invasion tactics that encompass behaviours that
violate victims’ privacy and are typically illegal when perpetrated in isolation,
such as burglary; (d) proxy pursuit/intrusion whereby perpetrators involve third
parties in the stalking, either by knowingly or unknowingly contributing to the
harassment of the victim or by providing the perpetrator with information about
the victim; (e) intimidation and harassment incorporating a range of ‘aggressive
verbal or nonverbal activities designed to bother, annoy or otherwise stress the
victim’ (Spitzberg and Cupach, 2007, p. 71) such as threats and abusive calls; (f)
coercion and constraint that involves using force to restrict or control the victim’s
behaviour such as coercing the victim into a date, extortion or kidnapping; (g)
aggression incorporating violence towards the victim or his/her property, pets or
loved ones. Those behaviours on the less severe end of the continuum (e.g. pur-
suit, proximity and surveillance behaviours and phone calls, letters/emails; Purcell
et al., 2001; Spitzberg, 2002; Strand and McEwan, 2012) are more common than
behaviours at the severe end of the continuum (e.g. aggression/physical violence;
Purcell et al., 2001; Spitzberg, 2002; Strand and McEwan, 2012). Similarly, less
severe acts of violence (e.g. pushing, slapping) are more likely than severe acts
of violence (e.g. attempted murder, rape; Meloy and Boyd, 2003; Mullen, Pathe,
Purcell and Stuart, 1999; Rosenfeld and Harmon, 2002).
Motives and feelings. A plethora of motives have been identified for stalking.
As researchers often identify feelings (e.g. anger, jealousy, love) as motives for
stalking (Meloy and Boyd, 2003; Spitzberg and Cupach, 2007), feelings will be
subsumed within motives in this section. After reviewing 24 studies on stalking
motives, Spitzberg and Cupach (2007) organised motives into four general factors;
intimacy (M = 32 per cent; e.g. issues associated with abandonment, jealousy, love,
obsession, reconciliation, sexual), aggression (M = 22 per cent; anger, revenge,
control, intimidation, harm), disability (M = 12 per cent; substance-related,
72 Carleen M. Thompson and Benoit Leclerc
mental health issues) and task conflict or issues (M = 13 per cent; e.g. disputes
or custody issues); as well as miscellaneous and unclassifiable reasons (e.g. no
reason or unsure). However, in many cases stalking is perpetrated for multiple,
often inconsistent, motives (e.g. love/reconciliation and anger/revenge; Spitzberg
and Cupach, 2007). Some of these conflicting motives may arise because stalk-
ing occurs over time and researchers do not disaggregate stalking motives across
violent and non-violent stalking behaviours. It is possible that some violent stalk-
ing behaviours are perpetrated for very different reasons to the initial stalking.
For example, amorous pursuit may escalate to violence following an argument
or confrontation (e.g. Thompson, 2009). Although motives for violence specifi-
cally have not been examined, stalking that is more broadly motivated by revenge
(Rosenfeld and Harmon, 2002), anger and/or jealousy (Dutton and Winstead,
2006; Morrison, 2008) is more likely to include violent behaviour.
Differences according to sex. The degree of disparity between rates of male-
and female-perpetrated stalking varies across sample sources. Research utilising
samples of convicted stalkers, or samples that require victims to experience fear,
suggest that males perpetrate the vast majority of stalking (e.g. 75 to 80 per cent;
Spitzberg, Cupach and Ciceraro, 2010). In contrast, research utilising samples of
unconvicted stalkers that also omit the fear requirement, report comparable rates
of stalking across sex (see Dutton and Winstead, 2006; Sinclair and Frieze, 2000;
Williams, Frieze and Sinclair, 2007). Despite these disparities, few sex differ-
ences have been identified between the types and nature of stalking behaviours in
both convicted samples of stalkers (e.g. Meloy et al., 2011; Purcell et al., 2001;
Strand and McEwan, 2012) and samples of unconvicted stalkers that omit the fear
requirement (e.g. Bjerregaard, 2000; Langhinrichsen-Rohling, Palarea, Cohen
and Rohling, 2000; Sinclair and Frieze, 2000). This pertains to similarities across
sex for most types of stalking behaviours, including violence, as well as com-
parable durations of stalking (Purcell et al., 2001; Strand and McEwan, 2012).
While the behaviours that differ across sex vary across studies, some of the few
differences that have been identified are that male stalkers are more likely to fol-
low (Purcell et al., 2001; Strand and McEwan, 2012) or physically approach the
victim (Meloy et al., 2011) and female stalkers are more likely to send letters/
faxes/emails (Meloy et al., 2011; Strand and McEwan, 2012) or call the victim
(Purcell et al., 2001). Together, this has led some researchers to conclude that
the perpetration of stalking behaviour is similar across sex but that female-per-
petrated stalking is underestimated in official records because females and males
respond to victimisation differently (i.e. males are less likely to self-identify as
victims of stalking, experience fear or report stalking victimisation to the police;
see Bjerregaard, 2000; Haugaard and Seri, 2000; Lyndon et al., 2012; Sheridan
and Lyndon, 2012). Despite perpetrating similar behaviours, it is possible that the
motives and feelings underlying stalking differ between male and female stalkers
(Langhinrichsen-Rohling et al., 2000).
To date, few researchers have disaggregated stalking motives and feelings
across sex. One study suggested that female stalkers were more likely to have
intimacy-seeking motivations than male stalkers (defined as an infatuation with
The phenomenon of stalking 73
establishing intimacy and the, sometimes delusional, belief that their feelings
are reciprocated; Purcell et al., 2001). Another study reported comparable rates
of such intimacy-seeking motives (Strand and McEwan, 2012). However, more
males than females were motivated by a desire for a romantic relationship, with-
out illusions of reciprocity or infatuation, but instead caused by a lack of social
skills (Strand and McEwan, 2012). Additionally, more females were motivated
by resentment than their male counterparts (defined as motivated to frighten or
distress the victim and to seek retribution for perceived wrongdoings; Strand and
McEwan, 2012). This study also suggested that similar stalking motives were
associated with violence for male and female perpetrators; however, the motives
for violence itself were not examined (Strand and McEwan, 2012). Lastly, a third
study indicated that females, like the majority of male stalkers, were most com-
monly motivated by anger and obsession (Meloy and Boyd, 2003). However,
in this study motives relating to entitlement, grandiosity and narcissistic inju-
ries, often identified for male stalkers, were not as relevant for female stalkers
(Meloy and Boyd, 2003). Importantly, the samples used in these three studies
predominantly comprised convicted offenders and therefore findings may reflect
sex differences in the types of stalking males and females report to the police.
Given these limitations and contrasting findings, replication seems necessary.

The rational choice perspective

The rational choice perspective is a conceptual framework for understanding
crime events and guiding situational prevention initiatives. Key to this perspective
is that criminal behaviour is purposeful (Cornish and Clarke, 1986; 2008). That
is, offenders perpetrate crime because it provides a means to attain their goals or
satisfy their needs or desires such as revenge, sexual gratification, power, money
and so on (Cornish and Clarke, 2008). Consequently, the correct identification
of offenders’ goals is paramount to understanding their involvement in crime.
In the case of stalking, some of the perceived benefits may be: the satisfaction
of avenging a wrongdoing, the pleasure associated with reconciling with a part-
ner, or the satisfaction of dominating, controlling or even frightening the victim
(e.g. Spitzberg and Cupach, 2007). Offenders’ goals are also proposed to interact
with situational factors and may vary according to opportunities or inducements
(Cornish and Clarke, 2008).
According to the rational choice perspective, offenders consider the costs (i.e.
risks and effort) and benefits (i.e. rewards) of different behaviours before acting
(Cornish and Clarke, 1986; 2008). Accordingly, criminal behaviour, like other
behaviour, is considered rational (Cornish and Clarke, 1986; 2008). Importantly,
however, their ‘rational’ decisions are constrained by the circumstances under
which these decisions are made. For example, offenders’ decisions are likely
to be influenced by biases in information processing as a result of limited time,
substance use, previous learning or intense emotions (Cornish and Clarke,
2008). In these circumstances, the offender may choose actions that are con-
sidered ‘good enough’ without evaluating all possible actions (i.e. satisficing
74 Carleen M. Thompson and Benoit Leclerc
principle; Cornish and Clarke, 2008). Additionally, ‘when the expected costs
of [not offending] or doing nothing are great, actors may feel they have little
choice but to [offend]’ (Tedeschi and Felson, 1994, p. 211). Consequently, deci-
sions to engage in criminal behaviour are considered to be made with ‘bounded
rationality’ and interact with situational factors (e.g. opportunities and induce-
ments; Cornish and Clarke, 2008).
Although the rational choice perspective has traditionally been applied to
property crimes, it was intended to be applicable across crime types (Cornish
and Clarke, 2008) and has been successfully applied to interpersonal and vio-
lent crimes that may not immediately seem rational, such as terrorism (Clarke
and Newman, 2006), crowd violence (Madensen and Knutsson, 2011) and sexual
offending (Beauregard and Leclerc, 2007). The present study extends this liter-
ature by applying this perspective to stalking. Although stalking has not been
examined from a rational choice perspective, a number of studies have examined
the types of behaviours perpetrated by stalkers and the motives and feelings asso-
ciated with these behaviours (reviewed above).
In the case of stalking, stalkers may engage in a range of stalking behaviours
because, in the context of their current circumstances (e.g. intense anger, jeal-
ousy, obsessional love, confusion over the break-up), he/she believes that these
persistent behaviours will help him/her achieve his/her goal (e.g. revenge, recon-
ciliation or establishing intimacy, gaining an understanding or explanation for the
break-up). In fact, the level of desperation and obsession associated with much
stalking may mean that stalkers feel they have little choice but to persist with
these behaviours. Given the repetition of behaviour associated with stalking, vic-
tims’ or third parties’ reactions to previous stalking behaviours may also interact
with stalkers’ decisions to continue, desist or escalate their pursuit.
In the context of stalking violence, the factors associated with the decision
to perpetrate violence itself may differ from those associated with other (non-
violent) types of stalking behaviours. Therefore, to understand why stalking
escalated to violence, the specific decisions, motives and environmental factors
associated with violence itself must be examined. Violent stalking behaviours
could be perceived as a way to achieve original stalking goals (e.g. revenge or
reconciliation) or as a way to ‘remove barriers’ to achieve these goals (e.g. the
victim’s new partner). Alternatively, violence may be unrelated to the stalkers’
overall goals and may be a response to a particular incident that occurred during
the course of stalking.
The rational choice perspective may also provide a useful conceptual frame-
work for understanding potential differences between male and female stalkers.
Not only could the purpose of stalking differ across male and female stalkers
(i.e. motives; Purcell et al., 2001; Strand and McEwan, 2012), but their deci-
sion-making processes may differ because the outcomes of similar behaviours
may be perceived differently according to the sex of the offender. Additionally,
the perceived risks associated with similar behaviours may vary across sex. For
example, given typical power imbalances between males and females physically,
males may believe that they can influence/control the victim’s behaviour with less
The phenomenon of stalking 75
intrusive behaviours than females may need to adopt for the same level of influ-
ence/control. Additionally, since the severity of female-perpetrated aggression
and violence is often underestimated by both victims and the justice system (e.g.
Archer, 2000), females may evaluate the costs of aggressive behaviours as less
risky than a male perpetrating similar behaviours (e.g. see Thompson, Dennison
and Stewart, 2012). Consequently, it is possible that males and females use very
different behaviours for similar purposes or may use similar behaviours for very
different purposes.

Situational precipitators
The rational choice perspective views criminal behaviour as purposeful. Before
deciding on whether or not to commit crime, the offender uses cues from the
environment to make the best decision possible (Cornish and Clarke, 1986;
2008). What matters from this point of view is how the immediate environment
enables criminal behaviour. However, as argued by Wortley (2008), the immedi-
ate environment can also encourage or induce individuals who would not have
contemplated crime in the first place – the environment can precipitate criminal
behaviour. As pointed out by Wortley (2008), the concept of situational precipita-
tors complements the rational choice perspective. The environment can generate
opportunities for crime for already motivated offenders, intensify existing moti-
vations to offend or create motivations to offend in individuals who may not have
offended in other situations. Wortley (2008) identifies four types of precipitators,
that is, prompts, pressures, permissions and provocations. The environment can:
1) prompt individuals to commit crime by bringing to the surface feelings and
desires that would not emerge in other situations (e.g. pubs with violent reputa-
tions), 2) exert pressure on individuals to offend (e.g. conform to peer pressure),
3) permit individuals to engage in normally proscribed behaviour such as crime by
weakening moral prohibitions (e.g. blaming alcohol for violent altercations), and
4) provoke a criminal response by creating a high level of stress in the individual
(e.g. road rage due to frustration). Situational precipitators may also interfere with
offenders’ abilities to make decisions, by limiting the availability or viability of
alternative courses of action. It is also possible that males and females may react
differently to similar precipitators. Therefore, different factors may precipitate
offending for males and females.
The importance of situational precipitators has been documented in the
stalking literature, although referred to as dramatic moments (Meloy, 1997) or
triggering events (Thompson, 2009; Thompson et al., in press). Although few
researchers have investigated situational precipitators, there is some evidence
of the importance of several situational precipitators on stalking (e.g. disputes;
Harris, 2000; custody battles; Meloy and Boyd, 2003; substance use, Logan,
Leukefeld and Walker, 2000; Meloy and Boyd, 2003; losses in work or love,
Meloy et al., 2011; Kienlen, Birmingham, Solberg, O’Regan and Meloy, 1997)
and stalking violence (substance use; McEwan et al., 2009; protection orders,
Brewster, 2003; Spitzberg, 2002; arguments, confrontations and conflict;
76 Carleen M. Thompson and Benoit Leclerc
Thompson, 2009; Thompson et al., in press; loss of employment and provo-
cation, Thompson, 2009). Additionally, one study indicated that precipitating
events were more likely to precede female-perpetrated stalking than male-per-
petrated stalking (i.e. predominantly losses; Meloy et al., 2011). However, it is
possible that male-perpetrated stalking is just associated with different types of
precipitating factors than those measured in this study (e.g. alcohol use; Logan
et al., 2000), possibly contributing to differences in male and female stalkers’
motives. Given the paucity of research, there is clearly a need for further inves-
tigation of situational precipitators, including an examination of differences
across sex.

The current study

The research reviewed above suggests that the rational choice perspective, together
with situational precipitators, may have some explanatory value for understand-
ing the phenomenon of stalking and how stalking may differ across sex. Indeed
even though a number of studies have examined the types of behaviours perpe-
trated by stalkers and the motives and feelings associated with these behaviours,
there is still a need to understand how motives and feelings vary across sex, as
well as the impact of situational precipitators on stalking behaviour. Moreover,
while researchers have compared the broader motivations for stalking across vio-
lent and non-violent stalkers, there is a need to examine the violent event itself.
This will enable researchers to understand how motives, feelings and situational
precipitators lead to violence and whether these factors differ across sex. This
study examines stalking that occurred after the dissolution of an intimate relation-
ship or in pursuit of a new intimate relationship to address these gaps.
Six research questions are addressed:

1 What types of behaviours do relational stalkers perpetrate and do these differ

across sex?
2 What reasons are reported for perpetrating relational stalking behaviour and
do these differ across sex?
3 What feelings lead to stalking behaviours and do these differ across sex?
4 What types of violent behaviours do violent relational stalkers perpetrate and
do these differ across sex?
5 What reasons are reported for perpetrating stalking violence and do these dif-
fer across sex?
6 What feelings lead to stalking violence and do these differ across sex?

By addressing these research questions, the present study aims to contribute to

the development of theories explaining stalking and associated violence across
sex. Understanding these dimensions of stalking may facilitate the identification
of strategies to prevent the continuation of stalking or the escalation to violence
according to the sex of the offender.
The phenomenon of stalking 77
The data used in this study were derived from a larger study examining the risk
factors for stalking violence (see Thompson et al., 2012; in press). Consequently,
the participants, materials and procedure have been described elsewhere in
detail (see Thompson et al., in press). While all of the relevant components of
the method will be described for this study, interested readers can consult these
papers for further information.

The sample comprised 703 participants classified as relational stalkers from
South-East Queensland (Australia). These relational stalkers were derived from
a convenience sample of 1,738 participants who completed a questionnaire
identifying stalking behaviour in the pursuit and/or dissolution of intimate rela-
tionships (1,035 participants did not meet the criteria for stalking; 59.6 per cent).
The sample of relational stalkers was disproportionately female (75.2 per cent
female; 19.1 per cent male, 5.6 per cent missing data), primarily because a dis-
proportionate number of females completed the questionnaire (70.2 per cent
female, 24.4 per cent male, missing data = 5.4 per cent). The reasonably com-
parable rates of stalking across sex in this study (φ = .067) are consistent with
previous research using similar samples and definitions of stalking (e.g. Cupach
and Spitzberg, 2000; Dutton and Winstead, 2006; Langhinrichsen-Rohling
et al., 2000; Sinclair and Frieze, 2005; Spitzberg, Nicastro and Cousins, 1998).
Relational stalkers primarily targeted persons of the opposite sex (n= 620,
88.2 per cent; missing n = 42, 6.0 per cent). Consequently, the majority of victims
were male (i.e. 75 per cent). Participants typically reported perpetrating stalking
after an intimate relationship dissolved (n= 638; 90.8 per cent), rather than dur-
ing the pursuit of an intimate relationship (n= 65, 9.2 per cent). Sixty per cent
of stalking persisted for longer than one month (n = 422) and approximately
20 per cent persisted for more than six months (n = 125; 17.8 per cent). The ages
of participants classified as relational stalkers ranged from 17 to 62 years (M
= 23.5 years, SD = 8.5, missing n = 44). Participants were treated according to
the ethical principles of the National Health and Medical Research Council and
the Griffith University Human Research Ethics Committee (Reference number

Participants were recruited for the questionnaire through a range of conveni-
ence sampling techniques. First, 600 brochures advertising the questionnaire
were distributed to residential letterboxes in four randomly selected suburbs
from South-East Queensland. Second, 800 brochures advertising the web-based
questionnaire were distributed in South-East Queensland’s central business dis-
trict (CBD) (including the city mall, busy CBD footpaths and inner-city train
78 Carleen M. Thompson and Benoit Leclerc
stations). Third, an email was distributed to the ‘Crimnet’ mailing list (includ-
ing academics, professionals, practitioners and students in the criminal justice
field) inviting participation. Fourth, Griffith University staff and students were
invited to participate through a university-wide email, an advertisement on the
Griffith University computer laboratory homepage and during undergraduate
lectures. Griffith University staff and students comprised 40 per cent of the total
sample (n = 712).
Participants were told that the purpose of the study was to examine how indi-
viduals behave when they pursue new intimate relationships or break up from
existing intimate relationships. Although describing the purpose of the study in
this way may have circumvented error associated with preconceived notions of
what constitutes stalking, it is possible that a relationship questionnaire is more
appealing to female participants and may have contributed to the disproportion-
ate number of female participants in this study. However, it is also possible that
females are more willing to participate in research in general. Regardless of the
reason, the procedure adopted in this study did identify substantially more female
than male participants.

A self-report questionnaire was used to measure (a) male and female par-
ticipants’ perpetration of relational stalking, their rationales for engaging in
relational stalking behaviours, situational precipitators and feelings and (b)
male and female participants’ perpetration of violence during the course of
stalking, their rationales for engaging in stalking violence, situational pre-
cipitators and feelings. The questionnaire took up to 60 minutes to complete,
depending on the answers provided by participants (i.e. whether more informa-
tion was required on the basis of answers to previous questions). Participants
who had not perpetrated any stalking-like behaviours completed the question-
naire in less than 15 minutes.

Relational stalking
Relational stalking behaviours. Relational stalking behaviours were measured
using an amended version of Spitzberg and Cupach’s (1997) ‘Relational Pursuit:
Pursuer Short Form’ (see Thompson, 2009 for scale items). This scale asks par-
ticipants to report whether they had ever engaged in any of 25 behaviours that
they knew were unwanted after they had broken up with an intimate partner or
pursued an intimate relationship (scored never, once, two or more, five or more,
or ten or more times; Cronbach’s alpha coefficient = .83). Minor changes were
made to the scale items to maintain consistency with stalking legislation (e.g.
omitting items such as making exaggerated expressions of affection; California
Penal Code, Section § 646.9; Criminal Code [Stalking] Amendment Act 1999
[QLD]; Canada’s Criminal harassment Act, 1993). Examples of stalking behav-
iours included in the checklist are following him/her, making unwanted telephone
The phenomenon of stalking 79
calls and making verbal threats. Although legislation often requires that the victim
did, or could reasonably, experience fear, this cannot be reliably assessed through
perpetrator self-reports (see Thompson and Dennison, 2008).
Participants who perpetrated five or more unwanted intrusions from the stalking
checklist against any one person were defined as relational stalkers. This cut-point
has been used in other research (e.g. Thompson et al., 2012; in press) and is based
on research conducted by Thompson and Dennison (2008) that indicated that a
cut-point of five balanced the issues associated with over- and under-attribution
of stalking in samples of unconvicted perpetrators. Specifically, although stalk-
ing legislation typically requires two or more behaviours (e.g. California Penal
Code, Section § 646.9; see Ogilvie, 2000) a cut-point of two is judged as too
encompassing in studies using self-report perpetration questionnaires that cannot
reliably assess victim fear (Thompson and Dennison, 2008). At the same time, a
cut-point of ten or more behaviours is much higher than legislative requirements
and excludes a lot of violent and threatening behaviour that may have, in fact,
induced fear. Therefore, relational stalking is operationally defined in this study
as five or more unwanted intrusions perpetrated against any one person after the
dissolution of an intimate relationship, or in the pursuit of an intimate relation-
ship (this definition was also adopted in Thompson et al., 2012 and Thompson
et al., in press). While it is unlikely that all participants classified as relational
stalkers would be convicted under stalking legislation, particularly given that vic-
tim fear could not be assessed, these participants will be referred to as relational
stalkers for simplicity.
Reasons for engaging in relational stalking behaviours. Participants who had
engaged in stalking-like behaviours were asked to report the reasons why they per-
petrated these behaviours. They were then provided with 30 reasons from which
they could select multiple responses. The reasons incorporated in the checklist
included those that could be considered general rationales as well as situational
precipitators. Examples of items that could be considered general rationales
include amorous rationales such as ‘you wanted to get back together with them’
as well as sinister rationales such as ‘you wanted to hurt this person physically’.
Examples of items that could be considered situational precipitators include
‘because he or she confronted you’ or ‘you were drunk/stoned/“wasted”’. Items
suggesting a loss of control were also included (e.g. ‘you lost control’ or ‘the
situation got out of control’). All of the reasons included in the list were derived
from previous stalking research as well as domestic violence research (Dennison
and Stewart, 2006; Follingstad, Bradley, Laughlin and Burke, 1999; Holtzworth-
Munroe and Stuart, 1994; Meloy and Boyd, 2003; Spitzberg and Cupach, 2007;
Wilkinson and Hamerschlag, 2005; Wright et al., 1996). Participants were also
given the opportunity to list any additional reasons for perpetrating stalking-
like behaviours that were excluded from the list. Although stalking occurs over
time and encompasses multiple behaviours, it was not possible to measure par-
ticipants’ reasons for perpetrating each individual stalking behaviour using a
questionnaire methodology, given that some participants could perpetrate hun-
dreds of behaviours over time.
80 Carleen M. Thompson and Benoit Leclerc
Feelings associated with stalking. Feelings associated with stalking behav-
iours were assessed by asking participants ‘How did you feel when you were
behaving this way?’ (i.e. stalking) and they were provided with 21 feelings
from which they could select all that were appropriate. The feelings included
in the list were derived from previous stalking and domestic violence research
(Dennison and Stewart, 2006; Follingstad et al., 1999; Holtzworth-Munroe and
Stuart, 1994; Meloy and Boyd, 2003; Spitzberg and Cupach, 2007). Feelings
included in the list ranged from love to rage to suicidal. Again, it was not pos-
sible to measure how participants felt prior to perpetrating each individual
stalking behaviour since some stalkers were likely to perpetrate hundreds of
stalking-like behaviours.

Stalking violence
Stalking violence behaviours. Violence during the course of stalking was meas-
ured using two subscales from the Revised Conflict Tactics Scale (CTS2) (physical
assault and sexual coercion subscales; CTS2, Straus, Hamby, Boney-McCoy and
Sugarman, 1996). This scale was also amended to include additional violent behav-
iours commonly reported in stalking violence research, such as running the victim’s
car off the road (amended scale is available in Thompson, 2009; Cronbach’s alpha =
.83). The final scale included 23 physically violent acts that were perpetrated against
the stalking victim or someone for whom this person cared, during the course of
stalking. Examples of behaviours included in the checklist are punching or hitting
with something that could hurt, choking, slapping, pushing or shoving. This scale
has been used to define violence in previous stalking violence research (Thompson
et al., 2008; Thompson et al., 2012; in press).
Reasons for perpetrating violence during the course of stalking. Participants
who had perpetrated stalking violence were asked to report the reasons for these
violent behaviours. Participants were provided with 30 reasons from which they
could select multiple responses. The reasons incorporated in the checklist were
identical to those used for stalking and included both general rationales as well as
situational precipitators. Additionally, participants were given the opportunity to list
any additional reasons for perpetrating stalking violence that were excluded from
the list.
Feelings immediately before perpetrating stalking violence. Feelings preced-
ing violence were assessed by asking participants ‘How did you feel immediately
before you acted this way? [i.e. violently]’. Participants were provided with 21 feel-
ings from which they could select all that were appropriate. The feelings included
in the list were identical to those included in the list for stalking in general.

Three dimensions of stalking were examined according to the sex of the par-
ticipant: (a) the stalking behaviours perpetrated by stalkers, (b) the rationales
for engaging in stalking behaviours (including situational precipitators), and (c)
The phenomenon of stalking 81
associated feelings. Subsequently, these three dimensions were examined for
stalking violence according to the sex of the participant. Chi square analyses were
used to investigate between-sex differences. Fisher’s exact tests were used when
the assumptions of the chi square tests were violated (i.e. expected frequencies
fell below five). Owing to missing data on sex (n = 42), the sample size for all
analyses is 661. Given the exploratory nature of this study, Bonferroni’s adjust-
ment will not be utilised (see Jaccard and Wan, 1996).

Relational stalking
Research question 1: What types of behaviours do relational stalkers
perpetrate and do these differ across sex?
Both male and female relational stalkers perpetrated stalking behaviours on an
average of 16 different occasions (i.e. different stalking events; female range = 5
to 118; male range = 5 to 67; t [659] = .56, p = .58). Additionally, both male and
female stalkers perpetrated an average of six different types of stalking behav-
iours across these events (e.g. loitering, following, phone calls; female range =
1 to 25; male range = 1 to 21; t [659] = .49, p = .62). There were few differences
between the individual types of stalking behaviours perpetrated by male and
female relational stalkers (see Table 5.1). For both males and females, the most
common stalking behaviours were pursuit, proximity and surveillance behaviours
(i.e. monitoring, showing up, watching, loitering and approaching) and contact-
ing the target (i.e. leaving messages, telephone calls). These are arguably the least
intrusive types of stalking behaviours. None of these behaviours differed across
sex. Male and female stalkers differed on just four types of stalking behaviours.
Male stalkers were more likely than female stalkers to follow the victim, leave/
send unwanted gifts or threaten people the victim cared about (although still only
10 per cent of males made these threats). Female stalkers, on the other hand, were
more likely to perpetrate physical violence than their male counterparts, whereby
violence was perpetrated by nearly half of the female stalkers compared to one
third of males. Threats were less common than physical violence for both males
and females and stalkers of both sexes were more likely to threaten to hurt them-
selves than the victim.

Research question 2: What reasons are reported for perpetrating

relational stalking behaviour and do these differ across sex?
Two stalkers did not report their reasons for stalking the victim and are excluded
from the following analyses. The reasons provided by all other stalkers are
reported in Table 5.2 according to sex. Although 36 (5.4 per cent) participants
provided additional reasons for stalking, most of these responses could be sub-
sumed under the existing 30 responses and thus were recoded accordingly (e.g.
‘rejected for someone else’ was coded as ‘because he or she rejected you’). Three
additional reasons were identified in these responses: to understand why the
82 Carleen M. Thompson and Benoit Leclerc
Table 5.1 Stalking behaviours in a convenience sample of relational stalkers in Queensland
Australia (N= 661)

Stalking behaviours Males n Females n Phi

(%) (%)

Monitoring him/her or his/her behaviour 101 (68.7) 365 (71.0) .021

Showing up at places 89 (60.5) 336 (65.4) .042
Watching him/her 95 (64.6) 305 (59.3) –.045
Leaving unwanted messages 85 (57.8) 304 (59.1) .011
Approaching him/her 91 (61.9) 279 (54.3) –.064
Making unwanted telephone calls to him/her 59 (40.1) 248 (48.2) .068
Loitering near him/her or places where he/she 69 (46.9) 235 (45.7) –.010
lives, works or visits
Physical violencea 46 (31.3) 234 (45.5) .120**
Covertly obtaining private information 51 (34.7) 181 (35.2) .005
Following him/her around 53 (36.1) 142 (27.6) –.077*
Threatening to hurt yourself 35 (23.8) 105 (20.4) –.034
Leaving/sending unwanted gifts 45 (30.6) 60 (11.7) –.215***
Verbally threatening him/her personally 12 (8.2) 69 (13.4) .067
Leaving unwanted threatening messages 15 (10.2) 62 (12.1) .024
Intimidating or harassing him/her 11 (7.5) 54 (10.5) .042
Invading his/her personal property 16 (10.9) 44 (8.6) –.034
Stealing or damaging his/her possessions 8 (5.4) 40 (7.8) .037
Threatening others she/he cares about 15 (10.2) 28 (5.4) –.080*
Ordering something on his/her behalf 10 (6.8) 19 (3.7) –.063
Engaging in regulatory harassment 4 (2.7) 20 (3.9) .026
Leaving/sending offensive or unusual materialb 6 (4.1) 10 (1.9) –.058
Other – To understand why the relationship 1 (0.7) 6 (1.2)
had endedc
Other – Seek forgiveness for wrongdoings in the 3 (2.0) 0 (0.0)
Other – Believed they were receiving mixed 1 (0.7) 3 (0.6)

a For simplicity and consistency, physical violence is measured using the stalking violence checklist
based on the CTS2.
b Fisher’s exact tests were used owing to expected frequencies below 5. All other comparisons were
made using chi square analyses.
c Comparisons were not made across ‘other’ responses as these reasons were not assessed equally
across all participants.
* p > .05 ** p > .01 *** p > .001
The phenomenon of stalking 83
Table 5.2 Reasons for perpetrating stalking behaviours in a convenience sample of
relational stalkers in Queensland Australia (N= 659)

Reasons for perpetrating stalking behaviours Males n Females n Phi

(%) (%)

Because you were very emotional 78 (53.1) 356 (69.3) .142***

Because you loved him or her 94 (63.9) 312 (60.7) –.028
To get back together with him/her 75 (51.0) 265 (51.6) .004
Because you felt betrayed by this person 55 (37.4) 259 (50.4) .108**
You wanted a relationship with him or her 67 (45.6) 211 (41.1) –.038
Because he or she rejected you 54 (36.7) 198 (38.5) .015
Because you were frustrated 46 (31.3) 198 (38.5) .062
You were angry with this person 32 (21.8) 200 (38.9) .149***
To show him/her you liked/wanted them 70 (47.6) 140 (27.2) –.182***
Because you were in an argument 23 (15.6) 137 (26.7) .107**
Someone else was coming between you and this 32 (21.8) 108 (21.0) –.008
Because you were humiliated 18 (12.2) 122 (23.7) .117**
You were retaliating against what they did to you 19 (12.9) 116 (22.6) .099*
The situation got out of control 19 (12.9) 115 (22.4) .098*
You were embarrassed that you were rejected 20 (13.6) 111 (21.6) .083*
Because you were in a dispute or in conflict with 25 (17.0) 106 (20.6) .038
him or her
You lost control 17 (11.6) 103 (20.0) .091*
You wanted to hurt this person emotionally 15 (10.2) 84 (16.3) .072
Because he or she was rude to you 16 (10.9) 78 (15.2) .051
To get them back for what they did to you 13 (8.8) 75 (14.6) .070
Because you were drunk/stoned/wasted 20 (13.6) 64 (12.5) –.014
To catch this person in the act or doing 24 (16.3) 60 (11.7) –.058
something wrong
You wanted to get revenge 10 (6.8) 65 (12.6) .077*
You were provoked 10 (6.8) 50 (9.7) .042
It was your last resort 13 (8.8) 42 (8.2) –.010
Because you were upset about something else, 9 (6.1) 44 (8.6) .037
for example, you were upset because of a
custody battle
Someone this person cared about was ruining 20 (13.6) 30 (5.8) –.122**
your plans to get together with him/her
Because he or she confronted you 7 (4.8) 32 (6.2) .026
You wanted to hurt this person physicallya 1 (0.7) 16 (3.1) .064
You wanted to frighten this persona 2 (1.4) 4 (0.8) –.026

a Fisher’s exact tests were used owing to expected frequencies below 5. All other comparisons were
made using chi square analyses.
* p > .05 ** p > .01 *** p > .001
84 Carleen M. Thompson and Benoit Leclerc
relationship had ended, seek forgiveness for wrongdoings in the relationship and
the participant believed they were receiving mixed signals. Most stalkers reported
multiple reasons for perpetrating stalking behaviours (92.2 per cent); with female
stalkers (M = 7.20; SD = 4.71) reporting significantly more reasons than male
stalkers (M = 6.12; SD = 4.32; t [659] = –2.43, p = .015). There were many and
varied combinations of reasons, suggesting both complexity and individual vari-
ability. Although approximately 20 per cent of females and 10 per cent of males
reported that the situation got out of control or that they lost control, all of these
stalkers also provided other reasons for their behaviour.
The most common reasons reported for both males and females were ‘you were
emotional’, ‘you loved him/her’, ‘to get back together’, ‘you felt betrayed’ and
‘you wanted a relationship’; however, females were significantly more likely to
report their reasons as ‘very emotional’ and ‘feeling betrayed’. Females were also
more likely to report stalking the victim because they were angry, humiliated, or
embarrassed they were rejected, they wanted revenge, the situation got out of con-
trol or they lost control. Additionally, females were more likely to report that their
stalking behaviour was a response to an argument or retaliation against something
the victim did. In contrast, males were more likely to report stalking the victim ‘to
show him/her that they liked/wanted them’ and were more likely to report stalk-
ing behaviour being a response to ‘someone ruining his/her plans to get together
with the victim’. Being drunk/stoned/wasted, provoked and confronted, played a
relatively low role in stalkers’ reasoning and did not differ across sex. Very few
stalkers of either sex said that they perpetrated stalking behaviours to frighten or
physically hurt the victim and the vast majority did not say that they intended to
hurt the victim emotionally either.

Research question 3: What feelings lead to stalking behaviours and do

these differ across sex?
Most stalkers reported a combination of feelings leading to their stalking
behaviours (91.7 per cent) (see Table 5.3). Both males and females reported an
average of seven feelings (female range = 1 to 16; male range = 1 to 19; t [659]
= –1.13, p = .26). In general, there were many and varied combinations of feel-
ings reported and at times these feelings seem contradictory. The most common
feelings reported for both males and females were hurt, sad, depressed, upset,
thinking about the victim a lot, not being able to get the victim off his/her mind
and frustration. Although more than 20 feelings were examined, males and
females differed on only three feelings. Specifically, females reported higher
rates of feeling upset and angry. Males, on the other hand, reported higher rates
of feeling in love. Jealousy was only reported by one third of relational stalkers,
both male and female.
The phenomenon of stalking 85
Table 5.3 Feelings leading to stalking behaviours in a convenience sample of relational
stalkers in Queensland Australia (N= 661)

Feelings leading to stalking behaviours Males n Females n Phi

(%) (%)

Hurt 80 (54.4) 301 (58.6) .035

Sad 86 (58.5) 291 (56.6) –.016
Depressed 90 (61.2) 291 (56.6) .039
Upset 68 (46.3) 299 (58.2) .100**
Thought about him/her a lot 81 (55.1) 276 (53.7) –.012
Couldn’t get him/her off my mind 83 (56.5) 269 (52.3) –.034
Frustrated 70 (47.6) 264 (51.4) .031
Vulnerable 52 (35.4) 226 (44.0) .072
Let down 62 (42.2) 212 (41.2) –.008
Angry 32 (21.8) 200 (38.9) .149***
Jealous 54 (36.7) 170 (33.1) –.032
In love 58 (39.5) 146 (28.4) –.099*
Anxious 39 (26.5) 163 (31.7) .047
Unable to cope 35 (23.8) 160 (31.1) .067
Tense 39 (26.5) 131 (25.5) –.010
Annoyed 29 (19.7) 137 (26.7) .066
Rage 10 (6.8) 64 (12.5) .074
Vengeful 11 (7.5) 60 (11.7) .056
Suicidal 13 (8.8) 58 (11.3) .033
Excited 10 (6.8) 24 (4.7) –.040
Happya 6 (4.1) 13 (2.5) –.039

a Fisher’s exact tests were used owing to expected frequencies below 5. All other comparisons were
made using chi square analyses.
* p > .05 ** p > .01 *** p > .001

Stalking violence
Research question 4: What types of violent behaviours do violent
relational stalkers perpetrate and do these differ across sex?
The majority of violence reported by both males and females in this sample could
be considered moderate in nature, and included pushing or shoving, grabbing,
slapping and throwing something that could hurt (see Table 5.4). The higher rates
of female-perpetrated violence observed in this sample (refer to Table 5.1) is due
to female stalkers perpetrating much higher rates of these types of violence than
male stalkers (with the exception of grabbing). Males were more likely to slam
the victim against a wall or grab someone the victim cared about, although these
were still only perpetrated by 20 per cent and 11 per cent of males respectively.
Although there was more symmetry across sex for potentially more serious types
of violence (i.e. forcing sexual contact, kicking, punching, biting, tackling), these
types of violence were uncommon in the sample. Additionally, very serious forms
of violence were rare. For example, of all 703 relational stalkers, one stalker used
86 Carleen M. Thompson and Benoit Leclerc
Table 5.4 Violent behaviours in a convenience sample of relational stalkers in Queensland
Australia (n = 280)

Violent behaviours Males n Females n Phi

(%) (%)

Push or shove this person 15 (32.6) 163 (69.7) .285***

Grab this person 25 (54.3) 116 (49.6) –.035
Slap this person 9 (19.6) 125 (53.4) .251***
Throw something at this person that could 7 (15.2) 108 (46.2) .233***
Force any sexual contact on this person, 4 (8.7) 20 (8.5) –.022
including kissinga
Kick this person 3 (6.5) 37 (15.8) .098
Punch or hit this person with something 3 (6.5) 37 (15.8) .098
that could hurt
Twist this person’s arm or hair 8 (17.4) 28 (12.0) –.060
Slam this person against a walla 9 (19.6) 20 (8.5) –.134*
Bite this persona 3 (6.5) 22 (9.4) .037
Push or shove someone this person cared 7 (15.2) 16 (6.8) –.113
Tackle this persona 2 (4.3) 18 (7.7) .048
Throw something at someone this person 2 (4.3) 17 (7.3) .043
cared about that could hurta
Grab someone this person cared abouta 5 (10.9) 9 (3.8) –.119*

Note: As the scale included 46 violent acts, of which 29 were perpetrated by less than 5 per cent of
stalkers, for clarity and simplicity, only behaviours perpetrated by 5 per cent or more were
included in this table.
a Fisher’s exact tests were used owing to expected frequencies below 5. All other comparisons were
made using chi square analyses.
* p > .05 *** p > .001

a knife or gun, six stalkers forced sexual intercourse with the victim, one stalker
dragged the victim along the road using their car, one stalker burned or scalded
the victim and seven stalkers used their car to run the victim’s car off the road.

Research question 5: What reasons are reported for perpetrating

stalking violence and do these differ across sex?
Three stalkers did not report their reasons for violence and are excluded from
the following analyses. The reasons provided by all other violent stalkers are
reported in Table 5.5 according to sex. Although three participants provided
additional reasons for perpetrating stalking violence, all of these responses
could be subsumed under the existing 30 responses and thus were recoded
accordingly (e.g. release anger was recoded as angry). Stalkers typically
reported multiple reasons for perpetrating violence (91.5 per cent); with com-
parable numbers of reasons across male (M = 7.24; SD = 5.18) and female
The phenomenon of stalking 87
Table 5.5 Reasons for using violence in a convenience sample of violent relational stalkers
in Queensland Australia (N= 277)

Reasons for perpetrating stalking violence Males n Females n Phi

(%) (%)

You were angry with this person 22 (47.8) 169 (72.2) .194***
Because you were very emotional 20 (43.5) 151 (64.5) .160**
Because you were in an argument 17 (37.0) 145 (62.0) .188**
Because you felt betrayed by this person 18 (39.1) 114 (48.7) .071
Because you were frustrated 18 (39.1) 114 (48.7) .071
Because you loved him or her 20 (43.5) 95 (40.6) –.022
You lost control 17 (37.0) 92 (39.3) .018
Because he or she rejected you 16 (34.8) 92 (39.3) .035
Because you were in a dispute or in conflict 10 (21.7) 95 (40.6) .144*
with him or her
The situation got out of control 14 (30.4) 87 (37.2) .052
To get back together with him/her 21 (45.7) 77 (32.9) –.099
Because he or she was rude to you 12 (26.1) 69 (29.5) .028
You were retaliating against what they did to 13 (28.3) 65 (27.8) –.004
You were provoked 12 (26.1) 58 (24.8) –.011
You wanted a relationship with him or her 16 (34.8) 52 (22.2) –.109
Because you were humiliated 5 (10.9) 63 (26.9) .139*
You wanted to hurt this person emotionally 5 (10.9) 58 (24.8) .123*
You wanted to get revenge 9 (19.6) 47 (20.1) .005
To get them back for what they did to youa 4 (8.7) 53 (22.6) .128*
Someone else was coming between you and 10 (21.7) 39 (16.7) –.049
this person
Because you were drunk/stoned/wasted 8 (17.4) 37 (15.8) –.016
You were embarrassed that you were rejected 6 (13.0) 37 (15.8) .028
Because he or she confronted you 5 (10.9) 37 (15.8) .051
To show him/her you liked/wanted them 12 (26.1) 26 (11.1) –.162**
Someone this person cared about was ruining 10 (21.7) 21 (9.0) –.151*
your plans to get together with him/her
Because you were upset about something else, 5 (10.9) 24 (10.3) –.007
for example, you were upset because of a
custody battlea
You wanted to hurt this person physicallya 0 (0.0) 28 (12.0) .148**
It was your last resorta 3 (6.5) 23 (9.8) .042
To catch this person in the act or doing 4 (8.7) 11 (4.7) –.066
something wronga
You wanted to frighten this persona 1 (2.2) 13 (5.6) .057

a Fisher’s exact tests were used owing to expected frequencies below 5. All other comparisons were
made using chi square analyses.
* p > .05 ** p > .01 *** p > .001
88 Carleen M. Thompson and Benoit Leclerc
stalkers (M = 8.51; SD = 5.69; t [278] = –1.41, p = .161). In general, many
and varied combinations of reasons were reported which, at times, may appear
contradictory (e.g. ‘you were angry at this person’ reported alongside ‘you
wanted to get back together with him/her’). The most common reasons for
violence reported by both males and females were ‘because I was angry’,
‘I was very emotional’, ‘I felt betrayed’ and ‘I was frustrated’. However,
females were significantly more likely to report their reasons for violence
as ‘being angry’ and ‘very emotional’; both of which were reported by more
than two thirds of violent females. Females were also more likely to be violent
because they were humiliated, they wanted to hurt the person emotionally, to
get the victim back for what they did to her and to hurt the victim physically
(although still just over 10 per cent reported this latter reason). Additionally,
females were more likely to be violent in response to an argument or a dis-
pute/conflict, whereby 70 per cent of female-perpetrated violence occurred in
one of these contexts compared to 43.5 per cent of male-perpetrated violence.
Interestingly, despite these aggressive reasons being reported for female-per-
petrated violence, still 40 per cent of females reported that they were violent
because they were in love with the victim. A similarly high proportion of
males reported this amorous rationale for violence. Compared to females,
males were more likely to be violent in response to ‘someone ruining his/her
plans to get together with the victim’ and more likely to be violent because he
‘wanted to show the victim he wanted/liked her’, along with trends towards
wanting reconciliation/intimacy (i.e. ‘wanted to get back together’ or ‘wanted
a relationship’). Few violent male stalkers were aiming to hurt the victim
emotionally and none reported aiming to hurt the victim physically. Being
drunk/stoned/wasted and confronted were reported as reasons for violence by
less than 20 per cent of both males and females. Very few stalkers of either
sex said that they perpetrated violence to frighten the victim.

Research question 6: What feelings lead to stalking violence and do

these differ across sex?
Most stalkers reported that they experienced a combination of feelings imme-
diately before they perpetrated violence (91.5 per cent). This was the case for
both male (M = 5.87; SD = 4.51) and female relational stalkers (M = 6.94;
SD = 4.52), t [278] = –1.48, p = .14). The most common feelings reported
for both males and females were being angry, upset, frustrated, depressed and
sad; however, females reported significantly higher rates of being angry, upset
and frustrated than males. Females also reported higher rates of feeling hurt.
Although anger was the most prevalent feeling for both sexes, more extreme
feelings of rage and vengeance were experienced by far fewer violent stalkers
(both male and female). Compared to violent females, violent males reported
higher rates of feeling ‘in love’. Jealousy was reported by one third of violent
male and female stalkers.
The phenomenon of stalking 89
Table 5.6 Feelings leading to violent behaviours in a convenience sample of violent
relational stalkers in Queensland Australia (n = 280)

Feelings leading to violent behaviours Males n Females n Phi

(%) (%)

Angry 21 (45.7) 149 (63.7) .137*

Upset 20 (43.5) 140 (59.8) .122*
Frustrated 19 (41.3) 144 (61.5) .152*
Hurt 12 (26.1) 142 (60.7) .258***
Depressed 21 (45.7) 106 (45.3) –.003
Sad 19 (41.3) 101 (43.2) .014
Let down 17 (37.0) 92 (39.3) .018
Tense 13 (28.3) 81 (34.6) .050
Couldn’t get him/her off my mind 15 (32.6) 70 (29.9) –.022
Unable to cope 10 (21.7) 77 (32.9) .089
Vulnerable 13 (28.3) 74 (31.6) .027
Annoyed 9 (19.6) 75 (32.1) .101
Jealous 14 (30.4) 68 (29.1) –.011
Anxious 12 (26.1) 68 (29.1) .024
Thought about him/her a lot 14 (30.4) 55 (23.5) –.060
Rage 8 (17.4) 60 (25.6) .071
In love 16 (34.8) 44 (18.8) –.144*
Vengeful 7 (15.2) 51 (21.8) .060
Suicidala 5 (10.9) 23 (9.8) –.013
Exciteda 3 (6.5) 4 (1.7) –.114
Happya 2 (4.3) 1 (0.4) –.141

a Fisher’s exact tests were used owing to expected frequencies below 5. All other comparisons were
made using chi square analyses.
* p > .05 *** p > .001

This study aimed to improve the current understanding of stalking behaviour by
examining factors related to the crime event/s and its immediate environment
and how these factors may differ according to sex. Guided by a rational choice
perspective and a situational precipitators framework, this study investigated
sex differences in the behaviours selected by stalkers, the reasons for engag-
ing in stalking behaviours and associated feelings. Subsequently, each of these
dimensions was examined for violence within stalking cases to understand how
motives, feelings and situational precipitators led stalkers to escalate to violence
and whether these factors differed across sex.
Prior to discussing the findings, it is important to note two implications of the
sample used in this study. First, the majority of stalking examined in this study
occurred after the dissolution of an intimate relationship. Partner stalking is often
regarded as a form of intimate partner violence (IPV) or an extension of IPV that
occurred during the relationship (Melton, 2007; Norris, Huss and Palarea, 2011)
and therefore may differ from stalking in other contexts (e.g. stranger stalking,
90 Carleen M. Thompson and Benoit Leclerc
celebrity stalking; James et al., 2009; 2010; Norris et al., 2011). Additionally,
there are likely to be overlaps between the stalking reported in this study and
IPV. Second, although the comparable rate of stalking across sex reported in
this study is consistent with research using similar methodologies (e.g. Cupach
and Spitzberg, 2000; Sinclair and Frieze, 2005), research using convicted stalk-
ers typically report lower rates of female-perpetrated stalking (often 80 per cent
male-perpetrated; Spitzberg et al., 2010). While these differences may be due to
findings that males are less likely to self-identify as victims, experience fear or
report stalking to the police (see Bjerregaard, 2000; Haugaard and Seri, 2000;
Lyndon et al., 2012; Sheridan and Lyndon, 2012), there are also likely to be dif-
ferences between the nature and severity of stalking across these methodologies
(e.g. see Spitzberg et al., 2010). These distinctions may resemble the differences
between ‘common couple violence’ and more serious, often male-perpetrated vio-
lence (i.e. ‘intimate terrorism’; Johnson, 2006; Johnson and Leone, 2005) reported
in the IPV literature (Archer, 2000; 2002; Johnson, 2006). Consequently, the find-
ings from this study may be generalisable to the less serious forms of stalking
identified in community and university samples, but potentially less generalisable
to more serious forms of stalking identified in convicted samples of stalkers.

Similarities and differences in the nature of stalking across sex

There were many similarities in the nature of stalking perpetrated by male and
female relational stalkers in this sample. Not only was the frequency of stalking
behaviours similar across sex, but there were few differences in the individual
types of stalking behaviours employed by male and female stalkers. Both males
and females most commonly perpetrated pursuit, proximity and surveillance
behaviours (i.e. monitoring, showing up, watching, loitering and approaching)
and contacting the target (i.e. leaving messages, telephone calls). Both the com-
monality of these behaviours and the identification of few differences across sex
are consistent with previous research (Meloy et al., 2011; Purcell et al., 2001;
Spitzberg, 2002; Strand and McEwan, 2012).
In this study, there were also few sex differences identified in relation to the
feelings that led to stalking behaviours, with stalking for both sexes typically pre-
ceded by depressive and obsessive feelings; including feeling hurt, sad, depressed,
upset, thinking about the victim a lot, not being able to get the victim off his/
her mind and frustration. While more sex differences were identified for the rea-
sons for perpetrating stalking, males and females both identified the key reasons
for stalking as being: they were emotional, they loved him/her, they wanted to
get back together with the victim, they felt betrayed and they wanted a relation-
ship. These feelings and reasons for stalking are frequently reported in stalking
research (e.g. Spitzberg and Cupach, 2007) and have been reported as important
for both male and female stalkers (e.g. Meloy and Boyd, 2003).
Despite similarities in the nature of stalking across sex, male and female rela-
tional stalking also differed in fundamental ways across the dimensions examined
in this study. Female-perpetrated stalking encompassed much higher rates of
The phenomenon of stalking 91
physical violence than male-perpetrated stalking; albeit predominantly ‘minor’
forms of violence (Archer, 2000; 2002) such as pushing or shoving, grabbing and
slapping. Although few females threatened the victim, physical violence was per-
petrated by nearly half of female stalkers and was just as common as phone calls
and loitering. Female-perpetrated stalking was also driven by aggressive ration-
ales and feelings of anger significantly more than male-perpetrated stalking. For
females, stalking was largely expressive and represented a mixture of aggressive
and amorous rationales although it could not be determined whether the compet-
ing rationales females reported co-occurred or fluctuated. Females also reported a
combination of angry, depressive and obsessive emotions, with amorous feelings
less prominent. Females were also more likely to report that they were reacting
to situational precipitators, such as arguments, and retaliating against the victim’s
actions. Therefore, female-perpetrated stalking in this sample seemed to be quite
reactive, emotionally driven and related to issues with their (current or previous)
relationship with the victim.
In contrast to female stalkers, male stalkers were more likely to follow the
victim or leave/send gifts but less likely to use violence. In line with these less
overtly aggressive behaviours, the rationales for stalking seemed to be less aggres-
sive and primarily amorous. For instance, males were more likely to stalk their
victim to show him/her he liked/wanted them. Similar to female stalkers, males
also reported a mixture of feelings including depressive and obsessive emotions,
with significantly more males reporting feeling in love and fewer males reporting
feeling angry than females. Although male stalkers were less overtly aggressive
towards the victim than female stalkers, they were more aggressive towards third
parties. Male stalkers were more likely to perpetrate stalking behaviours because
someone was interfering with their attempts at intimacy and were more likely to
threaten people the victim cared about or grab someone the victim cared about,
and there was a trend towards males being more likely to push or shove some-
one the victim cared about. Therefore, male-perpetrated stalking in this sample
seemed to centre around establishing or re-establishing intimacy with the victim
and thwarting others from obstructing these goals. For males, stalking was less
reactive against the victim’s behaviour and more reactive against interference by
third parties.
Similar sex differences were identified for stalking violence as were reported
for stalking. Compared to violent male stalkers, violent female stalkers were
significantly more likely to perpetrate violence to communicate their anger, frus-
tration and pain, and violence was more likely to occur in the context of situational
precipitators such as interpersonal conflict. In a number of cases, these aggressive
reasons and feelings were also mixed with rationales associated with loving the
target and feelings of depression. While anger was the most common rationale for
males, amorous rationales were almost as common, even though violence may
appear counterproductive to these goals. It is possible that male stalkers used
violence in an attempt to coerce females to reciprocate their feelings or reconcile
(e.g. see Brewster, 2003; Davis et al., 2012) although this cannot be confirmed
with this data. Males were also more likely to perpetrate violence against third
92 Carleen M. Thompson and Benoit Leclerc
parties to stop others from interfering with their attempts at intimacy. Therefore,
female stalkers’ violence (like their stalking behaviour in general) was very
reactive, emotionally driven and associated with interpersonal conflict with the
victim. Male stalkers’ violence, in contrast, appeared to be less emotionally and
interpersonally reactive towards the victim and more instrumental to achieving
reconciliation or establishing a new relationship. However, it was more reactive
to interference by third parties.
What is fascinating in this study is that there seem to be key differences between
males’ and females’ reasons for perpetrating relational stalking, and the use of vio-
lence in the context of relational stalking. It appears that male relational stalking is
primarily driven by the pursuit of love and the desperate need to either reconcile
with an ex-intimate or establish a new romantic relationship. On the other hand,
female relational stalking was more commonly driven by a combination of love
and resentment. Female stalkers were found to be more likely than males to feel
betrayed, humiliated and angry and females who used violence were more likely to
report that they wanted to hurt their partner emotionally and/or physically. These
findings are consistent with Strand and McEwan (2012) who also found higher
rates of romantic rationales in male stalkers (albeit socially unskilled expres-
sions of romance) and higher rates of resentment in female stalkers. Additionally,
although most research suggests that males and females perpetrate comparable
rates of stalking violence, females have been reported to perpetrate higher rates
of physically hurting the victim (Dutton and Winstead, 2006) and higher rates of
moderate violence (Sinclair and Frieze, 2000) than their male counterparts.
But why do these differences between female and male stalkers emerge?
Females reported stalking their partner and using violence more than males for
reasons such as having an argument or conflict/dispute. Therefore, one possibility
is that these contexts are, in some cases, more likely to precipitate females than
males towards stalking or an escalation to violence. Another possibility is that
the reason for the dissolution of the relationship differs across sex or has a dif-
ferential impact according to sex. Accordingly, the reason for the dissolution of
the relationship may be associated with the underlying causes of conflicts, argu-
ments, disputes and associated negative feelings experienced by female stalkers,
for example infidelity during the relationship. It is well established that infidel-
ity is more frequently committed by males than their female counterparts (e.g.
see Atkins, Baucom and Jacobson, 2001; Lalasz and Weigel, 2011; Waite and
Gallagher, 2000) and has been associated with both IPV (Nemeth, Bonomi, Lee
and Ludwin, 2012) and stalking (Mullen and Pathe, 1994). Perhaps females are
more likely than males to be enraged, resentful, argumentative and use violence
because their partner was unfaithful to them. In turn, perhaps males stalk their
partner to seek forgiveness, demonstrate their commitment to their relationship/
love because they feel responsible for the relationship going awry or to convince
(or even coerce) them back into the relationship? However, the reason/s why the
relationship dissolved was not covered in the questionnaire. Therefore, infidelity,
and other reasons for the breakdown of the relationship, could not be examined as
potential explanations for sex differences.
The phenomenon of stalking 93
Given that the sample predominantly comprised young adults, it is likely
that many of these stalkers are also inexperienced in establishing and dissolving
romantic relationships. This inexperience may contribute to their poor choices in
the methods they select to reunite with their ex-partner, seek retribution or estab-
lish an intimate relationship. Given that much female-perpetrated stalking and
stalking violence seemed to be linked to complex (sometimes competing) emo-
tions and motivations and conflict or perceived mistreatment, female relational
stalking may be driven by inadequate skills in dealing with these difficult social
contexts (i.e. similar to common couple violence; see Archer, 2000). Moreover,
the fact that female-to-male aggression is often trivialised (see Thompson et al.,
2012 for instance), females may believe that stalking and moderate forms of
stalking violence is acceptable or even justified in response to their perceived
mistreatment or interpersonal conflict.
Stalking perpetrated by young males in the context of establishing or re-estab-
lishing intimacy may reflect a lack of skills in courting and negotiating intimate
relationships. It is possible that males are actually using stalking tactics in an
(albeit unhealthy and misguided) attempt to be chivalrous. In fact, such exag-
gerated ‘romantic’ expressions are potentially reinforced by similarly persistent
behaviour by males in ‘romantic’ movies and songs, including ‘removing barriers’
to achieve reconciliation (see White, Kowalski, Lyndon and Valentine, 2000). Of
course, it is also possible that male stalkers’ desperate attempts at intimacy, which
sometimes included aggression towards the victim or others who interfered, may
be a reflection of males being more proprietary in relationships and an unwilling-
ness to ‘let the relationship go’ (e.g. Brewster, 2003; Duntley and Buss, 2012;
Lyndon et al., 2012). However, none of these explanations could be tested in the
present study and require additional research.

Utility of combining the rational choice perspective and situational

precipitators framework for understanding stalking and associated violence
Despite the need for further research to investigate the origins of the sex differences
in stalkers’ rationales, the present research illustrates the utility of the rational
choice perspective and situational precipitators framework for understanding
stalking and stalking violence. Almost all of the relational stalkers in this research
could provide rationales for their behaviour, highlighting the purposive role of
stalking and associated violence. Investigation of these rationales illustrates some
of the perceived costs and benefits associated with stalking, a crime for which
the offender can suffer dramatic consequences other than being arrested such as
‘losing’ the one he/she loves. In this context, the notion of bounded rationality
is evident because this crime often involves an important range of motivations
intertwined with a number of negative feelings between two persons who often
loved each other (and maybe still do). The offender decision-making process is
influenced by external circumstances but also by the nature of the relationship
between the two parties and the reactions of both the victim and third parties to
previous stalking behaviours.
94 Carleen M. Thompson and Benoit Leclerc
Given that stalking involves repeated behaviours that are likely to elicit some
form of reaction from the victim, it is perhaps not surprising that the environment
had an important role in precipitating the escalation of stalking behaviours. As
Wortley (2008) puts it, the immediate environment can encourage or induce indi-
viduals who would not have contemplated crime in the first place. In stalking, a
range of situational features could potentially fuel and lead to violence. Indeed, the
context of stalking is already characterised by a high level of stress – an additional
dispute, a conflict or the interference of a third party in the relationship could have
serious consequences for both parties. This was demonstrated by the influential
role of precipitators for stalking violence in this study, whereby stalkers reported
that arguments or conflict/disputes preceded 70 per cent of female-perpetrated vio-
lence and 43.5 per cent of male-perpetrated violence. Moreover, 25–30 per cent
of violent stalkers reported that they were retaliating against the victim or were
provoked and more than 15 per cent said that they were violent because they were
intoxicated. Given that stalking behaviours often bring the offender into close
physical proximity to the victim, these non-violent behaviours may also provide
opportunities for situational precipitators to occur such as a third party confronting
the stalker or an argument between the stalker and victim. Consequently, stalking
researchers must consider the influential role of situational factors as well as the
underlying reason for stalking the target in the first place.

Implications for interventions

The present findings have at least five implications for intervention strategies.
Most of these strategies may be useful across sex. First, a range of situational
crime prevention techniques could be adopted by victims to increase the difficulty
of perpetrating stalking behaviours. Briefly, situational crime prevention aims to
modify certain aspects of the immediate environment in which crime occurs by
reducing opportunities for crime or controlling precipitators of crime (Cornish
and Clarke, 2003). For example, victims may change their routines, avoid places
where the offender visits, works or lives, block the offender from electronic
accounts such as Facebook, enhance home security, reduce outside visibility to
their house, change their phone number or block the offenders’ calls. Strategies
that limit the stalker’s ability to approach the victim are also likely to reduce the
opportunities for violence, including planned attacks as well as violence escalat-
ing from non-violent interactions. Second, given the key role of interpersonal
conflict in the escalation to violence in stalking cases, victims should avoid situ-
ations that may lead to arguments and conflict/disputes with the stalker or be
construed as provocative (e.g. a female target’s new boyfriend confronting an
ex-partner stalker). Additionally, victims in these situations should attempt to dif-
fuse the situation or seek help. Third, in a small number of cases, people the
victims care about may be at risk of threats and/or violence, particularly if they
are perceived to be obstructing the stalkers’ goals. Consequently, it may be advis-
able for victims’ loved ones to also enhance their security measures, particularly
for female victims. Fourth, even though intoxication was associated with just
The phenomenon of stalking 95
15 per cent to 20 per cent of stalking and violence, victims should avoid contact
with the stalker when they are under the influence of drugs and/or alcohol. Fifth,
as the rationales and situational precipitators of stalking and violence vary across
sex, male- and female-perpetrated stalking may require some differences in inter-
vention strategies. For example, given that much male-perpetrated stalking was
targeted towards establishing or re-establishing intimacy, it may be useful for vic-
tims to clearly communicate to the offender that they do not want a relationship
to reduce misinterpretation and avoid sending mixed messages. Given the aggres-
sive nature of female-perpetrated stalking and the importance of interpersonal
conflict for the escalation of stalking, victims may be advised to avoid contact
with the offender to avoid inciting the offender through arguments or conflict.
Importantly, as much stalking behaviour is not reported to the police, or is only
reported once it has already escalated, community education strategies are vital to
educate victims about these preventative strategies.

Limitations and directions for future research

The findings in this study need to be interpreted according to the limitations of
this research. First, the present sample was attained through convenience sampling
techniques and is unlikely to be representative of relational stalkers in the com-
munity or in the criminal justice system. The sample comprised disproportionate
numbers of females and young adults. Moreover, the proportion of behaviour that
could have been prosecuted as unlawful stalking cannot be determined given the
difficulties associated with simulating legislative criteria such as victim fear in
self-report perpetration surveys (see Thompson and Dennison 2008). Therefore,
replication is necessary in a more representative community sample, as well as
samples of convicted stalkers, to establish generalisability. Second, the present
study examined relational stalking only and primarily partner stalking. As the
offenders’ decision-making processes are influenced by the nature of the intimate
relationship between the two parties and factors related to relationship pursuit/
dissolution, the current findings are unlikely to be applicable to stalking in other
contexts, including the sex differences identified. Future research should examine
sex differences in non-relational stalking contexts. Third, this study utilised a ret-
rospective self-report methodology. Although this methodology was necessary to
address the current research questions, the accuracy of the findings are contingent
on participants’ honesty and memory biases/limitations. Fourth, it was not possible
to determine which stalking behaviours were associated with specific rationales,
situational precipitators and feelings using a questionnaire methodology, particu-
larly given that some participants perpetrated more than 100 behaviours. As a
consequence, it was also difficult to determine whether rationales changed over
time (and how) or whether conflicting rationales co-occurred. Future research
could use interviews with stalking perpetrators to better understand how, and under
what conditions, rationales change over time.
96 Carleen M. Thompson and Benoit Leclerc
The findings of this study emphasise the need to distinguish between the con-
texts of male- and female-perpetrated stalking. Even among stalking similarly
perpetrated in response to rejection, there are important sex differences in the
reasons for their pursuit, the situations that precipitated their pursuit and the fac-
tors related to the escalation of such behaviour. Consequently, to best respond to
stalking, intervention strategies need to be tailored towards these gendered con-
texts. The findings also illustrate the importance of investigating factors related to
the offence of stalking (i.e. rationales, feelings, situational precipitators). Lastly,
these findings demonstrate the importance of separately examining the rationales,
feelings and situational precipitators for violent events within the context of stalk-
ing. This will foster a better understanding of how stalkers’ rationales change over
time and in response to situational factors and in turn facilitate the identification
of strategies to prevent the escalation to violence.

We would like to thank Associate Professor Susan Dennison for her helpful com-
ments and suggestions on an earlier draft. We would like to thank the participants
who kindly participated in this research. This research was funded by an Australian
Commonwealth Government scholarship, Australian Postgraduate Award.

Archer, J. (2000). Sex differences in aggression between heterosexual partners: A meta-
analytic review. Psychological Bulletin, 126, 651–680.
Archer, J. (2002). Sex differences in physically aggressive acts between heterosexual
partners: A meta-analytic review. Aggression and Violent Behavior, 7, 313–351.
Atkins, D. C., Baucom, D. H. and Jacobson, N. S. (2001). Understanding infidelity:
Correlates in a national random sample. Journal of Family Psychology, 15, 735–749.
Beatty, D. (2003). Stalking legislation in the United States. In M. P. Brewster (ed),
Stalking: Psychology, risk factors, interventions, and law (pp. 2.1–2.55). Kingston, NJ:
Civic Research Institute.
Beauregard, E. and Leclerc, B. (2007). An application of the rational choice approach to
the offending process of sex offenders: A closer look at the decision-making. Sexual
Abuse: A Journal of Research and Treatment, 19, 115–133.
Bjerregaard, B. (2000). An empirical study of stalking victimization. Violence and Victims,
15, 389–406.
Brewster, M. P. (2003). Power and control dynamics in prestalking and stalking situations.
Journal of Family Violence, 18, 207–217.
California Penal Code, Section § 646.9.
Canada’s Criminal Harassment Act, 1993.
Clarke, R. V. and Newman, G. R. (2006). Outsmarting the Terrorist. Westport, CT:
Praeger Security International.
Coleman, F. L. (1997). Stalking behavior and the cycle of domestic violence. Journal of
Interpersonal Violence, 12, 420–432.
The phenomenon of stalking 97
Cornish, D. B. and Clarke, R. V. (1986). The reasoning criminal. New York:
Cornish, D. B. and Clarke, R. V. (2003). Opportunities, precipitators and criminal
decisions: A reply to Wortley’s critique of situational crime prevention. In M. Smith
and D. B. Cornish (eds), Crime Prevention Studies: Theory for practice in situational
crime prevention (Vol. 16, pp. 41–96). Monsey, NY: Criminal Justice Press.
Cornish, D. B. and Clarke, R. V. (2008). The rational choice perspective. In R. Wortley
and L. Mazerolle (eds). Environmental Criminology and Crime Analysis (pp. 21–47).
Cullompton, UK: Willan Publishing.
Criminal Code [Stalking] Amendment Act 1999 [QLD].
Cupach, W. R. and Spitzberg, B. H. (2000). Obsessive relational intrusion: Incidence,
perceived severity, and coping. Violence and Victims, 15, 357–372.
Davis, K. E., Ace, A. and Andra, A. (2000). Stalking perpetrators and psychological
maltreatment of partners: Anger-jealousy, attachment insecurity, need for control, and
break-up context. Violence and Victims, 15, 407–425.
Davis, K. E, Swan, S.C. and Gambone, D. (2012). Why doesn’t he just leave me alone?
Persistent pursuit: A critical review of theories and evidence. Sex Roles, 62, 328–339.
Dennison, S. M. and Stewart, A. (2006). Facing rejection: New relationships, broken
relationships, shame and stalking. International Journal of Offender Therapy and
Comparative Criminology, 50, 324–337.
Dennison, S. M. and Thomson, D. M. (2005). Criticisms or plaudits for stalking laws?
What psycholegal research tells us about proscribing stalking. Psychology, Public
Policy and Law, 11, 384–406.
Duntley, J. D. and Buss, D. M. (2012). The evolution of stalking. Sex Roles, 66, 311–327.
Dutton, L. B. and Winstead, B. A. (2006). Predicting unwanted pursuit: Attachment,
relationship satisfaction, relationship alternatives, and break-up distress. Journal of
Social and Personal Relationships, 23, 565–586.
Fisher, B. S., Cullen, F. T. and Turner, M.G. (2002). Being pursued: Stalking victimization
in a national study of college women. Criminology & Public Policy, 1, 257–308.
Follingstad, D. R., Bradley, R. G., Laughlin, J. E. and Burke, L. (1999). Risk factors and
correlates of dating violence: The relevance of examining frequency and severity levels
in a college sample. Violence and Victims, 14, 365–380.
Harris, J. (2000). The Protection from Harassment Act 1997: An evaluation of its use and
effectiveness. London: Home Office Research, Development and Statistics Directorate.
Haugaard, J. J. and Seri, L. G. (2000). Stalking and other forms of intrusive contact in
adolescent and young-adult relationships. Kansas City Law Review, 69, 227–238.
Holtzworth-Munroe, A. and Stuart, G. L. (1994). Typologies of male batterers: Three
subtypes and the differences among them. Psychological Bulletin, 116, 476–497.
Jaccard, J. and Wan, C. K. (1996). LISREL Approaches to Interaction Effects in Multiple
Regression. Thousand Oaks, CA: Sage Publications.
James, D., McEwan, T., MacKenzie, R., Meloy, J., Mullen, P., Pathe, M., Farnham, F.,
Preston, L. and Darnley, B. (2010). Persistence in stalking: A comparison of associations
in general forensic and public figure samples. Journal of Forensic Psychiatry and
Psychology, 21, 283–305.
James, D. V., Mullen, P. E., Pathe, M. T., Meloy, J. R., Preston, L. F., Darnley, B. and
Farnham, F. (2009). Stalkers and harassers of royalty: The role of mental illness and
motivation. Psychological Medicine, 39, 1479–1490.
Johnson, M. P. (2006). Conflict and control: Gender symmetry and asymmetry in domestic
violence. Violence Against Women, 12, 1003–1018.
98 Carleen M. Thompson and Benoit Leclerc
Johnson, M. P. and Leone, J. (2005). The differential effects of intimate terrorism and
situational couple violence: Findings from the national violence against women survey.
Journal of Family Issues, 26, 322–349.
Kapley, D. J. and Cooke, J. R. (2007). Trends in antistalking legislation. In D. A. Pinals
(ed), Stalking: Psychiatric perspectives and practical approaches (pp. 141–163). New
York: Oxford University Press.
Kienlen, K. K., Birmingham, D. L., Solberg, K. B., O’Regan, J. T. and Meloy, J. R.
(1997). A comparative study of psychotic and nonpsychotic stalking. Journal of the
American Academy of Psychiatry and Law, 25, 317–334.
Lalasz, C. B. and Weigel, D. J. (2011). Understanding the relationship between gender and
extradyadic relations: The mediating role of sensation seeking on intentions to engage
in sexual infidelity. Personality and Individual Differences, 50, 1079–1083.
Langhinrichsen-Rohling, J., Palarea, R. E., Cohen, J. and Rohling, M. L. (2000). Breaking
up is hard to do: Unwanted pursuit behaviours following the dissolution of a romantic
relationship. Violence and Victims, 15, 73–90.
Logan, T. K., Leukefeld, C. and Walker, B. (2000). Stalking as a variant of intimate
violence: Implications from a young adult sample. Violence and Victims, 15, 91–111.
Lyndon, A. E., Sinclair, H. C., MacArthur, J., Fay, B., Ratajack, E. and Collier, K. E.
(2012). An introduction and review of gender and stalking. Sex Roles, 66, 299–310.
McEwan, T., Mullen, P. E., MacKenzie, R. D. and Ogloff, J. R. P. (2009). Violence in
stalking situations. Psychological Medicine, 39, 1469–1478.
Madensen, T. D. and Knutsson, J. (eds) (2011). Preventing Crowd Violence. Crime
Prevention Studies (Vol. 26). Monsey, NY: Criminal Justice Press.
Meloy, J. R. (1997). The clinical risk management of stalking: ‘Someone is watching over
me …’. American Journal of Psychotherapy, 51, 174–184.
Meloy, J. R. and Boyd, C. (2003). Female stalkers and their victims. Journal of the
American Academy of Psychiatry and Law, 31, 211–219.
Meloy, J. R., Mohandie, K. and Green, M. (2011). The female stalker. Behavioral Sciences
and the Law, 29, 240–254.
Melton, H. C. (2007). Closing in: Stalking in the context of intimate partner abuse.
Sociology Compass, 1/2, 520–535.
Miller, N. and Nugent, H. (2002). Stalking laws and implementation practices: A national
review for policymakers and practitioners (NCJ 197066). Washington DC: United
States Department of Justice.
Morrison, K. A. (2008). Differentiating between physically violent and nonviolent stalkers:
An examination of Canadian cases. Journal of Forensic Sciences, 53, 742–751.
Mullen, P. E. and Pathe, M. (1994). Stalking and the pathologies of love. Australian and
New Zealand Journal of Psychiatry, 28, 469–477.
Mullen, P. E., Pathe, M., Purcell, R. and Stuart, G. W. (1999). Study of stalkers. American
Journal of Psychiatry, 156, 1244–1249.
Nemeth, J. M., Bonomi, A. E., Lee, M. A. and Ludwin, J. M. (2012). Sexual infidelity as
trigger for intimate partner violence. Journal of Women’s Health, 21, 942–949.
Norris, S. M., Huss, M. T. and Palarea, R. E. (2011). A pattern of violence: Analyzing the
relationship between intimate partner violence and stalking. Violence and Victims, 26,
Ogilvie, E. (2000). Stalking: Legislative, policing and prosecuting practices in Australia.
Canberra: Australian Institute of Criminology.
Purcell, R., Pathe, M. and Mullen, P. E. (2001). A study of women who stalk. American
Journal of Psychiatry, 158, 2056–2060.
The phenomenon of stalking 99
Rosenfeld, B. (2004). Violence risk factors in stalking and obsessional harassment: A
review and preliminary meta-analysis. Criminal Justice and Behavior, 31, 9–36.
Rosenfeld, B. and Harmon, R. B. (2002). Factors associated with violence in stalkers and
obsessional harassment cases. Criminal Justice and Behavior, 29, 671–691.
Sheridan, L. and Lyndon, A. E. (2012). The influence of prior relationship, gender and fear
on the consequences of stalking victimization. Sex Roles, 66, 340–350.
Sinclair, H. C. and Frieze, I. H. (2000). Initial courtship behaviour and stalking: How
should we draw the line? Violence and Victims, 15, 23–40.
Sinclair, H. C. and Frieze, I. H. (2005). When courtship persistence becomes intrusive
pursuit: Comparing rejecter and pursuer perspectives of unrequited attraction. Sex
Roles, 52, 839–852.
Spitzberg, B. H. (2002). The tactical topography of stalking victimisation and management.
Trauma, Violence and Abuse, 3, 261–288.
Spitzberg, B. H. and Cupach, W. R. (1997). Relational pursuit: Pursuer short form.
Unpublished measure.
Spitzberg, B. H. and Cupach, W. R. (2007). The state of the art of stalking: Taking stock of
the emerging literature. Aggression and Violent Behavior, 12, 64–86.
Spitzberg, B. H., Cupach, W. R. and Ciceraro, L. D. L. (2010). Sex differences in stalking
and obsessive relational intrusion: Two meta-analyses. Partner Abuse, 1, 259–285.
Spitzberg, B. H., Nicastro, A. M. and Cousins, A. V. (1998). Exploring the interactional
phenomenon of stalking and obsessive relational intrusion. Communication Reports,
11, 33–47.
Strand, S. and McEwan, T. E. (2012). Violence among female stalkers. Psychological
Medicine, 42, 545–556.
Straus, M. A., Hamby, S. L., Boney-McCoy, S. and Sugarman, D. B. (1996). The revised
conflict tactics scale (CTS2): Development and preliminary psychometric data. Journal
of Family Issues, 17, 283–316.
Tedeschi, J. T. and Felson, R. B. (1994). Violence, Aggression and Coercive Actions.
Washington, DC: American Psychological Association.
Thompson, C. M. (2009). Developing and testing an integrated theoretical model of
stalking violence. (Doctoral dissertation). Available from Australasian Digital Theses
Program (Record No. 280553). Available from:
Thompson, C. M. and Dennison, S. M. (2008). Defining relational stalking in research:
Understanding sample composition in relation to repetition and duration of harassment.
Psychiatry, Psychology and Law, 15, 482–499.
Thompson, C. M., Dennison, S. M. and Stewart, A. (2012). Are female stalkers more
violent than male stalkers? Understanding gender differences in stalking violence using
contemporary sociocultural beliefs. Sex Roles, 66, 351–365.
Thompson, C. M., Dennison, S. and Stewart, A. (in press). Are different risk factors
associated with moderate and severe stalking violence? Examining factors from the
integrated theoretical model of stalking violence. Criminal Justice and Behavior.
Waite, L. J. and Gallagher, M. (2000). The Case for Marriage: Why married people are
happier, healthier, and better off financially. New York: Broadway Books.
White, J., Kowalski, R. M., Lyndon, A. and Valentine, S. (2000). An integrative contextual
model of male stalking. Violence and Victims, 15, 373–388.
Wilkinson, D. L. and Hamerschlag, S. J. (2005). Situational determinants in intimate
partner violence. Aggression and Violent Behavior, 10, 333–361.
100 Carleen M. Thompson and Benoit Leclerc
Williams, S. L., Frieze, I. H. and Sinclair, H. C. (2007). Intimate stalking and partner
violence. In J. Hamel and T. Nicholls (eds), Family Interventions in Domestic Violence:
A handbook of gender-inclusive theory and treatment (pp. 109–123). New York:
Springer Publishing Company.
Wortley, R. (1998). A two-stage model of situational crime prevention. Studies of Crime
and Crime Prevention, 7, 173–188.
Wortley, R. (2001). A classification of techniques for controlling situational precipitators
of crime. Security Journal, 14, 63–82.
Wortley, R. (2008). Situational precipitators of crime. In R. Wortley and L. Mazerolle
(eds). Environmental Criminology and Crime Analysis (pp. 48–69). Cullompton, UK:
Willan Publishing.
Wright, J. A., Burgess, A. G., Burgess, A. W., Laszlo, A. T., McCrary, G. O. and Douglas,
J. E. (1996). A typology of interpersonal stalking. Journal of Interpersonal Violence,
11, 487–502.
6 Interpersonal scripts and victim
reaction in child sexual abuse
A quantitative analysis of the
offender–victim interchange
Benoit Leclerc, Stephen Smallbone and
Richard Wortley

Although there is now a substantial body of research on the modus operandi of
child sexual abuse offenders (e.g., Kaufman et al., 1998; Smallbone and Wortley,
2000; for a review, see Leclerc, Proulx and Beauregard, 2009), little is still known
about the patterns of interchange between the offender and the victim—how the
behaviors of one shape the behaviors of the other. The offender may face resist-
ance when trying to involve the victim in sexual activity, but may also be able to
further exploit the victim’s compliance. The offender’s responses to the victim’s
reactions may also have an impact on outcomes of the abuse. Thus, the offender–
victim interchange in child sex abuse can take many routes depending on the
action of the offender to gain the cooperation of the victim in sexual activity,
the reactions of the victim, and the responses of the offender to victim reactions.
An important question emerging from an investigation into the offender–victim
interchanges is the following: What role does the victim’s compliance or noncom-
pliance play in the unfolding of a sexual offense committed against him or her?
This is a critical question especially in light of the popularity of self-protection
programs taught to children.
Building on the empirical child sexual abuse script developed by Leclerc,
Wortley and Smallbone (2011a) and using self-report data obtained from offend-
ers, the current study focuses on the offender–victim interchange where the
offender tries to gain the cooperation of the victim for sexual contact. This inter-
change is broken down into four sub-stages: 1) the offender’s action to gain the
cooperation of the victim; 2) the victim’s reactions; 3) the offender’s response;
and 4) the performance of sexual behaviors by the offender and victim. An inter-
personal script approach derived from cognitive psychology is adopted as an
extension of crime script analysis to facilitate our understanding of the offender–
victim interchange.

Crime scripts and interpersonal scripts

The rational choice approach focuses on the decision-making process underly-
ing the commission of crime (Clarke and Cornish, 1985; Cornish and Clarke,
102 B. Leclerc, S. Smallbone, R. Wortley
1986; 2008). Offender decision making takes place at each stage of the crime-
commission process, which means that the offender may choose different
methods to carry out the crime especially when s/he faces potential obstacles.
Cornish (1994) borrowed the script notion from cognitive psychology in order
to provide a framework to systematically investigate all of the stages of the
crime-commission process of a specific crime and in as much detail as possible.
Scripts represent knowledge of events and event sequences (Abelson, 1981;
Schank and Abelson, 1977). Applied to crime, scripts represent the complete
sequence of actions adopted prior to, during, and following the commission of
a particular crime.
Individuals have cognitive structures not only to guide them to accomplish a
certain goal or script but also to guide their interactions with others. In the 1990s,
based on models of cognitive processing, Baldwin (1992; 1995) developed the
concept of a relational schema that involves an interpersonal form of script (for an
application of this approach, see Fehr, Baldwin, Collins, Patterson, and Benditt,
1999). As discussed by Baldwin (1992), scripts for social situations include both
declarative knowledge (to help individuals think about and understand social situ-
ations) and procedural knowledge (to guide their social behaviors). Declarative
knowledge involves a descriptive summary of the characteristics of things or
people. Procedural knowledge is the individual’s repertoire of rules and skills
that are used in the processing of information, which includes knowledge about
how to attain goals. Individuals develop this knowledge through experiences.
Similarly, individuals also develop declarative and procedural knowledge about
their interactions with others through experiencing similar patterns of interaction
(Baldwin, 1992). This form of knowledge is organized under cognitive structures
called relational schemas. A relational schema captures the essence of interac-
tions between individuals in a specific situation and consists of an interpersonal
script with an associated self-schema (self-representation in this interaction) and
an associated other schema (representation of the other individual in this interac-
tion). According to Baldwin (1992, p. 468):

The interpersonal script can be defined as a cognitive structure representing

a sequence of actions and events that define a stereotyped relational pattern
….The interpersonal script will include declarative knowledge about the pat-
tern of interaction, specifically a summary statement about what behaviors
tend to be followed by what responses. This knowledge can be used to inter-
pret social situations and the behavior of others. As procedural knowledge,
the if–then nature of the script can be used to generate interpersonal expecta-
tions and to plan appropriate behavior.

As one may expect, such a script includes a complex sequence of behaviors in

which one individual’s choice of behaviors is contingent on the behaviors of the
other individual. This is an iterative and dynamic process whereby the outcome of
an interaction cannot be predicted by the initial behavior.
Victim reaction in child sexual abuse 103
Crime scripts and child sexual abuse
The first study to apply the rational choice approach to sexual offenders was
completed by Proulx, Ouimet and Lachaine (1995) who investigated the deci-
sions and behaviors of ten sexual offenders against children. Later, Leclerc,
Carpentier and Proulx (2006) examined strategies adopted by child sexual
offenders with a particular focus on cases where offenders seemed to act on the
spur of the moment without any apparent plan (i.e., non-persuasive approach).
Beauregard and Leclerc (2007) also examined the rationales underlying the
behaviors adopted by serial stranger sex offenders before, during and after the
commission of the crime.
Following empirical research on sexual offender decision making, two stud-
ies have applied the crime script framework to the crime-commission process
of sexual offenders. Using a sample of serial stranger sex offenders who abused
women and/or children, Beauregard, Proulx, Rossmo, Leclerc and Allaire
(2007) identified script tracks based on offender strategies to commit the crime
with a focus on the use of different locations by offenders throughout the crime-
commission process. For situational prevention purposes, Leclerc et al. (2011a)
focused on child sexual offenses and proposed an empirical script model detail-
ing a series of eight steps from the start to the end of the abuse. The script
contains the following steps: 1) entry to setting (i.e., setting where victim was
first encountered—institutional, public or domestic), 2) instrumental initiation
(e.g., gaining trust strategies), 3) continuation (i.e., proceeding to the crime
location), 4) location selection (i.e., location for sexual contact), 5) instrumental
actualization (e.g., isolation); 6) completion (e.g., gaining cooperation), 7) out-
comes (i.e., sexual behaviors), and 8) post condition (i.e., avoiding disclosure).
Consistent with Proulx et al.’s (1995) study, the script model shows that the
offender may have to go through a number of steps and make decisions at each
of these steps to perpetrate the offense.

Current study
While an event script is a cognitive template that breaks down an ordered set
of actions required to achieve a goal in a particular situation, an interpersonal
script focuses on the sequencing of interpersonal events and patterns of behav-
ioral interdependence. Applied to crime, the interpersonal script approach
reminds us to pay attention to the links or contingencies between offender
actions, victim reactions, offender responses, and specific outcomes. We frame
our study within this approach to examine the patterns of interchange between
the offender and the victim when the offender seeks to gain the cooperation of
a child for sexual contact.
Virtually all research on modus operandi has focused on strategies adopted
by offenders to gain the cooperation of the victim in sexual activity (e.g.,
Elliott, Browne, and Kilcoyne, 1995; Kaufman et al., 1998; Smallbone and
Wortley, 2000; for a review, see Leclerc, Proulx and Beauregard, 2009), or on
104 B. Leclerc, S. Smallbone, R. Wortley

Offender actions Victim reaction Offender response Sexual behaviors

for cooperation
of victim

Actions Actions Actions Actions

– Persuasive – Compliance – No procedural – Sexual behaviors

strategy – Resistance variation performed by
– Non-persuasive – No discernable – Procedural the offender
strategy reaction variation – Sexual behaviors
performed by
the victim
A = Script without interchange
B = Script with interchange

Figure 6.1 Approaches employed to investigate crime-commission processes in adult–

child sexual abuse.

the relationship between offending strategies and sexual behaviors (Leclerc,

Proulx, Lussier, and Allaire, 2009; Leclerc and Tremblay, 2007). Figure 6.1
presents this common approach (see scenario A). Under this scenario, the
interchange leading to sexual behaviors is ignored. The present study takes a
different approach. It focuses on scenario B where a more detailed interchange
is examined: 1) offender action, 2) victim reaction, 3) offender response, and 4)
sexual behaviors performed by the offender and the victim. More specifically,
using data obtained from offenders, a quantitative analysis of the offender–vic-
tim interchange in child sexual abuse is conducted. Regression analyses are
first performed on the intrusiveness of offender and victim sexual behaviors as
outcomes of the interchange. Then, regression analyses are performed on stages
within the script to examine the presence of indirect relationships between each
stage of the script.
Victim reaction in child sexual abuse 105
This study uses data from a research project on child sexual abuse in which events
and related circumstances were examined by administering a self-reported ques-
tionnaire to 104 adult male offenders. These offenders were incarcerated between
2007 and 2009 for having committed a sexual offense against a child under 16
years of age. Missing data on the offender–victim interchange was found for 14
participants. Therefore, a total of 90 offenders was included in this study. Mean
age of participants was 45.37 years (SD=11.85). The majority was Australian
born (84.9 percent). Other participants were either Aboriginal or Australian Torres
Strait Islanders (11 percent) or born in another country (4.1 percent). This infor-
mation was missing for 17 participants. Most of the participants did not achieve
an education level higher than secondary school (70.8 percent). On average, par-
ticipants were serving a sentence of 104.53 months (SD=74.19, Range=358).

Participants completed a 52-page self-report survey, only part of which is relevant
to the current study. All questions used in this study related to the circumstances
surrounding three specific offenses—the first, second, and most recent occasions
that the participant had sexual contact with a child. The questionnaire covered in a
detailed way what happened during the offense, including the offender’s actions,
the victim’s reactions, the offender’s responses, and the outcomes of the abuse.

Initially, individuals who were incarcerated for having committed a sexual offense
against a child (< 16 years) were identified and approached by local Corrective
Services staff to participate in a research project by completing a self-report ques-
tionnaire. Later in the project, to facilitate the recruitment process, individuals who
were identified were approached individually by a member of the research team.
Participants were told that their involvement in this study was strictly voluntary.
Each participant signed a consent form stating that the information would be used
for research purposes only. They were assured that the information would be kept
confidential and that records of names would be destroyed after data collection.

Variables and analytic strategy

Participants reported a total of 213 events involving sexual contact with a child—
the unit of analysis in this study. The sequence of the offender–victim interchange
was first reconstructed by examining four closed-ended questions answered by
offenders—the actions they performed to gain the cooperation of the victim;
the reactions, if any, of the victim; the action, if any, the offender adopted in
response to the victim reaction; and the sexual behaviors performed with the vic-
tim. Descriptive statistics for these variables can be found in Table 6.1.
106 B. Leclerc, S. Smallbone, R. Wortley
Actions performed by the offender to gain cooperation were categorized as per-
suasive and non-persuasive. Persuasive actions (n=122) included manipulative (i.e.,
give bribes, give love and attention, encourage sexuality, show pornography, give
alcohol and drugs; n=118) and coercive strategies (i.e., use threats of force, use
physical force; n=4). Non-persuasive strategies (n=91) involved cases in which the
offender acted apparently on the spur of the moment and in the absence of any obvi-
ous attempts to persuade the victim to comply (see Leclerc et al., 2006).
The victims’ reactions to the offenders’ initial behavior were categorized as
compliant and non-compliant. Victim compliance (n=116) was determined to be
present when the offender reported that the victim was sexually aroused (n=26),

Table 6.1 Descriptive statistics of variables in multivariate analyses

Variable Category(ies) Distributions

Offender action(s) (for cooperation of victim Percent each category

in sexual activity)
Non-persuasive strategy 42.7%
Persuasive strategy 57.3%
Base N= 213
Victim reaction(s) Percent each category
Non-compliance 45.5%
Compliance 54.5%
Base N=213
Offender response Percent each category
No procedural variation 81%
Procedural variation 19%
Base N = 213
Victim characteristics Mean (SD)
Age of victim 10.85 (3.17)
Base N =212
Gender of victim Percent each category
Male victim 32.5%
Female victim 67.5%
Base N =212
Duration of sexual contact Percent each category
Less than 5 minutes 49.5%
More than 5 minutes 50.5%
Base N = 208
Severity of sexual contact Mean (SD)
Index of intrusiveness of offender 3.79 (3.52)
sexual behaviours Base N =213
Index of intrusiveness of victim sexual Mean (SD)
behaviours 1.48 (2.31)
Base N=213
Transformed index of intrusiveness of Mean (SD)
victim sexual behaviours (Log 10) .25 (.33)
Base N=213
Victim reaction in child sexual abuse 107
affectionate toward him (n=37), or both sexually aroused and affectionate (n=53).
Non-compliance (n=97) included cases in which the victim actively resisted the
offender (i.e., hit/push away, yell for help, say no/tell to stop, cry, become fright-
ened; n=32) and those cases in which the offender reported that there was no
discernible reaction by the victim (n=65).
The actions adopted by offenders in response to victim reaction (if any) were
categorized in terms of whether or not procedural variation occurred. Procedural
variation is a concept employed by Cornish (1994) to illustrate the flexibility of
scripts when offenders encounter an obstacle and choose another strategy to over-
come that obstacle (see also Lacoste and Tremblay, 2003; Morselli and Roy, 2008).
We used the concept of procedural variation in this study in order to detect whether
the offender adopted a different action or not and whether this variation had an effect
on the performance of sexual behaviors by the offender and the victim. For instance,
an offender could use a non-persuasive strategy to first obtain the victim cooperation
in sexual activity but then adopt a persuasive strategy to overcome victim resistance.
It was found that 40 offenders changed their strategy while 173 did not.
The sexual behaviors performed with the victim—the dependent variables in
this study—consisted of two indices of the degree of intrusiveness. Participants
were asked, for each event, whether they had made their victim perform sexual
behaviors on them and if they had performed sexual behaviors themselves. To
create the index of the intrusiveness of sexual behaviors performed by the victim,
a score was assigned for each sexual behavior according to the level of intrusive-
ness as follows: 1=Fondling, 2=Masturbation, 3=Fellatio, 4=Digital penetration,
5=Anal penetration. The score of each sexual behavior performed by the vic-
tim was then summed. For instance, an offender who made his victim perform
masturbation and fellatio on him would obtain a score of 5 on this index. Thus,
for each event, a score of the intrusiveness of the sexual behaviors performed
by the victim in sexual episodes ranging from 0 (the victim did not perform a
sexual behavior) to 15 (all sexual behaviors were performed by the victim) was
obtained. As a result of a severely positive skewed distribution, this variable was
subsequently log-transformed for statistical analysis. The procedure to create the
index of the intrusiveness of sexual behaviors performed by offenders followed
the same pattern. A score was assigned for each sexual behavior according to the
level of intrusiveness as follows: 1=Fondling, 2=Masturbation, 3=Fellatio/cun-
nilingus, 4=Digital penetration, 5=Anal/vaginal penetration. The score of each
sexual behavior performed by the offender was then summed and the final score
obtained at the index for each event also ranged from 0 to 15.
A similar method was developed by Sellin and Wolfgang (1964) to evaluate
the degree of seriousness of crime. They designed a classification ranking 141
offenses in terms of a priori defined seriousness. Over the years, this method has
been challenged for many reasons including its level of measurement and additiv-
ity assumption—two issues that remain unresolved (Stylianou, 2003). Although
the indices used in the present study come with the same limitations (the order
of sexual behaviors is debatable especially for less serious acts and numerical
weights are somewhat arbitrary—e.g., masturbation is not necessarily twice as
108 B. Leclerc, S. Smallbone, R. Wortley
intrusive as fondling), they were constructed to tap into the varying degree of
intrusiveness and complexity of offender–victim interchanges in child sexual
abuse which would not be possible otherwise. In this context, indices offer a rea-
sonable approximation of outcomes of offender–victim interchanges in a real-life
situation because they have the advantage of capturing the qualitative nature of
crime events while allowing quantitative analysis. Similar indices have been used
previously in the sexual offending literature to measure the severity of abuse (e.g.,
Aylwin et al., 2000; Leclerc and Tremblay, 2007).
To get a clearer understanding of the offender–victim interchange in child
sexual abuse, we also included the duration of sexual contact as part of the
analysis and incorporated victim characteristics (age and gender) as control vari-
ables. The rationale underlying the inclusion of duration as part of the script
is that an event of substantial duration should be more likely to lead to intru-
sive sexual behaviors. As such, the performance of intrusive sexual behaviors
during the offense may be contingent upon duration of sexual activity. Victim
characteristics have been included in the analysis because they are statistically
related to strategies adopted by offenders (Leclerc et al., 2006) and penetration
by the offender (Leclerc et al., 2009). In addition, as children have been found
to use different strategies to resist victimization according to their age and gen-
der (Asdigian and Finkelhor, 1995; Leclerc, Wortley and Smallbone, 2011b),
one would also expect victim characteristics to play a role in the nature of the
interchange between the offender and the victim in child sexual abuse. Finally,
the analysis on the intrusiveness of offender sexual behaviors controls for the
intrusiveness of victim sexual behaviors and vice versa. This is because sexual
behaviors performed by the offender and the victim are interdependent events
(Leclerc et al., 2009; Leclerc and Tremblay, 2007).

Statistical analysis
As each offender may have been involved in several events for up to a maximum
of three (the average number of crime events involving the same offender is 2.3
or 2.4 depending on the dependent variable), the data are nested within offenders
and consequently, the independence of observations assumption is not met in this
study. If an inappropriate statistical technique is used with these data, parameter
estimates and their standard errors may be biased. This might lead, in turn, to an
incorrect interpretation of the findings (Guo and Zhao, 2000; see also Carjaval,
Baumler, Harrist and Parcel, 2001). In this study, depending on the nature of the
dependent variable, a particular technique is employed to correct for the prob-
lem of independence of observations. Using Stata (version 11.0), mixed-effects/
multilevel linear regression analysis is employed to investigate the intrusive-
ness of sexual behaviors of offenders and victims as the dependent variables.
Mixed-effects/multilevel logistic regression analysis is also performed on victim
reaction, offender response and duration of sexual contact as the dependent vari-
ables. In the presence of nested data, within-group homogeneity is permitted and
estimated by the inclusion of random effects in mixed-effects regression analysis,
Victim reaction in child sexual abuse 109
which corrects for the problem of dependent observations and prevents inflation
of Type I error rate in the analysis (Bickel, 2007).

Before estimating the intrusiveness of offender and victim sexual behaviors with
mixed-effects regression analysis, an unconditional model (i.e., a model with
no predictors or control variables) for each dependent variable was performed
(analysis not shown here). The variance component was significant (p < .05),
which indicates that significant variation occurs in the dependent variables across
offenders. Therefore, this finding supports the use of mixed-effects regression
over standard regression analyses.
As shown in Table 6.2, we then performed a hierarchical mixed-effects regres-
sion analysis of the intrusiveness of offender sexual behaviors following the
sequence of the interchange between the offender and the victim (i.e., offender
action–victim reaction–offender response–sexual behaviors). We also controlled
for duration of sexual contact, victim characteristics and the intrusiveness of vic-
tim sexual behaviors. In Model 1, the analysis shows that the offender action is
positively associated with the intrusiveness of offender sexual behaviors. This
relationship indicates that the use of a persuasive strategy to gain victim coop-
eration in sexual activity is associated with a higher level of intrusive sexual
behaviors performed by the offender than is the case when a non-persuasive strat-
egy is used. However, in Model 2, this relationship disappears when introducing
victim reaction. In turn, victim reaction is positively associated with the intrusive-
ness of offender sexual behaviors. This finding indicates that victim compliance
is associated with a higher level of intrusive sexual behaviors performed by the
offender than is the case for victim non-compliance. In Model 5, this relationship
disappears as well when introducing the control variables. Net of other variables,
duration of sexual contact, age of the victim and the intrusiveness of victim sex-
ual behaviors are all strongly and positively associated with the intrusiveness of
offender sexual behaviors.
In Table 6.3, we performed a hierarchical mixed-effects regression analysis
following the sequence of the interchange between the offender and the victim
(i.e., offender action-victim reaction-offender response) but this time on the intru-
siveness of victim sexual behaviors. We controlled for duration of sexual contact,
victim characteristics and the intrusiveness of offender sexual behaviors. Model 2
introduces victim reaction and shows a positive relationship between this variable
and intrusive sexual behaviors performed by the victim. Victim compliance is
positively associated with a higher level of intrusive sexual behaviors performed
by the victim on the offender. When introducing control variables in Model 5, this
relationship can still be observed. Victim reaction predicts the level of intrusive-
ness of sexual behaviors performed by the victim on the offender independently
of other variables of the offender-victim interchange. In addition, duration of
sexual contact and the intrusiveness of offender sexual behaviors are positively
associated with the intrusiveness of victim sexual behaviors.
Table 6.2 Mixed-effects linear regression models of the intrusiveness of offender sexual behaviors in child sexual abuse (n=205 events)

Fixed effects Model 1 Model 2 Model 3 Model 4 Model 5

Coefficient Conf. Coefficient Conf. Coefficient Conf. Coefficient Conf. Coefficient Conf.
(standard interval (standard interval (standard interval (standard interval (standard interval
error) error) error) error) error)

Offender actions 1.09** .05 – .598 –.46 – .626 –.447 – .296 –.742 – .162 –.869 –
Persuasive strategy (.531) 2.13 (.540) 1.66 (.547) 1.70 (.530) 1.33 (.526) 1.19
Victim reaction – – 1.63*** .709 1.63*** .708 .911* –.031 .654 –.287
Compliance (.469) – 2.55 (.470) – 2.55 (.480) – 1.85 (.480) – 1.60
Offender response – – – – –.182 (.612) –1.38 –.609 (.593) –1.77 – –.778 (.587) –1.93 –
Procedural variation – 1.02 .553 .372
Duration of sexual contact – – – – – – 2.46*** 1.46 2.02*** .991
(.508) – 3.45 (.524) – 3.05
Victim characteristics – – – – – – – – .222*** .071 –
Age of the victim (.077) .372
Female victim – – – – – – – – .097 –1.17
(.646) – 1.36
Index of logged intrusiveness – – – – – – – – 2.01*** .500
of victim sexual behaviors (.771) – 3.52

Random effect Variance-components (Standard error) (p-value)

Offender-specific random 2.16 (.306) (p < .05) 2.27 (.307) (p < .05) 2.26 (.311) (p < .05) 2.00 (.318) (p < .05) 2.06 (.312) (p < .05)

Note: *p≤.10, **p≤.05, ***p≤.01.

Table 6.3 Mixed-effects linear regression models of logged intrusiveness of victim sexual behaviors in child sexual abuse (n=205 events)

Fixed effects Model 1 Model 2 Model 3 Model 4 Model 5

Coefficient Conf. Coefficient Conf. Coefficient Conf. Coefficient Conf. Coefficient Conf.
(standard interval (standard interval (standard interval (standard interval (standard interval
error) error) error) error) error)

Offender actions .028 –.065 – .021 –.113 – .025 –.119 – .042 –.136 – .050 –.143 –
Persuasive strategy (.048) .121 (.047) .072 (.047) .068 (.048) .052 (.047) .043

Victim reaction – – .154*** .081 – .153*** .079 – .109*** .030 – .089** .010 –
Compliance (.037) .228 (.038) .227 (.040) .189 (.040) .168
Offender response – – – – .058 –.040 – .046 –.054 – .070 –.029 –
Procedural variation (.050) .155 (.051) .147 (.051) .169
Duration of sexual contact – – – – – – .139*** .053 – .098** .009 –
(.044) .225 (.046) .188
Victim characteristics – – – – – – – – –.011 (.007) –.024 –
Age of the victim .003
Female victim – – – – – – – – –.081 (.068) –.215 –
Index of intrusiveness of – – – – – – – – .016*** .005 –
offender sexual behaviors (.006) .028

Random effect Variance-components (Standard error) (p-value)

Offender-specific random .268 (.026) (p < .05) .271 (.026) (p < .05) .269 (.026) (p < .05) .256 (.025) (p < .05) .255 (.025) (p < .05)

Note: *p≤.10, **p≤.05, ***p≤.01.

112 B. Leclerc, S. Smallbone, R. Wortley
In Table 6.4, mixed-effects/multilevel logistic regression analysis is reported
on victim reaction, offender response, and duration of sexual contact as the
dependent variables. Victim characteristics are included as control variables in
each model. This analysis allows for the examination of indirect relationships
between the stages of the interpersonal script and, therefore, behavioral interde-
pendence within the script. The first analysis shows that the offender action is
positively associated with victim reaction indicating that the use of a persuasive
strategy is more likely to be associated with victim compliance than is the use of
a non-persuasive strategy. Victim characteristics are also associated with victim
reaction. First, as the victim gets older, the victim is more likely to be compliant
to the action of the offender. Second, male victims are more likely to be compliant
than are female victims when the offender tries to gain their cooperation in sexual
activity. The second analysis shows a positive association between offender action
and offender response. The use of a persuasive strategy by the offender is more
likely to lead to procedural variation (i.e., a modification of the offender action) in
his response to victim reaction than is the use of a non-persuasive strategy.
The third and final analysis focused on the inclusion of duration of sexual
contact to capture whether this variable mediates the effects that each stage of the
script can have on the intrusiveness of sexual behaviors. This analysis shows that
offender action, victim reaction and victim characteristics are all associated with
duration of sexual contact. First, the use of a persuasive strategy is more likely
to be associated with a longer duration of sexual contact than is the use of a non-
persuasive strategy. Second, victim compliance is more likely to be associated
with a longer duration of sexual contact than victim non-compliance. Third, as
the victim gets older, the sexual contacts are more likely to be of longer duration.
Finally, male victims are more likely to be involved in sexual contact of longer
duration than female victims.

The current study examined the offender–victim interchange when the offender
tries to gain the cooperation of the victim in sexual activity. We investigated
the following sequence: 1) offender action, 2) victim reaction, 3) offender
response, and 4) sexual behaviors performed by the offender and the victim.
We also included the duration of sexual contact and controlled for victim char-
acteristics during the analyses. Finally, we examined the presence of indirect
relationships between each stage of the script which included the duration of
sexual contact.
One interesting aspect to examine was the role played by the victim during
the interchange with the offender. The findings of this study suggest that victim
responses to sexual abuse situations affect the immediate outcomes of the abuse,
at least according to offenders’ perceptions. In fact, victim compliance was pre-
dictive of both the intrusiveness of offender and victim sexual behaviors when
examining the sequence offender action–victim reaction–offender response–
sexual behaviors (although the relationship with the former disappeared when
Table 6.4 Mixed-effects logistic regression models of interpersonal script stages

Fixed Effects Victim reaction Offender response Duration of sexual contact

(n=210 events) (n=210 events) (n=205 events)

Coefficient Conf. interval Coefficient Conf. interval Coefficient Conf. interval

(standard error) (standard error) (standard error)

Offender actions
Persuasive strategy 1.20*** .372 – 1.67** .097 – 1.29** .107 –
(.422) 2.03 (.800) 3.23 (.601) 2.46
Victim reaction
Compliance – – .422 –.846 – 1.73*** .471 –
(.647) 1.69 (.643) 2.99
Offender response
Procedural variation – – – – .978 –.333 –
(.669) 2.29
Duration of sexual contact – – – – – –
Victim characteristics
Age of the victim .151** .022 – .020 –.175 – .164* –.011 –
Female victim (.066) .279 (.100) .216 (.089) .339
–1.52*** –.247 – –.147 –1.79 – –1.95*** –3.40 –
(.487) –.561 (.841) 1.50 (.737) –.510

Random effect Variance-components (Standard error) (p-value)

Offender-specific random intercept 1.08 (.378) (p < .05) 2.37 (.645) (p < .05) 1.97 (.570) (p < .05)

Note: *p≤.10, **p≤.05, ***p≤.01.

114 B. Leclerc, S. Smallbone, R. Wortley
including control variables such as the age of the victim). Why is this so? A
possible explanation is that a child is likely to be more compliant with physi-
cal interactions that involve sexual contact if s/he first receives attention and
affection from the adult offender. To facilitate this, the offender can adopt
manipulative strategies. This explanation is supported by the positive correla-
tion observed between the use of a persuasive strategy and victim compliance
(see Table 6.4). In the end, the relationship between the use of a persuasive
strategy and the intrusiveness of sexual behaviors also disappears when intro-
ducing the victim reaction in the analysis. The offender action and the offender
response are not directly associated with the intrusiveness of sexual behaviors.
Therefore, what truly matters in explaining the intrusiveness of sexual behaviors
(or severity of abuse) is not so much what the offender does in the first place to
gain the cooperation of the victim—the offender could adopt persuasive or non-
persuasive strategies—but how the victim reacts in this context. If the victim is
compliant, the intrusiveness of sexual behaviors performed by the offender and
the victim increases. Note that the transition from day-to-day physical interac-
tions to sexual contacts may occur without the victim being conscious of what
the sexual contacts actually mean or the intentions of the offender (Berliner and
Conte, 1990).
Victim characteristics are also important to understand the offender–victim
interchange in child sexual abuse. First, the age of the victim was positively associ-
ated with the intrusiveness of offender sexual behaviors. This finding is consistent
with those of previous studies. For instance, Leclerc et al. (2009) found that the
age of the victim was positively associated with the occurrence of penetration.
Second, the investigation of indirect relationships showed that victim characteris-
tics were predictive of both victim reaction and duration of sexual contact. If the
victim is a male, victim compliance as well as a longer duration of sexual contact
is more likely than if the victim is a female. As the victim gets older, victim com-
pliance and a longer duration of sexual contact is also more likely. In turn, what
the full regression models showed is that if the victim is compliant and the sexual
contacts are of a long duration, the intrusiveness of sexual behaviors performed
by the victim on the offender increases.
Two important questions emerge in light of these findings: 1) Why is victim
compliance more likely as the victims get older, and 2) why are boys more
likely than girls to be compliant to the offender’s actions? One possibility is
that offenders are using particular strategies based on the age of the victim (see
Kaufman et al., 1996). Another possibility is that as children get older, they
become more receptive and willing to explore sexuality, which makes them more
vulnerable to abuse. In the human sexual development literature, DeLamater
and Friedrich (2002) point out that the capacity for a sexual response is present
from birth. They explain that between the ages of 3 and 7 there is a substantial
increase in sexual interest and activity which develops into sexual exploration
and learning between the ages of 8 and 12. Then, in adolescence, there is a surge
of sexual interest as a result of biological changes associated with puberty. In
addition, DeLamater and Friedrich (2002) note that the increase of testosterone
Victim reaction in child sexual abuse 115
levels during the adolescence period creates the possibility of sexual interac-
tions which may also explain why boys, compared to girls, are more likely to
be compliant in response to the action of the offender. In summary, consistent
with our findings, children could be easier to sexually exploit as they approach
puberty, especially boys. However, it should be kept in mind that interpretations
of victim reactions during the offense are based on offender self-reported data.
This study adds to the criminological literature in explaining how victims
react when the offender is seeking their cooperation for sexual contact and what
happens if the victim shows compliance. It appears imperative for therapists and
practitioners to use this knowledge to support victims and prevent sexual re-vic-
timization of children. In addition, as suggested elsewhere (e.g., Asdigian and
Finkelhor, 1995; Leclerc, Wortley and Smallbone, 2010; 2011b; Rispens, Aleman
and Goudena, 1997; Roberts and Miltenberger, 1999), these findings support the
necessity to design different prevention programs according to specific subgroups
of children (boys vs. girls, younger vs. older).
Apart from deconstructing the offender–victim interchange in child sexual
abuse for better understanding sexual crime events, this study was innovative in
three additional ways. First, we focused on the role of victim compliance dur-
ing the interchange with the offender. To our knowledge, no empirical study has
analyzed the impact of victim compliance during the offense. One of the reasons
for that is probably because child compliance during the abuse may be seen as
controversial. Child victims are not only likely to resist but also to comply with the
offender’s action to gain their cooperation for sexual contact. In fact, our findings
indicate that just over half of events involved some compliance by the victim dur-
ing the offense. Second, we operationalized the action of the offender as the use of
a persuasive strategy or non-persuasive strategy. This is because a large proportion
of events have been found to be characterized by the use of a non-persuasive strat-
egy (Beauregard et al., 2007; Leclerc et al., 2006). Third, consistent with a script
framework, a variable of procedural variation to account for the offender response
to victim reaction was included in this study. Despite its theoretical importance in
explaining how offenders adapt to situations, procedural variation was not found
to play an important role during the interchange in child sexual abuse. Procedural
variation was only associated with (and predicted by) the offender action, a finding
which is in accordance with a recent Canadian study completed on the offender
response in sexual offenses (Balemba and Beauregard, 2012).

Investigating the reactions of victims in this study was an important challenge.
First, out of 213 events, a total of 116 events were reported by offenders as involv-
ing victim compliance (i.e., being sexually aroused and/or affectionate toward the
offender) and 65 as involving no discernible reaction from the victim. Clearly,
the offender’s perception of how the victim reacted is characterized by limita-
tions and direct accounts from the victim would be more reliable. Indeed, some
offenders potentially reported that the victim complied or did not react in order
116 B. Leclerc, S. Smallbone, R. Wortley
to rationalize their behaviors and excuse themselves. Added to the potential of
offenders to downplay the seriousness of their behavior, we were also asking them
to recall events that may have occurred many years ago. Second, how the variable
of victim reaction was operationalized in this study is not perfect either. Even if
the focus was placed on victim compliance, distinguishing victim resistance from
no reaction would have been beneficial to further understand the theoretical rel-
evance of the role played by the victim during the offense. However, the sample
size used in this study was too small to reach that level of details in the analyses.
Regarding victim compliance specifically, it is possible that some children were
sexually aroused without being “compliant.” Similarly, just because a child is
affectionate to the abuser does not mean they are complying with the abuse. These
limitations clearly illustrate how difficult and complex it is to analyze the reac-
tions of victims during episodes of sexual abuse which is probably one of the main
reasons why this critical dimension of crime events has been ignored in the past.
In this study, the sequence of the offender–victim interchange may have been
more complex in some cases and involved a greater number of steps where the
victim reacted and the offender responded to victim’s reaction and so forth. As
a consequence, in these cases, a victim may have had different reactions at dif-
ferent points which could have influenced the nature of sexual activities between
the offender and the victim. In a context where a perfect account of this inter-
change could be obtained, this characteristic would limit our understanding of the
offender–victim interchange in child sexual abuse.
Our study is also characterized by the fact that two of the reported offenses
were the first and second offenses ever committed by offenders. In these cases,
offenders may not yet have developed procedural knowledge about patterns of
interaction in child sexual abuse. As pointed out by Baldwin (1992), individu-
als develop this knowledge through experiencing similar patterns of interaction.
Without experience, it would be difficult for offenders to interpret the reaction of
the victim and to plan appropriate responses to commit their offense. In this sense,
the findings may not be entirely representative of offender–victim interchanges
involving experienced offenders.

To our knowledge, the sequencing of interpersonal events and their interrela-
tionships has never been investigated in child sexual abuse prior to this study.
Here we were interested in examining the offender–victim interchange with a
focus on the stage where the offender seeks the cooperation of the child for sex-
ual activity. We drew from an interpersonal script framework to investigate the
sequencing of interpersonal events and patterns of behavioral interdependence
in child sexual abuse (Baldwin, 1992; 1995; Fehr et al., 1999). This approach
provides researchers interested in understanding the if–then nature of inter-
changes between the offender and the victim with a theoretical framework from
which to study interpersonal forms of crime such as sexual offenses and domes-
tic violence. In turn, as it focuses on the interchange between the offender and
Victim reaction in child sexual abuse 117
the victim not the offender himself, such an analysis more accurately reflects
the reality of child sexual abuse and as a result, increases the potential to inform
crime prevention initiatives and prevention programs administered to children.
Studies of the offender–victim interchange in violent crimes have been con-
ducted in the past on the assumption that the outcomes of a violent altercation
are determined by what occurs during the offense rather than by the individual
offender or his initial goals for committing the offense (e.g., Block, 1981; Felson
and Steadman, 1983; Luckenbill, 1977; 1982; see also Tedeschi and Felson, 1994).
These studies showed in fact that the offender and the victim adopt behaviors
shaped in part by the actions of the other which is instrumental to the outcomes of
this interchange. Consistent with these studies, our study showed that the sever-
ity of child sexual abuse in terms of sexual behaviors is determined in part by the
reaction of the victim and victim characteristics.

This paper was supported by a Researcher Grant from Griffith University. The
research project was supported by a grant awarded by the Australian Research
Council (#LP0668287).

1 Participants were also allowed to choose more than one answer. This is because more
than one action may be adopted during crime by the offender (and victim). For the first
question, a total of ten events involved the use of manipulative strategies and of a non-
persuasive strategy (or direct action). As at least one persuasive strategy was adopted,
these cases were included in the use of a persuasive strategy category. As for victim
reaction, a total of five events involved resistance as well as compliance from the victim.
As the focus was on victim compliance exclusively, these cases were included in the
category of non-compliance.

Abelson, R. P. (1981). Psychological status of the script concept. American Psychologist,
36, 715–729.
Asdigian, N. and Finkelhor, D. (1995). What works for children in resisting assaults?
Journal of Interpersonal Violence, 10, 402–418.
Aylwin, A. S., Clelland, S. R., Kirkby, L., Reddon, J. R., Studer, L. H. and Johnston,
J. (2000). Sexual offense severity and victim gender preference: A comparison of
adolescent and adult sex offenders. International Journal of Law and Psychiatry, 23,
Balemba, S. and Beauregard, E. (2012). Reactions to resistance: The role of contextual
factors in sex offending. Violence and Victims, 27, 148–165.
Baldwin, M. W. (1992). Relational schemas and the processing of social information.
Psychological Bulletin, 112, 461–484.
Baldwin, M. W. (1995). Relational schemas and cognition in close relationships. Journal
of Social and Personal Relationships, 12, 547–552.
118 B. Leclerc, S. Smallbone, R. Wortley
Beauregard, E. and Leclerc, B. (2007). An application of the rational choice approach to
the offending process of sex offenders: A closer look at the decision making. Sexual
Abuse: A Journal of Research and Treatment, 19, 115–133.
Beauregard, E., Proulx, J., Rossmo, D. K., Leclerc, B. and Allaire, J.-F. (2007). Script
analyses of the hunting process of serial sex offenders. Criminal Justice and Behavior,
34, 1069–1084.
Berliner, L. and Conte, J. R. (1990). The process of victimization: The victim’s perspective.
Child Abuse & Neglect, 14, 29–40.
Bickel, R. (2007). Multilevel analysis for applied research: It’s just regression, New York,
NY: The Guilford Press.
Block, R. (1981). Victim–offender dynamics in violent crime. The Journal of Criminal
Law and Criminology, 72, 743–761.
Carjaval, S. C., Baumler, E., Harrist, R. B. and Parcel, G. S. (2001). Multilevel models and
unbiased tests for group based interventions: Examples from the Safer Choices Study.
Multivariate Behavioral Research, 36, 185–205.
Clarke, R. V. and Cornish, D. B. (1985). Modeling offenders’ decisions: A framework for
research and policy. In M. Tonry and N. Morris (eds) Crime and justice: An annual
review of research, Vol. 6. (pp. 147–185). Chicago, IL: University of Chicago Press.
Cornish, D. B. (1994). The procedural analysis of offending and its relevance for situational
prevention. In R. V. Clarke (ed) Crime Prevention Studies: Vol. 3. (pp. 151–196).
Monsey, NY: Criminal Justice Press.
Cornish, D. B. and Clarke, R. V. (1986). The reasoning criminal: Rational choice
perspectives on offending, New York, NY: Springer-Verlag.
Cornish, D. B. and Clarke, R. V. (2008). The rational choice perspective. In R. Wortley
and L. Mazerolle (eds) Environmental Criminology and Crime Analyses (pp. 21–47).
Cullompton, UK: Willan.
DeLamater, J. and Friedrich, W. N. (2002). Human sexual development. The Journal of
Sex Research, 39, 10–14.
Elliott, M., Browne, K. and Kilcoyne, J. (1995). Child sexual abuse prevention: What
offenders tell us. Child Abuse & Neglect, 19, 579–594.
Fehr, B., Baldwin, M., Collins, L., Patterson, S. and Benditt, R. (1999). Anger in close
relationships: An interpersonal script analysis. Personality and Social Psychology
Bulletin, 25, 299–312.
Felson, R. B. and Steadman, H. J. (1983). Situational factors in disputes leading to criminal
violence. Criminology, 21, 59–74.
Guo, G. and Zhao, H. (2000). Multilevel modeling for binary data. Annual Review of
Sociology, 26, 441–462.
Kaufman, K. L., Hilliker, D. R. and Daleiden, E. L. (1996). Subgroup differences in the
modus operandi of adolescent sexual offenders. Child Maltreatment, 1, 17–24.
Kaufman, K. L., Holmberg, J. K., Orts, K. A., McCrady, F. E., Rotzien, A. L., Daleiden, E. L.
et al. (1998). Factors influencing sexual offenders’ modus operandi: An examination of
victim-offender relatedness and age. Child Maltreatment, 3, 349–361.
Lacoste, J. and Tremblay, P. (2003). Crime and innovation: A script analysis of patterns in
check forgery. In M. J. Smith and D. B. Cornish (eds) Crime Prevention Studies: Vol.
16. Theory for practice in situational crime prevention (pp. 169–196). Monsey, NY:
Criminal Justice Press.
Leclerc, B. and Tremblay, P. (2007). Strategic behavior in adolescent sexual offenses
against children: Linking modus operandi to sexual behaviors. Sexual Abuse: A Journal
of Research and Treatment, 19, 23–41.
Victim reaction in child sexual abuse 119
Leclerc, B., Carpentier, J. and Proulx, J. (2006). Strategies adopted by sexual offenders
to involve children in sexual activity. In R. Wortley and S. Smallbone (eds), Crime
Prevention Studies: Vol. 19, Situational prevention of child sexual abuse (pp 251–270),
Monsey, NY: Criminal Justice Press.
Leclerc, B., Proulx, J. and Beauregard, E. (2009). Examining the modus operandi of
sexual offenders against children and its practical implications. Aggression and Violent
Behavior, 14, 5–12.
Leclerc, B., Proulx, J., Lussier, P. and Allaire, J.-F. (2009). Offender–victim interchange
and crime event outcomes: Modus operandi and victim effects on the risk of intrusive
sexual offenses against children. Criminology, 47, 595–618.
Leclerc, B., Wortley, R. and Smallbone, S. (2010). An exploratory study of victim
resistance in child sexual abuse: Offender modus operandi and victim characteristics.
Sexual Abuse: A Journal of Research and Treatment, 22, 25–41.
Leclerc, B., Wortley, R. and Smallbone, S. (2011a). Getting into the script of adult child
sexual offenders and mapping out situational measures. Journal of Research in Crime
and Delinquency, 48, 209–237.
Leclerc, B., Wortley, R. and Smallbone, S. (2011b). Victim resistance in child sexual
abuse: A look into the efficacy of self-protection strategies based on the offender’s
experience. Journal of Interpersonal Violence, 26, 1868–1883.
Luckenbill, D. F. (1977). Criminal homicide as a situated transaction. Social Problems,
25, 176–186.
Luckenbill, David (1982). Compliance under threat of severe punishment. Social Forces,
60, 810–825.
Luckenbill, D. F. (1980). Patterns of force in robbery. Deviant Behavior, 1, 361–378.
Morselli, C. and Roy, J. (2008). Brokerage qualifications in ringing operations.
Criminology, 46, 71–98.
Proulx, J., Ouimet, M. and Lachaine, N. (1995). Criminologie de l’acte et pédophilie.
Revue Internationale de Criminologie et de Police Technique, 48, 294–310.
Rispens, J., Aleman, A. and Goudena, P. P. (1997). Prevention of child sexual abuse
victimization: A meta-analysis of school programs. Child Abuse & Neglect, 21,
Roberts, J. A. and Miltenberger, R. G. (1999). Emerging issues in the research on child
sexual abuse prevention. Education and Treatment of Children, 22, 84–102.
Schank, R. and Abelson, R. P. (1977). Scripts, plans, goals and understanding: An inquiry
into human knowledge, Hillsdale, NJ: Erlbaum.
Sellin, T. and Wolfgang, M. (1964). The measurement of delinquency. Montclair, NJ:
Patterson Smith (Reprinted, 1978).
Smallbone, S. and Wortley, R. (2000). Child sexual abuse in Queensland: Offender
characteristics and modus operandi (full report), Brisbane, Australia: Queensland
Crime Commission.
Stylianou, S. (2003). Measuring crime seriousness perceptions: What have we learned and
what else do we want to know. Journal of Criminal Justice, 31, 37–56.
Tedeschi, J. T. and Felson, R. B. (1994). Violence, aggression, and coercive actions,
Washington, DC: American Psychological Association.
7 Drug dealing
Amsterdam’s Red Light District
Scott Jacques and Wim Bernasco

Amsterdam has many lively spots, including Leidseplein, Rembrandtplein, and

an area that locals call de Wallen. Foreigners know it as the Red Light District.
The sex trade gives this area its name but it is also a hot spot to procure drugs,
having been called everything from a “drug paradise” to “Mecca” (Grapendaal,
Leuw, and Nelen, 1995: 82). This chapter uses crime script analysis to address
three questions: (1) What are the general actions involved in drug dealing? (2)
How does street dealing in Amsterdam’s Red Light District unfold? (3) What
opportunities does this afford prevention oriented formal control?
In what follows, we first situate the offense by providing background on drug
policy and law enforcement practice in Amsterdam, the Netherlands. Then the
orienting perspectives—opportunity theory, rational choice, and crime script
analysis—are reviewed (Cohen and Felson, 1979; Cornish, 1994). This is fol-
lowed by a description of our method, namely a mixed-method study of active
drug dealers. That information is used to determine and describe which steps are
involved in each drug sale. The chapter concludes by suggesting preventative
intervention points in that process.

Formal control of drug trade in Amsterdam

In the Netherlands the guiding legal framework for drug-related issues is the
Opium Act of 1976 (Leuw, 1991). It divides drugs into two categories differen-
tiated by their health risk to users. Substances with “unacceptable health risk”
are placed in Category I and referred to as “hard drugs.” This category includes
cocaine, amphetamines, heroin, ecstasy, and LSD. Within Category II are “soft
drugs” deemed to be significantly less harmful; this includes cannabis (marijuana
and hash) and psychedelic mushrooms. In essence, the Dutch government decided
that the health risks of soft drugs are acceptable but those of hard drugs are not.
There is medical evidence to support this decision; in the recent past, there are
approximately 100 deaths per year in the Netherlands related to hard drug use but
almost none tied to soft drug use (Trimbos Institute, 2009).
Regardless of whether a drug is categorized as hard or soft it is illegal to
possess, sell, transport, or produce it (Netherlands Ministry of Foreign Affairs,
2008).1 “The penalties for drug-related offences [i.e., sanctions] and the priority
Drug dealing: Amsterdam’s Red Light District 121
given to those offences in law enforcement policy [i.e. detection] depend on the
nature of the offence and on the type of drug involved” (Netherlands Ministry
of Foreign Affairs, 2008: 6; also see Leuw, 1991). In other words, law enforce-
ment priorities correspond with the risk per category (Netherlands Ministry of
Health, Welfare, and Sport, 2003). “The sale of any quantity of hard drugs has
[the] high[est] priority in investigative and prosecution policies and will receive
harsh penalties” (Netherlands Ministry of Health, Welfare, and Sport, 2003: 17),
whereas the “[s]ale and possession of cannabis for personal use are much lower
priorities” (Netherlands Ministry of Foreign Affairs, 2008: 10).2
There are several government agencies that manage Dutch and, by extension,
Amsterdam’s drug policy and law enforcement practice. Because the drug “prob-
lem” is primarily seen as a health issue rather than in moralistic terms (Verdurmen,
Ketelaars, and van Laar, 2004: 5), policy is coordinated by the Ministry of Health,
Welfare and Sport. The Ministry of Justice—especially the Public Prosecutor—han-
dles matters related to law enforcement. Local matters are managed by the Ministry
of the Interior and Kingdom Relations. Within municipalities, policy and law
enforcement practices are determined—to a degree—by the so-called “three way
consultations” between the mayor, field officer (i.e., Public Prosecutor), and chief
of police (Netherlands Ministry of Foreign Affairs, 2008). Thus, a large degree of
the official stance on drugs in the Netherlands is determined at the national level, but
municipalities—such as the city of Amsterdam—have powers of their own.
The measures taken by police to detect and thereby prevent or sanction hard
drug-related activities have evolved over the years. As relates to the street scene
in the Red Light District, prior research reports policing techniques such as ever-
present patrolling, secret observations posts, and undercover officers.3 Preventive
ordinances related to the drug trade have restricted the assembling of individuals,
possession of knives, and access to public areas (Grapendaal et al., 1995; Leuw,
1991; van Gemert and Verbraeck, 1994). The modern era of policing highlights
an array of cameras that are always on the lookout.

Scripts, rationality and opportunity

To help us create a script of street dealing in Amsterdam’s Red Light District,
this chapter uses the rational choice and opportunity perspectives. This latter per-
spective holds that if the necessary conditions for a behavior are not present in a
given situation at a specific place and time, then it is impossible for that behavior
to occur then and there (see Cohen and Felson, 1979; Cohen, Kluegel, and Land,
1981: Gottfredson and Hirschi, 2003). Applied to crime, opportunity theorists
try to identify the necessary conditions for offenses. Practitioners may use that
knowledge to manipulate environmental conditions to reduce crime.
A comrade of opportunity theory is the rational choice perspective. It has
three main features (Clarke and Cornish, 1985). One is an emphasis on agency
and choice. The rational choice perspective on criminal behavior assumes that
criminal behavior is goal oriented. Our analysis of offender decision making
will emphasize both the negative consequences and positive incentives of selling
122 Scott Jacques and Wim Bernasco
drugs. Second is a focus on offense-specific analyses. This follows from the prin-
ciple that different offenses involve different types of decisions and different
types of incentives. We focus on a specific kind of drug dealing, street-level, in a
specific urban environment, Amsterdam’s Red Light District. Last, this perspec-
tive distinguishes between criminal involvement (processes of criminal initiation,
continuance, and desistance) and criminal events (processes of choice that relate
to immediate circumstances and situations). We make this distinction too and will
focus solely on the criminal event. We address the question: What necessary and
facilitating steps are involved in making a drug sale?
Script analysis is particularly well suited to this kind of question, which focuses
on the unfolding of events (Cornish, 1994). Rather than see any given offense as
a single action, script analysis describes and explains the array of goal-oriented
sub-actions that occur before, during, and even after a crime. “[S]cripts generally
are event sequences extended over time, and the relationships have a distinctly
causal flavor, that is, early events in the sequence produce or at least ‘enable’ the
occurrence of later events” (Nisbett and Ross, 1980: 34).
Scripts may be thought of as hierarchically differentiated and organized
according to their level of abstraction (Cornish, 1994). General scripts are meant
to describe the sequence of a specific offense (e.g., burglary or robbery) across
all contexts (e.g., in lower- and upper-class areas or rural and urban communi-
ties). Track scripts provide greater detail about the steps for a specific offense in
a given context. For instance, there may be a general script for drug dealing that
is academically useful but practically speaking inept because it hardly describes
what concrete actions happen in any market. It may be warranted, therefore, to
develop a track script for a specific market.
Script analysis has practical utility for reducing crime. “The unfolding of a
crime involves a variety of sequential dependencies within and between elements
of the action. Crimes are pushed along or impeded by situational contingencies
…. Each, several, or all of these may have to be present if the action is to be car-
ried forward at that time” (Clarke, 1994: 156). By determining the steps involved
in an offense, script analysis suggests multiple intervention points for crime pre-
vention measures.

The present study: data and methods

Previous work in crime script analysis has examined drug manufacturing in clan-
destine laboratories (Chiu, Leclerc, and Townsley, 2011), cigarette smuggling
(von Lampe 2010), and sexual crimes (Beauregard, Proulx, Rossmo, Leclerc,
and Allaire, 2007; Beauregard, Rossmo, and Proulx, 2007; Leclerc, Wortley, and
Smallbone, 2011). The task for the present paper is to uncover a general script
of drug dealing and a track script for street dealing in Amsterdam’s Red Light
District. The script is written, so to speak, by drawing on data obtained from a
study involving quantitatively oriented interviews with 50 drug dealers and quali-
tatively oriented independent observations of the scene. Below we describe the
data and explain how they were collected and analyzed.
Drug dealing: Amsterdam’s Red Light District 123
As will be discussed in the findings section, street dealers in this part of
town are known to solicit potential customers, such as tourists, through various
methods. At the time of the study, one method is saying such things as “coke,”
“heroin,” and/or “ecstasy” as people walked by (also see Grapendaal et al., 1995;
van Gemert and Verbraeck, 1994). Another technique is to use eye contact or
head-nods to suggest the possibility of trade. Accordingly, we initially sought to
recruit dealers for our study by making use of their soliciting techniques. This
process would begin when the lead author entered the Red Light District. He
walked the streets and stood in dealing “hot spots.” On occasion a dealer would
initiate contact by attracting his attention in the ways described above. Then he
would inform the dealer of the research project and ask for their participation. If
the dealer agreed to be interviewed, they would immediately go to a nearby public
place to conduct the interview. Participants recruited through purposive sampling
were asked to refer other dealers for the study. Of the 50 dealers we interviewed,
19 were recruited by purposive sampling and 31 were recruited via snowball sam-
pling. All participants were remunerated €50 for their cooperation.
Although in the absence of a sample frame we cannot be certain about how
representative the sample is for the population of hard drug street dealers in
Amsterdam’s Red Light District, we briefly describe the main characteristics of
this group of 50 dealers. Except for two women, all interviewed dealers were male
(96 percent). They were 38.7 years old on average. Only two of them (4 percent)
were married. The majority of the dealers (72 percent) were born outside the
Netherlands. Many of the dealers were school dropouts, as 55 percent had not
completed their secondary education. About one third (32 percent) was employed
and thus had another source of income in addition to drug dealing. Many of the
drug dealers were users too: 70 percent used cannabis daily and 30 percent used
cocaine daily themselves.
The interviews with dealers consisted of fixed-choice survey-type items. The
fixed-choice items collected information on participant’s characteristics, busi-
ness, victimization, use of social control, dealing history, and style. To obtain
detailed information on the tangible aspects of drug dealing, the dealers were
asked a number of specific questions in their last drug sale. The items discussed
included whether they were dealing alone or with somebody else, where and when
the customers were solicited and where and when the transaction took place, the
characteristics of the customers, the communication with the customers, and the
quantities of drugs and money that were exchanged.
As with any data collected through self-report, our participants may have
resorted to lying for various reasons. They may, for instance, have been con-
cerned about the risks involved in telling a stranger about their criminal activities.
To maximize the validity of our interview data, we employed the following strate-
gies: the lead author—who is American and thus unlikely to be wrongly perceived
as an undercover police officer by dealers—made clear to participants his goals
and interests in conducting an interview; participants were promised confiden-
tiality; they were told that it is better to refrain from answering a question than
to distort the answer; and, comments initially judged strange or baseless were
124 Scott Jacques and Wim Bernasco
probed further to reveal and, if possible, resolve inconsistencies. Despite these
steps, however, it is possible that some of the participants stretched the truth.
In addition, independent observations of how drug dealers operate were made
by the lead author from September 2008 through June 2010. These observations
were made informally (i.e., without a systematic observation checklist or at speci-
fied times and places) between eight o’clock in the morning and midnight. The
kind of information recorded during observations includes where solicitations and
sales occurred, the methods of doing so, aspects of formal control (e.g., where
and how police patrolled in the area) and its effect on dealing. Because all of the
observations were unobtrusive and occurred in public space as well as in plain
sight (i.e., not from a hidden position), no persons—such as dealers, customers,
police, or others—were informed that they were being observed.

The script of street dealing in the Red Light District

Five key steps can be distinguished in the general script of drug dealing. Two of
the actions are logically required for a drug sale to occur; these necessary steps
are (1) agreeing on trade terms and (2) making an exchange. There are three other
actions that are not required but nonetheless very common owing to serving some
function; we term these facilitating steps. This category includes (A) searching
for a customer, (B) soliciting, and (C) agreeing on and going to the trade locale.
Necessary steps apply to all trades, but facilitating steps do not.
Although script analyses typically lay out a fixed order of actions involved in
committing a crime, our analysis suggests this might be slightly more complex
for drug dealing in general. The characteristics of drug markets—including who
is involved, where it is, and the legal climate—can affect the action sequence. For
example, when dealers have regular clients who know how to contact them, deal-
ers may be able to largely skip searching for potential trade partners (step A) and
soliciting them (step B). In such cases a customer can call the dealer to arrange
a place and time to trade (step C), or could temporally truncate the facilitating
steps by going to a dealing location (e.g., a crack-house) and asking for a trade.
Figure 7.1 is a graphic illustration of the script of drug dealing, and demonstrates
the possible variations in the number of steps and their order of execution.
These many possibilities show it is difficult to pin down the sequence of drug
dealing’s general script because this offense requires coordination with other crimi-
nals—namely customers. Indeed, the general script of drug buying reflects that for
selling: customers may have to search for and solicit dealers, agree on and go to trade
places, come to terms with what is owed and when, and make the exchange. Unlike
other offenses that only require a criminal’s unilateral action (e.g., assault or bur-
glary), drug trade involves bilateral action—give and take (see Jacques and Wright,
2008a); therefore, it is particularly tricky to sequentially script its general steps.4
Such difficulties are partially circumvented by moving from a general to track-
level analysis. Rather than provide a description intended to cover all contexts,
track scripts narrow the focus in order to provide greater specification with regard
to what, how, and why actions unfold. In what follows, we present a “best fit
Drug dealing: Amsterdam’s Red Light District 125

A. Search customer:
Walk to spot lone with white
male, 30–49 years of age

B. Solicit customer:
Say type of drug(s) for sale
when face-to-face

1. Agree on terms: C. Arrange place/time:

Agree to provide 1 gram Tell customer to walk together
of cocaine for €50 to nearby place ASAP

2. Make exchange:
Fulfill agreement

Figure 7.1 General drug dealing script and track level script for street dealing in
Amsterdam’s Red Light District.

Note: steps A, B, and C are facilitating steps; steps 1 and 2 are necessary steps. Italicized script
represents most common track of street dealers in Amsterdam’s Red Light District. Failure may occur
at each step and create feedback loop to a previous step (often to step A). For reasons of parsimony the
corresponding arrows are omitted from the figure.

script” of street dealing in the Red Light District. This model describes the ways
in which street dealers typically accomplish each step, including the sequence.

Facilitating steps
A facilitating step is an action that is functionally important for an ultimate goal
to occur. Drug sales do not necessarily involve finding and soliciting customers,
or arranging and travelling to a place to do the deal. Without such facilitating
actions, however, drug sales would be far less common or altogether nonexistent.
At the most basic level, drug sales have three facilitating steps: searching for a
customer; soliciting the customer; agreeing on a place to make the sale and going
there. Street dealers in the Red Light District have a script characterized by walk-
ing and/or standing to find a customer, who is solicited with a whisper, glance,
or nod, and they jointly agree to do a small-scale exchange in the immediate or a
nearby place that is travelled to on foot.
126 Scott Jacques and Wim Bernasco
A. Search for a customer
Before a dealer and customer can arrange the terms of agreement and make an
exchange they must locate each other through search methods: ways of finding a
specified item. In the drug world, the principal methods are retrieving potential
trade partners’ contact information (i.e., phone number or email address) and,
sometimes related to this, going to places where such persons might be present.
There are two key decisions involved in searching for a customer: “Who do I look
for?” and “How do I look?”
The answer to each question depends on the other. It is easy and common to
search for friends by calling their phone number, but that can be more difficult
and unusual with strangers. Thus a dealer who sells to friends will likely have a
different method of finding customers than a seller to strangers. Unlike dealers
who have friends as customers or fixed selling locations (e.g., in a bar), dealers
selling to unknown, nomadic, or electronically disconnected customers are espe-
cially reliant on tracking down their clientele through more traditional methods.
Dealers should be expected to use the search method they find most rational,
given their range of opportunities. In theory this search method will be one
that provides a satisfactory or ideal rate of (1) “path crosses” but (2) not sub-
stantially increase the risk of victimization or detection by police. A path cross
refers to the intersection of a dealer and seller in a “communication channel”
where it is possible to signal. The rate of path crosses will increase as does
the method’s ability to identify who they are and where they congregate. The
process of searching for customers must be balanced against concerns for adver-
saries (see Jacques and Reynald, 2012). For example, some styles of finding
customers, such as standing around in known hot spots, are likely to result in
path crosses but also prone to detection by police. Dealers do face negative
consequences if they overestimate the pool of potential customers. It may result
in lost time, formal punishment, or victimization. This suggests accuracy in
predicting customers’ characteristics increases the ability to sell and reduces the
likelihood of incurring harms.
For street dealers in the Red Light District, success or failure in searching for
customers depends on three decisions: Who to look for? Where to find them?
How to not attract police attention? First, we can expect dealers to search for
persons with whom they have had previous success in making a sale. For the
dealers in our sample, and based on their accounts of their last deal, the typi-
cal customer was alone (74 percent), a stranger (56 percent), between 30 and 49
years old (70 percent), white (78 percent), and male (100 percent). To the degree
this finding characterizes the “average” drug buyer in the Red Light District, we
should expect dealers to focus their searches on such persons. For example, they
are probably unlikely to pay much attention to females (at least for the purpose
of selling) and will ignore very old or very young persons. This discrimination
process is rational by saving energy and increasing efficiency.
There are two more attributes that characterize buyers in the Red Light District:
addicts (for whom drug use is a necessary aspect of daily routines) and “drug
Drug dealing: Amsterdam’s Red Light District 127
tourists” (who come from abroad and for whom drugs use is occasional and rec-
reational only):

The phenomenon of foreigners coming to the Netherlands to use or purchase

drugs is known as “drugs tourism.” Nuisance caused by drug tourists is espe-
cially evident in Dutch border cities and in large cities …. Cannabis tourists
create nuisance at coffee shops; hard drug tourism often goes hand in hand
with aggressive recruitment methods (drug runners). This results in unaccep-
table nuisance in residential areas and city centres.
(Netherlands Ministry of Health, Welfare, and Sport, 2003: 22)

Dealers are aware that drug tourists frequent the Red Light District, and also that
addicts find the area attractive. The flipside to this is that “normal Dutch people”
and certain kinds of foreigners/tourists are perceived as uninterested in buying
drugs off the street. Accordingly, dealers develop a “likely customer schema.”
When searching for potential trade partners, dealers focus their efforts on finding
and identifying persons who match the likely customer schema. In practice, this
is done by doing such things as looking at persons’ style of dress and personal
hygiene (e.g., that of addicts is notoriously poor), and listening to whether Dutch
or another language is spoken.
Rationality and opportunity theory help us to understand the likely customer
schema and dealing script.5 Opportunity wise, drug tourists and addicts are not the
only persons in the Red Light District, but are the most likely to be seeking drugs.
Unlike integrated and better-to-do Amsterdamers, addicts and especially tourists
are unlikely to have stable house or phone dealers to obtain drugs from. Without
such contacts, customers’ opportunity to trade is narrowed to known public mar-
kets—a famous one being the study area. The irony here is that customer’s lack of
options provides opportunity and rationale for street dealing. Knowing customers
are found in this area, dealers travel there to search for them.
Once in the area, dealers must decide on whether to walk, stand, or do both.
It seems like a simple decision, but it has far reaching consequences. Rationally
speaking, the optimal choice is to stand still. This is for several reasons. Compared
to walking, standing increases the chance of crossing paths (assuming potential
customers walk every street in the area). Over time, customers may learn that this
is a likely place to find a dealer. Perhaps least pertinent, less physical energy is
expended when standing than walking.
With that said, dealers must be careful not to spend too much time in any one
spot or the whole of the Red Light District because police officers or concerned
citizens may begin to suspect and investigate wrongdoing. “Hanging around,” or
what others call loitering, is both a cultural activity and rule violation depending
on who is doing it and why. This activity—or lack thereof—may be perceived
as facilitating drug dealing for the rational reasons mentioned directly above.
Therefore, law enforcement officials have good reason to deter this activity. Our
field observations suggest that while dealers stand on occasion they are much
more prone to walk around when looking for customers (also see Grapendaal
128 Scott Jacques and Wim Bernasco
et al., 1995). When they do stand it is typically for only a minute or two. The
dealers are conscious of the police presence and are apt to forgo the benefits of
standing for the reduced risks associated with walking.

B. Solicit the customer

After a potential customer and dealer have linked up, the next facilitating step is
solicitation of this person. To solicit is to let others know that a particular good
or service is available or wanted for purchase (see Gambetta, 2009). This action
requires the dealer to make two decisions: What to signal? How to communicate?
The “how” of solicitation is more specifically a question of “In what way to
send signals?” Different modes of communication have different benefits and
costs associated with their use. Nowadays, drug market participants typically
solicit each other by talking in person or over the phone and through written
word used in email or text messages. These types differ in the convenience, risk,
and limiting preconditions. For example, written messages sent electronically are
often unwittingly saved in one way or another, making them vulnerable targets
for evidence gathering by police. It would seem spoken words are less susceptible
in this regard but not entirely so (e.g., phone taps are a well-known tool of law
enforcement officials). Maybe the least risky way to communicate is also the least
convenient, verbally in-person, but doing so with an undercover police officer is
a hazardous possibility. And one precondition for communicating over the phone
or through the internet is that the would-be sender has the receiver’s number
or address—a facet of opportunity.6 Without such information these signaling
devices cannot be used, which makes face-to-face communication a necessity.
The “what” of solicitation is in some respects a straightforward issue. “Drugs
are for sale” is what a dealer hopes to communicate to potential customers. There
are many conceivable ways of relaying that information, each of which comes
with its own benefits and costs. To literally say or write “drugs are for sale” is
a very direct approach with a low margin of error (i.e., miscommunication), but
for that same reason it exposes the dealer to considerable risk of detection by law
enforcement officials. A less direct approach is to say/write something in code;
for example, “white” or “brown” will be interpreted by knowledgeable persons
as meaning “coke” or “heroin” respectively (also see Grapendaal et al., 1995; van
Gemert and Verbraeck, 1994). Coded signals have less obvious meanings and are
therefore less likely to lead to formal detection or the successful solicitation of
a customer. An even more abstract and less blatant signal involves literally say-
ing and writing nothing at all; this is where the “how” and “what” of solicitation
intersect. In the streets, at least, potential dealers and customers show interest in
each other through body language, like eye glances and head nods; such a method
has the benefit of keeping the dealer’s intentions hidden from police but does so
at the cost of losing business.
Among street dealers in the Red Light District, solicitation principally involves
verbal face-to-face communication (52 percent) and body language or facial
expressions (15 percent).7 We also find that a fair number of solicitations involve
Drug dealing: Amsterdam’s Red Light District 129
a phone (26 percent). For face-to-face communication, the range of what is com-
municated extends from blatant whispers to discrete physical movements. As
dealers walk or stand in the area and cross paths with a potential customer they
will sometimes say in a light voice, “ecstasy” or “coke.” In other cases dealers
will use glances with the eyes or head nods to make initial contact with a cus-
tomer. On occasion a customer will initiate contact with a dealer by acting on
their presumptions about what dealers look like (e.g., black, alone, and dressed in
a certain style of clothes) and approach such persons despite them being strangers.
Dealers may also first arrange a deal through the phone; these cases principally
involve friends or trusted customers.
Street dealing has a particular solicitation script for rational and opportunistic
reasons.8 It is possible to imagine otherwise, but it would seem the nature of an
illicit street market based on tourist customers is unlikely to rely on phone num-
bers or email addresses to send messages. After deciding that in-person signaling
is best, the next question is what exactly to communicate. Dealers have seemingly
selected a middle-ground option where they either quietly say the name of a drug
or give off body language that suggests “something is up.” Dealers perform a
combination of signaling strategies depending on the present environment. These
middle-road techniques are effective in so far as they provide moderate benefits
owing to their ability to successfully solicit customers without overtly inviting the
attention of formal control agents.

C. Agree on and go to trade locale

Once a dealer and customer have passed through the solicitation stage they must
implicitly or explicitly agree on where to trade and, in some cases, travel to that
place.9 The questions facing a dealer are “How to communicate? What to signal?
Where to meet? How to get there? And when?” Here, how to communicate is a
question of language (e.g., spoken, written), distance (e.g., miles vs. in person),
and tool use (e.g., whether to phone/message or talk to face-to-face). What to
communicate involves figuring out an exact place and time to trade. If this can be
agreed on then the customer and dealer are able to meet, travelling if necessary.
Agreeing on and going to a trade locale has important implications for the
sale and safety from police. These kinds of choices are guided by rationality and
opportunity. Different languages, distances, and tools of communication vary in
their availability and ability to facilitate signaling to potential customers without
attracting police attention. For example, some directions are more precise but take
longer to explain, thereby increasing exposure. Some places are closer but riskier
because of CCTV. A distant time may be safer from police but inconvenient for
a trade partner. And some forms of travel, such as bicycling versus walking, are
faster but—for their own reasons—more likely to lead to detection by police.
These are the kinds of cost–benefit analyses that dealers must engage in, con-
sciously or not.
A key decision for Red Light District drug traders is how to travel to do a
deal. This decision requires an agreement between buyer and seller, but whether
130 Scott Jacques and Wim Bernasco
the decision is guided jointly or by one party is a separate issue. In all but one of
the sales we examined, the dealer made the decision about where to go, not the
customer. There might be a number of explanations for this finding. One is that
the market relies heavily on tourists, who are unfamiliar with the city; it is only
rational that local dealers will decide where to go.
A sub-decision in deciding how to travel is how far to travel. Dealers in our
sample perceive further distance to be a cost not worth the benefits. For the sales
that began through communication in the Red Light District, two-thirds were
completed in the same area and a quarter took place just outside it. These areas
are fairly small; for example, a person walking from one corner to the other of the
Red Light District would take about three minutes.
There are other common actions of street dealers that belong in their script.
After the solicitation but before the trade, the dealer uses (1) words or body lan-
guage to tell the customer that the exchange will be in the immediate or a nearby
place, (2) at the immediate or nearby time. If necessary, they travel to the place by
foot.10 The physical and temporal immediacy of the event reduces risks associated
with detection by police. Words are beneficial owing to their directness, but body
language is useful when a verbal communication barrier exists. When travelling
together, the slowest form of travel, walking, is popular as most everyone can do
it (i.e., the opportunity exists).

Necessary steps
A necessary step is an action that must logically occur for an ultimate goal, such
as a drug sale, to happen. The definitional properties of “drug sale”—namely that
it is a fair (i.e., agreed upon) exchange of resources between parties—requires
it to involve two steps: a dealer and customer must (1) agree on trade terms and
(2) make an exchange of drugs for something else.11 Whether in Amsterdam or
Bangkok, America or Britain, among the rich or poor, all drug deals will involve
agreeing on trade terms and the transfer of wealth. The script used by street deal-
ers in the Red Light District involves face-to-face verbal agreements to make a
small-scale trade.

1. Agree on trade terms

After a successful solicitation (and either before or after agreeing on going to a
trade locale), a dealer and customer find themselves in a position to discuss trade
terms. This involves communication—whether written, verbal, or otherwise—
between a customer and seller (or in some cases an intermediary) about what
resource(s) and/or service(s) each party is obliged to give the other, and when.
No matter the mode of communication, the seller must ask him or herself, “What
should I agree to?” Every economic transfer has three principal components: how
much of what the customer owes the seller; vice versa; and, when the payments
are due (see Jacques and Wright, 2008a). For a trade to occur, the dealer and
buyer must come to agreement on these matters.12 This will raise decision points
Drug dealing: Amsterdam’s Red Light District 131
for the dealer: “How much should I agree to charge (and what forms of payment
to receive)? Should I agree to give credit and, if so, for how long?”
Whichever answer has the perceived greatest utility should, in theory, be the
one chosen by the dealer. The dealer must weigh, for instance, the potential ben-
efits of trying to charge an excessive price versus the risks of having the customer
walk away. Or customers may counter-offer or altogether turn down the deal. At
that point the dealer can pull out, counter-offer the counter-offer, or accept the
customer’s proposal. This (joint-)decision-making process—including whether
to continue or abort it—must account for not only the direct costs and benefits
of the exchange but also include, at a minimum, consideration of the legal risks
associated with open air drug markets. Although communication is beneficial by
facilitating trade, it is costly when it draws the attention of government officials.
The time component is important in another way as well. The dealer may have to
agree or reject selling on credit. Giving credit is beneficial if it facilitates a trade
that would otherwise not occur, but it does so at the cost of delaying income or
opening up the possibility of never being paid at all.
As discussed above, the relationship between a Red Light District seller and
customer initially begins when one signals to the other that trade is a possibility.
Following this, the seller and buyer immediately talk out a deal, which takes little
time and effort (thereby reducing risk) and hastens the transfer of goods (i.e., quick-
ens benefits). In the streets, agreements are made swiftly or the process aborted. This
process usually lasts seconds or at most a couple of minutes (also see Kraan, 1994).
Failed attempts to agree on trade terms may spring from the inability to com-
municate—a facet of opportunity. The Red Light District market involves not
only Dutch (immigrants) but also a considerable number of foreigners, few of
whom speak the native tongue. Street dealers do have competence in Dutch and
English; this bilingualism facilitates drug trade by allowing for trade terms to be
discussed. Potential customers who cannot speak at least one of these languages
are more likely to be uncomfortable with negotiations or unable to communicate
and agree on trade terms.
Street markets based primarily on stranger-traders, especially tourists, are
unlikely to involve agreements for credit (but see Jacobs, 1999). This would be
risky business by opening up the possibility of fraud. Obviously a customer who
will be leaving the country in a short time may be apt to renege on promises to
pay debt. Related to this, but only in part, is that persons who are not familiar with
each other may lack the necessary trust involved in giving now and getting later.

2. Make an exchange
The final necessary step in all drug trades is the actual exchange: a movement of
wealth (i.e., drugs for cash, sex, etc.) between parties (see Jacques and Wright,
2008a). This step involves a decision on what to actually trade and when, as com-
pared to merely agreeing on what should be done (which is step 1).
The quantitative information we collected from Red Light District sellers is
useful in deciphering a script for this step in street dealing and the previous one.
132 Scott Jacques and Wim Bernasco
First, consider what the sales were supposed to involve according to agreements
between the seller and customer: the vast majority were over cocaine (88 per-
cent), but a few were about heroin (8 percent) and cannabis (4 percent). If we
exclude two outliers, the amount to be traded was no more than 3 grams, as little
as two-tenths of a gram, and the average was about 1 gram; this was also the most
commonly traded weight. The most common price was €20 but the average was
€50, and the price ranged from €20 to €150.
A related but separate issue is how much weight and money was actually
involved. What a person says they will do and actually do is not always the
same13; such discrepancies that involve trade are known as fraud (see Jacques
and Wright, 2008a). Like many street-level drug markets (Jacobs, 1999), the one
in the Red Light District is thought to be a hotbed of rip-offs. Blickman, Korf,
Siegel, and Zaitch (2003: 26) suggest the dealers here “hardly ever really sell what
they offer.”

The risk of buying a pig in a poke is very real [and customers are] ripped off
sooner or later: thinking they are a purchasing heroin or cocaine, they buy
cough tablets, ground catlitter, Epsom salt, or lidocaine …. [which] is a bit
like cocaine, in that both drugs produce a slight numbness if a small amount
is rubbed against tongue or gum. But here the likeness ends.
(Grapendaal, Leuw, and Nelen, 1995: 93)

Our data suggest the majority of trades are fair (88 percent), but in three cases
(12 percent) the dealer or customer gave less than promised. In two of the cases
the dealer defrauded the customer by providing fake rather than real cocaine; the
other instance involved a customer who turned over €100 instead of the €120
that was owed. In large part, we see that exchange in the Red Light District’s
illicit drug market is small scale and usually but not always fair.
The reasons for this state of affairs are largely related to rationality and oppor-
tunity. Typically, bigger sales come with greater financial incentive per unit of
energy expended and time spent, but it is far harder to find a customer for, say, 1
kilo of cocaine than it is for 1 gram of cocaine. This suggests that the small-scale
trades characterizing the Red Light District street market under study are affected
not so much by rationality as by opportunity. It may be true that the dealers we
studied would prefer to move from small to large trades and profits, but they
cannot do so without the necessary means (e.g., money, suppliers, or customers).
In such a position, the most rational option is not a possible action and thereby
relegates the seller to the bottom division: the retail-level market.
Fraud is a rational action for dealers in so far as it increases profits by reducing
the amount of drugs provided to buyers. Based on previous research (Blickman
et al. 2003; Grapendaal et al., 1995), we might have expected fraud to occur
far more often than in approximately one-tenth of trades. Nonetheless, this is a
fairly high rate. We would be surprised, for example, to find that bars selling a
legal albeit regulated drug, namely alcohol, rip off one in ten customers.14 It has
been argued that illicit markets involve more predation than licit/regulated ones
Drug dealing: Amsterdam’s Red Light District 133
for the reason the latter group has access to law whereas the former one does not
(Black, 1983; Jacobs, 1999). When a trade is illegal, a defrauded exchange partner
is unlikely to report the victimization to police which would eventually lead to
arrest, prosecution, or punishment (Black, 1976). Thus, there is a low likelihood
of law punishing fraud among illicit drug traders and this makes them attractive
targets for predation.
That theory begs the question, “Then why are there not more rip-offs?” Formal
control is not the only way to handle wrongdoing (Black, 1998). Informal control
is another option (Cooney, 1998; 2009). This includes everything from retalia-
tion to less aggressive response such as negotiation, avoidance, and toleration
(Jacques and Wright, 2008a, 2011). Each of these forms of social control has
their own unique costs and benefits for both the controller and the controlled
(Jacques and Wright, 2011). Deciding whether to defraud a trade partner involves
weighing the monetary benefit against the cost of being informally controlled. For
example, giving a customer fake rather than real cocaine may increase a dealer’s
profits, but doing so may not be worth the risk of being robbed as retaliation for
the transgression. Or, a rip-off may risk future business if the victimized buyer
thereafter avoids the dealer. Dealers who are aware of such possibilities may
forgo the benefit of defrauding in order to avoid the costs of popular justice (see,
e.g., Jacques and Wright, 2011).

A script provides drug dealers with a plan of action for how to complete drug
sales. As alluded to above, however, the script is not always successful. Each step,
both facilitating and necessary, can fail to produce a drug sale.
For step A, a dealer may be unable to find and identify anyone who matches their
likely customer schema. Unless a customer locates and solicits the dealer, the only
outcome of this failure is for the dealer to give up the search for the time being.
Regarding step B, even when a potential customer is discovered, the solicitation
of that individual may fail. This occurs when the dealer is unable to capture the
person’s attention, unsuccessful in communicating the intent to trade, or simply
because the solicitee is uninterested in purchasing drugs (of that particular type or
from that particular seller). When this happens, the dealer will return to searching
for a customer, give up, or retry solicitation with the already identified individual
if the reason for failure was the inability to capture the person’s attention.
As with step B, the inability to communicate may also prevent potential trade
partners from agreeing on and going to a trade locale, which is step C, or from
agreeing on trade terms, which is step 1. Other factors—such as anxiety about the
suggested time and place or the exchange rate—may also deter success at this point
in the script. This type of breakdown will lead a dealer to search anew or call it quits.
The exchange of drugs for another item of wealth, typically money, is the final
step in the script. Despite agreeing on trade terms, a buyer may pull out at the last
second if they come to believe the substance is fake or “short” (i.e., less than the
agreed upon quantity). A dealer may do the same if they perceive, for instance, the
134 Scott Jacques and Wim Bernasco
buyer’s money is counterfeit. In addition, factors beyond the immediate control
of the traders may deter their exchange, such as a police officer walking by at an
inopportune time. When the exchange fails to occur, the reaction may be to rene-
gotiate a time and place to trade, the trade terms, or to simply start over the search
for an exchange partner.
But there is another, more costly, way that dealers experience failure in
exchange. Dealers’ adversaries include defrauders and police. Not all rip-
offs are equally troublesome. So long as a dealer makes some profit off of an
exchange, it might not be deemed a complete failure. For example, if a customer
pays €40 instead of the full €50 owed for a gram of coke, the exchange is typi-
cally still profitable for the seller despite being fraudulent. Yet, if the dealer is
ripped off on such a scale that the trade actually costs them money, as happens
when counterfeit money is accepted, then the exchange is no doubt a failure.
The other way a completed exchange can be a failure is when it leads a seller to
be formally sanctioned. A dealer, for instance, may be arrested, fined, impris-
oned, or otherwise punished by the law for selling to an undercover officer or
in sight of a CCTV camera. Depending on the circumstances, these costs can,
on the one hand, lead a dealer to desist or reduce the frequency at which they
act out the dealing script, or, on the other, escalate in criminal activity by work-
ing harder in the drug underworld or retaliating against the victimizer or law
enforcement officials (Jacobs and Wright, 2006; Jacques and Wright, 2008b;
Topalli, Wright, and Fornango, 2002).

Implications for prevention

The analysis above presents a general script of drug dealing and, nested within
it, a track level script of street dealing in the Red Light District of Amsterdam.
The two necessary steps are agreeing on trade terms and making an exchange.
To reach those necessary steps, dealers must often search for customers, solicit
them, agree on when and where to meet for trade, and travel there. Among the
dealers we studied, their modus operandi involves face-to-face verbal agreements
about trade terms that lead to small-scale drug–cash trades. To facilitate the ulti-
mate goal of making a sale, dealers walk or stand in the area while looking for
males, especially suspected tourists or addicts. They solicit potential customers
with whispers, nods, and glances. This act may lead to an agreement about doing
a deal in the immediate or a nearby place or time travelled to by foot. Each of
these steps in the street dealing sequence is thought to happen for rational or
opportunistic reasons.
This raises the question: How do we lower the rationality of or remove the
opportunity for each step in the script? Rather than focus on punishing crimes
that have already occurred, how might they be deterred and prevented? “Crime
scripts can help to design prevention measures by revealing potential ‘weak
spots’ where the crime-commission process could be disrupted by reducing
opportunities and resources and increasing risks associated with the commission
of crime” (Chiu et al., 2011: 359). More specifically, what are the implications
Drug dealing: Amsterdam’s Red Light District 135
of the general and track level scripts we described for reducing the frequency
of drug sales?
It should be stressed at this point that drug dealing is a consensual crime. Unlike
predatory crime, which involves a victim motivated to prevent crime and a target
that can be guarded, consensual crimes are very difficult to prevent because they
involve two parties who both are willing to enter into an illegal transaction. Thus,
the two parties directly involved are not motivated to prevent drug dealing—it is
only third parties (parents, school administrations, business owners, neighbor-
hood residents) that have stakes in preventing drug dealing.
The fact that general scripts have greater breadth at the cost of less precision
means they are useful for orienting initial thinking about crime prevention but
less valuable for creating specific intervention measures. Essentially, the general
script for drug dealing described in this chapter suggests that law enforcement
practices should aim to reduce the utility of and remove the necessary elements
inherent in each step of drug dealing. One way this is done is to more severely
penalize the necessary steps: agreeing on trade terms or making the exchange.
Less attention is paid to adjusting the rationality of facilitating steps, but this
is another possibility. For example, searching for customers or soliciting them
could be given greater penalties. Penalties are not required, however, if the
opportunity for various steps in the crime commission process are reduced. If
it is possible to take away the ability to carry out each facilitating or necessary
action, then the government could save itself and offenders from the costs of
penalizing people.
Obviously those possibilities are quite abstract and could lead to a variety of
law enforcement practices. Track level scripts allow for the abstract to be brought
down to the concrete. Because the steps of street dealing in the Red Light District
have unique traits, there are specific strategies that can be used to disrupt them.
Based on our findings, here are some possibilities:

● Dealers typically sell to and, by logical extension, search for and solicit
males, tourists, and addicts. This suggests the government could focus their
attention on pushing addicts away from this area; warn tourists of the risks
involved in making trades—including not only formal sanctioning but also
victimization; and, largely ignore interactions involving females in order to
focus on males. These tactics would make law enforcement more efficient
and, in turn, less costly.
● Whispers, nods, and glances are the most common ways of communicating to
potential customers that drugs are for sale. Undercover police officers could
make use of this information by walking or standing in the area and initiating
or reciprocating these styles of body language to see if they lead to commu-
nication about trade terms and an exchange. The chance of being solicited in
the first place would be increased by having male officers act as addicts or
tourists (e.g., by dressing as they do or speaking in English). Should dealers
learn that police are mimicking their actions and customer base then at least
some may be deterred from offending.
136 Scott Jacques and Wim Bernasco
● More intensive surveillance (in person, or CCTV) may hinder dealers from
soliciting customers and make the exchange. Surveillance is not limited to
police controls. Public space can be designed to making searching and solic-
iting customers and exchanging money for drugs more difficult. The design
would imply maximization of indivisibility (preventing hidden corners, blind
alleyways, etc.) and minimization of opportunities for loitering.

Reducing drug use and sales may be seen as meritoriously beneficial actions, but
what must be kept in mind is that law enforcement practices have costs for society
as a whole and for individual citizens—including offenders. As researchers we are
not in a position to decide what is ultimately right and wrong; justifiable or inex-
cusable. It is our duty, however, to inform governments and the public of not only
the benefits but also the costs of various law enforcement practices. For example,
the earlier suggestion of focusing on males and ignoring females is empirically
based and would be efficient, but this is a form of gender-based discrimination.
We did not suggest that undercover agents should employ the “whisper” strategy
of dealers to solicit them because this is—or at least approaches—entrapment.
Scaring dealers and users away from the Red Light District might come with
the cost of having a replacement market “pop up in another” area that is more
disruptive and difficult to control (Grapendaal et al., 1995: 85; Leuw, 1991; but
see Weisberg et al., 2006). Imprisoning drug dealers may prevent future offenses
through incapacitation, but whether the greater harm is caused by drug sales or
this formal sanction is an open question. Perhaps the legal policy with the greatest
utility would be to have no drug law at all, which might increase consumption,
but would also open avenues for taxation, regulation, and formal control—includ-
ing quality control and providing access to law to resolve conflicts. Research can
lead toward “objective” answers about these issues, but it is inevitable that what
and how we describe, count, weigh, and analyze is inevitably situated within a
particular moral context.

We thank Richard Wright for his comments on an earlier draft. The research on
which this paper is based was funded in part by a Research Grant Award from the
National Science Foundation, Division of Social and Economic Sciences, Law
and Social Sciences Program (NSF ID #: 0819090).

1 Note that drug use per se is not illegal. Also, there are exceptions for the medical
2 Possession of less than half a gram of hard drugs has a lower law enforcement priority
than the sale of hard or soft drugs (other than that done legitimately by coffee shops)
(Netherlands Ministry of Foreign Affairs, 2008). Typically no punishment is meted
out, other than the confiscation of the substance (Leuw, 1991).
Drug dealing: Amsterdam’s Red Light District 137
3 “The use of undercover agents to incite people into trafficking is not allowed by Dutch
law” (Blickman et al., 2003: 77–78), “unless the person involved was already predis-
posed to do so. The presumed predisposition of a person is based on his criminal record
and on facts from the operational reports of the undercover agent” (Silvis, 1994: 52).
4 This is probably true for other consensual crimes as well, such as prostitution.
5 Our discussion of path crosses is by no means complete. There is far more to consider,
on which subsequent research may want to focus. Potential lines of inquiry include
how dealers’ search patterns are shaped by conflict with residents, business employ-
ees/owners, or competitors.
6 Another, simpler, requirement is that they have access to these technologies; not eve-
ryone can afford such things.
7 Seven percent of the dealers did not remember or would not answer this question.
8 For practical reasons, we cannot delve too deeply into the complexities of how and
why signals are used by dealers to arrange and execute trades. Criminology would
benefit if future work devoted more explicit attention to this behavior, both among
drug dealers and other types of offenders (see, e.g., Gambetta, 2009).
9 Drug dealers do not always have drugs on their person and cannot always trade imme-
diately. This caveat opens up a series of complexities that are not dealt with here.
Future work should specify how this complication affects the script for drug dealing.
10 The first necessary step in the script, agreeing on trade terms, may come before, after,
or be intertwined with this step.
11 It could reasonably be argued that “obtaining drugs to sell” is a necessary stage as well
but we leave it aside because it is in fact a separate crime than the one that is our focus.
In effect, we assume dealers have the ability to obtain drugs for sale.
12 The rate of exchange does not need to be agreed upon in precise terms, and often
can be quite loose. For example, one end of the specificity continuum is charac-
terized by trades where the exact amount of drugs and money is agreed upon by
both the customer and seller; the other end has exchanges where the customer has
only communicated which drug they want and how much they will pay but not how
much they expect to receive. Thus the agreement process involved in making a sale
can be exacting and drawn out, sloppy and quick, or something between those two
13 Admittedly, what interviewees say they did is not necessarily always what they actually
did. It seems unlikely, however, that dealers would speak about so many (incriminat-
ing) aspects of their offending behavior and at the same time lie about a well-known
issue in their trade.
14 To be clear, what matters is not the nature of the good/service but rather the (il)legal
nature of it.

Beauregard, E., Proulx, J., Rossmo, K., Leclerc, B. and Allaire, J. (2007). Script analysis
of the hunting process of serial sex offenders. Criminal Justice and Behavior, 34,
Beauregard, E., Rossmo, D. and Proulx, J. (2007). A descriptive model of the hunting
process of serial sex offenders: A rational choice perspective. Journal of Family
Violence, 22, 449–463.
Black, D. (1976). The Behavior of Law. New York: Academic Press.
Black, D. (1983). Crime as Social Control. American Sociological Review, 48, 34–45.
Black, D. (1998). The social structure of right and wrong, rev. edn. San Diego, CA:
Academic Press.
138 Scott Jacques and Wim Bernasco
Blickman, T., Korf, D. J., Siegel, D. and Zaitch, D. (2003). Synthetic drug trafficking in
Amsterdam. in Synthetic drug trafficking in three European cities: Major trends and the
involvement of organized crime. Turin, Italy: TNI/Gruppo Abele/IECAH, pp. 19–94.
Chiu, Y., Leclerc, B. and Townsley, M. (2011). Crime script analysis of drug manufacturing
in clandestine laboratories: Implications for prevention. British Journal of Criminology,
51, 355–374.
Clarke, R. V. and Cornish, D. B. (1985). Modeling offenders’ decisions: A framework for
research and policy. In M. Tonry (ed) Crime and justice: A review of research, Vol. 6
(pp. 147–185). Chicago, IL: University of Chicago Press.
Cohen, L. and Felson, M. (1979). Social change and crime rate trends: A routine activity
approach. American Sociological Review, 44, 588–608.
Cohen, L., Kluegel, J.R. and Land, K.C. (1981). Social inequality and predatory criminal
victimization: An exposition and test of a formal theory. American Sociological Review,
46, 505–524.
Cooney, M. (1998). Warriors and peacemakers: How third parties shape violence, New
York, NY: New York University Press.
Cooney, M. (2009). Is killing wrong? A study in pure sociology. Charlottesville, VA:
University of Virginia Press.
Cornish, D. B. (1994). The procedural analysis of offending and its relevance for situational
crime prevention. In R. V. Clarke (ed) Crime Prevention Studies, Vol. 3 (pp. 151–196).
Monsey, NY: Criminal Justice Press.
Gambetta, D. (2009). Codes of the underworld: How criminals communicate, Princeton,
NJ: Princeton University Press.
Gottfredson, M. R. and Hirschi, T. (2003). Self-control and opportunity. In C. L. Britt and
M. R. Gottfredson (eds), Control theories of crime and delinquency (pp. 151–160).
New Brunswick, NJ: Transaction Publishers.
Grapendaal, M., Leuw, E. and Nelen, H. (1995). A world of opportunities: Lifestyle and
economic behavior of heroin addicts in Amsterdam, Albany, NY: State University of
New York Press.
Jacobs, B. A. (1999). Dealing crack, Boston, MA: Northeastern University Press.
Jacobs, B. A. and Wright, R. (2006). Street justice: Retaliation in the criminal underworld,
New York, NY: Cambridge University Press.
Jacques, S. and Reynald, D. (2012). The offenders’ perspective on prevention: Guarding
against victimization and law enforcement. Journal of Research in Crime and
Delinquency, 49, 269–294.
Jacques, S. and Wright, R. (2008a). The relevance of peace to studies of drug market
violence. Criminology, 46, 221–253.
Jacques, S. and Wright, R. (2008b). The victimization–termination link. Criminology, 46:
Jacques, S. and Wright, R. (2011). Informal control and illicit drug trade. Criminology,
49, 726–765.
Kraan, D. J. (1994). An economic view on Dutch drugs policy. In E. Leuw and I. H.
Marshall (eds) Between prohibition and legalization: The Dutch experiment in drug
policy, Amsterdam/New York: Kugler Publications.
Leclerc, B., Wortley, R. and Smallbone, S. (2011). Getting into the script of adult child
sex offenders and mapping out situational prevention measures. Journal of Research in
Crime and Delinquency, 48, 209–237.
Leuw, E. (1991). Drugs and drug policy in the Netherlands. In M. Tonry (ed) Crime and
Justice, Vol. 14 (pp. 229–276). Chicago, IL: University of Chicago Press.
Drug dealing: Amsterdam’s Red Light District 139
Netherlands Ministry of Foreign Affairs. (2008). Q & A Drugs. The Hague, NL: Ministry
of Foreign Affairs.
Netherlands Ministry of Health, Welfare, and Sport. (2003). Drug policy in the Netherlands:
Basic principles and enforcement in practice, The Hague, NL: VWS.
Nisbett, R. E. and Ross, L. (1980). Human inference: Strategies and shortcomings of social
judgment, Englewood Cliffs, NJ: Prentice-Hall.
Silvis, J. (1994). Enforcing drug laws in the Netherlands. In E. Leuw and I. Haen Marshall
(eds) Between prohibition and legalization: The Dutch experiment in drug policy
(pp. 41–58). Amsterdam/New York: Kugler Publications.
Topalli, V., Wright, R. and Fornango R. (2002). Drug Dealers, Robbery and Retaliation:
Vulnerability, Deterrence, and the Contagion of Violence. British Journal of
Criminology, 42, 337–351.
Trimbos Institute (2009). Netherlands national drug monitor: NDM annual report 2009,
Utrecht, NL: Trimbos Institute.
van Gemert, F. and Verbraeck, H. (1994). Snacks, sex, and smack: The ecology of drug
trade in the inner city of Amsterdam. In E. Leuw and I. H. Marshall (eds) Between
prohibition and legalization: The Dutch experiment in drug policy (pp. 145–467).
Amsterdam/New York: Kugler.
Verdurmen, J. E. E., Ketelaars, A. P. M. and van Laar, M. W. (2004). The Netherlands
national drug monitor. Fact sheet drug policy, Utrecht, NL: Trimbos Institute.
von Lampe, K. (2010). Preventing organized crime: The case of contraband cigarettes. In
K. Bullock, R. V. G. Clarke and N. Tilley (eds) Situational prevention of organized
crimes (pp. 35–57). Cullompton, UK: Willan Publishing.
Weisburd, D., Wyckoff, L. A., Ready, J., Eck, J. E., Hinkle, J. C. and Gajewski, F. (2006).
Does crime just move around the corner? A controlled study of spatial displacement
and diffusion of crime control benefits. Criminology, 44, 549–592.
8 Human trafficking for sexual
exploitation in Italy
Ernesto U. Savona, Luca Giommoni and
Marina Mancuso

This chapter focuses on human trafficking for sexual exploitation in Italy. Two
Italian case studies are analyzed, applying the crime script approach. They have
been provided by state prosecutors of two Italian provinces (Campobasso and
Ancona) and are referred retrospectively to as indoor and outdoor prostitution.
Starting from the literature on this topic, the different stages of this crime (recruit-
ment, transportation, exploitation and aftermath) are identified.
From the analysis of the two arrest warrants, information on the actions and
decisions of both traffickers and victims is linked to the stages of the crime. The
added value of this work is the application of a methodological tool (script) not
often used to examine human trafficking. This allows a concrete analysis of the
vulnerabilities associated with the crime-commission process and the search for
situational crime prevention measures that might target these vulnerabilities. Both
research and policy implications are drawn at the end of the chapter.

Human trafficking is often associated with a new form of slavery because it
implies the exploitation of victims and the systematic violation of their human
rights (Adepoju 2005: 77; Savona and Stefanizzi 2007: 5). Because of its serious-
ness, in the last decades this crime has received great attention by international
authorities and by the research and academic world with the aim of better under-
standing its features and its dynamics.
A definition of this global phenomenon is contained in the “Protocol to pre-
vent, suppress and punish trafficking in persons, especially women and children,”
a document drafted by the United Nations in 2000 that supplements the United
Nations Convention against transnational organized crime. According to Article
3 of the Protocol:

[Human trafficking] shall mean the recruitment, transportation, transfer, har-

boring or receipt of persons, by means of the threat or use of force or other
form of coercion, of abduction, of fraud, of deception, of the abuse of power
or of a position of vulnerability or of giving or receiving of payments or ben-
efits to achieve the consent of a person having control over another person,
for the purpose of exploitation.
Human trafficking for sexual exploitation 141
The Article also specifies that “exploitation shall include, as minimum, the
exploitation of the prostitution of others or other forms of sexual exploitation,
forced labour or services, slavery or practices similar to slavery, servitude or the
removal of organs”. The main economies in which the highest incidence of this
crime has been registered are the conventional market economies (people traf-
ficked working in bars, restaurants, construction companies, factories and so on),
the legitimate domestic service economy (people used as carers and domestic
servants) and finally the criminal economies of the sex industry (both indoor
and outdoor prostitution) (Aronowitz 2001: 172). Following the above defini-
tion, human trafficking is a crime that involves different phases: recruitment,
transportation and exploitation of the victims. These phases can be analyzed as a
criminal process1 carried out without the informed and complete consent of the
people involved as victims2 (Aronowitz 2009). This process involves many peo-
ple: on the one hand the traffickers, and on the other the victims. The former are
goal-oriented individuals able to identify specific market demands and provide
a corresponding supply of people that can satisfy them. They are more or less
organized according to the extent of the traffic: the more victims are moved from
one area to another, the more organized the criminal group involved will be and
vice versa. In many cases criminal networks of people connected to each other
are the main structure identified3 (Iselin 2003; Picarelli 2009; Vermeulen, Van
Damme and De Bondt 2010). The latter group involved in trafficking, the victims,
are associated with high levels of risk: quite often they live in poor countries,
have no job opportunities and a low level of education that make them vulnerable
and hence easily controllable by traffickers (Limanowska 2002; Okojie, Okojie,
Eghafona, Vincent-Osaghae and Kalu 2003). Once transferred and entered in the
exploitation chain they suffer many physical and psychological problems. They
are employed without any assistance and protection and are considered only as
property by their exploiters (Beeks and Amir 2006).
The abuse of victims happens mostly with sex trafficking, which is “the best
known type of exploitation in Europe” (Van Liemt 2004: 20). Many studies have
focused their attention on the general analysis of the phenomenon while oth-
ers on its specific features associated with specific countries (Lebov 2010; Liu
2011; Van Duyne and Spencer 2011). In both cases the main aspects considered
have been the phases required to commit the crime (Aronowitz 2009; Transcrime
2003; Vermeulen et al. 2010), the traffickers and the organizations involved in
this crime (David and Monzini 2000; Picarelli 2009; UNODC 2009), the vic-
tims and the problems associated with them (Limanowska 2002; Okojie et al.
2003), the routes used to carry out the illicit transportation of the victims (IOM
2005; Ministry of the Interior 2007; Motta 2008, 2010; Shelley 2010; Transcrime
2010; Unicri 2005), the causes and the consequences of this crime (Shelley 2010;
US Department of State 2005; Zimmerman 2003) and finally the strategies to
combat this crime (Aronowitz 2009; Bosco, Di Cortemiglia, Serojitdinov 2009;
Friesendorf 2007).
Despite this wide range of studies, there remain significant gaps in our knowl-
edge around precise details of human trafficking (IOM 2009). Data collected
142 E. U. Savona, L. Giommoni, M. Mancuso
are quite generic. The main data sources at a global level are produced by inter-
national organizations that, using different points of view, give an idea of the
diffusion of this crime, the gender of the victims, their nationality and the kind
of exploitation most detected.4 For example, according to the US Department of
State (2008), about 800,000 people are trafficked every year across international
borders while according to the UNODC 70,000 victims are yearly trafficked in
Europe for sexual exploitation (UNODC 2010: 16). These are mainly estimates,
in some cases associated with illegal migration in general or with smuggling
of migrants (Laczko and Gramegna 2003: 181). Also, when data or estimates
focused on human trafficking have been produced, they present some limits con-
nected to the nature of the crime.
The first limit is the “dark figure” of under-reported crime. The fear of retal-
iation for collaborating with law enforcement authorities and the fear of being
considered an illegal migrant by these authorities produce low reporting levels
(IOM 2005: 234). When the victims are included in the data it is because they have
been registered by law enforcement and/or they have come in contact with social
services and NGOs to receive assistance and protection (Kleemans 2011: 1).
The second limit relates to data comparability. Data collected by different
agencies are not comparable even if they have been collected by agencies operat-
ing in the same country (in many countries a central data collection agency does
not exist) (IOM 2009). This is caused by the different methodologies, languages
and goals of the agencies.5 A low level of collaboration and a corresponding inad-
equate sharing of information among them also adds to the problem (Bosco et al.
2009: 37; Salt 2000: 37).
The lack of precise and comparable secondary data (macro) on the phenom-
enon makes the script approach (micro) proposed in this chapter more valuable. It
could help the focusing on different opportunities/vulnerabilities and the finding
of situational prevention remedies. This chapter is an example of how macro and
micro analyses could be combined for a better understanding of the phenomenon
and finding more effective solutions against it.

Human trafficking and crime script

Human trafficking is a continuous process involving different and multiple deci-
sions taken by a motivated and rational offender who weighs the costs, benefits
and efforts required to accomplish the crime (Cornish and Clarke 2008).
Both the rational choice perspective and human trafficking have their roots in
the economic field: the former translated the homo economicus concept into the
reasoning criminal concept (Cornish and Clarke 1986), whereas the latter con-
siders economic factors (e.g. unemployment, underemployment and economic
difficulties) as the main push factors causing crime (Aronowitz 2001 and 2009;
Bosco et al. 2009; Van Liemt 2004). Consistent with the rational choice per-
spective, considering trafficking from an economic point of view emphasizes the
high level of rationality of the traffickers involved. This crime seems to be ruled
by free market laws: poor countries export (both legally and illegally) abundant
Human trafficking for sexual exploitation 143
unemployed resources and rich countries import and employ them in specific
market sectors in which a high unsatisfied demand is registered (Shelley 2010).
In order to understand the rationality behind any crime, Cornish (1994) pro-
posed the concept of crime script. Thanks to this concept, borrowed from the
cognitive sciences (Schank and Abelson 1977), it is possible to “identify every
stage of the crime-commission process, the decisions and actions that must be
taken at each stage, and the resources required for effective action to each step”
(Cornish and Clarke 2008: 31). According to this definition, crime is an action
that occurs at a specific time and place but it is only one event within the com-
plete crime commission process. Before and after the criminal act there are other
events and decisions vital to the crime (Cornish 1994). The analytical framework
offered by the concept of crime scripts is a useful tool to organize the knowledge
of a complex and structured behavioral sequence. Breaking down the complete
crime process into scenes, and identifying actions, decisions and alternatives, cast
(actors involved in the scene), responsible agencies and legislation involved in
each scene, it is possible to identify the details associated with this crime.
An in-depth analysis applying the rational choice perspective and script anal-
ysis can help reveal both complexity and flexibility of the crime-commission
process. The complexity is shown by dividing the crime into a series of steps and
identifying the prerequisites and the facilitators of the single actions. To represent
the flexibility in the crime commission (a concept that emerged especially in the
study carried out by Lacoste and Tremblay 2003), Cornish adopted a series of
cubes strung out along an axis. Each cube corresponds to a stage and each cube
face corresponds to possible ways to carry out a stage. For example, a crime may
be committed as a sequence of 3 × 3 × 3 × 3 (4 stages with 3 possible ways to
carry out each stage) or by a sequence of 4 × 4 × 4 (3 stages and 4 ways to play
each stage) (Tremblay, Talon and Hurley 2001: 561). Examining the stages of a
crime, ways to play stages, actors and actions in each “cube face” (stage) leads to
the identification of vulnerabilities in the crime-commission process and possible
adaptations of the offender.
Since the introduction of the crime script concept in 1994, several authors
have applied this scheme to different criminal phenomena, including complex
and organized crimes such as: infiltration of mafia organizations in the public
construction industry (Savona 2010), stolen vehicles (Morselli and Roy 2008;
Tremblay 2001), illegal waste smuggling (Tompson and Chainey 2011), hunting
pattern of serial sex offenders (Beauregard, Proulx, Rossmo, Leclerc and Allaire
2007), employee computer crime (Willison and Siponen 2009), organized crime
(Hancock and Laycock 2010), drug manufacturing (Chiu, Leclerc and Townsley
2011), and child sex offenders (Leclerc, Wortley and Smallbone 2011). The popu-
larity of this analytic method can be explained in a number of ways: it can be
easily applied by all practitioners without requiring any form of training, particu-
lar software or specialization; it helps in structuring and deepening knowledge
and developing an organizational memory; it can provide a conceptual framework
in cases where there are limited data; it permits a bottom-up and a top-down visu-
alization of the crime commission process; it may facilitate the identification of
144 E. U. Savona, L. Giommoni, M. Mancuso
opportunities for detection and prevention; and it may help in the individuation of
flexible “spots” in the crime-commission process that enables quicker adaptation
to law enforcement responses.
It is proposed in this chapter that crime scripts can be successfully applied also
to the study of human trafficking. Until now only one study has applied crime
script to human trafficking.6 This is surprising, given the shortage of quantitative
data and focused analysis. Script analysis can provide the “zoom-in” necessary to
develop a micro-analysis of this phenomenon, highlighting its complex dynamics
and identifying vulnerabilities and opportunities exploited by traffickers. This is
what this chapter attempts to do. In the following section the sources of data and
the methodology adopted to carry out the analysis are described.

Application of the script analysis to trafficking in human

beings for sexual exploitation: data sources and methodology
A crime script analysis must fulfill two conditions: first, it should be crime spe-
cific and should not encompass the general offense category (i.e. not car theft, but
car theft for resale, joyriding, etc.); second, it should analyze in depth the crime
commission process providing an insight into it. To accomplish these conditions,
suitable and proper data sources are needed. Reviewing the literature on crime
script, the traditional data sources adopted are:

● in-depth interviews with offender or victim of the offense;

● CCTV footage;
● police investigations (wiretappings, warrants of arrest, judicial decisions).

The sources of data used in this chapter are two arrest warrants associated with
Italian sex trafficking cases. This judicial material has been chosen because it
contains a lot of information about the dynamics of the crime and the different
phases to commit it. The two arrest warrants describe in detail all the passages and
all the actions carried out by the traffickers and the victims in order to success-
fully achieve the illicit goal. Thanks to these documents it is possible to obtain
a relatively objective reconstruction of events, minimizing the distortions that
might be contained in interviews of the criminals or victims involved. The wire-
tappings reported give also an idea of the rules and the decisions taken from the
beginning to the end of the criminal act. The arrest warrants considered refer to
two recent judicial cases:

● Case study 1 deals with trafficking in human beings for sexual exploitation
carried out in night clubs involving victims from Eastern Europe (especially
Poland and Romania). This case was indicted by the Court of Campobasso
(Molise region) in 2010.
● Case study 2 deals with trafficking in human beings for sexual exploita-
tion carried out on the street involving victims from Nigeria. This case was
indicted by the Court of Ancona (Marche region) in 2009.
Human trafficking for sexual exploitation 145
Both arrest warrants consist of approximately 500 pages describing the organiza-
tions in detail and containing wiretap transcriptions, witnesses’ examination and
other police activities. These cases have been considered because they refer to
two main sex trafficking entrance flows in Italy: from Eastern Europe and from
Nigeria (Transcrime 2010).
Despite being both focused on sex trafficking, the two cases are different in a
number of ways. They show two different modalities of sex trafficking developed
in Italy and elsewhere. In the first case trafficking is carried out in nightclubs,
so the focus is on indoor prostitution. In the second case trafficking is carried
out on the street, so it refers to outdoor prostitution. Secondly, the nationalities
of the victims and of the traffickers are different: in one case they are people
from Eastern Europe and the offenders are principally Italians having contacts
with Polish and Romanian brokers; while in the other case both victims and traf-
fickers are Nigerian. A major female offender involvement is registered in this
second case, given the high relevance of the Nigerian “madams,” women who, as
those described in the paper written by Zhang, Chin and Miller (2007) on female
participation in the Chinese criminal associations involved in human smuggling,
manage the entire human trafficking process thanks to their wide social and rela-
tional resources, to the reduced use of violence that characterizes this crime and
to their gender affinity with the victims (Ministry of the Interior 2007). Finally
the cases involve a different number of offenders (respectively 17 and 67). These
differences can lead to diverse ways of committing the crime.
The two cases have been analyzed using the script analysis proposed by
Cornish (1994). More labels, not directly connected to those suggested by
Cornish, have been selected using a particular graphic visualization that adds
value to the understanding of the analysis. This will help to understand the
continuous flow of actions that characterize crime script approach: when one
finishes, another starts. This approach avoids the risk of representing the crime-
commission process in a rigid and static way and the offenders as passive actors
not directly connected to specific decisions and choices. In order to overcome
these limits and to highlight the rationality of the subjects involved, the solution
proposed here is to show all the activities committed, the decisions taken by
offenders and the environmental factors influencing these decisions (opportuni-
ties). This means that not only the track is described (single cubes concatenated
to represent the minimum detail level of the crime process), but also the differ-
ent ways used to carry out each scene. This aspect is important because in some
cases, facilitators and impeding factors can influence the normal sequence of
actions requiring fewer activities to be carried out in the same scene or forcing
the offenders to implement additional actions to perpetrate crime. The steps con-
sidered by this analysis are the following:

● identification of the stage process (what was defined as the cubes and that
hereafter will be defined as stages);
● drawing for each stage the cast, activities, decisions, alternatives, comprising
a complete process;
146 E. U. Savona, L. Giommoni, M. Mancuso
● determination of opportunities and vulnerabilities within the crime commis-
sion process exploited by the offenders;
● application of situational crime prevention techniques aiming at reducing the
above mentioned vulnerabilities.

The first step is to identify the stages constituting the crime. As already mentioned,
there are no settled terms even if most authors (Beauregard et al. 2007; Hancock
and Laycock 2010; Morselli and Roy 2008; Savona 2009; Tremblay et al. 2001;
Willison and Siponen 2009) adopted the subdivision proposed by Cornish (1994)
and derived from cognitive sciences. This division usually entails the following
stages: preparation, entry, pre-condition, instrumental pre-condition, instrumental
initiation, instrumental actualization, doing, post-condition and exit. This subdivi-
sion has been adapted case by case to different crime phenomena analyzed fixing
the stages to the crime commission flow.
Recently some authors (Brayley, Cockbain and Laycock 2011; Tompson and
Chainey 2011) have argued that following the cognitive sciences terminology
could be misleading, and have proposed instead an ad hoc subdivision that does
not take into account the items advanced by Cornish. This new proposal does not
alter the script process but redistributes the action flow within the script with a
terminology focused on the problem considered. Thus Brayley et al. (2011), for
internal child sex trafficking, divided the crime commission process into find,
groom and abuse. Tompson and Chainey (2011), for the illegal waste smuggling,
divided the process into creation, storage, collection, treatment, transport and dis-
posal. This review does not affect the meaning of script analysis but just proposes
a different way to organize the knowledge. In this chapter existing terms describ-
ing the different stages of the trafficking in human beings for sexual exploitation
have been used. These are: recruitment, transportation and exploitation. To these
three stages a fourth stage has been added, the so-called aftermath. This stage
considers the actions taken after the crime commission that, as Cornish writes,
“may include further crimes and their sequel” (Cornish 1994: 155). Both case
studies will be divided according to these stages.

Trafficking in human beings for sexual exploitation in Italy

In this section the two crime scripts are presented and represented in a flow chart.
This approach highlights the conditions allowing crime to be committed, the alter-
natives, and the spots where the offenders can readapt quicker. This means a deep
focus on offender involvement in the crime commission process, as main actor,
but also the role played by the others involved in the scenes, as supporting actors
or counterparts. Unlike other authors (Brayley et al. 2011), we also considered the
actions undertaken by the victims. Although in the majority of cases these are not
criminal actions, they are critical to the crime commission. Without the decisions
of victims (even if not fully informed) crime would not take place.
Human trafficking for sexual exploitation 147
Case study 1: Sex trafficking from Eastern Europe to Italy
The script analysis referring to the first case study7 is represented as a flow of
concatenated actions producing the different phases of the crime (see Figure 8.1
and Table 8.1). As emerges from the script represented in Figure 1.1, crime starts
with the recruitment. This phase is principally carried out in the source countries
(Poland and Romania) and coordinated from Italy by the group leaders. The head
of the group gets into contact with a broker (already known to him/her) acting in
the country of origin and asks him to advertise a job in local newspapers and on
the Internet. These advertisements are for young women to be employed as wait-
resses, barmaids or hostesses in Italy. The job is presented as completely legal and
as offering a good income (about 1,300/1,500 euros per month). Since the offer
seems promising, many women respond to this advertisement and get in con-
tact with the broker. The first request from the broker is to send him an updated
curriculum vitae together with some photos. In this way a first selection of the
possible victims to be trafficked is made: only good looking women are selected.
To these girls, the broker gives some indications on the job, reassuring them about
the legality of the activities and specifying that they will not be involved in pros-
titution once they arrive in Italy. These explanations succeed in convincing some
girls to leave. Others, reluctant to depart, are contacted by one of the leaders (the
boss or his mistress) or, in some cases, another girl of the same nationality already
involved in the job. They give further explanations and reassurances to potential
victims. The recruitment is based on the deception of the victims, who receive
fake information about their future job. According to the indications obtained
and the need to have a job, the victims decide whether to leave or not to leave. In
the first case, the transportation phase starts, while in the second case the process
finishes here.
The transportation stage consists principally of the journey from the country
of origin to the country of destination. The journey may be carried out by car, bus
or plane. The choice of transport depends on the event considered. In general,
the traffickers organize the trip in detail, trying to have victims leave as early as
possible in order to minimize the chances of them changing their minds. A car
is chosen when accomplices have to go to Italy for other reasons and can offer
them a lift. When the transport is by bus or plane the victims travel alone follow-
ing the orders of the traffickers.8 The cost of the trip is borne by the women, who
are reassured that they will quickly recoup their money once they arrived at their
Once the victims arrive in Italy, one of the group leaders (the boss, his mistress
or his brother) meets the women at the bus stop or at the airport to take them to
the boss’s house. During the transport toward the apartment, some police routine
controls might take place. This does not constitute a problem for the offenders
because the victims are from EU countries and therefore they have the right to
stay in Italy for three months without any particular document.
At this point, the third stage starts. From the first evening the women are
involved in the new job. They are told to put on scanty clothes and high-heeled
148 E. U. Savona, L. Giommoni, M. Mancuso
shoes. A man (the barman of the night club) brings the women to the night club.
Here they learn that the club is an entertainment club and they are figuranti di
sala (girls that have to entertain the clients). They also receive information on the
job and some rules of conduct to be respected. At first these rules seem to refer
to legal activities: the women should spend time with clients having a drink with
them and entertaining them without carrying out any sexual behavior. To enforce
this position, the women are invited to inform the boss or his brother if anybody
tries to touch them in an inappropriate manner. In return they receive a fixed pay
from 50 to 70 euros per day with free accommodation and transport. On the sur-
face, clients entering this club can spend an evening together with some girls and
can hobnob with one of them, chatting and dancing with her in common spaces
or in a privé. But despite this reassuring information, the girls’ freedom of move-
ment is limited. They are prevented from leaving the club or the house unescorted
and they are permitted just a few moments in the toilet.9
After a few nights’ work, respecting the rules and receiving the corresponding
fixed wage, the girls understand what the real nature of the job is. In order to have
a high number of clients, they have to excite them by providing sexual services,
induce them to pay for many drinks and try to seduce them so that they ask the
boss or his brother about the possibility of having a serata with the girls outside
the club.10
The system develops through mechanisms of incentives/disincentives involv-
ing the girl together with her boss and club manager: girls can move to prostitution
by force or by choice because they are aware that if they do not accept the advances
made by the clients in the privé and if they do not spend a serata with them, their
wages are reduced. In the club the clients pay to obtain the physical company
of a girl and the cost of the service offered changes according to the time spent
together: if she is not able to attract the client and to satisfy his requests, he stops
drinking and consequently stops giving money to the club manager. Also, if she
does not play pretty, the client does not invite her for a serata, thus producing less
profit for the boss. This causes a corresponding reduction in the money designated
for the girl (her salary becomes directly proportional to the number of drinks her
clients have). In order to receive the payment for their work, these women have to
undergo this exploiting mechanism. The constriction–violence process is oriented
towards those girls who do not accept the rules imposed on them by the boss: from
being friendly and solicitous, the bosses become strict and exacting. In particular,
they withhold the victims’ documents (officially required to draw up the employ-
ment contract), add rental and transport expenses initially not charged, and so on.
A delicate psychological violence is carried out with a progressive reduction of
the material benefits.
In this situation, some women decide to escape and return to their country of
origin, while others get involved in prostitution.11 The latter victims can be sold
to other exploiters, obtain some privileges12 or become the boss or his brother’s
mistress, thus joining the criminal group and gaining a promotion. They pass
from being prostitutes to bar girls, or collaborators in Italy or in their country
of origin.
Recruitment Transportation Exploitation Aftermath

Group Phone
leaders contact Journey
traffickers arrest

Broker Fake
Phone Yes information
contact Bus Car Plane
No sex No sex No payment of Known/trusted
Deceptive Girls Forced into at home nightclub the serata at the clients
prostitution nightclub
Italy Control
Phone (Campobasso)
Prostitutes contact Legend
reduction Sold
Action performed

Increase Process ending

costs Privilages
Multiple action
No Visa
retirement Decision
Join group
Flow line
Eventual flow line

Figure 8.1 Crime script case study 1: from Eastern Europe to Italy.
Source: Authors’ elaboration of the information contained in the arrest warrant referred to case study 1.
150 E. U. Savona, L. Giommoni, M. Mancuso
In order to ensure the trafficking goes smoothly and to avoid being identified and
arrested by the law enforcement agencies, the leaders of the group adopt several
strategies (the “aftermath” stage of the script). They do not allow clients to have
sex with the girls inside the club in order to avoid the discovery of illegal practices
during possible police controls. Similarly they do not permit the women to have
sex in the house in which they live, since it is owned by the boss and might link
him to illegal activities committed on the premises.13 A third strategy is to prohibit
the payment of the serata inside the nightclub, because the owner of the club
might be accused of pimping: in general the client meets the girl outside the club
and then pays her for the service she will provide. The girl only passes the money
back to the boss on the second occasion she is with the client. This last measure
is to give the boss or his brother the opportunity to check out the client and to
ensure that he is not a plain clothes policeman. These strategies demonstrate how
traffickers take into account all the risks connected with trafficking girls and take
prudential measures for minimizing them.

Case study 2: Sex trafficking from Nigeria to Italy

The second case considered has also been divided into four main stages (see
Figure 8.2). While the first case involved indoor prostitution, the second case refers
to outdoor prostitution. The different phases occur between Nigeria and Italy.
In this case a pivotal role is played by the madam, a Nigerian woman who is the
leader of the criminal group involved in this crime. According to her requests, the
members of the criminal group acting in Nigeria recruit young and good-looking
girls to be transported and exploited in Italy. They focus their attention on people
with a low level of education and who are part of big and poor families. The first
contact happens between these people, the girl and her family. The recruiters offer
a regular job in Europe and offer to pay for their travel expenses (40,000 to 60,000
euros according to the place of origin, the economic possibilities of the family,
and the cost of the trip), which the girls will have to reimburse to the madam using
part of the wages earned. These conditions are in general accepted by the girl and
by her family who hope thereby to solve their economic problems. A contract of
sorts is agreed upon by the parties. Then, the girl has to participate in a voodoo rit-
ual in which she has to swear she will respect the contract. This ritual is celebrated
by a magician who intimidates the victim by warning of future misfortunes and
misery if she walks away from her responsibilities. This ritual is very important
for Nigerians: they believe in the magic power of the magician, and therefore they
are willing to do what the madam orders in order to avoid negative consequences
for the victim and her family. In this case the soliciting of new girls is carried out
by creating a debt that binds the victims to the madam until it is fully repaid. This
link is cemented by the celebration of the ritual.
After the oath, the transportation stage starts. The victims are forced to
move into some houses owned by the criminal group for the period necessary
to organize the trip and to obtain fake documents. This a delicate phase since it
involves the crossing of European borders. The success of this trip depends both
Recruitment Transportation Exploitation Aftermath

Madam Broker Joint Girls are sold Girls

to other idendification by
No groups law enforecment
members in Yes
documents Avoid
Nigeria Paid by girls
Forced into
Girls Oath/wodoo Journey
Material Legal Threat/
assistance assistance intimidation
Contract/debt Magician Violoence
Threats Violence against
the family
Lagos Kano
No No
Prostitution No
The Netherlands Turkey

Intermediate Greece


Figure 8.2 Crime script case study 2: from Nigeria to Italy.

Source: Authors’ elaboration of the information contained in the arrest warrant referred to case study 2.
152 E. U. Savona, L. Giommoni, M. Mancuso
on the strategies adopted by the madams (i.e. the routes of the travel and the
means of transport chosen), and on the national legislations of the transit coun-
tries, which generally criminalize this activity. The arrest warrant records that
some victims are transferred by air, stopping over in several European coun-
tries, and others arrive in Italy by land or sea. For air travel, the most common
transfer method, the victims leave from the airport of Lagos or Kano using fake
documents. If there is a problem passing through immigration controls, the traf-
fickers bribe the immigration officers. In general the first destination is Turkey
or the Netherlands.
The Netherlands is the best point of entrance in Europe thanks to its legisla-
tion concerning the protection of human trafficking victims. According to the
Dutch legislation, minors who declare being a victim of this crime can receive
temporary residence documentation and can be accommodated in a specific reha-
bilitation center where they have also the possibility of using a telephone. Once
a visa is obtained, and consequently the possibility to continue the trip towards
Italy, the victims get in contact with the local representative of the group (of
whom they have the telephone number) who gives indications on how to escape
from the center. Then, the victims meet this representative who arranges the
travel towards the next destination of the trip. He is responsible for the victims
from their entrance in his country to their departure.14 The trip is very long (it can
last weeks) with many intermediate stops before arriving in Italy; during these
stops the victims can spend one or more days under the protection of people
acting as local representatives in the respective country. The choice of the stops
depends on the contacts activated in the different areas and on the relationships
with the madam.
The second route is through Turkey: in this country a very active cell has
been established and has good success in corrupting officials and facilitating
the entrance of the victims. The victims are transferred from Turkey to Greece
and then on to Italy. When the victims reach their destination—in this case, in
Ancona, Macerata and Ascoli Piceno—the exploitation stage begins. The girls
are entrusted to the madam who acquired them. She pays the travel expenses
to the traffickers and takes the girls’ passports in order to give them to the traf-
fickers. With no documents, work or knowledge of the Italian language, the
victims are completely dependent on the madams, who give them accommoda-
tion and manage all logistical issues connected to their activities (e.g. finding
phone cards). The debt payment binds the victims to the madams, who consider
them as property.
In this case, an important precondition connected to outdoor sexual exploita-
tion is the “joint”. This term refers to the place (footpath or stretch of road) where
the prostitute works: those who do not have the possibility of using a joint can-
not stay in that place and cannot work according to the code of conduct existing
among the criminals involved in outdoor prostitution. Since prostitution is the
source of the money required by the girls to compensate the madam, if there are
no joints available the victims have two alternatives: they can be sold to another
madam who can engage them immediately in prostitution or, to use the joint, they
Human trafficking for sexual exploitation 153
have to pay rent for it. The cost of the joint, together with the cost of board and
lodging, is added to the initial debt.
When the joint is available, threats and violence against the girls or their fami-
lies are made to force them to prostitute themselves. The increase of the debt and
the commitment taken with the voodoo ritual are also two important factors con-
tributing to enforce the slavery status of the victims.15 Continuous and pervading
control is exercised over the victims: the exploiters supervise the time dedicated
to work as well as spare time, the freedom of movement and access to external
Some victims become exploiters themselves in order to use their earnings
to pay the debt more quickly: the madams give them the possibility of recruit-
ing a girl (generally a sister or a friend) to be exploited in prostitution so that
they can obtain the money in a shorter time. This is an important feature of the
modalities used to carry out the exploitation because in this way the victims
are directly involved in the criminal activities and less disposed to denounce
the crime to law enforcement authorities. Since outdoor prostitution is more
obvious than indoor prostitution, the arrest warrant contains many examples
of victims identified by police during routine controls and held in Centers of
Temporary Permanence. In order to avoid victims’ deportation or prosecution,
madams offer legal and material assistance to the identified girls. They also use
threats and intimidation: the fear of suffering retaliation, and the dependency
and the bond with the exploiter, as a consequence of the voodoo ritual, can
discourage the girls from collaborating with the authorities and force them to
provide fake personal details. The girls reassure the madam of their faithfulness
and their respect for the conspiracy of silence, telling her they wish to escape
from the center and return to work in order to pay their debt. Only a few girls
report to the police their status of victims and even fewer are able to exit from
the illegal exploitation.

Vulnerabilities/opportunities and possible situational crime

prevention techniques
The cornerstone of situational crime prevention and the main idea introduced
by script analysis is the shift of attention from “remote” causes of crime (moti-
vation) to “near situational causes of crime” (motive) (Clarke 2008: 181–182).
Whereas changing so-called “root causes” of crime (such as poverty, parental
love, psychological problems etc.) may produce results in the long term, inter-
ventions focused on immediate factors of crime, namely opportunities, can bring
immediate benefits and results (Ibid.). Despite this, most of the studies on human
trafficking focus on long-term factors that make victims more vulnerable to such
crime (namely, poverty, underdevelopment, etc.) and propose the improvement
of economic opportunities as the main solution. We are not denying that such
a response may help reduce human trafficking in the long term, but the meth-
odological procedure adopted in this chapter shows what it is like to wear the
offender’s shoes and to thereby understand the modus operandi of the crime
154 E. U. Savona, L. Giommoni, M. Mancuso
(Ekblom 1995: 128), identifying the immediate vulnerabilities/opportunities
behind the offender’s decisions. Breaking down the complexity of the complete
crime commission process through script analysis shows up the weak points in
current attempts to control trafficking which are exploited by traffickers. The next
step here is to make more difficult the crime commission by acting upon the
benefits and costs associated with offending. This involves the application of the
25 situational crime prevention techniques. These techniques are grouped in five
categories which are: increase in the efforts, increase in the risks, reduction of
the rewards, reduction of provocations and removal of excuses. Five situational
prevention techniques correspond to each category and, as the entire chapter, they
focus on immediate responses to reduce crime. Although this classification can
be useful to understand the aim of each situational measure (each category cor-
responds to the aim of the measure), in practice these may often overlap and work
for more purposes (Clarke and Eck 2003).
As has already been discussed, even if the analysis is more focused on the
crime commission from the offender point of view, the role of victims and of
their decisions is crucial for the crime to be accomplished. Thus, situational crime
prevention techniques may also be applied to victims, with the aim of increasing
the risks and reducing rewards associated with their participation in trafficking.
This approach is not designed to blame victims, but it aims at explaining that
crime is, as defined by Felson (2008: 75), “a kind of social chemistry” where the
simultaneous presence of offenders and victims (and of their decisions) in a given
time/place and under certain circumstances is needed for the crime commission.
For these reasons situational crime prevention techniques are suitable for victims
as well.
Some possible and useful situational crime prevention techniques aimed at
reducing trafficking in human beings are presented below. These possible solu-
tions, based on the two scripts discussed in this chapter, can be extended to other
cases of human trafficking. The suggested strategies follow the four different
script phases: recruitment, transportation, exploitation, aftermath. The situational
crime prevention measures proposed in this section are not intended to be exhaus-
tive but are illustrative of the sorts of strategies that might be applied in these and
similar cases.

Preventing recruitment
The involvement of women in sex trafficking is associated with many risk fac-
tors. Some refer to the personal conditions experienced by these people in the
country of origin, while others are connected to their socio-economic and political
background. As concerns their personal conditions, the literature underlines that
women are more vulnerable and they can be easily manipulated because they are
less able to protect themselves and to claim their rights. In addition to this, they
are generally not as educated as men and this results in high rates of unemploy-
ment forcing them to look for job opportunities abroad. Besides these personal
conditions, the environment in which they grow up plays an important role. These
Human trafficking for sexual exploitation 155
countries are characterized by poverty, job discrimination and underdevelopment
(Limanowska 2002; Okojie et al. 2003; Wijers and Lap-Chew 1999).
The application of preventive measures to human trafficking at the recruit-
ment stage means acting upon the supply of services provided. The measures
should aim to reduce the push factors that lead young girls to leave their own
country by giving a more accurate picture of risks and benefits: limited free-
dom, being abused, mistreated, beaten and threatened. Raising consciousness
through campaigns about the risks and dangers of such overseas trips is the first
step toward reducing the number of recruited girls. This could be achieved by
several means: through newspapers, TV advertisements, local leaders and so
on. Potential victims may also be targeted by creating anti-trafficking watch-
dog committees (UNESCO 2006: 59). In particular, watchdog committees would
focus on rural areas in the African countries (less exposed to media sensitization)
and on young girls from Eastern countries working in the entertainment indus-
try (who are more vulnerable). Finally, the recruitment of girls may be reduced
by monitoring advertisements on local/national newspapers, websites etc. These
advertisements give little or no information about the job, the workplace and the
entrepreneurs. Since case study 1 showed advertisements to be one of the main
lures to “capture” victims, reducing advertisement anonymity (requiring further
information) may increase risks for whoever (traffickers or brokers) publish
them. For example, the policies on the information needed to publish journals
or dedicated public advertisements should be revised and should oblige whoever
publishes them to require more details on workplace, place (address) and manag-
ers’ personal details.

Preventing transportation
The application of situational crime prevention to the transportation stage is dif-
ficult within the EU countries owing to the principle of freedom of movement
within them. The main problems to face in the transportation stage are two con-
nected criminal phenomena, namely corruption and forged documents. The role
of corruption in trafficking of human beings has been previously highlighted as
a crucial factor (Zhang and Pineda 2008). Corruption not only occurs within the
source countries, but it takes place in destination and transit countries as well.
Illegal trafficking of people across countries requires negotiating a number of
checks: visa/documents/resident permit checks and embassy/border patrol checks.
Traffickers need to avoid or overcome these checks so as not to interrupt the traf-
ficking process. Criminal organizations use corruption to obtain forged documents
and to facilitate border crossings. As Van de Bunt and Van der Schoot (2003)
pointed out in their recommended situational crime measures against organized
crime, forged documents are important tools exploited by criminals, and conse-
quently decreasing their availability can help reduce human trafficking/smuggling.
Countermeasures identified by Van de Bunt and Van der Schoot (2003: 23), aim-
ing at limiting the availability of such tools, include improving controls where
documents are stored and improving technical instruments to reduce their forgery
156 E. U. Savona, L. Giommoni, M. Mancuso
(such as biometric technologies). Moreover, airline companies and officials across
all European countries should increase their commitment to controlling passen-
gers’ identity documents, acknowledging (thus removing excuses and assisting
compliance) that their negligence could favor criminal activities.
Case study 2 demonstrated a clear case of offender adaptation (defined as cir-
cumventing the preventive measures implemented with the aim to tackle criminal
activities), namely the response of traffickers to the B9-regulation of the Dutch
Code (Chapter 9 of the Aliens Act Implementation Guidelines). The B9-regulation
allowed to whoever declared having been a victim or witness of trafficking to
reside legally in the Netherlands upon condition that they collaborated with the
authorities. During their stay victims can take advantage of a place to stay, wel-
fare benefits and medical assistance. Traffickers exploit this regulation to let
victims enter the Netherlands and, once in, they help them to escape from refugee
centers. Respecting everyone’s human rights, B9 application should require some
further evidence of being victims of trafficking and, moreover, victims should be
informed of the risks (being beaten, exploited, threatened, etc.) they run if they
want to escape from the refugee centers. This duty should be covered with the
help of NGOs in the Netherlands.

Preventing exploitation
The application of preventive measures to the exploitation stage of human traf-
ficking mainly involves acting on the demand side for such service, and this
requires attempts to reduce requests for illegal sex workers. There are various
available interventions. The most extreme, the legalization of prostitution as
a profession, was applied for the first time in the Netherlands in 2000. Other
practical actions may be designed to disrupt the market. One strategy is to con-
centrate on those points where the legal and illegal world come into contact since
these are the places that facilitate the exploitation of prostitution. They include
restaurants, hotels, motels, discos, nightclubs, taxis, beauty centers and so on.
Suspending licenses or imposing high costs on those operators who facilitate the
exploitation of illegal immigrants would deter entrepreneurs from investing in
such activities. Finckenauer and Chin (2010: 75) also suggested targeting unan-
nounced raids on the above-mentioned businesses as a way to break down the
illegal sex worker market.
Information campaigns have already been mentioned as measures to reduce
the supply of illegal workers, thus increasing risks (or represented risks) for
potential victims of exploitation by traffickers. But the purpose of information/
sensitization campaigns may go further and help to crack down on the market for
alien workers. In fact, demand (the clients) and supply (the illegal sex workers)
meet in a “grey area” halfway between the hidden (to law enforcement) and the
known (by clients). Information campaigns should focus on those people (for
instance postal workers, doorkeepers, etc.) who may have access to such a “grey
area” and would be willing to report to law enforcement possible cases of sexual
workers’ exploitation.
Human trafficking for sexual exploitation 157
Another counteraction to decrease demand for victims of human trafficking is
to act directly on buyers of such services by deterring clients (by increasing costs)
of illegal sex workers. Measures aimed at acting on this side of the sexual market
are controversial and no agreement has been reached yet in Western countries.
In Italy prostitute clients are not criminalized, but other Western countries apply
criminal sanctions for soliciting prostitutes. In France and Sweden clients of pros-
titutes are criminally liable, whereas some cities in the US apply car confiscation
programs for soliciting prostitutes (Finkenauer and Chin 2010). Car confiscation
programs, driving license withdrawal and other administrative measures offer
ways to act upon costumers without criminalizing them.
One of the vulnerabilities exploited by criminal organizations that was iden-
tified in the two case studies is the misuse of non-banking financial institutes.
These have been defined by Van de Bunt and Van der Schoot (2003: 13) as the
“weak spot” in the fight against money laundering. Case study 2 showed that
money is sent via Western Union Money Transfer. Van de Bunt and Van der
Schoot argue that such entities, even if they are not banks, should be ruled by anti-
money laundering legislation to make the payments traceable and to close off the
vulnerability gap exploited by criminal organizations.

Preventing aftermath
The last pinch-point identified where it is possible to apply situational prevention
measures is the collaboration between law enforcement and victims of traffick-
ing. Usually this collaboration is quite poor owing to the victims’ lack of trust
in the law enforcement and their fear of being deported from the country (since
they are illegal immigrants). The victims’ collaboration with the law enforcement
agency can be disruptive for the criminal organization for two main reasons: (1)
the loss of the sexual services offered by the victims for whom traffickers have
paid travel expenses to the destination country represents an increase in the costs
(this is shown in case study 2, where traffickers paid legal assistance for those
victims identified by law enforcement); (2) as any form of collaboration with
police forces, this can reveal information for targeted law enforcement measures.
Policies to reinforce collaboration should be enforced. In particular, residence
permits for those immigrants who are exploited by criminal organizations should
be made available in all the European countries as provided for in existing Italian
legislation (i.e. Legislative Decree 286/1998 called “Immigration Law”—in
particular Article 18 of this Decree “Residence for social protection motives”).
Usually victims are not aware of the existence of such measures and their mistrust
of the police means that these measures are rarely applied. NGOs are the first line
in victims’ collaboration and can inform victims of their rights and facilitate their
collaboration, protection and assistance.
All the situational crime prevention techniques identified in order to reduce sex
trafficking are shown in Table 8.1.
158 E. U. Savona, L. Giommoni, M. Mancuso
Table 8.1 Proposed situational crime prevention techniques

Situational crime prevention Related situational Aim Crime script

measure prevention technique stage

Awareness campaigns and Alert conscience Increase risks Recruitment

anti-trafficking watchdog and remove
committees excuses
Reducing announcement Reduce anonymity Increase risks Recruitment
Improving controls over Control access to Increase efforts Transportation
documents store facilities
Improving technical Target harden Increase efforts Transportation
instruments against
documents forgery
Improve commitment to Assist compliance Reduce excuses Transportation
control passengers documents
Unannounced raids and Disrupt market Reduce rewards Exploitation
suspended licenses
Information campaigns Alert conscience Reduce excuses Exploitation
Car confiscation and driving Deny benefits Reduce rewards Exploitation
license withdrawal
Traceable payments Reduce anonymity Increase risks Exploitation
Favor victims collaboration Assist compliance Remove excuses Aftermath

The analysis proposed in this chapter represents a new approach to human traf-
ficking. The shortage of data on human trafficking led the authors to shift from a
macro analysis to an in-depth study applying script analysis. This chapter follows
and develops this line of research, aiming to present the crime commission pro-
cess in terms of environmental opportunities/vulnerabilities. The breaking down
of the complex components of the crime through script analysis helped to identify
those vulnerabilities/opportunities exploited by criminals and to suggest possible
crime prevention measures to address such vulnerabilities.
This chapter provides examples of how script analysis could be applied to
crimes related to the phenomenon of human trafficking. The scripts represent
the two main exploitation typologies of alien sexual workers: indoor and outdoor
In highlighting the main situational prevention interventions to address the
different phases of human trafficking we are not suggesting that we have neces-
sarily added any new strategies to those already known from past experience.
Governments, law enforcement agencies and NGOs have known about and prac-
ticed some of these interventions for a long time, on and off. The added value of
Human trafficking for sexual exploitation 159
this analysis is to incorporate these interventions into a broader and more coherent
strategy for addressing the phenomenon. Script analysis provides a framework for
monitoring the impact of each intervention and consequently for understanding
and modifying interventions if they do not work. We argue that script analysis can
improve the effectiveness of the entire situational prevention effort.
More generally, this chapter has been an opportunity to develop, building on
previous research (Savona 2009), the crime script approach to organized crime.
Crime scripts and organized crime do not go easily together. Organized crime,
owing to its international nature, complexity and connections with legal economy,
cannot be simplified into steps without naming properly these steps in relation
to the content of the crime considered and without pointing out how these steps
are in relation to other variables. The adoption of flow chart terminology and
computer-based semiotics represent the next step (and the next challenge) for a
more precise approach to scripting that permit the analysis of crimes with a high
level of complexity. If this ambition represents the main meal for the future of
crime scripts applied to complex crimes, this chapter has been an appetizer.

1 To each phase different collateral crimes can be associated, many of them strictly
connected to the commission of the crime. These crimes can be carried out against
victims (e.g. kidnapping, simple and aggravated assault, rape, extortion, threat and so
on) or against the State (e.g. document forgery, corruption of governmental officials,
abuse of immigration law, tax evasion, money laundering) (Aronowitz 2009; Comitato
parlamentare per la sicurezza della Repubblica 2009).
2 According to Paragraph b of Article 3, the consent of the victims should be considered
irrelevant because it is obtained using deception, violence or other means and for this
reason it is not a valid consent. In many cases, consent is given without knowing the
real conditions of exploitation that will be suffered in the destination country (Goodey
3 These networks can be informal or organized. The informal networks are associated
with small-scale trafficking and the members are connected by ethnic and/or familiar
bonds (Bosco et al. 2009; Väyrynen 2003) while the organized networks operate on a
large scale with criminals specialized in each phase of the trafficking. They are very
flexible, horizontal and able to change their action according to legislative variations
and to the affirmation of new profitable opportunities (Schloenhardt 1999).
4 There are three main international organizations that produce global data on human
trafficking: the International Labour Organization, the International Organization for
Migration and the United Nations Office on Drugs and Crime (Bosco et al. 2009).
5 The law enforcement agencies collect data on the victims identified and registered by
the police, the border guards and other control structures while the social services and
the NGOs are focused only on the victims involved in specific assistance and protec-
tion programs (Savona and Stefanizzi 2007; Vermeulen et al. 2010; IOM 2005).
6 Only one study based on this technique has been carried out, by Cockbain et al. in
2011, in order to analyze UK internal child sex trafficking.
7 It concerns trafficking in human beings for sexual exploitation carried out in night-
clubs involving victims from Eastern Europe (especially Polish and Romanian). This
case was indicted by the Court of Campobasso (Molise region) in 2010.
8 The traffickers give the victims precise instructions concerning the departure and the
arrival time, the bus stops or the airport chosen. This is done in order both to control
160 E. U. Savona, L. Giommoni, M. Mancuso
the victims and to assure their arrival, considering that they do not know the language
and have no experience in travelling.
9 Victims can go out only if accompanied by the boss or the boss’s brother and at certain
times of the day in order to avoid having contact with other people.
10 The term serata is used to indicate the services given to some clients, involving spend-
ing time with the girls in the club and having sex with them outside the night club (this
service costs about 200/250 euros). The possibility of serata depends on the drinks
paid for by the client: only those buying a large number of drinks and consequently
ensuring high earnings to the club can have sexual intercourse with the girl.
11 Many women are not psychologically and materially able to escape because they live
in a very vulnerable condition: they have no money and they need to work, they do not
speak Italian and do not know the place where they live; they depend on the traffickers
for all their primary requirements, and they have no documents.
12 The wishes of those girls who ensure a high profit are often met. For example, they can
stay at home some evenings and can wear trousers.
13 If the girls receive the clients at home they might attract the attention of the neighbors,
who could advise the police.
14 Each lap of the trip is coordinated and managed by the local representative who is
directly connected to the madam.
15 The increase of the debt is arbitrarily chosen by the madams and is given as sanction
to complaints or violation of agreements.

Adepoju, A. (2005). Review of research and data on human trafficking in sub-Saharan
Africa. In International Organization for Migration (ed) Data and research on human
trafficking: A global survey (pp. 75–98). Geneva: International Organization for
Aronowitz, A. (2001). Smuggling and trafficking in human beings: The phenomenon,
the markets that drive it and the organizations that promote it. European Journal on
Criminal Policy and Research, 9, 163–195.
Aronowitz, A. (2009). Human trafficking, human misery: The global trade in human
beings, Westport, CT: Praeger Publishers.
Beauregard, E., Proulx, J., Rossmo, K., Leclerc, B. and Allaire, J. F. (2007). Script analysis
of the hunting process of serial sex offenders. Criminal Justice and Behavior, 34,
Beeks, K. and Amir, D. (eds.) (2006). Trafficking and the global sex industry, Lanham,
MD: Lexington Books.
Bosco, F., Di Cortemiglia, V.L. and Serojitdinov, A. (2009). Human trafficking patterns.
In C. Friesendorf (ed) Strategies against human trafficking: The rule of the security
sector (pp. 35–82). Vienna: National Defence Academy and Austrian Ministry of
Defence and Sports.
Brayley, H., Cockbain, E. and Laycock, G. (2011). The value of crime scripting:
Deconstructing internal child sex trafficking. Policing, 5, 132–143.
Chiu, Y. N., Leclerc, B. and Townsley, M. (2011). Crime script analysis of drug
manufacturing in clandestine laboratories: Implications for strategic intervention.
British Journal of Criminology, 51, 355–374.
Clarke, R. V. (2008). Situational crime prevention. In R. Wortley and L. Mazerolle (eds)
Environmental Criminology and Crime Analysis (pp. 178–194). Cullompton, Devon:
Willan Publishing.
Human trafficking for sexual exploitation 161
Clarke, R. V. and Eck, J. (2003). Become a problem solving crime analyst in 55 small
steps, London: Jill Dando Institute of Crime Science.
Cockbain, E., Brayley, H. and Laycock, G. (2011). The value of the crime scripting:
Deconstructing internal child sex trafficking. Policing, 5, 132–143.
Comitato parlamentare per la sicurezza della Repubblica (2009). Relazione. La tratta degli
esseri umani e le sue implicazioni per la sicurezza della Repubblica, Doc. XXXIV, n. 2,
XVI Legislatura.
Cornish, D. (1994). The procedural analysis of offending and its relevance for situational
prevention. In R. Clarke (ed), Crime Prevention Studies, Vol. 3 (pp. 151–196). New
York: Criminal Justice Press.
Cornish, D. and Clarke R. V. (eds) (1986). The reasoning criminal: Rational choice
perspectives on offending, New York: Springer-Verlag.
Cornish, D. and Clarke R. V. (2008). The rational choice perspective. In R. Wortley
and L. Mazerolle (eds) Environmental criminology and crime analysis (pp. 21–47).
Cullompton, Devon: Willan Publishing.
David, F. and Monzini, P. (2000). Human smuggling and trafficking: A desk review on
the trafficking in women from the Philippines. Paper presented at the Tenth United
Nations Congress on the Prevention of Crime and the Treatment of Offenders, Vienna,
April 10–17, 2000; Doc A/Conf.187/CRP.1 15 April 2000.
Ekblom, P. (1995). Less crime, by design. Annals of the American Academy of Political
and Social Sceince, 539, 114–129.
Europol—European Law Enforcement Agency (2002). Crime assessment: Trafficking in
human beings into the European Union, Aja: Europol.
Felson M. (2008). Routine activity approach. In R. Wortley and L. Mazerolle (eds)
Environmental criminology and crime analysis (pp.70-77). Cullompton, Devon: Willan
Finckenauer, J. O. and Chin, K. (2010). Sex trafficking: A target for situational crime
prevention? In K. Bullock, R. V. Clarke, and N. Tilley (eds) Situational prevention of
organized crime (pp. 58–80). Cullompton, Devon: Willan Publishing.
Friesendorf, C. (2007). Pathologies of security governance: Efforts against human
trafficking in Europe. Security Dialogue, 38, 379–402.
Goodey, J. (2004). Sex trafficking in women from Central and East European countries:
Promoting a ‘victim-centered’ and a ‘woman-centered’ approach to criminal justice
intervention. Feminist Review, 76, 26–45.
Hancock, G. and Laycock, G. (2010). Organized crime and crime scripts: Prospects for
disruption. In K. Bullock, R. V. Clarke, and N. Tilley (eds), Situational prevention of
organized crime (pp. 172–192). Cullompton, Devon: Willan Publishing.
IOM—International Organisation for Migration (2005). Data and research on human
trafficking: A global survey, Geneva: International Organisation for Migration.
IOM—International Organisation for Migration (2009). Guidelines for the collection
of data on trafficking in human beings, including comparable indicators, Geneva:
International Organization for Migration.
Iselin, B. (2003). Trafficking in human beings: New patterns of an old phenomenon. Paper
presented at Trafficking in Persons: Theory and practice in regional and international
cooperation, Bogota, Colombia, November 19–21.
Kleemans, E. R. (2011). Expanding the domain of human trafficking research: Introduction
to the special issue of human trafficking. Trends in Organized Crime, 14, 95–99.
Lacoste, J. and Tremblay, P. (2003). Crime and innovation: A script analysis of patterns in
check forgery. In M. Smith and D. B. Cornish (eds) Crime prevention studies: Theory
162 E. U. Savona, L. Giommoni, M. Mancuso
for practise in situational crime prevention, Vol. 16 (pp. 169–196). Monsey, NY:
Criminal Justice Press.
Laczko, F. and Gramegna, M.A. (2003). Developing better indicators of human trafficking.
Brown Journal of World Affairs, X(1), 179–194.
Lebov, K. (2010). Human trafficking in Scotland. European Journal of Criminology, 7,
Leclerc, B., Wortley, R. and Smallbone, S. (2011). Getting into the script of adult child
sexual offenders and mapping out situational measures. Journal of Research in Crime
and Delinquency, 48, 209–237.
Limanowska, B. (2002). Trafficking in Human Beings in South Eastern Europe, Geneva:
Liu, M. (2011). Migration, prostitution and human trafficking: The voice of Chinese
women, New Brunswick, NJ and London: Transaction Publishers.
Ministero dell’Interno (2007). Rapporto sulla criminalità in Italia, Analisi, prevenzione,
contrasto, Roma, 18 giugno.
Morselli, C. and Roy, J. (2008). Brokerage qualifications in ringing operations.
Criminology, 46, 71–98.
Motta, C. (2008). Contrasto alla tratta di persone: normative, investigazioni, esigenza di
azioni integrate. Relazione del dott. Cataldo Motta, Procuratore distrettuale antimafia
di Lecce, al two day meeting di Zagreb 16 e 17 maggio 2008 nell’ambito del progetto
E.n.a.t.—European network against trafficking.
Motta, C. (2010). La tratta delle donne e lo sfruttamento della prostituzione, in Consiglio
Superiore della Magistratura—Ufficio per gli incontri di studio, Incontro di studio sul
tema: Violenza di genere, mobbing e stalking , Roma 17–19 maggio 2010 Ergife Palace
Okojie, C.E.E., Okojie, O., Eghafona, K., Vincent-Osaghae, G. and Kalu, V. (2003).
Trafficking of Nigerian girls to Italy. Report of field survey in Edo State, Nigeria, Turin:
United Nations Interregional Crime and Justice Research Institute.
Picarelli, J. (2009). Human trafficking and organized crime in the US & Western Europe.
In C. Friesendorf (ed) Strategies against Human Trafficking: The rule of the security
sector (pp. 115–136). Vienna: National Defence Academy and Austrian Ministry of
Defence and Sports.
Salt, J. (2000). Trafficking and human smuggling: A European perspective, Oxford:
Savona, E. U. (2010). Infiltration of the public construction industry by the Italian
organized crime. In K. Bullok, R. V. Clarke, and N. Tilley (eds) Situational Prevention
of Organized Crimes (pp. 130–150). Cullompton, Devon: Willan Publishing.
Savona, E. U. and Stefanizzi, S. (eds) (2007). Measuring human trafficking: Complexities
and pitfalls, Dordrecht: Springer.
Schank, R. C. and Abelson R. P. (1977). Script, plans, goals and understanding: An inquiry
into human knowledge, Hillsdale, NJ: Erlbaum.
Schloenhardt, A. (1999). Organized crime and the business of migrant trafficking. Crime,
Law and Social Change, 32, 203–233.
Shelley, L. (2010). Human trafficking: A global perspective. New York, NY: Cambridge
University Press.
Tompson, L. and Chainey, S. (2011). Profiling illegal waste activity: Using crime scripts
as a data collection and analytical strategy. European Journal on Criminal Policy and
Research, 17, 179–201.
Human trafficking for sexual exploitation 163
Transcrime (2003). Trafficking and smuggling of migrants into Italy. Analyzing the
phenomenon and suggesting remedies, report realized in collaboration with Direzione
Nazionale Antimafia for Italian Ministry of Justice and Italian Ministry for Equal
Transcrime (2010). Servizio di expertises e competenze per il monitoraggio, raccolta dati,
ricerche sperimentali, elaborazione ed implementazione di un sistema informatico
per supportare l’attivazione dell’Osservazione sul fenomeno della tratta degli esseri
umani, DPO Project, Dipartimento per le Pari Opportunità.
Tremblay, P., Talon, B. and Hurley, D. (2001). Body switching and related adaptations in
the resale of stolen vehicles. British Journal of Criminology, 41, 561–569.
UNESCO (2006). Human trafficking in Nigeria: Root causes and recommendations,
Policy Paper Poverty Series N. 14.2 (E).
UNICRI—United Nations Interregional Crime and Justice Research (2005). Programme
of action against trafficking in minors and young Women from Nigeria into Italy for the
purpose of sexual exploitation.
UNODC—United Nations Office on Drugs and Crime (2009). Global report on trafficking
in persons.
UNODC—United Nations Office on Drugs and Crime (2010). The globalization of crime:
A transnational organized crime threat assessment.
US Department of State (2005). Trafficking in persons report.
US Department of State (2008). Trafficking in persons report.
Van de Bunt, H. and Van der Schoot, C. (2003). Prevention of organized crime: A
situational approach, Research and Documentation Centre of the Ministry of Justice in
the Netherlands (WODC).
Van Duyne, P. and Spencer, J. (2011). Flesh and money. Trafficking in human beings,
Nijmegen: Wolf Legal Publishers.
Van Liemt, G. (2004). Human trafficking in Europe: An economic perspective, Geneva:
International Labour Organization.
Väyrynen, R. (2003). Illegal migration, human trafficking, and organized crime, Discussion
Paper N. 2003/72, United Nations University, World Institute for Development
Economics Research.
Vermeulen, G., Van Damme, Y. and De Bondt, W. (2010). Perceived involvement of
‘organised crime’ in human trafficking and smuggling. International Review of Penal
Code, 81, 247–273.
Wijers, M. and Lap-Chew, L. (1999). Trafficking in women, forced labour and slavery-like
practices in marriage, domestic labour and prostitution, Utrecht: STV.
Willison, R. and Siponen, M. (2009). Overcoming the insider: Reducing the employee
computer crime through Situational Crime Prevention. Communication of the ACM,
52(9), 133–137.
Zhang, S. X. and Pineda, S. L. (2008). Corruption as a casual factor in human trafficking.
In D. Siegel and H. Nelen (eds), Organized crime: Culture, markets and policies
(pp. 41–55). New York, NY: Springer.
Zhang, S., Chin, K., and Miller, J. (2007). Women’s participation in Chinese transnational
human smuggling: A gendered market perspective. Criminology, 45, 699–733.
Zimmerman, C. (2003). The health risks and consequences of trafficking in women and
adolescents, London: London School of Hygiene and Tropical Medicine.
9 Script analysis of corruption in
public procurement
Marco Zanella

Corruption in public procurement (CPP) has often been addressed as a crime that
is committed by rational agents who act when rewards are high and costs are low.
The majority of the research on CPP has explicitly or implicitly taken its cue from
the principles of the rational choice perspective as developed in the criminological
domain, which addresses the offender as a boundedly rational agent and the offense
as the result of involvement and event decisions. Involvement decisions are con-
cerned with the offender’s criminal career and deal with the initiation, habituation
and desistance from crime. Event decisions, on the other hand, are crime-focused,
and deal with the choices made when “preparing for, carrying out and concluding
the commission of a particular type of crime” (Cornish and Clarke 2008: 27).
Moving within the borders of the rational choice perspective, scholars have
examined the costs associated with CPP and how they influence corrupt agents’
decision making. Researchers have focused on the effect that official costs, such
as penal (Williams-Elegbe 2010), administrative (Nell 2007) and civil (Young
2009) sanctions, have on corrupt deals. Similarly, Rose-Ackerman (1997; see also
Pizzorno 1992) has analyzed the influence of informal costs, such as moral sanc-
tions, on corrupt agents’ reckoning. In addition, through the perspective of the
new institutional economics, scholars (e.g. Lambsdorff 2005) have focused on
the transaction costs of corruption, that is, the costs incurred by corrupt agents to
establish, maintain and enforce a corrupt deal and their impact on corrupt agents’
strategic behavior.
Alongside the formal and informal costs, studies have been carried out with
the aim of establishing how a rational agent makes decisions about initiating a
corrupt career. Culture, religion, gender and informal behavioral rules have been
addressed as background elements shaping the decision of a rational agent about
becoming involved in CPP (Barr and Serra 2010; Frank, Lambsdorff and Boehm
2011; Paldam 2001). In addition, attention has been focused on the “contem-
porary” experiences that may affect decisions of corrupt agents concerning the
continuation of a corrupt career. For example, Abbink (2000) analyzed the effect
of public sector salaries on civil servants’ corruptibility to test the hypothesis
put forward by Rijckeghem and Weder (2001) that higher salaries lead to lower
Script analysis of corruption in public procurement 165
corruption. In addition, research has investigated the effect that a greater or lower
presence of the state in the private sector has on CPP (La Palombara 1994), as
well as the relationship between this crime and administrative decentralization or
bureaucracy (Lessman and Markwardt 2010). Moreover, scholars have examined
the decisions about the withdrawal from CPP (Benno and Valev 2006) by exam-
ining whether older offenders’ decision making is associated with an increased
perception of the potential costs of the crime and by a reduced acceptance of
justifications for offending.
A few economic studies that are partially focused on event decisions in CPP
have benefitted from the explanatory power of game theory (Neumann and
Morgenstern 1944). Based on positivist formalism, game theory has been used in
the field of corruption research to help explain the strategic behavior of govern-
ment officials administering tests to grant permits (Cadot 1987) or bribe-initiated
corrupt transaction and the role of intermediaries in it (Guzin 2009). Despite their
usefulness, these studies have been often criticized since they describe actions that
corrupt agents may theoretically make but do not specify the moves that corrupt
agents do make (Osborne and Rubinstein 1994). Consequently, research on CPP
examining event decisions of corrupt agents typically lacks a detailed accounting
of the corruption-commission process and of the event decisions involved.
To study event decisions criminologists have recently borrowed the concept
of script from the cognitive sciences (Anderson 1983; Hill 1990), where the
concept was used to address “coherent sequence of events expected by the indi-
vidual, involving him either as a participant or as an observer” (Abelson 1976:
33). Moving from the idea that a crime is a contingent outcome of “sequences
of unit-actions in which actors are engaged in a particular behavior toward a
target” (Felson and Steadman 1983: 63), the script has been used in crimino-
logical research to examine in detail the crime-commission process. It has been
employed as an analytic tool for looking at behavioral routines (Cornish 1994a)
and for deconstructing a crime according to the moves and decisions followed
by the criminal agent (Clarke and Cornish, 2000). In this regard, the script
analysis has been exploited as a powerful tool to enhance understanding of the
crime commission process “involving stages in which resources and locations
are required and decisions are made” (Chiu, Leclerc and Townsley 2011: 358).
Since 1994, when Cornish (1994a, 1994b) first introduced script analysis to the
crime prevention field, criminologists have employed this tool to investigate
crimes such as: stolen vehicle resale (“ringing”) (Tremblay, Talon and Hurley
2001); cheque forgery (Lacoste and Tremblay 2003); vandalism (Smith 2005);
robbery (Copes, Hochstetler and Cherbonneau 2011; Petrosino and Brensilber
2003); employee computer crime (Willison 2009); and sex offenses (Leclerc,
Wortley and Smallbone 2011). More recently, script analysis has been applied
to organized crimes (Brayley, Cockbain and Laycock 2011; Bullock, Clarke and
Tilley 2010; Morselli and Roy 2008; Tompson and Chainey 2011). None of the
studies in this domain, however, applied script analysis to the event decisions of
corrupt agents.
166 Marco Zanella
Aims of the study
In a recent study Vannucci (2011) highlighted the need to break down corrupt
deals into “simpler phases, as a succession of different prearranged activities” (p.
26). Such an exercise would allow a more detailed understanding of the corrup-
tion-commission process and would help in the design of situational corruption
prevention measures to break the weak points of the process. This study addresses
this need, the overall aim being to understand the corruption-commission pro-
cess by using crime script analysis to delve into event decisions of corrupt agents
involved in public procurement.

Identification of the “specific corruption”
Crime is the result of a decision-making process that is highly crime-specific. For
this reason the analysis conducted in this chapter is crime-specific in its nature.
First it focuses exclusively on the Italian legal, administrative and bureaucratic
system and deals only with “domestic” cases of corruption, so it does not aim at
studying transnational or international corrupt deals. In addition, taking its cue
from the script exercises proposed by Cornish (1994a), this study moves from
the general crime of corruption in the public sector (protoscript) to the crime of
CPP (script) and focuses on the track of CPP related to works contracts, that is,
public contracts having as their object either the execution, or both the design and
execution, of works such as streets, bridges and other constructions a public con-
tracting authority may need. Such a selection allows the delving into a sector of
public intervention that is shaped by ad hoc administrative and legal regulations1
and the modeling of the so-called procurement cycle, that is, the “sequence of
related activities, from needs assessment, to the award stage, up until the contract
management and final payment” (OECD, 2008). In addition, this selection allows
a focus on a sector vulnerable to corruption because of limitations on competition
deriving from: asymmetric information flows in the national and local markets;
barriers to entry; fragmentation in the supply and demand structure; inefficient
public monitoring and lack of reputational requirements (Bentivogli, Casadio,
Cullino and Fabrizi 2007).

Procedure and sample

Court records and key informants have been selected as sources of data concerning
the corruption commission in the field of CPP related to works contracts. Owing to
the intrinsic nature of CPP, where direct access to information is not allowed, court
records have been selected as a first source of data (e.g. Natarajan and Belanger
1998). Court records provide a detailed summary of actions and moves of criminal
agents which are described through the lens of multiple sources of evidence. This
study relies on court records from two different Italian courts concerning—though
from different perspectives—the topic addressed: Criminal Justice Courts (CJC)
Script analysis of corruption in public procurement 167
and Court of Auditors (CA). CJC make decisions on whether criminal proceedings
instituted by a public prosecutor against an individual are well founded and provide
specific data concerning the tactics and the modi operandi of corrupt agents, the
steps of the corruption-commission and the circumstances that facilitated the cor-
ruption-commission. In contrast, the CA2 deals with issues regarding management
of the public budget and is “competent to prosecute and convict public officials
who have caused financial damage to public administrations through fraudulent or
faulty conduct” (GRECO 2009: 32); thus it provides an in-depth description of the
administrative environment where corrupt deals occur.
Court records were searched using two Boolean search operators, namely:
Dejure and that used by the CA.3 The search yielded 53 cases for the CJC and 39
for the CA in a time frame of May 2006 to May 2011. To qualify for inclusion
in the study, the cases had to: (1) deal with corrupt deals reached in the context
of public procurement related to public works contracts; (2) deal with facts that
occurred between May 2006 and May 2011 and (3) be related to at least one of the
five phases of the procurement cycle, namely: planning phase, preparation phase,
selection phase, implementation phase, assessment phase.
Owing to these selection criteria, only 22 cases qualified for inclusion in the
study, 14 from the CJC and eight from the CA.
Data collected through analysis of court records have been augmented with
information from questionnaires administered to 45 key informants with in-depth
expertise on CPP in works contracts. Key informants were selected through offi-
cial correspondence with law enforcement authorities and through the snowball
sampling method, and included police officers, officials of the Guardia di Finanza,
lawyers, prosecutors of the CJC and of the CA as well as judges of the CJC, the
CA and of the Regional Administrative Courts (RAC). Judges of the RAC assess
the lawfulness of public administrative decisions and for this reason they have an
in-depth knowledge of the details and steps that characterize public procurement
activities, and of the “niches” (Felson 2006: 128) of the “administrative environ-
ment” that facilitate corrupt agents when negotiating and executing corrupt deals
(Mariuzzo 2005; Merloni and Vandelli 2010).

Data and analytic strategies

The content analysis was conducted bearing in mind that descriptions of corrupt deals
are often biased by individual defensive strategies (della Porta and Vannucci 1999).
For these reasons, descriptions of a corrupt deal provided by corrupt agents involved
in the same case have been triangulated in order to verify their reliability. Moreover,
data sources have been classified according to their reliability, giving priority to
depositions, seized documents, eavesdropping activities, as well as to wiretappings,
to guarantee a high degree of objectivity (Paoli 2000). Data have been collated in
synoptic tables where the information concerning the chain of corruption-commis-
sion has been systematized and the elements that facilitated corrupt agents identified.
Even though only part of the results of the research are discussed in this chapter, it
is worth mentioning that synoptic tables systematize the data regarding each corrupt
168 Marco Zanella
deal so as to facilitate the development of the script, and for each case they focus on:
(1) the “overall framework,” i.e., the overall economic and administrative context in
which the crime was committed; (2) the “location of the works,” i.e., the geographi-
cal area where the works were to be carried out; (3) the “administration,” i.e., the
public administration regulating the public procurement (State, Region, Province,
Municipality); (4) “how the case came to the attention of the police,” whether
because of a police investigation or internal weaknesses in the corrupt deal; (5) the
“number of persons involved”; (6) the “roles” of the persons involved, i.e., their
position within (e.g., as political actors or bureaucratic actors) or outside the public
administration (e.g., as entrepreneurs); (7) the “setting,” i.e., the phase of the public
procurement cycle in which the corrupt deal was concluded; (8) the “entry condi-
tion,” i.e., the conditions that led to the corrupt deal; (9) the “results,” i.e., the goal
that the corrupt agents attained as a result of the corrupt deal; (10) the “scenes/func-
tions,” i.e., the data concerning the chain of commission of the crime, divided into
key stages and moves; (11) the “facilitators that enabled commission of the crime,”
i.e., the information concerning the opportunities that facilitated corrupt agents when
committing the crime; and (12) the “contributory crimes,” i.e., those crimes commit-
ted by the corrupt agents to facilitate the corrupt deal and its execution.

The result section of this chapter is organized as follows. First the key stages of
CPP in works contracts are presented and the moves of corrupt agents are singled
out and described to identify the procedural aspects of the crime. This allows the
highlighting of the ways in which corrupt deals take place in public procurement,
the tactics employed by corrupt agents and the choices behind them. Second, the
procedural requirements of the crime are explored, by identifying the facilita-
tors of CPP, that is, all those tangible and intangible assets that facilitate corrupt
agents when performing the crime.

The key stages of CPP in works contracts

The analysis of the data shows that CPP in works contracts is a process made up
of fixed key stages, namely:

KEY STAGE 1 (KS1) identification of the “target/need for” and of the

KEY STAGE 2 (KS2) meeting, negotiation and conclusion of the corrupt
KEY STAGE 3 (KS3) performance guarantee and execution of the cor-
rupt deal
The case of CPP in works contracts exhibits an interesting variation on the key
stages already singled out for “one-person crimes.” In the latter, the criminal
agent acts according to his own plan and needs. In contrast, the case of corruption
Script analysis of corruption in public procurement 169
is intrinsically based on a dialogue between at least two corrupt agents who must
identify each other, coordinate their actions, find a deal suitable for both of them,
and execute that deal. For this reason the event schema of CPP comprises two
key stages (KS1 and KS2) whose object is identification of the “target/need for”
and of the partner (KS1) and the approach of the parties, as well as the dynamics
of negotiating and concluding the corrupt deal (KS2). These key stages are com-
pleted by a third key stage (KS3) concerning the performance guarantees on the
deal and its execution. In this regard, CPP in works contracts is the result of the
above-mentioned key stages as expressed in the following formula
CPP = KS1 + KS2 + KS3
Following the above-mentioned key stages, CPP may be shaped by the presence
of an additional key stage (KS4) concerning the emergence of conflicts and their
resolution, which due to space limits is not dealt with in this chapter.

From the key stages to the moves

Each corrupt deal is the result of specific behavioral routines and contracting strat-
egies that may emerge on shifting the perspective from the key stages (macro) to
the moves (micro) that can be identified within each key stage. On “zooming in”
from the macro to the micro, the above-mentioned formula concerning the crime
may be reframed as follows:
CPP = (KS1m) + (KS2m) + (KS3m)
where “m” denotes the possible moves that the corrupt agents may make within
each key stage. These moves, as shown in Table 9.1, are discussed in the following

Table 9.1 CPP—key stages and moves

KSI – Identification of the “target/need for” and of the partner

Move 1 – The identification of the “target/need for”

Move 2 – The identificatioon of the corrupt partner
Move 3 – The identification of the mediator

KS2 – Meeting, negotiation and conclusion of the corrupt deal

Move 4 – The organization of the meeting

Move 5 – The identification of the object of the corrupt deal
Move 6 – The identification of the timing of the corrupt deal

KS3 – Performance guarantee and execution of the corrupt deal

Move 7 – The performance guarantees of corrupt agents 1 and 2

Move 8 – The execution of the corrupt deal
170 Marco Zanella
Moves under key stage 1 (KS1): identification of the “target/need for”
and of the partner.
Moves under KS1, as graphically represented in Figure 9.1, are exploratory in
nature. They do not consist of material actions, but rather of immaterial decisions.
A corrupt deal needs two initial conditions to be in place: (1) a target and (2) the
presence of at least two agents. One agent, referred to as the outsider, is external
to the contracting authority (e.g. an entrepreneur) and wants to negotiate with
it in regard to one or more works contracts. The second actor, referred to as the
insider, is internal to the contracting authority (e.g. a public official) and plays a
role in its structure with reference to one or more works contracts. Alongside the
terms outsider and insider in this chapter, the term corrupt agent 1 is employed
when identifying the “active” actor of the deal, that is, the outsider or the insider
who concretely starts the corrupt deal in an outside-in process or in an inside-in
process, searching for corrupt agent 2, who plays a “passive” role since he4 is not
the originator of the deal.


The corrupt deal arises when a corrupt agent identifies a “target,” that is, a goal
to be reached by means of a corrupt deal. The corrupt agent’s reckoning is shaped
by that target because he perceives the difficulty of achieving it without a corrupt
deal. The corrupt agent therefore perceives the “need for” such a deal. The “tar-
get” may take different forms and contents depending on the actor (outsider and
insider) and on the different phases of the public procurement cycle. Identification
of the “target” presupposes a choice by the corrupt agent, who selects it in order
to pursue his interests. From the outsider perspective, the “target” may relate, for
example, to his desire to be awarded a specific contract because of its economic
value or to fix an administrative problem that has emerged during execution of the
contract. From the insider perspective, the “target” may relate to his willingness
to gain an illicit reward while administering a specific public procurement proce-
dure or to maintain a web of social and economic relations with entrepreneurs and
other private actors revolving around the public construction industry.

Identification of the

Corrupt agent 1 Identification of the

target/need for

Identification of
corrupt agent 2

Figure 9.1 (KS1m) moves in KS1.

Script analysis of corruption in public procurement 171
Data have demonstrated that the corrupt agent tries to identify a target suited to his
needs and capabilities and which is placed in an administrative and bureaucratic
environment with which he is familiar, in a sort of circuit where the corrupt agent
selects a target on the basis of the possible partner that he might find. Moreover,
data showed that in “second-time” corrupt deals, that is when the corrupt agents
have negotiated previous corrupt deals, identification of the “target” is not a uni-
lateral process. It is rather the result of collaboration between the insider and the
outsider, who by performing outside-in or inside-in recruitment processes lays the
bases for several corrupt deals.


As soon as a corrupt agent 1 has identified the “target” and has perceived the
“need for” a corrupt deal, he proceeds with identification of his partner (corrupt
agent 2). Selection of the corrupt partner is a crucial moment in CPP related to
works contracts which is a highly “parochial” form of corruption, since it is based
on transactions “with few potential contractors, and thus, restricted competition”
(Lambsdorff, 2002, p. 222). In this regard it was found that a corrupt agent inter-
ested in identifying a corrupt partner undertakes a strict selection procedure in
order to find a partner, the key criteria being: (1) fit for the purpose, (2) corrupti-
ble, (3) trustworthy and (4) available. These criteria apply both to the insider and
to the outsider.
The insider must be fit for the purpose in the sense that he is the right person in
the right place at the right time, having specific decisional power over the specific
target in which the outsider is interested. The insider must be corruptible in the
sense that he is willing to receive an illicit benefit from the outsider. The corrupt-
ibility of the insider is of importance especially in a “first-time” corrupt deal—that
is, when corrupt partners bargain a corrupt deal for the very first time—where the
outsider must explore the insider’s corruptibility and avoid contacting a public
official who may prefer to report the corrupt offer to the police. The insider must
be trustworthy in the sense that he will provide the requested service. The reliabil-
ity of the insider varies between “first-time” and “second-time” corrupt deals. In
“second-time” corruption—that is, when the corrupt agents are already part of cor-
rupt deals—corrupt agents are locked into each other. Finally, the insider must be
available in the sense that he is not part of a corrupt deal with another outsider with
reference to the same contract. In this case, his reliability could be compromised.
Along the same line, the outsider should be fit for the purpose by operating in
a specific commercial sector and being concretely interested in a corrupt deal to
be negotiated in the area of works contracts. As far as the other characteristics are
concerned, the insider seeks a partner who is corruptible, trustworthy and avail-
able. Also the insider must choose wisely, because he must avoid contacting an
outsider who is ready to report the corrupt deal to the police, is working under-
cover for the police as an “agent provocateur,” or is involved with other public
officials in a corrupt deal related to the same work contract in the same contract-
ing authority.
172 Marco Zanella
Having the characteristics of corrupt agent 2 in mind, corrupt agent 1 now
needs to identify and contact him. Identification of the partner is not a theoretical
exercise. It requires a particular type of know-who and the acquisition of informa-
tion on corrupt agent 2, which is often difficult to obtain in a context where public
advertisements are not available (Lambsdorff, 2007, p. 140; Lambsdorff, 2002, p.
In this regard, data show that three situations may arise. In the first situation,
corrupt agent 1 already knows corrupt agent 2 because of previous corrupt deals.
Thus, corrupt agent 1 easily identifies his potential partner. This is the case for
“second-time” corruption deals or when corrupt agents 1 and 2 are linked to each
other by friendship or by family ties.
In the second situation, corrupt agent 1 knows corrupt agent 2 in the sense that
he can identify him as a potential partner but is unable to determine whether or
not he meets all the criteria, and he does not have any links with him. This is the
case, for example, when corrupt agent 1 has already had previous licit business
relations with corrupt agent 2, but he does not know whether he meets all the other
In the third situation, corrupt agent 1 does not know corrupt agent 2. He cannot
identify him, and, as a logical consequence, he does not know whether or not he
fulfills the characteristics.
Both in the second and in the third situations, corrupt agent 1 is a “blind
corrupt agent” who needs specific information concerning his possible partner.
In the above-mentioned situations, the identification and the collection of infor-
mation about corrupt agent 2 as well as the link with him are facilitated by the
intervention of mediators, that is, third parties who, knowing the actors and the
administrative and local context, intervene and act as links between the outsider
and the insider.


Identification of the mediator is as delicate as identification of the partner. Data

reveal that a corrupt agent when searching for a mediator looks for an individual
who is (1) fit for the purpose; (2) corruptible; (3) trustworthy; (4) available; (5)
willing to run the risk and (6) influential.
Also the mediator must be the right person in the right place at the right time.
He must be corruptible, trustworthy and available. However, besides these charac-
teristics, the mediator must be willing to run the risk and be influential. Mediation
is often risky, especially when the mediator is involved in “dirty work” such as the
material exchange of cash payments. Moreover, the mediator must be influential
with respect to corrupt agent 2. In other words, he must be able to contact corrupt
agent 2 and reassure him about corrupt agent 1. He must also be able to assure
corrupt agent 1 that corrupt agent 2 is willing to engage—directly or through
mediators—in the corrupt deal and is able to execute it.
Script analysis of corruption in public procurement 173
Moves under key stage 2 (KS2): meeting, negotiation and conclusion of
the corrupt deal
Having selected the target, decided to commence a corrupt deal, and identified
the suitable partner, corrupt agent 1 must begin action by contacting corrupt agent
2. He then needs to ensure the availability of corrupt agent 2 and to negotiate the
object of the corrupt deal and the modes and timing of its execution. These moves
are graphically summarized under Figure 9.2.


Once corrupt agent 1 has identified corrupt agent 2 and determined that he
fulfills the criteria, he must contact him. In this regard it has been found that
in the specific field of CPP related to works contracts, corrupt agent 1 rarely
resorts to direct investigation and direct contact with corrupt agent 2. This only
happens in “second-time” corrupt deals or when corrupt agents 1 and 2 already
know each other. It rarely happens that, in “first-time” corrupt deals, corrupt
agent 1 resorts to abboccaggio (literally “bait-casting”). Bait-casting relates
to meetings during which corrupt agent 1 puts out feelers as to the availability
of corrupt agent 2, using suggestive phrases or expressions with double mean-
ings, and, like an angler, leaving it to the interlocutor whether or not to take
the bait. Bait-casting tends to be used in corrupt deals reached in sectors other
than that of public procurement as such. CPP related to works contracts is too
“complex” in terms of the length of the period for which corrupt agents 1 and
2 are bound to each other, and of the variety of the possible objects of the cor-
rupt deal. In this regard, data show that meetings between corrupt agents 1 and
2 are meticulously organized and prepared by the actors themselves or through
their mediators.
The parties tend to use technological devices such as telephones or Internet
(Skype) for discussions of a general nature and to organize their meetings, using
cryptic language and expressions. However, when the details of the agreement
are discussed, corrupt agents usually meet personally and in public places, and
therefore not within the administration’s offices, where the actors are aware that
there is greater risk of attracting attention.

Meeting and Identification of Identification of

bargin the object the timing

Figure 9.2 (KS2m) moves in KS2.

174 Marco Zanella

During meetings corrupt agents identify the object of their deal. Corruption is an
“instrumental crime” that assumes the features of a “contract by consideration”
based on specific gains that are closely related to and shaped by the “target” that
corrupt agents have identified at the beginning of the process (Move 1).
In this regard the data collected allowed identification of a wide range of
objects of the corrupt deal that are well summarized in the Latin concepts of do
ut des, do ut facias, do ut non facias (I give so you may give, I give so you may
do; I give so you may omit). These gains, which represent the “basis of reciprocal
relations in past and present societies” (Bijsterveld 2007: 7), are the basis of all
the corrupt deals where the outsider promises and gives money or other assets of
value to the insider, while the insider protects the outsider by giving him informa-
tion or advantages, by intervening (doing) to facilitate his actions or by omitting
controls over the activities of the insider.


During meetings, corrupt agents do not only negotiate on the object of the cor-
rupt deal. They also agree on the timing of the deal’s execution, which involves
defining when each agent will fulfill his specific “obligation,” that is, execute the
object of the corrupt deal. This move in key stage 2 (KS2) completes the moves
to do with “meeting, negotiation and conclusion of the corrupt deal” and links
KS2 with key stage 3 (KS3) because it leaves the door open for the “performance
guarantee and execution of the corrupt deal.”

Moves under key stage 3 (KS3): performance guarantee and execution

of the corrupt deal
Having agreed upon the corrupt deal, both corrupt agent 1 and corrupt agent 2
must protect their investment and must execute the deal itself. Moves performed
under KS3 are graphically summarized in Figure 9.3.

corrupt agent 1

Execution of the
corrupt deal

corrupt agent 2

Figure 9.3 (KS3m) moves in KS3.

Script analysis of corruption in public procurement 175

In the shadowy environment of corrupt deals, “words can be easily spent, promises
formulated, agreements negotiated, but sometimes the most profitable decision is
to renege them, whenever possible” (Vannucci 2011: 2). For these reasons, cor-
rupt agents try to create performance guarantees in order to protect their interests
against total or partial breach of the corrupt deal.
Corrupt agents must protect the deal not only from possible external aggression
(by law enforcement or by other entrepreneurs or public officials not involved in
it), but also from possible internal breaches. To this end, both the insider and the
outsider use performance guarantees, which can be divided into three main types:
(1) “natural” performance guarantees, (2) “artificial” performance guarantees and
(3) “protector” performance guarantees.
“Natural” performance guarantees result from the illicit nature of corruption,
which is its weakness but also its strength. The illicit nature of the corrupt deal
binds the parties to each other. They are locked into each other because both have
a common interest: not only execution of the corrupt deal but also the avoid-
ance of law enforcement. This need is binding upon corrupt agents because they
have a mutual interest in maintaining the corrupt deal in the shadow of the law.
A further “natural” performance guarantee is linked to the reputational assets
of the corrupt agents. If one of the two agents breaks his promises, he will lose
future opportunities for collaboration with other corrupt agents because he will
be branded as unreliable.
Besides these “natural” performance guarantees, corrupt agents also resort to
“artificial” performance guarantees. These performance guarantees result from
specific strategies undertaken during key stage 2 (KS2) and produce their effects
in key stage 3 (KS3). The “artificial” performance guarantees exploited by the
insider are related to the payment of the “commodities” and to his role within the
public administration. The payment guarantee resides in the ability of the insider
to obtain an advance payment, which is frequently used when the payment is in
cash. Second, the insider can leverage his “administrative funnel” role as an addi-
tional “artificial” performance guarantee. Because the insider is fit for the purpose
(i.e. the right person at the right place at the right time) he often has long-standing
decisional power over several contracts. In this regard, the data show that corrupt
public officials are vested with key roles in the process, and that these roles are
always performed by the same subjects over a long period of time. The insider
therefore forms the narrowest part of the “administrative funnel.” Partial or total
breach of the deal by the outsider would result in his complete exclusion from the
market of works related to public procurement contracts in at least one specific
geographic area.
The “artificial” performance guarantees exploited by the outsider are almost
exclusively associated with the methods of payment, with reference both to the
financial instruments used and the mode of cash delivery. Everything revolves
around the need to delay payment until after intervention by the insider and
to ensure traceability of the payment either through bank records or through
176 Marco Zanella
witnesses (e.g. mediators) who could guarantee payment. These traces are par-
ticularly important because they can be used to blackmail the insider. Besides
these “artificial” performance guarantees, recent cases have shown the emergence
of so-called “gooey payments.” These payments have a particular advantage for
the outsider because they establish long-standing relations. Consultancy contracts
for the insider, job contracts for his family members, or the renovation of his
buildings are payments that take time to be collected—the time to conclude the
consultancy contract, the time to earn the salary (job contract), the time to com-
plete renovation of a building.
Both the insider and the outsider may use a further performance guarantee
(“protector” performance guarantee) represented by the mediators who operate as
protectors of corrupt agents. These may be members of organized crime groups,
or political actors or high-level administrative actors. They may not be directly
involved in the corrupt deal but their presence next to the insider or the outsider,
or both, ensures the smooth flow of the corrupt deals.


Key stage 3 (KS3) concludes with the final step in the corruption-commission
chain: execution of the deal. In this step, corrupt agents 1 and 2 achieve the target
identified at key stage 1 (KS1). In the aftermath of execution of the corrupt deal,
corrupt agents 1 and 2 may permanently separate or commence a long-standing
corrupt relationship.
Merging all the key stages of the corrupt deal together, and including all the
moves identified in the previous sections, it is possible to visualize the script
of corruption in public procurement in public works contracts as showed in
Figure 9.4.

The facilitators of CPP in works contracts

Script analysis of CPP shows that a corrupt deal does not consist solely of moves.
The moves are accompanied by tangible and intangible assets that facilitate cor-
rupt agents when moving within the key stages.
In each key stage, corrupt agents benefit from specific facilitators, which are
functional to the need of corrupt agents to: (1) minimize the transaction costs of
the crime—that is, all those costs related to the establishment and maintenance of
property rights (Allen 1999: 898) —and, as a consequence, (2) establish a smooth
corrupt deal that enables them to achieve their identified “target.”
Three typologies of facilitators of CPP have been identified, namely: (1)
cognitive and social facilitators related to the knowledge (know-how, know-
what, know-who) of corrupt agents and to their (bad) social capital; (2) legal
and administrative facilitators related to the “niches” (Felson 2006: 128) in the
“administrative environment” created by decisional powers allocated to mono-
cratic agents with monopoly positions operating with discretion and reduced or
no accountability; and (3) physical facilitators related to the items and facilities
Key stage 1 (KS1m) Key stage 2 (KS2m) Key stage 3 (KS3m)
Identification of the target/need Meeting, negotiation and Performance guarantee and
for and of the partner conclusion the corrupt deal execution of the corrupt deal

corrupt agent 1
Identification of
the mediator

Corrupt Identification of the Meeting and Identification of Identification of Execution of the

agent 1 target/need for bargin the object the timing corrupt deal

Identification of
corrupt agent 2
corrupt agent 2

Figure 9.4 The script of CPP in works contracts.a

a This chapter, owing to space limits, does not deal with the case of the so-called “structural” corrupt deals that emerge when corruption is systemic within a certain
public administration or sector. This nature of the “structural” corrupt deal, however, does not “delete” or change the event schema of the crime, which is still based on
key stages 1, 2 and 3 (KS1, KS2, KS3) but it greatly reduces the moves to be made.
178 Marco Zanella
exploited by corrupt agents to meet each other and communicate, as well as to the
“commodities” to be paid in the course of the corrupt deal. These facilitators play
different roles in the corruption-commission chain, as shown by Table 9.2, where
facilitators are combined with the key stages KS1, KS2 and KS3 and related to
their impact on the phases of the corrupt deal.

Table 9.2 The facilitators of CPP in work contracts

Faciliators Impact on the corrupt deal

Key Stage 1 (KS1) 1. Cognitive and social facilitators – search and
✓ know-what information costs
✓ know-who – risks (e.g. law
✓ (bad) social capital = privilaged enforcement)
contacts, concrete personal + cooperation among
relations, bonds of trust, agents
“gooey” systems of mutual + distribution of
obligations and reputational information aong
assets corrupt agents
Key Stage 2 (KS2) 1. Cognitive and social facilitators – bargaining and
✓ know-who
✓ know-how

2. Legal and administrative

} decision costs
– risks (e.g. law
+ cooperation among
– external control/risks
✓ layout of the administrative
environment (monocratic
agents, monopoly positions,
discretion, reduced (no)
} + “reward”

3. Physical facilitators – bargaining an

✓ commodities easily concealed
from law enforcement
✓ instruments (mobile phones,
phones, email, Skype)
✓ facilities (semi-public places)
Key Stage 3 (KS3) 1. Cognitive and social facilitators
} decision costs
– risks (e.g. law

– policing and
✓ know-who
✓ know-how
} enforcement costs
– risks (law
– risks (internal

2. Legal and administrative – external control/risks
layout of the administrative
environment (monocratic agents,
monopoly positions, discretion,
reduced (no) accountability)
Script analysis of corruption in public procurement 179
Table 9.2 shows that the facilitators from which corrupt agents benefit during the
corrupt deal are partially tangible in their nature (e.g. physical facilitators). The
combination of procedural aspects and procedural requirements, however, shows
that corrupt agents make extensive use of intangible facilitators. In this regard,
Table 9.2 shows how knowledge and (bad) social capital help corrupt agents under
key stage 1 (KS1). They reduce the efforts by minimizing search and information
costs and increasing the opportunities of cooperation among corrupt agents.
Also in key stage 2 (KS2), corrupt agents are facilitated by cognitive and social
facilitators, as well as by legal and administrative ones. On the one hand, cogni-
tive and social facilitators assist corrupt agents by reducing the bargaining and
decision costs, and they help corrupt agents reduce the external dangers of the
corrupt deal by concealing it from the law enforcement agencies and from other
competitors on the corruption market. On the other hand, legal and administrative
facilitators resulting in monocratic and monopoly positions, as well as discretion-
ary and non-accountable powers, are of particular importance in facilitating the
corrupt deal, because they reduce the risks related to the crime-commission and
may be exploited to increase the rewards of the corrupt deal.
The same applies under key stage 3 (KS3) where cognitive and social facilita-
tors affect the corrupt deal by reducing the policing and enforcement costs. In
addition, the above-mentioned facilitators strengthen the bonds of trust among
corrupt agents and reduce the risks of the deal’s internal failure owing to its total
or partial breach by one of the partners.
What emerges from this analysis is that cognitive and social facilitators play
a crucial role in each key stage. The knowledge and the (bad) social capital of
corrupt agents are the cornerstones of their decision making (Egidi and Rizzello
2003). The exploitation of these specific facilitators during each key stage of a
corrupt deal may be tacit—in the sense that the corrupt agents may not be aware
of all the mental processes involved (Friedman 1953)—and it may consist (espe-
cially in “second-time” corrupt deals) in a process lasting a “few seconds.” But
these cognitive and social assets shape the capacity of corrupt agents to move in
each key stage, to plan each move, and to benefit as much as possible from the
other facilitator (i.e., physical and legal and administrative ones).

The aim of this chapter has been to understand the tactics, dynamics and modi
operandi of corrupt agents, with the purpose of highlighting how CPP unfolds. On
the one hand, the script analysis sheds light on event decisions by addressing the
procedural aspects of the crime. In this regard, the corrupt deal was divided into
key stages, and therefore into the macro-activities carried out by the protagonists
of the pactum sceleris. This decomposition laid the basis for discussion of the
specific moves made by corrupt agents in each key stage in order to minimize
risks and efforts while increasing the rewards.
The analysis of the moves highlighted the particular nature of corruption.
A corrupt deal is embodied in an encounter between two parties engaged in a
180 Marco Zanella
dialogue characterized by a tension between trust and fear. Trust binds the two
protagonists of the deal, who have to work and take risks together in order to
fulfill their personal interests. On the one hand, fear relates to the external risks of
the corrupt deal—such as law enforcement—while on the other hand it depends
on internal risks related to possible turnaround or “lemons” that one partner may
hold to the other one.
This tension between trust and fear inevitably marks all the moves made by
corrupt agents in each key stage, and it heightens the efforts and risks related to
the corruption-commission. Corrupt agents must seek to control and manage the
deal in order to protect their investment and to ensure the success of the deal itself.
Trust and fear result in transaction costs that affect the corrupt deal. These
costs, which arise in each exchange of property rights, are particularly acute in
the case of a deal of this kind, which by its very nature is illegal and is not legally
enforceable. Starting from these considerations, the study of event decisions
focused on the procedural requirements of corruption-commission. This consisted
in identification of those items, tools and situations that facilitate action and mini-
mize transaction costs. The decomposition of the corrupt deal highlighted that
corrupt agents benefit from three types of facilitators: cognitive and social facili-
tators, legal and administrative facilitators and physical facilitators.
This study—albeit with limitations particularly related to the small sample of
data, which reduces the generalizability of the results, and to the complexity of the
topic addressed, which mixes several disciplines—has started to build knowledge
on a topic neglected for a long time: the “how” of corruption in public procure-
ment. This is the beginning of a process of understanding corruption, not the end.
Interdisciplinary research in this domain is still necessary.
On the one hand, the application of script analysis to study event decisions
could be extended to other forms of corruption. The analysis of the procedural
aspects and requirements for commission of specific corruptions would be of a
particular importance. On the other hand, future research could take a step for-
ward by moving from the analysis and description of event decisions to policy
implications. Detailed understanding of the procedural aspects and requirements
of corruption-commission provides important insights for both corruption detec-
tion and prevention.
In this regard the analysis conducted in this chapter shows several areas of
intervention for situational prevention measures. The deconstruction of the cor-
rupt deal reveals that prevention measures could be framed within two main lines
of intervention: one relating to the external risks and the other relating to the
internal risks.
The external risks, which relate to law enforcement and to the possibility that
the corrupt deal comes to the “public attention,” could be targeted by enforcing
specific intervention points, such as: increasing the internal supervision on spe-
cific offices of the public administration where decisions concerning the public
procurement are taken; increasing and sustaining external controls by publishing
all the data and documents regarding each procurement procedure on the web so
as to allow rapid control by law enforcement and by the private enterprises or
Script analysis of corruption in public procurement 181
actors operating within each specific market; implementing situational risk audits
within public administration; monitoring gifts and privileges by increasing judi-
cial and financial controls on cash payments or on the assets of public officials;
increasing the presence of collegial bodies, in lieu of monocratic agents, in the
procurement process; and changing the “administrative environment” through the
amendment of laws (e.g. crime proofing) that create the specific “niches” (Felson
2006: 128) that facilitate corrupt agents.
Alongside external risks, situational prevention measures could also increase
internal risks of the corrupt deal, that is, increase the pressure of the transac-
tion costs. This goal might be achieved through the adoption of policies aimed
at hampering trust among corrupt agents in all the identified key stages, from
the identification of the partner till the execution of the corrupt deal. For exam-
ple, staff rotation within public bodies prevents a public official from continuing
to serve in a given office for a long time. Constant staff rotation forces corrupt
agents to pay the transaction costs typical of “first-time” corrupt deals. It obliges
them to run greater risks when identifying and contacting a “new” partner, nego-
tiating the object of the deal and executing it. Changing the partner would oblige
corrupt agents to rebuild their trust relationships. In addition, an amendment to
the Criminal Code, precluding, under particular circumstances, the possibility of
punishment for the corrupt agent who first denounces a corrupt deal would fuel
uncertainty among corrupt agents. It would affect trust between them by increas-
ing the internal risks of the deal and, as a consequence, its transaction costs
(Spagnolo 2010).
The above-mentioned prevention measures are not meant to be “one size fits
all” solutions or an exhaustive list of policies to be enforced. However, these
examples show the potential of script analysis in the elaboration and identifica-
tion of “situational corruption prevention policies” (Zanella 2011), that could be
enforced to curb corruption and to reduce its discrètes vertus (Koenig 2009).

1 The sector of public work contracts has been shaped by regulations closely related to
the complexity of the procedures and of the design and execution of the works to be car-
ried out. This has been the case of the Law, 11 February 1994, n. 109 and the Decree of
the President of the Republic, 21 December 1999, n. 554 that after Tangentopoli—the
scandal of the early 1990s that exposed the pervasive corruption in the Italian politi-
cal system—attempted to reduce the discretionary powers of public actors and thereby
reduce the opportunities for corruption.
2 Only decisions linked to criminal proceedings conducted against a public official for
corruption were selected. Considering that the focus is on the administrative circum-
stances even facts which took place before 2006 have been selected.
3 The following keywords were inserted in the Boolean search operators: “works AND
corruption AND extortion by public official AND rigging AND public procurement.”
The crimes of extortion by a public official and rigging were included in the research
formula because defensive strategies adopted by the parties at court may try to change
the count from corruption to extortion by a public official and/or rigging. For this rea-
son, also, these crimes were included, but only corruption cases qualified for inclusion
in the study.
182 Marco Zanella
4 The masculine pronoun is used throughout the chapter. This choice is only partially for
stylistic reasons, since it also depends on the gender of the players identified during the
analysis of the cases, who were all male. In the knowledge that the cases collected lack
“generalizability,” their figures, also associated with responses to the questionnaires—
which contained a specific question on the point—seem to confirm the hypothesis in
the literature that women are less likely to be involved in corrupt deals. Only in one
case of those identified was a woman partially involved in a corrupt deal. She was not
directly involved in corrupt activities but was the assistant of an entrepreneur—who was
directly involved in the corrupt deal—and worked only to help and control the intraneus
involved in the writing of the tailored calls for tender.

Abbink, K. (2000). Fair Salaries and the Moral Costs of Corruption. Discussion Paper
1/2000. Bonn Econ Discussion Papers.
Abelson, R.P. (1976). Script Processing in Attitude Formation and Decision Making. In
J.S. Carroll and J. Payne (eds) Cognition and Social Behavior (pp. 33–46). Hillsdale,
NJ: Lawrence Erlbaum Associates.
Anderson, J.R. (1983). The Architecture of Cognition, Cambridge, MA: Harvard University
Allen, D.W. (1999). Transaction Costs. In B. Bouckaert and G. de Geest (eds) Encyclopedia of
Law and Economics. Available at: (last accessed 25 July 2011).
Barr, A. and Serra, D. (2010). Corruption and Culture: An Experimental Analysis. Journal
of Public Economics, 94, 862–869.
Benno, T. and Valev, N.T. (2006). Corruption and Age. Journal of Bioeconomics, 8:
Bentivogli, C., Casadio, P. and Cullino, R. (2010). I Problemi nella Realizzazione delle
Opere Pubbliche: Le Specificità del Mezzogiorno. Rivista Economica del Mezzogiorno,
1/2, 21–62.
Bijsterveld, A.J. (2007). Do ut Des: Gift giving, Memoria, and Conflict Management in the
Medieval Low Countries, Uitgeverij: Verloren.
Brayley, H., Cockbain, E. and Laycock, G. (2011). The Value of Crime Scripting:
Deconstructing Internal Child Sex Trafficking. Policing, 5, 132–143.
Bullock, K., Clarke, R.V. and Tilley, N. (2010). Situational Prevention of Organized
Crimes, Cullompton, Devon: Willan Publishing.
Cadot, O. (1987). Corruption as a Gamble. Journal of Public Economics, 23, 223–244.
Chiu, Y., Leclerc, B. and Townsley, M. (2011). Crime Script Analysis of Drug
Manufacturing in Clandestine Laboratories: Implications for Prevention. British
Journal of Criminology, 51, 355–374.
Clarke, R.V. and Cornish, D.B. (2000). Rational Choice. in R. Paternoster and R. Bachman
(eds.) Explaining Crimes and Criminals: Essays in Contemporary Criminological
Theory (pp. 23–42). Los Angeles, CA: Roxbury Publishing Company.
Copes, H., Hochstetler, A. and Cherbonneau, M. (2011). Getting the Upper Hand: Scripts
for Managing Victim Resistance in Carjackings. Journal of Research in Crime and
Delinquency. Available online:, doi: 10.1177/002242781
Cornish, D.B. (1994a). The Procedural Analysis of Offending and Its Relevance for Situational
Prevention. In R.V. Clarke (ed) Crime Prevention Studies Vol. 3 (pp. 151–196). Monsey,
NY: Criminal Justice Press.
Script analysis of corruption in public procurement 183
Cornish, D.B. (1994b). Crime as Scripts. In D. Zahm and P. Cromwell (eds) Proceedings of
the International Seminar on Environmental Criminology and Crime Analysis (Vol. 1),
Tallahassee: Florida Statistical Analysis.
Cornish, D.B. and Clarke R.V. (2008). The rational choice perspective. In R. Wortley
and L. Mazerolle (eds) Environmental Criminology and Crime Analysis, Cullompton,
Devon: Willan Publishing.
della Porta, D. and Vannucci, A. (1999). Corrupt Exchanges: Actors, Resources, and
Mechanisms of Political Corruption, New York, NY: Aldine de Grutyer.
Egidi, M. and Rizzello, S. (2003). Cognitive Economics: Foundations and Historical
Evolution. Working Paper, n. 200304. Turin: University of Turin.
Felson, M. (2006). Crime and Nature, Thousand Oaks, CA: Sage.
Felson, M. and Steadman, J.J. (1983). Situational Factors in Disputes Leading to Criminal
Violence. Criminology, 21, 59–74.
Frank, B., Lambsdorff, J.G. and Boehm, F. (2011). Gender and Corruption: Lessons from
Laboratory Corruption Experiments. European Journal of Development Research, 23,
Friedman, M. (1953). Essays in Positive Economics, Chicago, IL: Chicago University
GRECO (2009). Joint First and Second Evaluation Round: Evaluation Report on Italy,
Strasbourg: GRECO.
Guzin, B. (2009). Corruption and Intermediaries: A Game Theoretical Approach. Middle
East Technical University Studies in Development, 36, 25–49.
Koenig, G. (2009). Les Discrètes Vertus de la Corruption, Paris: Grasset.
Hill, W.F. (1990). Learning: A Survey of Psychological Interpretations, New York, NY:
La Palombara, J. (1994). Structural and Institutional Aspects of Corruption. Social
Research, 61, 325–350.
Lacoste, J. and Trembley, P. (2003). Crime and Innovation: A Script Analysis of Patterns
in Check Forgery. In M.J. Smith and D.B. Cornish (eds) Theory for Practice in
Situational Crime Prevention. Crime Prevention Studies, Vol. 16 (pp. 41–96). Monsey,
NY: Criminal Justice Press.
Lambsdorff, J.G. (2002). Making Corrupt Deals: Contracting in the Shadow of Law.
Journal of Economic Behavior and Organization, 48, 221–241.
Lambsdorff, J.G. (2005). The New Institutional Economics of Corruption, New York, NY:
Lambsdorff, J.G. (2007). Institutional Economics of Corruption and Reform, Cambridge:
Cambridge University Press.
Leclerc, B., Wortley, R. and Smallbone, S. (2011). Getting into the Script of Adult Child
Sex Offenders and Mapping out Situational Prevention Measures. Journal of Research
in Crime and Delinquency, 48, 209–237.
Lessman, C. and Markwardt, G. (2010). One Size Fits All? Decentralization, Corruption,
and the Monitoring of Bureaucrats. World Development, 38, 631–646.
Mariuzzo, F. (2005). La Tutela Cautelare negli Appalti. In M.A. Cabiddu (ed) Appalti e
Responsabilità: Da Tangentopoli agli Attuali Scenari (pp. 163–171). Milano: Franco
Merloni, F. and Vandelli, L. (2010). Prevenzione e Repressione della Corruzione: Aprire
una Nuova Pagina. In F. Merloni and L. Vandelli (eds) La Corruzione Amministrativa:
Cause, prevenzione e rimedi (pp. 1–22). Roma: Astrid.
184 Marco Zanella
Morselli, C. and Roy, J. (2008). Brokerage Qualifications in Ringing Operations.
Criminology, 46, 71–98.
Natarajan, M. and Belanger, M. (1998). Varieties of Upper-Level Drug Dealing
Organizations: A Typology of Cases Prosecuted in New York City. Journal of Drug
Issues, 28, 1005–1026.
Nell, M. (2007). Contracts Induced by Means of Bribery: Should They Be Void or Valid?
Working Paper, n. V-53-07, Passau: Universität Passau, Wirtschaftswissenschaftlichen
Neumann von, J. and Morgenstern, O. (1944). Theory of Games and Economic Behavior,
Princeton, NJ: Princeton University Press.
OECD (2008) Recommendation of the Council on Enhancing Integrity in Public
Procurement [C(2008)105], Paris: OECD.
Osborne, M.J. and Rubinstein, A. (1994). A Course in Game Theory, Boston, MA: MIT
Paldam, M. (2001). Corruption and Religion: Adding to the Economic Model? Working
Paper, n. 1999–21, Aarhus: School of Economics and Management at the University
of Aarhus.
Paoli, L. (2000). Qualitative Methodology at the Edge: Assessing Italian Mafia
Groups’ Involvement in Drug Trafficking. In G. Greenwood and K. Robertson (eds)
Understanding and Responding to Drug Use: The Role of Qualitative Research
(pp. 191–196). Luxembourg: EMCDDA.
Petrosino, A. and Brensilber, D. (2003). The Motives, Methods and Decision Making
of Convenience Store Robbers: Interviews with 28 Incarcerated Offenders in
Massachusetts. In M.J. Smith and D.B. Cornish (eds) Theory for Practice in Situational
Crime Prevention. Crime Prevention Studies, Vol. 16 (pp. 237–265). Monsey: Criminal
Justice Press.
Pizzorno, A. (1992). La Corruzione nel Sistema Politico. in D. della Porta (ed.) Lo Scambio
Occulto: Casi di Corruzione Politica in Italia (pp. 13–72). Bologna: Il Mulino.
Rijckeghem van, C. and Weder, B., (2001). Bureaucratic Corruption and the Rate of
Temptation: How Much Do Wages in the Civil Service Affect Corruption? Journal of
Development Economics, 65, 307–331.
Rose-Ackerman, S. (1997). The Political Economy of Corruption. In K. Elliot (ed)
Corruption and the Global Economy (pp. 31–61). Washington, DC: Institute for
International Economics.
Smith, M.J. (2005). Robbery of Taxi Drivers, Problem-Oriented Guides for Police
Problem-Specific Guides Series No. 34. Available at: (accessed
20 July 2010).
Spagnolo, G. (2010). Divide et Impera: Combattere la Corruzione con Collaboratori di
Giustizia e Whistle-blowers. Paper presented at the seminar Integrità e trasparenza
nella PA strumenti anti-corruzione, Roma: Scuola Superiore della Pubblica
Tompson, L. and Chainey, S. (2011). Profiling Illegal Waste Activity: Using Crime Scripts
as a Data Collection and Analytical Strategy. European Journal of Criminal Policy and
Research, 17, 179–201.
Tremblay, P., Talon, B. and Hurley, D. (2001). Body Switching and Related Adaptations in
the Resale of Stolen Vehicles, British Journal of Criminology, 41, 561–79.
Vannucci, A. (2011). The Informal Institutions of Corruption. A Typology of Governance
Mechanisms and Anti-Corruption Policies. Working Paper, no. 3. Roma: Presidenza
del Consiglio dei Ministri, Scuola Superiore della Pubblica Amministrazione.
Script analysis of corruption in public procurement 185
Williams-Elegbe, S. (2010). The Limitations of Penal Mechanisms to Fight Corruption
in Public Procurement. In M. Kreutner (ed) Practice Meets Science: Contemporary
Anti-Corruption Dialogue IACSS 2009 (pp. 143–151). Vienna: Manzsche Verlags- und
Willison, R. (2009). Overcoming the Insider: Reducing Employee Computer Crime
through Situational Crime Prevention. Available at: http://northumbria.openrepository.
com (accessed 25 July 2011).
Young, S.M.N. (2009). Why Civil Actions against Corruption? Journal of Financial
Crime, 16, 144–159.
Zanella, M. (2011). La Prevenzione Situazionale della Corruzione nella Pubblica
Amministrazione. Blog integrità, Roma: Presidenza del Consiglio dei Ministri, Scuola
Superiore della Pubblica Amministrazione.
10 Cigarette smuggling and
terrorism financing
A script approach
Alexandra Hiropoulos, Joshua D. Freilich,
Steven Chermak and Graeme R. Newman

The present chapter uses the script method developed by Cornish (1994) and
applies situational crime prevention (SCP) techniques to a complex scheme of
cigarette smuggling committed by Hezbollah supporters in the United States
(U.S.). We build upon von Lampe’s (2010) study that applied SCP to cigarette
smuggling by “routine” offenders (i.e., non-extremists) in Germany during the
1990s. Our study makes important contributions to the literature. First, few stud-
ies have applied the script method and SCP strategies to profit-driven crime, such
as cigarette smuggling, committed by political extremists. In fact, we have uncov-
ered no published research that has used the script approach to study Hezbollah
or similar extremists1 and financial terrorism.2 Some studies have used the script
approach to examine traditional terrorism, ideologically motivated violent acts
(Clarke and Newman, 2006; Freilich and Newman, 2009) or financial terrorism
committed by far-right extremists (Belli and Freilich, 2009). Clarke and Newman
(2006) examined suicide bombing in Israel and used the script method to identify
the resources and tools that were needed for successful attacks. Using this knowl-
edge, they were able to devise possible intervention strategies. Additionally,
Freilich and Chermak (2009) applied the script method to deadly encounters
between the police and American far rightists, uncovering possible intervention
points that could be used to prevent these events from turning deadly.
Second, terrorism studies usually focus solely on ideologically motivated vio-
lence and ignore possible associations between terrorism and other crime forms.
Indeed, most American terrorism databases and definitions, such as those of the
Federal Bureau of Investigation (FBI), require “terrorist acts” to use “force or vio-
lence” and thus exclude non-violent offenses such as financial crimes (Belli and
Freilich, 2009; Freilich, Chermak, and Simone, 2009; Hewitt, 2003; Schmid and
Jongman, 1988; Smith, 1994; Weinberg, Pedahzur, and Hirsch-Hoefler, 2004).
Currently, many extremists commit crimes that are typically associated with
profit-driven crime (Belli, 2011; Belli and Freilich, 2009; deKieffer, 2008;
Dishman, 2005; Hamm, 2007; Horgan and Taylor, 2003; Passas, 2003). Al Qaeda,
Hezbollah, Hamas, and other similar extremists have committed crimes such as
credit card and financial fraud to raise funds for their ideological goals (Belli, 2011;
Kane and Wall 2005; Picarelli and Shelley, 2007; Smith and Damphousse, 2003).
Also, far-right extremists associated with the tax protest movement employ and
Cigarette smuggling and terrorism financing 187
endorse tax evasion and related fraudulent activities as a form of anti-government
protest (Belli, 2011; Belli and Freilich, 2009; Pitcavage, 1999).
This chapter applies Cornish’s (1994) “script” analysis to a case study of ciga-
rette smuggling committed by an extremist ring of Hezbollah supporters which
involved contraband cigarette trafficking from North Carolina to Michigan
whose revenues were transferred to Hezbollah sites in Lebanon (Shelley and
Melzer, 2008).
Hezbollah is a Shiite military, political, and social organization based in
Lebanon that is regarded as a terrorist group by the U.S. Hezbollah committed
acts of terrorism against American targets in Lebanon in the 1980s and 1990s.
These attacks included separate bombings of the U.S. military’s marine bar-
racks, the American embassy and the embassy annex, as well as the kidnapping
of Americans, including the Central Intelligence Agency’s (CIA) bureau chief
who was subsequently murdered. Collectively, these strikes claimed over 300
American lives. Hezbollah has also committed attacks against Israeli targets
(Jaber, 1997).
First, we describe the danger posed to the United States by financial terrorism,
especially crimes committed by Hezbollah, Hamas, Al Qaeda, and similar extrem-
ists, and identify the threat posed by cigarette smuggling. Second, we outline the
case study of the cigarette smuggling scheme. After Cornish’s “script” method is
explained, we analyze the open source information on the situational conditions
of criminal behaviors in the present case. Our goal is to uncover the facilitating
factors that provided the opportunities for these crimes to be committed and to set
forth specific SCP techniques that could be implemented to block such opportuni-
ties. Finally, the paper concludes with suggestions for future research.

Importance of the problem

Worldwide, illegal cigarette trafficking is a multibillion dollar a year crime
phenomenon (von Lampe, 2010). According to U.S. Immigration and Customs
Enforcement (ICE) and the Bureau of Alcohol, Tobacco, Firearms and Explosives
(ATF), cigarette smuggling is a significant problem in the U.S., though its extent
is impossible to measure with certainty due to its clandestine nature.
Owing to the fact that smuggled cigarettes are not taxed, federal and state
revenues are lost. In the U.S., tobacco product manufacturing is concentrated in
Georgia, Kentucky, North Carolina, and Virginia, with about 500 billion genu-
ine cigarettes produced in 2003 alone (GAO, 2004). Tobacco diversion within the
country is profitable because of the disparity among the states’ excise taxes. With
each tax rate hike imposed by federal, state, and local governments, the incentive to
profit by evading payment of taxes rises (ATF, 2009). Many states have increased
cigarette taxes, resulting in a large difference between the wholesale price and the
price paid by consumers at the retail level. This disparity creates potential illicit
profits of $7 to $13 per carton of cigarettes (GAO, 2004). High state tobacco excise
taxes make it profitable for individuals and groups to risk crossing state borders
to smuggle and engage in other illegal sales activities such as smuggling across
188 A. Hiropoulos et al.
international borders, avoiding taxes by pretending to export products but illegally
selling them in the United States, producing counterfeit products, selling products
without tax stamps or with counterfeit stamps, and selling products illegally over
the Internet. For example, solely by purchasing cigarettes in a low tax state and
reselling them in a high tax state, a seller can make a profit up to $23,000 on ten
cases of cigarettes (a car load), up to $90,000 on 50 cases (a van load), and up to
$465,000 for 200 cases (a small truck load) (ATF, 2009).
Indeed, it is estimated that in the U.S., tobacco diversion leads to over $5
billion in lost revenue annually due to unpaid excise taxes (GAO, 2004). The
diversion of tobacco can occur anywhere on the production or supply chain—
consequently, manufacturers, wholesalers, and retail outlets have been involved
in diverting tobacco products. Counterfeit and authentic contraband tobacco prod-
ucts are available through illegal “black market” sources, through the Internet,
and at legally operated retail locations (GAO, 2004). Importantly, as in the pre-
sent case study, some cigarette smugglers have ties to terrorist groups. Domestic
extremists, including supporters of Hezbollah and Al Qaeda, pose a threat to the
economic system in the U.S. (Belli, 2011; Freilich, Chermak, Belli, Gruenewald,
and Parkin, 2012). Recent research suggests that terrorists increasingly resort to
profit from crimes for self-financing purposes (e.g., smuggling of counterfeit
or stolen products, gemstone trade, and money laundering; see: Hamm, 2007;
Picarelli and Shelley, 2007; and Shelley and Picarelli, 2005).
Terrorist organizations need funds for various reasons even though specific
terrorist actions are not necessarily expensive. For example, the 9/11 Commission
estimated that the costs for the 9/11 attacks were between $400,000 and $500,000
(National Commission on Terrorist Attacks upon the United States, 2004).
However, it is costly to build and maintain an organizational capacity that, in
turn, determines what type of targets and criminal activity terrorists can choose
(Canadian Centre for Intelligence and Security Studies, 2006). Complex terrorist
networks must finance security, operations, intelligence, propaganda, recruitment,
training, communications, bribery, weapons, travel, forged documents, and living
expenses. Al Qaeda’s operating budget is estimated to be between 20 and 30 mil-
lion dollars per year with 10 percent used for specific operations and 90 percent
for infrastructure. Al Qaeda also uses funds to support other groups and enhance
their presence in key regions (Noble, 2003).
To capture the complexity of cases of financial crime by domestic extrem-
ists, the U.S. Extremist Crime Database (ECDB) created the concept of “financial
scheme.” The financial scheme is comparable to the concept of a violent terrorist
“incident” as it is the main unit of analysis. The financial scheme is defined as an
illicit financial operation involving a set of activities (i.e., techniques) carried out
by one or more perpetrators to obtain unlawful gain or other economic advantage
through the use of deliberate deception (Belli, 2011). The ECDB has identified
over 600 financial schemes that were committed by far-rightists and other extrem-
ists in the U.S. since 1990. Since data collection is ongoing and will then need to
be validated and cleaned, these numbers may change. The total financial loss of
these schemes is over $600,000,000. Far-rightists committed over 500 schemes
Cigarette smuggling and terrorism financing 189
and their most common scheme type was tax avoidance (68 percent) (Freilich
et al., 2012).
Al Qaeda, Hamas, Hezbollah, and supporters of similar extremist movements
perpetrated over 100 schemes that involved a variety of financial and non-financial
offenses such as smuggling, money-laundering, material support to terrorists, iden-
tity theft, and immigration fraud (Freilich et al., 2012). These schemes involve
cross-border activities, such as theft and smuggling of high-value commodities
(e.g., pharmaceuticals, infant baby formula, computer hardware, and cigarettes) to
jurisdictions where they can be sold at higher prices. We next discuss the source of
the data for our case study and provide a brief summary of this case.

Data: contraband cigarette smuggling case study

The present cigarette smuggling case was selected from the ECDB3 (see Belli,
2011; Chermak, Freilich, Parkin, and Lynch, 2012; Freilich, Chermak, and Caspi
2009; Freilich et al., 2012; Gruenewald, 2011; Gruenewald and Pridemore, 2012)
to illustrate the efficacy of the script approach to combat these crime types by
political extremists (von Lampe, 2010). The ECDB uses open sources to track
ideologically motivated and routine/non-ideological violent and financial crimes
committed by far-rightists, supporters of Al Qaeda, Hezbollah, Hamas and similar
extremist groups, and animal and environmental rights extremists in the United
States. It collects information on multiple units of analysis (events/schemes, sus-
pects, victims, targets, and business entities).
The selected case study was “ideal for getting a clear fix on the relevant empir-
ical and theoretical issues” (Snow and Trom, 2002, p. 158; see also Creswell,
2007). Data were collected from open sources, researched in 22 search engines
to uncover all publicly available information (1) Lexis-Nexis; (2) Proquest; (3)
Yahoo; (4) Google; (5) Copernic; (6) News Library; (7) Infotrac; (8) Google
Scholar; (9) Amazon; (10) Google U.S. Government; (11) Federation of American
Scientists; (12) Google video; (13) Center for the Study of Intelligence; (14) Surf
Wax; (15) Dogpile; (16) Mamma; (17) Librarians’ Internet Index; (18) Scirus;
(19) All the Web; (20) Google News; (21) Google Blog; and (22) Homeland
Security Digital Library.
These searches uncovered a large amount of information from a variety of
sources, which is not surprising considering the notoriety of the present case. We
reviewed over 40 documents such as court records including docket proceedings
and bills of indictments (n=3), scholarly publications (n= 2), books (n= 2), news/
journalistic accounts (n= 27), watch-group publications (n= 4), and other non-
scholarly Internet sites (n=4). The variety of sources from which information
was uncovered allowed us to use data triangulation and thus increase the likeli-
hood that there was no systematic bias one way or the other in the descriptions
of this specific case (Freilich and Pridemore, 2006, 2007; Yin, 2003). The open
source information was reviewed and used to write narratives on the activities
and crimes that were committed to carry out this scheme. A brief summary of the
case follows.
190 A. Hiropoulos et al.
Mohamad Youssef Hammoud led a cigarette smuggling ring that operated in
North Carolina and trafficked cigarettes between North Carolina and Michigan.
Exploiting the difference in taxation rates, the cell amassed more than $2 million
in illegal funds. This money was used to provide financial support to Hezbollah in
Lebanon as well as to purchase night vision goggles, cameras and scopes, survey-
ing equipment, global positioning systems, mine and metal detection equipment,
video equipment, advanced aircraft analysis and design software, laptop comput-
ers, stun guns, radios, mining, drilling and blasting equipment, radars, ultrasonic
dog repellers, and laser range finders.
In early 1995, a sheriff’s deputy moonlighting as a security guard in Iredell
County, North Carolina, became curious as cars with out-of-state license plates
approached a tobacco store. Three Arab-speaking men each purchased 299
cartons of wholesale cigarettes, just under the limit for a legal cigarette sale
transaction, paid for with cash removed from paper shopping bags. The officer
became suspicious and alerted the ATF of the suspicious transaction (Shelley
and Melzer, 2008).
This led to a four-year FBI Joint Terrorism Task Force investigation that uncov-
ered a cigarette smuggling enterprise involving two-dozen people, some of whom
had connections to Hezbollah operatives in Lebanon, ultimately resulting in a
Rackateer Influenced and Corrupt Organizations Act (RICO) and terrorist financing
indictment (Breinholt, 2003). The use of the RICO statute, traditionally used against
criminal organizations, highlights the convergence between organized crime and
terrorist organizations. The charges of conspiring to provide material support to
Hezbollah under 18 U.S.C. § 2339B were premised on funds the defendants sent
to Hezbollah, and a military procurement program in which Hezbollah operatives
in Beirut tasked North America-based adherents to purchase and ship a variety of
dual-use items purchased in the United States and Canada. Following a series of
guilty pleas by the other defendants, Mohamad Hammoud and his brother Chawki
Hammoud were convicted in the first ever jury trial of material support to designated
foreign terrorist organizations (18 U.S.C. § 2339) in American history (Breinholt,
2003). Based on the terrorism sentencing enhancement, Mohamad Hammoud was
sentenced to 155 years imprisonment on February 28, 2003. His brother Chawki
Hammoud received 51 months imprisonment.
Overall, Mohamad Hammoud was charged with sale of contraband cigarettes,
various immigration violations, money laundering, mail fraud, credit card fraud,
racketeering, and conspiracy to provide material support to a foreign terrorist
organization. Other crimes linked to the cell include bank fraud, bribery, wire
fraud, identify theft, interstate transportation of stolen property, money launder-
ing, and defrauding the Small Business Administration.

Cornish’s (1994) “script” approach, specifically the “universal script,” was uti-
lized to identify intervention points that could be used to implement strategies
to prevent cigarette smuggling. Cornish and Clarke (2002: 47) explain that “all
Cigarette smuggling and terrorism financing 191
crimes … involve … chains of decisions and actions, separable into interdepend-
ent stages, involving the attainment of sub-goals that serve the overall goals of the
crime.” Cornish’s universal script consists “of scenes arranged into a sequential
order which further the overall action, offer standardized guidelines for construct-
ing scripts at the track-level, whatever the state of knowledge about the offense in
question” (pp. 160) (see Table 10.1). We reviewed the open source data to create
chronological listings of the steps that were necessary to carry out the key compo-
nents of the scheme in our case study (Yin, 2003). This is especially useful where
the goal is to identify “how” the crime was successfully committed as opposed to
focusing on “why” it was committed, which is more likely to implicate methodo-
logical concerns (Freilich et al., 2009).
The script method is a logical outcome of the rational choice approach to under-
standing and explaining crime. It has been used in a variety of crimes, including
suicide bombing (Clarke and Newman, 2006), deadly encounters between law
enforcement and American far-rightists (Freilich and Newman, 2009), credit card
and check fraud (Lacoste and Tremblay, 2003), and child sexual abuse (Leclerc,
Wortley and Smallbone, 2011). Importantly, the script approach has been applied
to profit motivated cigarette smuggling, committed by non-extremists in Germany
(von Lampe, 2010). Again, the advantage of the script method is that it allows for
the detailed examination of the operational aspects of the case and offers points
of intervention and tactical suggestions for intervention. It highlights the oppor-
tunities for altering the outcome and illustrates where strategies could be used to
prevent such incidents (Freilich and Chermak, 2009).

Situational crime prevention (SCP)

The script method is the tool that many researchers and policymakers use to iden-
tify the specific SCP strategies that could be implemented to prevent or reduce
a specific crime problem. Rather than examining the so-called “root” causes of
crime, SCP analyzes factors through a prevention lens. Intervention techniques
are devised to manipulate situational factors to reduce crime (Wortley, 2002).
SCP is based on the assumption that criminals are rational goal-oriented individu-
als who seek to satisfy specific needs and desires by maximizing their personal
gains while minimizing the costs (Cornish and Clarke, 2003). It involves 25
techniques that are aimed at reducing opportunities for offenders to commit their
crimes. Though originally developed to deal with traditional crimes such as bur-
glary and theft, the techniques have been successfully applied to a broad range
of crimes, including identity theft, e-commerce crime, and child sexual abuse
(Newman and Clarke, 2003). The aim of the techniques is to broaden the reper-
toire of possible preventive responses in dealing with a specific form of crime.
They seek to modify the circumstances that encourage offenders to commit crime
in specific situations, and that make it possible for them to carry through their
crimes to completion.
The five main approaches that have been shown to change the decision-making
process of the offender or would-be offenders are: increase the effort, increase
192 A. Hiropoulos et al.
the risks, reduce the rewards, reduce provocations, and remove excuses. The first
three approaches are most pertinent to our case of cigarette smuggling and relate
to “hard” SCP strategies. These techniques seek to prevent the offender from
accomplishing an illegal act (Clarke, 1997; Wortley, 1997; 2002). Considerations
of risks, efforts, and rewards provide the basis for the decision-making process
(that could lead a potential offender to choose to commit a crime) and these strat-
egies are essential for the development of effective crime prevention strategies
(Clarke, 1997). Increasing the effort seeks to make criminal acts more difficult to
accomplish. If tasks are more difficult to accomplish, criminals may either give
up or take much longer to accomplish their goals (Clarke and Newman, 2006).
Increasing the risks of crime makes it easier to detect the offender. Reducing the
rewards of crime involves implementing as many protective measures as possible
to make criminal activities less unsuccessful than they would otherwise be.
The latter two approaches, reducing provocations and removing excuses, pre-
vent prompts and cues that increase a person’s proclivity to commit a crime. These
may also prevent a crime from occurring during a specific incident (Wortley, 1997;
1998; 2001; 2002). Wortley has called such techniques “soft” techniques and has
developed further techniques outside of the situation itself that can be implemented
to prevent crime, such as community outreach or use of community policing.

As Clarke and Newman (2006) noted, appropriate techniques can be applied only
after specific crime types are identified. This will insure that the correct degree of
specificity required for each situation is achieved (Clarke and Cornish, 1985; von
Lampe, 2010). Our review of the open source data demonstrated that the cigarette
smuggling scheme was complex (involving multiple acts, actors, agencies, and
jurisdictions over an extended time period) and involved multiple components.
Following Clarke and Cornish’s adage, we treated each component separately
and created a timeline of the steps necessary to carry out the crimes, viewing them
as a series of logistical steps that are instrumental in the commission of cigarette
We next discuss the timeline of each component followed by our application
of the techniques. Table 10.1 maps the steps, actions, and methods used in the
cigarette smuggling onto the universal script. We note though that this table is
schematic because, as is often the case in real life, the sequence of decisions is
more circular as opposed to linear. Our review of the open sources generated the
following chronology for this case.

Timeline of events
Mohamad Hammoud made several visa applications in Damascus, Syria which
were denied by the U.S. State Department. He then traveled to Venezuela in 1992
Table 10.1 Application of script and situational controls

Script scenes/functions Step Action Tools (actually used) Facilitating conditions Situational controls
Entry Enter the U.S. Get a visa Go to Venezuela and get a Network of terrorist Increase the
fraudulent visa stamp operatives in South Effort: Tamper-proof visas;
America Tough ID authentication
Precondition Obtain access Residency Fraudulent visa Lax ID vetting Increase the
Fraudulent marriages (3) Lax INS monitoring Effort: Tamper-proof visas;
Tough ID authentication
Instrumental Acquire adequate Use fraudulent Stolen credit cards Availability of returning Increase the Effort: ID
precondition funding bank accounts Student IDs foreign students IDs requirements for bank
accounts; Legislation
Instrumental Establish business Acquisition of Shell companies Availability of cigarettes Reduce Rewards: Eliminate
precondition goods Aliases Disparity in state taxing taxes or make uniform;
Cash and stolen credit cards levels for cigarettes Labeling

Instrumental Transportation Establish supply Rented vehicles Network of friends and Removing Excuses:
Actualization chain Plausible females as relatives Community outreach
“vacationing” drivers programs; Alert
Instrumental Storage Obtain Credit cards No monitoring of Increase the Risk: Increase
Actualization warehouses Rented mini-warehouses warehouse use by monitoring; Reward
landlord vigilance: publicity/
educational campaign
Script scenes/functions Step Action Tools (actually used) Facilitating conditions Situational controls
Instrumental Distribution Operate Fraudulent loan from Small Family ties Removing Excuses:
Actualization businesses Business Administration Community outreach
programs; Alert
Doing Cash smuggling Move money to Religious, political and Lax monitoring of Reducing Rewards:
terrorist groups family networks import/export of cash Removing target with
enhanced customs
Post-Condition Remain in U.S. Maintain a legal Use accounts separated
façade from illegal operations
Exit Leave U.S. [Not carried out]
Cigarette smuggling and terrorism financing 195
where he obtained a fraudulent visa stamp on his Lebanese passport, and flew to
the United States. In some South American countries, groups such as Hamas and
Hezbollah, considered to be terrorists by the U.S., have established support bases
that sustain their worldwide operations (Shelley and Melzer, 2008). According
to the Drug Enforcement Agency (DEA), terrorist operatives associated with
Hezbollah generate significant income from contraband, including drugs in several
Latin American countries, to support their organization in Lebanon (GAO, 2003).
When Hammoud landed at New York’s JFK airport, authorities detected the
counterfeit stamp and detained him. Hammoud immediately applied for asylum,
claiming persecution from Hezbollah. In December 1993, an immigration judge
denied asylum and Hammoud filed an appeal. Shortly after filing his appeal, he
moved to North Carolina.
As had several other cell members, Mohamad Hammoud attempted to improve
his immigration status and establish residency by entering into three different
fraudulent marriages. The first marriage was to a Sabina Edwards in December,
1994. The Immigration and Naturalization Services (INS) denied the marriage
because the documentation presented by Hammoud and a person claiming to be
Ms. Edwards was determined to be fraudulent. His second fraudulent marriage
was to Jessica Wadel in May, 1997. His third fraudulent marriage was to Angela
Tsioumas on September 12, 1997, even though he was not yet divorced from his
second wife. Unaware of his previous fraudulent marriage, the INS granted him
the status of Conditional Lawful Permanent Residence based on his marriage to
Tsioumas. The couple moved to Charlotte, North Carolina, where Hammoud
was reunited with his brother Chawki Youssef Hammoud (who, after marry-
ing Jessica Yolanda Fortune in 1994, was also granted permanent residency)
and with Said Mohamad Harb, a childhood friend from Lebanon (Shelley and
Melzer, 2008).

The cell led by Mohamad Hammoud was bound together by family ties, reli-
gion, criminal activities, and an association with, and sympathy for, Hezbollah.
Mohamad Hammoud first became involved with Hezbollah when he joined the
group’s militia at the age of 15 in Beirut, Lebanon. In fact, many of the North
Carolina cell members were linked to the same neighborhood in Beirut and
most knew each other long before immigrating to the U.S. Once in the United
States, members met weekly in private homes where Mohamad Hammoud
played speeches of the late Iranian cleric Ayatollah Khomeini, distributed
propaganda materials, showed videos, and encouraged attendees to donate
money to Hezbollah. It was during these meetings that members developed
the complex scheme to traffic cigarettes from North Carolina to Michigan.
Moreover, these family, community, and religious ties enabled the smugglers
to sell illegal cigarettes at a network of small convenience stores in Michigan
(GAO, 2003).
196 A. Hiropoulos et al.
Purchase of contraband cigarettes
A member of the cell would purchase bulk cigarettes from a wholesale distributor
in North Carolina, using the account of a shell company or an alias. Cell members
paid for the cigarettes in cash or used credit cards to make purchases through the
alias and shell companies. Payments for the trafficked cigarettes were later wired to
accounts held by Mohamad Hammoud and other members, then disbursed to other
accounts held by aliases. Different credit cards were used to purchase the cigarettes
and to pay the transport and storage expenses, thus not directly linking the purchase
of the cigarettes to their transportation or storage (Shelley and Melzer, 2008).

Transportation, storage, and distribution of contraband cigarettes

During the transportation phase of the scheme, active members of the terrorist
cell, their “spouses” and additional recruited drivers transported the cigarettes
from North Carolina to Michigan in rented vans and trucks. To evade detection
and cover up their illegal activities, the cell decided to hire American women
as drivers and placed bicycles on the back of vehicles to make it look as if the
women were vacationing.
To store the large quantities of cigarettes acquired, mini-warehouses were
rented in North Carolina and Michigan. The cell also operated a variety of busi-
nesses as fronts for the cigarette sales. The businesses would either account for
the large quantities of cigarettes purchased or provide a means to move the illicit
money into the legal economy. In 2000, Hammoud and Tsioumas applied for and
obtained a fraudulent $1.6 million loan from the Small Business Administration
to build a BP gas station in North Carolina. The gas station and other legitimate
businesses were essential for the cigarette smuggling operations. The gas station
provided a distribution point to sell the contraband cigarettes, legitimate revenues
for the group, and a way to launder the illegal profits.

Financing mechanisms
Financial investigations linked more than 800 bank accounts and multiple identi-
ties to the cell led by Mohamad Hammoud. Many of the identities were adopted
from departing international students who had acquired driver’s licenses, social
security cards, credit histories, and the necessary immigration documents (Shelley
and Melzer, 2008). Said Harb, a cell member who turned state’s evidence, admit-
ted that he would ‘max out’ credit cards that he obtained with a particular identity,
declare bankruptcy, and then refrain from using the identity for another seven
years until the credit record was expunged. Harb testified at trial that he made at
least $150,000 in profit per identity. However, not all of the stolen identities were
used for criminal purposes. Cell members kept some of their identities crime-free
as a means for them to disappear into society, evade detection, and appear as if
they were law-abiding individuals. The illegal profits were then transferred to
Hezbollah by wire transfers or by cell members carrying cash to Lebanon.
Cigarette smuggling and terrorism financing 197
Application of SCP techniques
The use of the script method allowed us to analyze the situational conditions
of the criminal activities in the present case. As can be seen in Table 10.1, our
analysis uncovered the tools and facilitating factors that provided the opportuni-
ties for these crimes to be committed. We were able to identify specific SCP
strategies that could be implemented to prevent the crimes. One advantage of the
script approach is that it usually uncovers a variety of techniques that can be used
(Felson and Clarke, 1997). Each community should select for implementation the
techniques they are most comfortable with, taking into account concerns such as
public safety, community rights, and individual liberty. Communities should cre-
ate standards as to the risks involved and the degree of harm posed.
The following section describes the SCP strategies that were identified as
useful for the development of effective crime prevention strategies for cigarette
smuggling committed by political extremists.

It is clear that there is a need for better information sharing. This is a complex
scheme involving multiple agencies and individuals. Because the response could
come from law enforcement, federal agencies or even the private sector, there is
a need for good information sharing among them. “Hard” SCP techniques can be
applied to make it more difficult for suspects to obtain fraudulent visas and mar-
riage licenses. Increasing the effort can be accomplished by utilizing tough ID
authentication procedures. These interventions could be costly though because of
the novel technologies involved and the costs for training personnel in these new
procedures. Another strategy involves tightening procedures for issuing docu-
ments such as marriage licenses and having more effective data systems where
information can be efficiently shared. These measures will increase the effort
needed to obtain such documents but also increases the risks of getting caught.
It is important to note though, that offenders can be innovative and some will
adapt—this is a longer-term process than displacement—and may take advan-
tage of new criminal opportunities presented by these interventions. For example,
the increased use of effective data-sharing systems over time could translate into
adaptations of offenders into hacking. Decisions about tightening procedures for
marriage licenses should consider gains in public safety in reducing these crimes,
and possible costs to society such a tightening of procedures could produce.

Terrorist networks represent a significant challenge for U.S. efforts to monitor
terrorists’ activities. In particular, these networks are difficult to identify and pen-
etrate because they are close knit and based on trust. The closeness and high
degree of trust between parties to terrorist financing networks are often based on
long-standing ethnic, family, religious, or organized criminal ties. Investigators
198 A. Hiropoulos et al.
who seek to monitor such networks rely on developing inside sources of infor-
mation. But the high degree of trust within the networks poses challenges for
recruiting informants and conducting undercover operations. Also, language and
cultural barriers can increase the difficulty of accessing such networks by imped-
ing communication between government officials and parties to the networks
(GAO, 2003).
Though some immigrant communities in the U.S. have been found to facilitate
terrorist activities, as was the case with Mohammad Hammoud’s cell, immi-
grant communities are also victims of terrorism. Arab and Asian communities
with Muslin ties especially are at risk of victimization and report being fearful
of being targeted by authorities with additional checks as well as being treated
with hostility by local populations. Protecting these communities while simulta-
neously ensuring that they do not support terrorism is undoubtedly a difficult task
(Newman and Clarke, 2008).
According to Wortley (2001), “soft” SCP techniques can prevent prompts and
cues that increase a person’s proclivity to commit a crime. In this case, commu-
nity outreach tactics can be especially useful in “reducing excuses” to commit
crime. There are various ways that community policing can be successfully
implemented in communities, including migrant ones. For example, community
policing officers can be designated to work exclusively with immigrant communi-
ties. To reach a larger constituency, ethnic radio and television as well as religious
institutions and employers can be used to communicate with immigrant commu-
nities. Police can also employ more translators and ensure that police materials
are available in foreign languages. Additionally, police officers should be trained
to effectively communicate with the different elements in immigrant communi-
ties. Immigrant leaders can be involved in designing and implementing training
programs for police officers in order to make training more culturally relevant.
Community advocates can also be informed of police departments’ roles and the
media should accurately report the policies and the dialog between police and
immigrants (Freilich and Chermak, 2009; Newman and Clarke, 2008). Finally,
public information/awareness campaigns could reduce excuses and appeal to peo-
ples’ consciences by highlighting the negative consequences of the illegal acts
(Freilich and Chermak, 2009; von Lampe, 2010).

Purchase, transportation, storage, and distribution

The general tobacco industry pattern is as follows: First, manufacturers deposit
cigarettes in warehouses and pay the federal cigarette excise tax of 39 cents per
pack of 20 cigarettes. Next, the cigarettes move to a wholesaler stamping agent.
Stamping agents have authority from states to affix to cigarette packs evidence
of the payment of the appropriate state tax. Once the stamping agent receives
the unstamped cigarettes, it usually does not redistribute them until the state tax
stamp has been affixed. After this is done, the stamping agent usually sells the
cigarettes to other wholesalers (non-stamping agents), distributors, and retailers.
As of 2004, the state excise tax rates for a pack of 20 cigarettes ranged from
Cigarette smuggling and terrorism financing 199
2.5 cents in Virginia to $2.05 in New Jersey. Importantly however, despite 47
states requiring that tax stamps be placed on cigarette packages as evidence that
state cigarette taxes were paid, North Dakota, North Carolina, and South Carolina
do not require tax stamps (GAO, 2004).
When Michigan raised its cigarette tax in 1994, its new 75 cents per pack rate
was the highest state cigarette tax in the country. More importantly, Michigan was
also the only high-tax state that did not have a cigarette tax stamp. In other words,
no stamp or other indicia was put on the packs once the distributor paid the related
cigarette tax to the state and there were no stamps on the packs sold to consum-
ers in Michigan to show that they were legal packs on which the state cigarette
tax had already been paid, as required by law. Soon afterwards, North Carolina,
with a cigarette tax of only 5 cents per pack, eliminated the tax stamps on packs
of cigarettes sold within its boundaries. As a result, it was easy for smugglers to
take low-cost cigarettes purchased in North Carolina, bring them into Michigan
for illegal resale, and pocket a profit per pack of up to the difference between the
two states’ tax rates. There was no way to distinguish the smuggled, contraband
cigarettes from legal cigarettes; and retailers in Michigan could easily sell smug-
gled cigarettes to unsuspecting customers with little risk of being caught by law
enforcement officials (Boonn, 2011).
An obvious solution would be to eliminate taxes on cigarettes altogether. This
is highly unlikely, however. As von Lampe (2010: 36) noted:

One might be tempted to reduce the problem to one of political will and argue
that all it would take to reduce or to entirely eliminate the illicit cigarette
trade is to reduce or eliminate the taxes on cigarettes …. [but] the problem
is more … difficult … Political support for high tobacco taxes seems to be

Another obvious solution, though also difficult to implement, is to encourage

governments and/or states to adopt uniform taxing regimes, thereby eliminating
the opportunity to profit from cigarette smuggling (von Lampe, 2010).
The possibility of profits through cigarette smuggling has attracted crimi-
nals, including international and domestic organized crime groups, because it is
considered to be a relatively low risk crime with penalties that are lower than the
penalties for smuggling drugs (GAO, 2004). The World Health Organization has
recommended aggressive enforcement of anti-smuggling legislation and con-
sistent application of penalties to deter smugglers. Through these tight controls
on smuggling, governments and states can improve revenue yields from tobacco
tax increases (WHO, 2011). In the U.S., the Prevent All Cigarette Trafficking
(PACT) Act has been effective since June 29, 2010. This act broadened the
definition of what constitutes contraband cigarettes by lowering the Contraband
Cigarette Trafficking Act (CCTA) quantity for nontaxed cigarettes to be con-
sidered contraband from 60,000 to 10,000. Lowering the contraband cigarette
threshold allows ATF to open more investigations and seek more federal felony
prosecutions of cigarette smugglers (GAO, 2004).
200 A. Hiropoulos et al.
The SCP method of reducing the rewards of illegal behavior can be accom-
plished through the technique of identifying property. An effective labeling
system would assist authorities in determining the origin of tobacco products and/
or their legal status (von Lampe, 2010: 52). For example, states could require
basic labeling such as, “Sales only allowed in state X” or Fiscal marking/tax
stamp (e.g., Duty Paid or Tax Paid) or color-specific mark. These marks or stamps
are registered with appropriate authorities to identify the product’s domestic legal
or tax-paid status (ATF, 2002). The cigarette companies themselves are in the best
position to lobby for these stamp laws, perhaps with some assistance from federal
Additionally, legally requiring a tax stamp on cigarettes will increase the
effort because a number of licenses are required to become a stamping agent.
Cigarettes then cannot be redistributed until the state tax stamp has been affixed.
Alternatively, the collection of tobacco taxes at the time of manufacture, includ-
ing on tobacco products destined for export, can enhance enforcement efforts
(ATF, 2002).
Increasing the efforts can also be accomplished through the technique of
controlling tools, in this case, further requirements for distributor licenses. For
example, to be granted a license, an applicant must demonstrate a record of busi-
ness integrity or, as was often described, be a “fit and proper” person. This could
include bonding, a record of good business practices (including the sale and
distribution of products within the normal channels of trade), and the proper col-
lection of revenue. Also, a past record in a tobacco related crime could exclude
Record keeping is another example of controlling the tools used to commit
crime. Record keeping is an important element of a licensing regime and can
increase the efforts as well as the risks of illegal behavior. Strict record keeping
can become essential to qualifying for, and maintaining, a license since it can help
to monitor and assess tobacco growing and manufacturing output and formulating
and implementing tobacco control policies (GAO, 2004).
In addition to licensing and record keeping, other mechanisms to control tools
can be utilized to increase the effort of illegal behavior. Permits and registration
could be required in which the privileges and burdens on the permit holder or reg-
istrant would not be as great as those for licensees (ATF, 2002). There would be a
need to provide appropriate education and outreach, as well as relevant interven-
tion programs to help support any record-keeping system. In other words, because
some individuals in the production/distribution system may have different levels
of experience with respect to bookkeeping requirements, part of the regulatory
scheme should provide for adequate training of these individuals. Some essential
elements of a record-keeping system include access to information and regular
auditing by competent Government authorities, requiring credible and verifiable
records for which the record holder would be held strictly accountable, and stand-
ardization of the type of information to be reported (ATF, 2002).
Finally, the risks of such illegal behavior can be decreased through monitor-
ing, documenting, and controlling the movement of tobacco products. Exercising
Cigarette smuggling and terrorism financing 201
some level of control over the movement and distribution (points of origin, transit,
and destination) of tobacco products is a key element to managing contraband
tobacco. An effective track and trace system would permit authorities to inde-
pendently trace the movement of tobacco products from the time of manufacture
through to the point where all duties and taxes have been collected, and identify
the customers throughout the distribution chain (ATF, 2002; von Lampe, 2010).

Financing mechanisms
Increasing the effort of obtaining false identities can be accomplished by requir-
ing extensive ID requirements for opening bank accounts. However, the law
enforcement community has long suspected that some terrorist organizations use
bulk cash smuggling to move large amounts of currency. Bulk cash smuggling is
an attractive financing mechanism because U.S. dollars are accepted as an inter-
national currency and can always be converted. Since there is no traceable paper
trail or a third party (such as a bank official) to become suspicious of the transac-
tion, the terrorist has total control of the movement of the money. In the U.S., bulk
cash smuggling is a money laundering and terrorism financing technique that is
designed to bypass financial transparency reporting requirements. Often the cur-
rency is smuggled into or out of the U.S. concealed in personal effects, secreted in
shipping containers, or transported in bulk across the border via vehicle, vessel,
or aircraft (GAO, 2003).
The USA PATRIOT Act of 2001 significantly expanded U.S. law enforce-
ment’s ability to deter, investigate, and prosecute cases of terrorist financing. In
2002, the United States enacted the Suppression of the Financing of Terrorism
Convention Implementation Act, which implements the requirements of the 1999
International Convention for the Suppression of the Financing of Terrorism.
Among its provisions, this act makes it a crime to provide or collect funds with
the intention of using the money for terrorist activities. Deterring terrorists’ use
of alternative financing mechanisms falls within the overall U.S. interagency
framework of plans, agency roles, and interagency coordination mechanisms
designed to combat terrorism. In general, the National Security Council man-
ages the overall interagency framework. The National Security Council heads the
Counterterrorism Security Group, which is composed of high-level representatives
(at the Assistant Secretary level) from key federal agencies that combat terror-
ism. To implement directives and strategies, various federal agencies are assigned
key roles and responsibilities based on their core missions. Numerous compo-
nents of the Departments of Justice, the Treasury, State, Homeland Security, and
other agencies participate in efforts to combat terrorist financing. In addition, the
intelligence community plays a significant role (GAO, 2003). However, commu-
nication and cooperation between the various arms of government are essential to
enhancing the effectiveness of tobacco control measures.
Implementation of the 2002 National Money Laundering Strategy, which
ostensibly directs the U.S. government’s resources against money laundering and
terrorist financing, has proven to be challenging, partially owing to the number
202 A. Hiropoulos et al.
of competing priorities. Increased emphasis on combating terrorism and terror-
ist financing since the September 11 terrorist attacks has placed greater urgency
on pre-existing responsibilities for some agencies. New laws such as the USA
PATRIOT Act are generally recognized as assisting U.S. law enforcement efforts
but also increase the workload of agencies. While the FBI is the lead agency on
terrorist financing investigations, all agencies have an inherent responsibility to
aid in this effort. However, some agency officials note that new tasks sometimes
compete with traditional roles or increase workloads, creating a strain on their
resources, which could slow the sharing of potentially useful information. As a
result, agencies may fail to fully utilize existing laws or fully implement strategic
efforts in a timely manner (GAO, 2003).
Owing to the criminal nature of terrorists’ use of alternative financing
mechanisms and the lack of systematic data collection and analysis, the extent
of terrorists’ use of alternative financing mechanisms in the U.S. is not known.
United States law enforcement agencies, and the FBI specifically, do not system-
atically collect and analyze data on alternative financing mechanisms. The lack of
such data hinders the FBI from conducting systematic analysis of trends and pat-
terns focusing on alternative financing mechanisms. Without such an assessment,
the FBI does not have analyses that could aid in assessing risk and prioritizing
efforts (GAO, 2003). ICE, which participates in terrorist financing investigations
in coordination with the FBI, also does not systematically collect and analyze data
on terrorists’ use of alternative financing mechanisms (GAO, 2003).
Finally, communication and cooperation between the various arms of govern-
ment and ministries (i.e., law enforcement, taxation and revenue administrations,
customs services, health ministries, agricultural departments etc.) as well as
appropriate regional and intergovernmental organizations (e.g., WHO, the World
Customs Organization, Interpol, the European Anti-Fraud Office) are essential
to enhancing the effectiveness of tobacco control measures. Regardless of the
measures used, they lose their effectiveness if governments, states, and competent
authorities cannot share information on timely or real-time bases and cooperate in
other ways (ATF, 2002).

This chapter used the script approach to study a financial scheme that involved
an extremist group of Hezbollah supporters. This case is interesting because of
the scope and duration of its operation. After the scheme was broken down into
sequential order, we provided numerous techniques that, if implemented, would
have prevented the scheme from maturing. The value of this application is that
not only is it clear that such techniques provide significant opportunities for effec-
tively responding to financial crimes, but also that many could be implemented
relatively easily. We conclude this chapter with a discussion of the problem of
financial crimes and terrorism and highlight how the potential for using such
techniques to respond to financial crime can be more carefully understood by
conducting additional research in this area.
Cigarette smuggling and terrorism financing 203
Understanding how terrorists finance operations and organizations has
certainly not been neglected by scholars, but the evaluation of sound financial
counterterrorism strategies is an important gap. Most terrorist and extremist
groups are relatively unsophisticated and have short life spans. Typically a group
of individuals come together driven by some ideological commitment, organize
and begin to recruit members, establish a mandate and agenda, and then dissolve
in under a year. Like any organization, terrorists learn that sustainability is dif-
ficult. One problem that terrorist organizations face is how to garner funds to
sustain the organization and purchase the material necessary to attack a specific
target. Most terrorist organizations are ineffective financial planners and are una-
ble to develop a strategy that provides adequate funding for the organization.
Some terrorist organizations, however, are able to use various strategies that
result in increased longevity. Organizations may be funded through the sale
of literature, videos, and propaganda, others charge fees or receive donations
from various supporting organizations, and many commit crimes that allow the
organization to thrive. These crimes include selling drugs, committing robberies,
burglaries, thefts, various financial frauds, and illegally counterfeiting products.
The most successful organizations have multiple streams of funding and as they
grow the funds are laundered through the creation and support of legal businesses.
The proceeds from these businesses are considered returns on their investment
and the funds are funneled back through the organization.
The potential of the techniques discussed here (e.g., controlling the tools for
crime and identifying property) and similar techniques could be of great use for
the prevention of future acts of terrorism. The problem is that there is still a signif-
icant need to better understand the nature of financial crimes and also how best to
respond. Future research should also develop strategies to measure the nature of
the financial crimes committed by terrorist organizations (and other large organi-
zations, such as organized crime groups). Financial crimes are difficult to study
because the number and duration of their component parts pose challenges to
identify the nature of the problem to be studied. Recently, Belli (2011) conducted
an important first step in more effectively studying financial crimes. Instead of
identifying specific incidents as the unit of analysis, Belli studied financial crime
from the perspective of the entire scheme and thus was able to capture all the
associated incidents and the perpetrators connected to that scheme. This approach
led to many revelations about the nature of financial crimes committed by terrorist
organizations, including that such groups are well connected and rely on individu-
als who are not ideologically committed to a cause but are involved for profit.
Future research would benefit by examining the nature of these network ties and,
importantly, how terrorist groups were initially connected to non-ideologically
supporting members. The identification of these links might help identify other
crime prevention techniques that could be used to undermine terrorism.
Perhaps the most significant lesson to be learned from the script approach is
that offending, whether by terrorists or criminals, is a dynamic process that can
be thwarted by interventions. Nevertheless, because these scripts are always com-
plex, skilled or committed offenders will find ways of adapting to interventions,
204 A. Hiropoulos et al.
so that a kind of “arms race” may ensue (Ekblom, 2002). Therefore, it must be
acknowledged, as von Lampe (2010: 35) found in his examination of cigarette
smuggling in Germany, that one intervention at a particular point in time is rarely
sufficient over time. Constant monitoring and adaptation on the part of security
responders and planners is essential.
Finally, critical to understanding how to best respond to terrorism is to con-
duct evaluations of implemented prevention strategies. Little research examines
the effectiveness of the specific crime prevention strategies discussed here on
cigarette smuggling or on counterterrorism financing strategies generally. There
would be great value in knowing what works in responding to terrorism and how
to best implement these techniques. Most of the techniques discussed here seem
like logical responses and straightforward to put into practice, but implementa-
tion may be a different story. There are potential social, legal, organizational, and
financial hurdles that must be addressed to ensure that the techniques work as they
are designed. Future research would benefit by evaluating the implementation of
these strategies and identifying the impact of such hurdles.

1 While these movements’ beliefs diverge sharply at times there are also similarities.
(Examples include that Al Qaeda supporters embrace radical Sunni beliefs, while
Hezbollah adherents subscribe to extreme Shiite views. Hezbollah supports Iran while
many Al Qaeda supporters view Iran as an enemy. Some of these extremists endorse a
global campaign against the West in general and the United States in particular while
others do not and instead focus on a specific conflict (see below)).
We define extremist supporters of Hezbollah, Hamas, Al Qaeda, and similar movements
as individuals who adhere to aspects of the following beliefs. They believe that only
acceptance of the Islamic faith promotes human dignity and affirms God’s authority.
The extremists reject the traditional Muslim respect for “People of the Book,” (i.e.,
Christians and Jews). They believe that “Jihad,” meaning to struggle in the path of God
in the example of the Prophet Muhammad, is a defining belief in Islam and includes
the “lesser Jihad” that endorses violence against the “corrupt” other. These extrem-
ists believe that the Islamic faith is oppressed in both “local and nominally Muslim”
Middle-Eastern/Asian governments that are corrupt and authoritarian, as well as in non-
Islamic nations (e.g., Israel/Palestine, Russia/Chechnya) that occupy indigenous Islamic
populations. In addition, the West, and the U.S. in particular, support the corruption,
oppression, and humiliation of Islam, and exploits the region’s resources. They believe
that the hedonistic culture of the West and the U.S. (e.g., gay rights, feminism, sexual
permissiveness, alcohol abuse, racism, etc.) has a corrosive effect on Muslim social and
religious values. These extremists believe that the people of the West and the U.S. are
responsible for the actions of their governments and culture. It is a religious obligation
to promote a violent Islamic revolution to combat this assault on Islam by targeting
nonbelievers. They believe that Islamic law—Sharia—provides the ideal blueprint for a
modern Muslim society and should be implemented in all “Muslim” countries by force.
Some extremists are most concerned with combating the West and the United States
in particular, while other extremists, such as supporters of Hamas, Hezbollah, or the
Taliban, focus on specific or regional conflicts (Freilich et al., 2012).
2 By financial terrorism we mean non-violent financial crimes whereby political extrem-
ists commit to aid in carrying out violent attacks or furthering other goals of their
movement or organization (Belli and Freilich 2009; Clarke and Newman 2006).
Cigarette smuggling and terrorism financing 205
3 The ECDB identifies incidents from: (1) terrorism databases such as the American
Terrorist Study, the RAND-MIBT database, and the Global Terrorism database; (2)
official sources such as Congressional hearings, and the FBI’s Terrorism in the United
States annual report; (3) scholarly and journalistic accounts; (4) watch-group reports
published by the SPLC, the Anti-Defamation League, the militia watchdog listserv and
other organizations; and (5) systematic media searches.

ATF (U.S. Department of Justice Bureau of Alcohol, Tobacco, Firearms and Explosives)
(2009). The Bureau of Alcohol, Tobacco, Firearms and Explosives’ Efforts to Prevent
the Diversion of Tobacco (Report Number I-2009-005). Available at:
ATF (Bureau of Alcohol, Tobacco and Firearms) (2002). International Conference on
Illicit Tobacco Trade. Co-Chairpersons’ Report. New York City July 30–August 1
2002. Available at:
(accessed December 9, 2011).
Belli, R. (2011). Where Political Extremists and Greedy Criminals Meet: A comparative
study of financial crimes and criminal networks in the United States (No. NCJ 234524).
Final report submitted to the National Institute of Justice, in fulfillment of Grant No.
Belli, R. and J.D. Freilich (2009). Situational crime prevention and non-violent terrorism:
A ‘soft’ approach against ideologically motivated tax-refusal. Crime Prevention
Studies, 25, 173–206.
Breinholt, J. (2003). Terrorist financing. United States Attorneys’ Bulletin. Available at: (accessed December 11,
Boonn, A. (2011). The Michigan Story: Cigarette tax increases, new revenues, and the
value of tax. Available at:
(accessed December 9, 2011).
Canadian Centre for Intelligence and Security Studies (2006). Actual and potential links
between terrorism and criminality. In Trends in Terrorism Series sponsored by the
Integrated Threat Assessment Centre. Available at:
Chermak, S.M., Freilich, J.D., Parkin, W. and Lynch, J.P. (2012). American terrorism and
extremist crime data sources and selectivity bias: An investigation focusing on homicide
events committed by far-right extremists. Journal of Quantitative Criminology, 28, 191–218.
Clarke, R.V. (1997). Situational Crime Prevention: Successful case studies (2nd edn),
Monsey, NY: Criminal Justice Press.
Clarke, R.V. and Cornish, D. (1985). Modelling offenders’ decisions: A framework for
research and policy. In M. Tonry and N. Morris (eds) Crime and Justice: An annual
review of research, Vol. 6 (pp. 147–185). Chicago, IL: University of Chicago Press.
Clarke, R.V. and Newman, G.R. (2006). Outsmarting the Terrorists, Westport, CT:
Greenwood Publishing Group.
Cornish, D. (1994). The procedural analysis of offending and its relevance for situational
prevention. Crime Prevention Studies, 3, 151–196.
Cornish, D. and Clarke, R.V. (2002). Analyzing organized crimes. In A.R. Piquero and
S.G. Tibbetts (eds), Rational Choice and Criminal Behavior: Recent research and
future challenges (pp. 41–63). New York: Routledge.
206 A. Hiropoulos et al.
Cornish, D.B. and Clarke, R.V. (2003). Opportunities, precipitators and criminal
decisions: A reply to Wortley’s critique of situational crime prevention. In M. Smith
and D.B. Cornish (eds) Theory for Situational Crime Prevention. Crime Prevention
Studies,Vol. 16 (pp. 41–96). Monsey, New York: Criminal Justice Press.
Creswell, J.W. (2007). Qualitative Inquiry and Research Design: Choosing among five
approaches, Thousand Oaks, CA: Sage Publications.
deKiefer, D.E. (2008). Trade diversion as a fundraising and money laundering technique of
terrorist organizations. In T.J. Biersteker and S. Eckert (eds), Countering the Financing
of Terrorism (pp. 150–173). New York: Routledge.
Dishman, C. (2005). The leaderless nexus: When crime and terror converge. Studies in
Conflict and Terrorism, 28, 237–252.
Ekblom, P. (2002). Future imperfect: Preparing for the crimes to come. Criminal Justice
Matters, 46 Winter 2001/02: 38–40. London: Centre for Crime and Justice Studies,
Kings College.
Felson, M. and Clarke, R.V. (1997). The ethics of situational crime prevention. In
G. Newman, R.V. Clarke and S. Shoham (eds) Rational Choice and Situational Crime
Prevention, Aldershot: Ashgate Press.
Freilich, J.D. and Chermak, S.M. (2009). Preventing deadly encounters between law
enforcement and American far-rightists. Crime Prevention Studies, 25, 14–172.
Freilich, J.D. and Newman, G.R. (2009). Introduction .Crime Prevention Studies, 25, 1–7.
Freilich, J.D. and Pridemore, W.A. (2006). Mismeasuring militias: Limitations of advocacy
group data and of state-level studies of paramilitary groups. Justice Quarterly, 23,
Freilich, J.D. and Pridemore, W.A. (2007). Politics, culture and political crime: Covariates
of abortion clinic attacks in the United States. Journal of Criminal Justice, 35, 323–336.
Freilich, J.D., Chermak, S.M. and Caspi, D. (2009). Critical events in the life trajectories
of domestic extremist white supremacist groups: A case study analysis of four violent
organizations. Criminology and Public Policy, 8, 497–530.
Freilich, J.D., Chermak, S.M, Belli, R., Gruenewald, J.A. and Parkin, W.S. (in press).
Introducing the United States Extremist Crime Database (ECDB). Terrorism and
Political Violence.
Freilich, J.D., Chermak, S.M. and Simone, Jr., J. (2009). Surveying American state police
agencies about terrorism threats, terrorism sources, and terrorism definitions. Terrorism
and Political Violence, 21, 450–475.
GAO (United States General Accounting Office) (2003). Terrorist Financing. Report
to congressional requesters (GAO-04-163). Available at:
d04163.pdf (accessed December 9, 2011).
GAO (United States General Accounting Office) (2004). Cigarette Smuggling. Report to
the chairman and ranking minority member, committee on government reform, House
of Representatives (GAO-04-64). Available at:
(accessed December 9, 2011).
Gruenewald, J.A. (2011). A comparative examination of far-right extremist homicide
events. Homicide Studies, 15, 177–203.
Gruenewald, J.A. and Pridemore, W.A. (2012). A comparison of ideologically motivated
homicides from the new Extremist Crime Database and routine homicides from the
Supplementary Homicide Reports using multiple imputation by chained equations to
handle missing values. Journal of Quantitative Criminology, 28, 141–162.
Hamm, M.S. (2007). Terrorism as Crime: From Oklahoma City to Al-Qaeda and beyond,
New York, NY: New York University Press.
Cigarette smuggling and terrorism financing 207
Hewitt, C. (2003). Understanding Terrorism in America: From the Klan to Al Qaeda, New
York, NY: Routledge.
Horgan, J. and Taylor, M. (2003). Playing the green card: Financing the Provisional IRA:
Part 2. Terrorism and Political Violence, 15, 1–60.
Jaber, H. (1997). Hezbollah: Born with a vengeance, New York, NY: Columbia University
Kane, J. and Wall, A. (2005). Identifying the Links between White Collar Crime and Terrorism,
NIJ sponsored study Final Report, Grant no 2003-IJ-CX-1018, National Institute of Justice,
Office of Justice Programs, US Department of Justice. September 2004.
Lacoste, J. and Tremblay, P. (2003). Crime innovation: A script analysis of patterns in
check forgery. Crime Prevention Studies, 16, 171–198.
Leclerc, B., Wortley, R. and Smallbone, S. (2011). Getting into the script of adult child
sex offenders and mapping out situational prevention measures. Journal of Research in
Crime and Delinquency, 48, 209–237.
National Commission on Terrorist Attacks Upon the United States (2004). The 9/11
Commission Report. Available at:
Newman, G.R. and Clarke, R.V.G. (2003). Preventing E-Commerce Crime, Portland, OR:
Willan Publishing.
Newman, G.R. and Clarke, R.V.G. (2008). Policing Terrorism: An Executive’s Guide,
Washington DC: Center for Problem Oriented Policing (POP), Community Oriented
Policing Services (COPS), US Department of Justice. Available at: www.popcenter.
org/library/reading/pdfs/PolicingTerrorism.pdf (accessed January 3, 2012).
Noble, R.K. (2003). The links between intellectual property crime and terrorist financing.
Testimony before the United States House Committee on International Relations,
July 16, 2003.
Passas, N. (2003). Informal Value Transfer Systems, Money Laundering and Terrorism,
Washington DC: Report to the National Institute of Justice and the Financial Crimes
Enforcement Network.
Picarreli, J.T. and Shelley, L.I. (2007). Organized crime and terrorism. In J.K. Giraldo
and H.A. Trinkunas (eds) Terrorism Financing and State Responses: A comparative
perspective (pp. 39–55). Stanford, CA: Stanford University Press.
Pitcavage, M. (1999). Old Wine, New Bottles: Paper terrorism, paper scams and paper
“redemption”, last modified, November 8, 1999. A Militia Watchdog Bulletin.
Available at:
Schmid, A.P. and Jongman, A.J. (1988). Political Terrorism: A new guide to actors,
authors, concepts, databases, theories and literature. Oxford: North Holland.
Shelley, L.I. and Melzer, S.A. (2008). The nexus of organized crime and terrorism: Two
case studies in cigarette smuggling. International Journal of Comparative and Applied
Criminal Justice, 32, 43–63.
Shelley, L.I. and Picarelli, J. (2005). Methods and motives: Exploring links between
transnational organized crime and international terrorism. Trends in Organized Crime,
9, 52–67.
Smith, B.L. (1994). Terrorism in America: Pipe bombs and pipe dreams, Albany, NY:
State University of New York Press.
Smith, B.L. and Damphousse, K.R. (2003). The American Terrorism Study, Oklahoma
City: Memorial Institute for the Prevention of Terrorism.
Snow, D.A. and Trom, D. (2002). The case study and the study of social movements.
In B. Klandermans and S. Staggenborg (eds) Methods of Social Movement Research
(pp. 146–172). Minneapolis, MN: University of Minnesota Press.
208 A. Hiropoulos et al.
von Lampe, K. (2010). Preventing organized crime: The case of contraband cigarettes.
in K. Bullock, R.V. Clarke and N. Tilley (eds.) Situational Prevention of Organised
Crime (pp. 35–57). Portland, OR: Willan Publishing.
Weinberg, L., Pedahzur, A. and Hirsch-Hoefler, S. (2004). The challenges of conceptualizing
terrorism. Terrorism and Political Violence, 16(4), 777–794.
WHO (World Health Organization) (2011). Smuggling. Available at:
tobacco/research/economics/smuggling/en/ (accessed December 3, 2011)
Wortley, R. (1997). Reconsidering the role of opportunity in situational crime prevention.
In G. Newman, R.V. Clarke and S.G. Shoham (eds), Rational Choice and Situational
Crime Prevention (pp. 65–81). Aldershot: Ashgate.
Wortley, R. (1998). A two-stage model of situational crime prevention. Studies on Crime
and Crime Prevention, 7, 173–188.
Wortley, R. (2001). A classification of techniques for controlling situational precipitators
of crime. Security Journal, 14, 63–82.
Wortley, R. (2002). Situational Prison Control: Crime prevention in correctional
institutions. Cambridge: Cambridge University Press.
Yin, R.K. (2003). Case Study Research: Design and methods, 3rd edn. Thousand Oaks,
CA: Sage.
11 Script analysis of the
transnational illegal market
in endangered species
Dream and reality
William D. Moreto and Ronald V. Clarke

To be effective, situational prevention must be focused upon a closely defined,
specific form of crime or disorder. It depends critically upon: (1) the identification
of the opportunity structure (or the facilitating conditions) for the form of crime
in question and (2) a detailed understanding of how the crime is accomplished,
or what is sometimes called the modus operandi. Script analysis is designed to
assist with understanding the modus operandi by laying out the sequential steps
in the completion of a crime. It does this by identifying the players involved at
each step, and by describing their tasks, their skills and the tools they need. This
information allows preventive measures to be directed to “pinch points” that will
yield the best preventive results.
Cornish’s (1994) original applications of script analysis dealt with what seemed
to be fairly simple crimes, such as breaking into cars, but which turned out to
involve a number of sequential steps—acquiring the necessary tools, identifying
suitable locations, choosing the target vehicles, removing the desired contents or
components, escaping the scene and disposing of the stolen goods. For each of
these steps it was possible to think of preventive measures that might be introduced
to thwart the completion of the crime. Since then, many subsequent applications
of script analysis have dealt with a range of more complex crimes, though not
always with prevention in mind. It has been used to analyze serial sex crimes and
offenders (Beauregard et al. 2007); child sex offences (Leclerc et al. 2011); the
operation of clandestine drug laboratories (Chiu et al. 2011); check frauds (Lacoste
and Tremblay 2003); suicide bombings (Clarke and Newman 2006); car theft rings
(Morselli and Roy 2008; Tremblay et al. 2001); illegal waste disposal (Tompson
and Chainey 2011); and child sex trafficking (Brayley et al. 2011). Many of these
crimes, or particular steps in these crimes, depend for their completion on a degree
of organization and it has been suggested that script analysis could be usefully
employed in studying what are normally thought to be organized crimes, that is,
crimes that depend upon the coordination of efforts among members of criminal
organizations or networks (Cornish and Clarke 2002; Hancock and Laycock 2010).
It is against this background of research that we embarked on our attempt to
think through how a script analysis could be undertaken of the Transnational
210 William D. Moreto and Ronald V. Clarke
Illegal Market in Endangered Species (or TIMES). The trade in endangered spe-
cies is regulated by international agreement under CITES, the Convention on
International Trade in Endangered Species of Wild Fauna and Flora, which to
date has been signed by 175 countries. CITES categorizes all listed species under
three appendices: Appendix I lists species that are prohibited from any form of
commercial trade; species listed under Appendix II can be traded commercially
under strict monitoring by CITES; those listed under Appendix III can be traded
so long as this trade conforms to the laws of the trading countries.
The TIMES, which refers to the international trade in contravention of
CITES, is just one component of the broader field of wildlife crime, but it has
attracted considerable attention from conservation agencies (e.g. TRAFFIC—
Trade Records Analysis of Flora and Fauna in Commerce) and international law
enforcement authorities such as INTERPOL. The name itself suggests a highly
complex process requiring the coordination of many steps, which are accom-
plished by different people or groups. These include initial acquisition (poaching)
of the endangered animal or plant (often from a developing country), extracting
the value of that species (for example, roe from a sturgeon or the skin and bones
of a tiger) and smuggling the species (or its parts) out of the country into another,
usually more developed country, where it is sold. A detailed description of this
process, or a “master script,” derived from script analysis, should provide the
framework for identifying the pinch points referred to above. This is the “dream”
that motivated our efforts; unfortunately, reality soon intruded and we were left in
doubt that master scripts of the TIMES were really feasible.

Reality intrudes
We preface our discussion in this section of the paper by noting that whatever
the impediments we discuss to developing master scripts of TIMES, we remain
convinced, for reasons given in the penultimate section of the paper, that script
analysis is a useful tool for achieving an understanding of its constituent steps.

The enormous task

As mentioned above, situational crime prevention must be focused upon highly
specific forms of crime because the facilitating conditions and modus operandi
for any specific form of crime or disorder will differ in important respects from
those of even apparently similar forms. A frequently cited example is that of resi-
dential burglary which one British study (Poyner and Webb 1991) found took two
quite different forms in the same city. One form involved offenders operating on
foot in an inner city area. Because the residences were row houses, they kicked in
the front door or pried open a ground floor window. They generally stole cash or
small valuables. The other form involved offenders who used cars to reach outly-
ing suburbs comprised of single-family residences. They got into the homes from
rear doors or windows and they generally targeted electronics such as TVs and
audio equipment. The scripts for these two offences would be quite different and
The illegal market in endangered species 211
would yield quite different pinch points for intervention. For example, the subur-
ban burglars would depend much more on “fences” than the inner city burglars,
and the latter would spend less time in the houses because they were more likely
to be seen entering by neighbors or passers-by.
The need for specificity almost immediately dooms the prospects for under-
taking comprehensive script analyses of the TIMES because it comprises many
dozens if not hundreds of different crimes. This is not just a matter of the different
species involved and the fact that the transnational illegal market for pangolins
will be very different from that for alligator skins, elephant ivory, rhino horns,
tiger bones, bear bladders, sea turtle eggs, mahogany, teak, Mexican cacti, par-
rots, blue fin tuna and a host of animals and plants that are illegally traded across
international borders. It is also partly because these species or their constitu-
ent parts are put to a great variety of uses: medicinal purposes, subsistence and
consumption, ornaments and furnishings, clothing, pets, ornamental plants, and
manufacturing and construction (Broad et al. 2003; Reeve 2002).
The international trade in particular species may also differ from country to
country or region to region. For example, the illegal trade in neo-tropical parrots
involves different species in Mexico from in Bolivia (Pires and Clarke 2011) and
the facilitating conditions for their theft also differ between the two countries
because Mexico has a tradition of licensing trappers (some of whom take par-
rots) while Bolivia does not. Another example concerns elephant ivory which is
traded domestically in some African countries, but not in others: the existence of a
domestic market facilitates the international trade (Lemieux and Clarke 2009). In
fact, the domestic illegal wildlife market can rival its international counterpart in
size and importance, which is the case for the caviar market in the former Soviet
countries surrounding the Caspian Sea (TRAFFIC 2007). In such cases, it may be
difficult to determine the impact and influence of local demand on international
A further complicating factor in understanding TIMES is that the same spe-
cies in the region of the world might attract different groups of poachers. For
example, two different groups of poachers appear to be involved in illegally
taking sturgeon for the roe (caviar) in the Caspian Sea: independent poach-
ers and organized poaching squads. Independent poachers are small scale and
opportunistic in their activities and they often take other species, apart from
sturgeon, for personal consumption or for secondary income (Raymakers 2002).
They tend to use simple equipment (e.g. rowing boats, small mesh gillnets),
operate at night within the Volga delta or coastlines near their villages and
process the caviar themselves. Organized poaching squads depend on poaching
as a primary source of income and they specifically target female sturgeon for
their roe. They invest in more sophisticated equipment (e.g. powerful motor
boats, large mesh gillnets and global positioning systems), which enables them
to operate in open sea. They typically do not process the roe themselves but
hand it over to middlemen who take care of the processing. Unlike independ-
ent poachers, organized poaching squads are directly part of the illegal caviar
network (Raymakers 2002).
212 William D. Moreto and Ronald V. Clarke
Another example of different groups of poachers operating in the same local-
ity is provided by Shepherd and Magnus (2004), who found that three different
groups of poachers were involved in the illegal tiger trade in Indonesia: oppor-
tunistic, professional and accidental. The “accidental” group comprised hunters
who incidentally kill tigers when seeking a different species and those who exact
retribution against problem tigers that have killed livestock or even people.
Because TIMES may usually evoke images of greedy poachers hunting, trap-
ping and killing wildlife, recognizing the role of “accidental” poachers is vital in
understanding the situational circumstances surrounding the illegal tiger trade.
Thus, Shepherd and Magnus (2004) found that retributive hunters are more likely
to kill tigers and sell their body parts when no government scheme is in place to
recompense villagers for lost livestock or, indeed, lost lives.
The multitude of the specific forms of crime that comprise TIMES far outnum-
ber the criminologists in the world who would have the capability or interest in
undertaking script analysis of these crimes. Indeed, most studies of wildlife crime
to date have been produced by conservation scientists and biologists (Pires and
Moreto 2011; Schneider 2012) and it is only recently that criminologists have
begun to take an interest in this field. Given the parallels between wildlife crime
and other transnational crimes, such as labor trafficking (EJF 2010; Sophal and
Sovannarith 1999) and the drug trade (Cook et al. 2002; Reeve 2002; South and
Wyatt 2011), the criminological neglect of wildlife crime is surprising; but what-
ever the reasons, it is clear we are far from accumulating enough script analyses
to make a discernible impact on TIMES.

The lack of research funds

Script analysis of any particular form of TIMES would generally require the
following steps to be studied: acquiring (poaching) the species, processing (i.e.
extracting the value for immediate use or further trade and/or disguising it for
trade), transporting (including smuggling) and selling to the ultimate user. In
many cases the master script will be more complicated than this. For example,
processing caviar for transport can occur immediately upon poaching the sturgeon
(extracting the roe from the fish) and also, but in a different form, at a later stage
when it becomes necessary to forge customs documents to falsify the status of
the caviar.
Since these are transnational crimes, the sequential steps would need to be
studied in the source country, the transition country—or transition countries in
those cases where species are transported across several international borders—
and finally in the destination country. To undertake the necessary fieldwork in
these different countries would require substantial sources of research funding,
which, like the qualified personnel to undertake script analysis, do not exist. The
mostly poor countries in which the species are poached lack the funds to prevent
the poaching, let alone the funds to study it. Funds for crime and criminal justice
research in destination countries tend to be reserved for crime occurring within
the borders of those countries, not outside the country, and in any case, wildlife
The illegal market in endangered species 213
crimes are not seen as high priority anywhere. As Cook et al. (2002) have argued,
wildlife crime is serious, but it is a crime that is not taken seriously.
Where government funds are available for research on transnational crimes,
these, understandably, tend to be reserved for more seemingly harmful forms
of crime such as trafficking in humans, in human organs, in guns or in drugs.
International agencies and organizations concerned with TIMES, such as
INTERPOL, CITES, ICCWC (International Consortium on Combating Wildlife
Crime) and TRAFFIC are either starved of official funds for their work or depend
on donors for their very survival. Various private foundations do provide (gener-
ally small) grants for wildlife research, but none could fund the volume of needed
research on TIMES. The same could be said of the US Fish and Wildlife Service
and the UK’s Department of Environment, Food and Rural Affairs (DEFRA).
Apropos the last point, when limited funds are available, such agencies or
organizations typically focus on the end stage of the market because this tends to
be more visible and open to study. Conducting market surveys, while time con-
suming, is not as difficult or dangerous as investigating how species are originally
harvested, or smuggled through borders. Thus, researchers may be able to gener-
ate a market profile simply by investigating a few stalls or vendors. Moreover, as
vendors may expect questions or bargaining they may be more willing to answer
questions in a more forthright and honest manner.

No agreed model of the market

The resource limitations discussed above, both in terms of qualified person-
nel and the funds available to permit script analysis, are compounded by some
inherent difficulties of conducting script analysis of the TIMES. A problem
of the chicken-and-egg variety is that so little is known about the nature of
the TIMES and much of what is said to be known is of doubtful validity. A
prime example is the claim frequently made that “organized crime,” such as the
Mafia, is heavily involved in TIMES (Elliot 2009; Gray 2012; UNODC 2010;
Zimmerman 2003). The claim is bolstered by the identification of supposed
indicators of organized crime, including the use or threat of violence, forgeries
of permits and certifications, and the management of international shipments
(Sellar 2007). While it is possible that organized crime is involved in some par-
ticularly lucrative forms of the TIMES, perhaps the illegal international trade
in caviar or rare timber, or at specific stages such as transnational movement
(Shaw and Kemp 2012), the claim tends to be made by enforcement organiza-
tions who might be seeking to obtain additional funds or to bolster their public
image. After all, it is more impressive to be fighting organized criminals making
huge profits through ruthless exploitation of the environment than to be pursu-
ing hapless peasants seeking to augment their miserable incomes by poaching
parrots or pangolins in the places where they live.
The “organized crime” model tends to assume that any particular form of
TIMES is masterminded by a “Mr Big” who arranges the entire process from ini-
tial poaching of the endangered species in an impoverished source country, right
214 William D. Moreto and Ronald V. Clarke
through to its ultimate sale to a privileged consumer in a much richer country. We
have scoured the literature, but have found little reliable evidence of this model
in place. Even when it is clear that particular steps in the process are well organ-
ized, this does not necessarily mean that organized crime groups are involved.
The individuals involved in each of these steps will likely want to complete their
tasks and maximize their profits with minimal effort and expenditure of resources,
which will inevitably require a degree of organization. It might also require them
to bribe officials or mete out punishment to members of their group, activities that
are not the sole prerogative of organized crime groups.
Indeed, we are increasingly coming to the belief that each of the steps in the
sequence are undertaken by people, however well-organized, who have little con-
tact with those responsible for earlier and subsequent steps, and who might have
little understanding of the inner workings of the overall market (cf. Bullock et al.
2010). Thus, the Bolivian peasant who sells his captured parrot to a middleman
for a couple of dollars might not know whether the middleman will sell it at a
street market in the city, or if it is rare enough, to a dealer for a much larger mark-
up. The dealer might have links with a legitimate pet exporting business, which
can conceal the parrot among larger consignments of animals to be sent overseas.
Those receiving the consignments might also be primarily legitimate dealers who
can sell the bird to a collector of rare parrots.
Under this model, it might be enough therefore for each of the separate groups
involved in the chain of commerce to conduct transactions that profit them with-
out their being linked in any way to those undertaking more distant steps in the
process. This “network model” (Hayman and Brack 2002; Natarajan and Clarke
2004) is clearly very different from the “organized crime” model. Not know-
ing which of these models is closer to the generality of TIMES is a significant
impediment to planning any script analysis since the dangers and difficulties of
the research might be significantly greater for the “organized crime” model than
for the other.

The difficulties and dangers of fieldwork

Even if the dangers of fieldwork may be greater when studying “organized
crime,” researchers might still encounter violence or the threat of violence when
studying the TIMES. Thus it would be unwise to study illegal pet markets too
openly in some cities (especially where law enforcement is weak or corrupt)
as the researcher could be suspected of being a government informant. Other
dangers of fieldwork in source countries could come from contracting tropical
diseases or being bitten by poisonous insects or snakes—and the more remote the
location, the more serious the consequences of these medical emergencies. These
kinds of risks rarely have to be considered when undertaking other kinds of crime
research, but they must be taken into account when researching wildlife crime,
though this is of course true whether or not the focus is script analysis.
Some more mundane difficulties of undertaking fieldwork in remote areas
include transport and communication problems (despite the ubiquity of mobile
The illegal market in endangered species 215
phones, service may not be available outside the city), extremes of climate, poor
food and uncomfortable accommodation. Any researcher willing to endure these
privations would have to be uncommonly resilient. Another important impedi-
ment to studying wildlife crime scripts can be language. Large parts of the
world understand English, Spanish or French, but this may not be the case in the
more rural settings where interviews may need to be conducted. Moreover, any
researcher conducting such research would need to be familiar with the cultures
encountered in the course of the work as bartering, trading and movement of
products at early stages may be heavily influenced by local contexts and customs.
In some cases, it might be possible to partner with local researchers to undertake
the interviews or to hire local research assistants, but again this adds an additional
level of difficulty to the research. Finally, gaining the cooperation and trust of
those interviewed, while at the same time satisfying institutional review board
conventions of informed consent and proof of remuneration, could be particularly
challenging under these conditions.
All these difficulties are multiplied many times over when fieldwork has to be
conducted, not merely in the source country, but also in transition and destination
countries. Together with the funding difficulties discussed in an earlier section,
it would indeed be surprising if a master script could ever be completed for the
TIMES, even for just one of its specific forms.

The dynamic nature of the TIMES

While some specific forms of the TIMES might be stable and long-lasting it
seems that others quite quickly appear and might substantially impact the viability
of the species concerned in a brief period of time. Equally quickly, the problem
might diminish for a variety of reasons—the networks of offenders might break
up for internal reasons or they might be disrupted through action taken by the
authorities. Subsequently, a different form of largely the same predation might
develop again as a new set of offenders find fresh ways to accomplish the crimes
or ways to defeat the preventive measures the authorities have put in place. One
example is the CITES ban on the international trade in elephant ivory, the over-
all effect of which was to reverse the African continent’s decline in elephants
(Lemieux and Clarke 2009). It has recently been reported that elephant poaching
is increasing again possibly because illegal traders have found ways to exploit the
(legal) domestic markets in ivory that exist in some African countries. This kind
of “offender adaptation” has frequently been identified in the situational crime
prevention literature (Ekblom 1997). Additionally, the TIMES is impacted by
other factors including growing regional populations and wealth, with the result
that demand for particular products may rapidly increase with resulting shifts
between source, transition and destination countries (Felbab-Brown 2011). One
example is the demand for tiger bones, used for traditional Chinese medicine,
which has increased in China in step with increasing prosperity (Verheij et al.
2010). Because of the length of time taken to conduct the detailed research needed
for a script analysis, offender adaptation and changes in global demand could
216 William D. Moreto and Ronald V. Clarke
result in the analysis providing only “yesterday’s answers to the questions of
today” (Cornish and Clarke 2002: 55).

A less ambitious role for script analysis

From a purely academic point of view, the impediments we have discussed to
achieving a full understanding of the TIMES through the development of master
scripts is disappointing, but it may be less disappointing from the perspective
of situational prevention, which might not in fact be helped by viewing the
TIMES as consisting of an integrated process from initial poaching to final use.
The reason for this is that no single agency exists with the jurisdiction, or indeed
the resources, to act upon the results of a script analysis of the entire process
involved in the TIMES. The agencies and organizations able to take preventive
action against the TIMES can only act at specific steps in the process. Thus, the
managers of nature reserves have little control over the markets for endangered
species, but they might be able to prevent the initial acts of poaching. They do
not need to know, therefore, which are the best pinch points overall for interven-
ing in TIMES, but only for the steps in the process for which they have some
responsibility. Similarly, customs officials might be able to impede the export or
smuggling of poached species, and they need to know how best to do this, not that
even more effective interventions might be taken at earlier or later steps. Indeed,
being told that more effective action could be taken at earlier or later stages might
work to absolve them of responsibility for acting themselves.
A rider to the above argument is that the only preventive value of developing
a master script for a particular form of the TIMES would be to describe the range
of pinch points that might exist in theory by outlining “the contours of organ-
ized crimes in enough detail for the purpose of guiding preventive approaches
and interventions” (Cornish and Clarke 2002: 57). This could be done through
desk research, through discussions with various authorities and through “thinking
thief” (Ekblom 1995), that is, by putting oneself in the minds of the offenders and
viewing the task from their perspective. A more formal investigation might some-
times be justified, but one that stops short of conventional methods of research and
uses instead the “rapid assessment” techniques advocated by Cornish and Clarke
(2002). These techniques include undertaking a script analyses of published stud-
ies; analyzing the police and prosecution case papers of known crimes (Natarajan
2006; Natarajan and Belanger 1998); interviewing experts and offenders about
script elements and linkages; studying the reports of undercover operations; inter-
viewing those involved in the legal analogues of the TIMES (in our case the legal
international market in endangered species) in order to understand the means by
which legal business is transacted and the loopholes available to offenders.
Having identified the range of possibly effective pinch points, it might be then
necessary to undertake a script analysis of only the step or steps in the process
encompassing these pinch points. This would be much easier to do than devel-
oping a master script of the entire market and, if the analysis were conducted in
partnership with the immediately responsible agencies, it could be expected that
The illegal market in endangered species 217
effective action would result from the analysis. At the very least, such analyses
can be useful in identifying features that may be typically ignored or areas that
require further investigation.
Even if script analysis were focused on specific steps in the process, it would
still be necessary to think about the possible ramifications of preventive action at
that step for earlier or later steps. One instructive example is the 1992 US ban on
the import of wild birds that was enacted primarily to reduce poaching of parrots
in Central and South America. Because pet shops selling these parrots in the US
could be easily monitored for compliance, this ban was followed by a substantial
and immediate reduction in the number of parrots brought into the US (cf. Pires
and Clarke 2011). However, two recent studies have found that substantial num-
bers of neo-tropical parrots are still being poached in the source countries, not for
the international trade but for domestic markets (Cantu et al. 2007; Herrera and
Hennessey 2007). That parrots continue to be poached, despite the international
ban, might suggest that the poachers taking the birds from the wild were unaf-
fected by the ban, indeed might never have heard of it. They continued to poach
parrots because they found it was still possible to sell them to middlemen, who in
turn sold them to feed the very large domestic market for parrots in these Central
and South American countries.

We began this paper by assuming that developing master scripts of particular
forms of the TIMES would help to identify the best preventive pinch points for
situational crime prevention. The ensuing discussion revealed that this was an
unrealistic goal—the TIMES is too diverse, too complex and too dangerous and
difficult to study to allow script analysis of the entire process involved in the
TIMES, except perhaps in very rare cases. Furthermore, a master script revealing
the pinch points for the entire process of the TIMES would only be useful if there
were an agency that was free to take action at any of the pinch points revealed.
But this is not the case. Rather, the responsibility and capacity for action will lie
with different agencies at different steps in the process and each of these agencies
needs to know where it should direct its energies to achieve the greatest preven-
tive effect. That question could be answered by a script analysis confined to the
step in the process where the agency has some leverage. To learn that another
agency, somewhere else in the process, could act with greater preventive effect
would be at best irrelevant, and at worst demoralizing.
We believe that our prescription for a more limited role for script analysis, sum-
marized in the paragraph above, applies to the generality of crimes that together
make up the TIMES, but we do not know how far it applies to other kinds of com-
plex crimes. We suspect, however, that it applies not just to the TIMES, but to
all forms of illegal trade across international borders including trafficking in arms,
drugs, humans or human organs. There seem to be no published script analyses of
the entire process involved in these crimes and, while they might be less diverse
than the TIMES, the difficulties and dangers of collecting the data needed for script
218 William D. Moreto and Ronald V. Clarke
analysis could be equally great. In addition, just as with the TIMES, there seem to
be no agencies with the responsibility, or the capacity, to take preventive action
at every step in the commission of the crimes. Nor does there seem to be much
hope for effective coordination of all the different agencies in the various countries
involved with an interest in the crimes concerned (Shaw and Kemp 2012).
Even if we offer a more limited role for script analysis of complex transna-
tional crimes, this still leaves an enormous amount of challenging research to be
done in undertaking script analysis of particular steps. Indeed, it is difficult to see
how these serious forms of crime could ever be consistently and effectively tack-
led unless the agencies responsible for enforcement in at least some of the steps in
the process had access to the kind of detailed knowledge of modus operandi that
script analysis can provide.

Beauregard, E., Proulx, J., Rossmo, K., Leclerc, B. and Allaire, J. (2007). Script Analysis
of the Hunting Process of Serial Sex Offenders. Criminal Justice and Behavior, 34,
Brayley, H., Cockbain, E. and Laycock, G. (2011). The Value of Crime Scripting:
Deconstructing Internal Child Sex Trafficking. Policing, 5, 132–143.
Broad, S., Mulliken, T. and Roe, D. (2003). The Nature and Extent of the Legal and
Illegal Trade in Wildlife. In S. Oldfield (ed), The Trade in Wildlife: Regulation for
Conservation (3–22). Sterling, VA: Earthscan Publications Ltd.
Bullock, K., Clarke, R.V., and Tilley, N. (eds). (2010). Situational Prevention of Organized
Crimes. Portland, OR: Willan Publishing.
Cantu, J.C.G., Saldana, M.E.S., Grosselet, M. and Gamez, J.S. (2007). The Illegal Parrot
Trade in Mexico: A Comprehensive Assessment. Mexico and Washington, DC:
Defenders of Wildlife.
Chiu, Y., Leclerc, B. and Townsley, M. (2011). Crime Script Analysis of Drug
Manufacturing in Clandestine Laboratories. British Journal of Criminology, 51,
Clarke, R.V. and Newman, G.R. (2006). Outsmarting the Terrorist. Westport, CT: Praeger
Security International.
Cook, D., Roberts, M. and Lowther, J. (2002). The International Wildlife Trade and
Organised Crime: A Review of the Evidence and the Role of the UK. Godalming,
Surrey: WWF-UK.
Cornish, D.B. (1994). The Procedural Analysis of Offending and Its Relevance for
Situational Prevention. In R.V. Clarke (ed), Crime Prevention Studies, Vol. 3 (151–196).
Monsey, NY: Criminal Justice Press.
Cornish, D.B. and Clarke, R.V. (2002). Analyzing Organized Crimes. In A.R. Piquero
and S.G. Tibbetts (eds), Rational Choice and Criminal Behavior: Recent Research and
Future Challenges (41–63). New York, NY: Routledge.
EJF. (2010). All at Sea: The Abuse of Human Rights Aboard Illegal Fishing Vessels.
London: Environmental Justice Foundation.
Elliot, L. (2009). Combating Transnational Environmental Crime: “Joined Up” Thinking
about Transnational Networks. In K. Kangaspunta and I.H. Marshall (Eds), Eco-Crime
and Justice: Essays on environmental crime (55–77). Turin: UNICR.
The illegal market in endangered species 219
Ekblom, P. (1995). Less Crime, by Design. Annals of the American Academy of Political
and Social Science, 539, 114–129.
Ekblom, P. (1997). Gearing up Against Crime: A Dynamic Framework to Help Designers
Keep up with the Adaptive Criminal in a Changing World. International Journal of
Risk, Security and Crime Prevention, 2, 249–265.
Felbab-Brown, V. (2011). The Disappearing Act: The Illicit Trade in Wildlife in
Asia. Working Paper Number 6. Foreign Policy. Washington, DC: The Brookings
Gray, D.D. (2012). Like a Bad Hollywood Cop Movie: Asia’s Biggest Wildlife Traffickers
Mostly Untouchable. Associated Press. August 15. Retrieved on August 16, 2012 from:
Hancock, G. and Laycock, G. (2010). Organised Crime and Crime Scripts: Prospects for
Disruption. In K. Bullock, R.V. Clarke, and N. Tilley (eds), Situational Prevention of
Organised Crimes (172–192). Portland, OR: Willan Publishing.
Hayman, G. and Brack, D. (2002). International Environmental Crime: The Nature and
Control of Environmental Black Markets. Workshop Report. London, UK: Royal
Institute of International Affairs.
Herrera, M. and Hennessey, B. (2007). Quantifying the Illegal Parrot Trade in Santa
Cruz de la Sierra, Bolivia, with Emphasis on Threatened Species. Bird Conservation
International, 17, 295–300.
Lacoste, J. and Tremblay, P. (2003). Crime Innovation: A Script Analysis of Check
Forgery. In M.J. Smith and D.B. Cornish (eds), Theory for Practice in Situational
Crime Prevention, Vol. 16 (171–198). Monsey, NY: Criminal Justice Press.
Leclerc, B., Wortley, R. and Smallbone, S. (2011). Getting into the Script of Adult Child
Sex Offenders and Mapping out Situational Prevention Measures. Journal of Research
in Crime and Delinquency, 48, 209–237.
Lemieux, A.M. and Clarke, R.V. (2009). The International Ban on Ivory Sales and Its
Effects on Elephant Poaching in Africa. British Journal of Criminology, 49, 451–471.
Morselli, C. and Roy, J. (2008). Brokerage Qualifications in Ringing Operations.
Criminology, 46, 71–98.
Natarajan, M. (2006). Understanding the Structure of a Large Heroin Distribution Network:
A Quantitative Analysis of Qualitative Data. Quantitative Journal of Criminology, 22,
Natarajan, M. and Belanger, M. (1998). Varieties of Upper-Level Drug Dealing
Organizations: A Typology of Cases Prosecuted in New York City. Journal of Drug
Issues, 28, 1005–1026.
Natarajan, M. and Clarke, R.V. (2004). Understanding and Controlling Organized Crime:
The Feasibility of a Situational Approach. Paper presented at the annual Environmental
Criminology and Crime Analysis Seminar, Wellington, New Zealand, July.
Pires, S.F. and Clarke, R.V. (2011). Sequential Foraging, Itinerant Fences and Parrot
Poaching in Bolivia. The British Journal of Criminology, 51, 314–335.
Pires, S.F. and Moreto, W.D. (2011). Preventing Wildlife Crimes: Solutions that Can
Overcome the “Tragedy of the Commons.” European Journal on Criminal Policy and
Research, 17, 101–123.
Poyner, B. and Webb, B. (1991). Crime Free Housing. Oxford: Butterworth Architect.
Raymakers, C. (2002). Study on the Social and Economic Aspects of Illegal Fishing in the
Caspian Sea. TRAFFIC Europe: WWF.
Reeve, R. (2002). Policing International Trade in Endangered Species: The CITES Treaty
and Compliance. London: Earthscan Publications Ltd.
220 William D. Moreto and Ronald V. Clarke
Schneider, J.L. (2012). Sold into Extinction: The Global Trade in Endangered Species.
Santa Barbara, CA: Praeger.
Sellar, J.M. (2007). International Illicit Trafficking in Wildlife. The Police Chief, 74,
Shaw, M. and Kemp, W. (2012). Spotting the Spoilers: A Guide to Analyzing Organized
Crime in Fragile States. New York, NY: International Peace Institute.
Shepherd, C.R. and Magnus, N. (2004). Nowhere to Hide: The Trade in Sumatran Tiger.
TRAFFIC Southeast Asia.
Sophal, C. and Sovannarith, S. (1999). Cambodian Labour Migration to Thailand: A
Preliminary Assessment. Working Paper 11. Phnom Penh: Cambodia Development
Resource Institute.
South, N. and Wyatt, T. (2011). Comparing Illicit Trades in Wildlife and Drugs: An
Exploratory Study. Deviant Behavior, 32, 538–561.
Tompson, L. and Chainey, S. (2011). Profiling Illegal Waste Activity: Using Crime Scripts
as a Data Collection and Analytical Strategy. European Journal on Criminal Policy and
Research, 17, 179–201.
TRAFFIC. (2007). Sturgeons and the 14th Meeting of the Conference of the Parties to
CITES, The Hague, Netherlands, 2007. A TRAFFIC and WWF Briefing Document.
Retrieved on January 12, 2010 from:
wildlifetrade/WWFBinaryitem6577.pdf. Tremblay, P., Talon, B., and Hurley, D.
(2001). Body Switching and Related Adaptations in the Resale of Stolen Vehicles.
British Journal of Criminology, 41, 561–579.
UNODC. (2010). What Does Environmental Crime Have in Common with Organized
Crime? Retrieved on August 25, 2010 from:
Verheij, P.M., Foley, K.E. and Engel, K. (2010). Reduced to Skin and Bones. An Analysis
of Tiger Seizures from 11 Tiger Range Countries (2000–2010). Cambridge: TRAFFIC
Zimmerman, M.E. (2003). The Black Market for Wildlife: Combating Transnational
Organized Crime in the Illegal Wildlife Trade. Vanderbilt Journal of Transnational
Law, 36, 1657–1689.
12 New developments in script
analysis for situational
crime prevention
Moving beyond offender scripts
Benoit Leclerc

In 1994, Derek Cornish introduced the concept of crime scripts to the field of
environmental criminology with a paper entitled ‘The procedural analysis of
offending and its relevance for situational prevention’ (Cornish, 1994a). This
work had one clear objective: to provide academics and practitioners working in
the field of crime prevention with a clear, innovative and applicable theoretical
framework to the study of crime-commission processes. Drawing from cognitive
psychology, Cornish retraced the origins of scripts and outlined its potential for
better understanding crime-commission processes. By applying scripts to crime,
Cornish developed a procedural framework to identify the complete sequence
of actions adopted prior to, during and following the commission of a particular
crime and thereby to offer a fuller range of intervention points to disrupt crime
commission. Because offenders can learn from their mistakes when committing
crimes, Cornish also discussed how crime scripts can evolve when disrupted by
obstacles that prevent their completion. Finally, Cornish discussed how lifestyle
and routine activities may affect scripts and how the enactment of particular
scripts may depend on requirements such as the presence of co-offenders. Despite
a recent increase of interest in the application of script analysis to various forms of
crime, the full value of scripts has not been realized. In particular, current applica-
tions of script analysis have largely ignored the behaviours of other actors present
during crime events, which limits the reach of situational crime prevention.
The objective of this chapter is to push the boundaries of script analysis. After
briefly reviewing what is currently known about crime scripts, I argue that script
analysis should be also applied to victims and guardians against crime. Then I pre-
sent the interpersonal script approach which has the potential to guide the analysis
of the interchange between actors involved in the crime event. Finally, I discuss the
flexibility of scripts.

Scripts and crime script analysis

In order to have a clear understanding of what scripts are, the concept of sche-
mas first needs to be defined and explained. A schema is a cognitive structure
222 Benoit Leclerc
that serves to organize our representations of past behaviour and experience. This
structure contains assumptions and expectations about the social world that guide
a person in interpreting future experiences. In other words, every person has accu-
mulated knowledge about him- or herself, other people, what role to adopt with
regard to particular positions in society, and what sequence of actions to follow
in order to achieve desired goals. This knowledge, in turn, assists and guides the
person in everyday life. Because the social world is complex, schemas are used
to simplify reality. There are different structures for different types of schemas
(e.g. self schemas, person schemas, role schemas, event schemas) (Augoustinos,
Walker and Donaghue, 2006; Fisk and Taylor, 1991). Self schemas refer to the
knowledge structure people have of themselves. This structure is linked to a
person’s traits and past behaviours and thus provides this person with a sense
of identity. A person schema is a knowledge structure that contains personality
prototypes of other people the person knows either personally or by reputation.
This structure enables the person to categorize others according to their domi-
nant personality traits and then to interact with them according to his or her own
personality. A role schema is a knowledge structure that refers to the norms and
expected behaviours of specific role positions in society. These structures sum-
marize the key features of the achieved roles that a person acquires through effort
and training. For instance, doctors and police officers are expected to have certain
characteristics and to behave in certain ways that are consistent with their specific
roles. There are also ascribed roles that are defined at birth and are associated with
factors such as age, race and gender. Finally, an event schema, commonly referred
to as script, is a knowledge structure that organizes the sequence of actions to
adopt in a particular context. According to Schank and Abelson (1977), a person
has a repertoire of behavioural sequences stored in memory ready to be activated
Schank and Abelson (1977) illustrated the concept of scripts by considering
the sequence of actions that must be taken by the customer of a restaurant (the
so-called restaurant script). The customer must: enter the restaurant; wait to be
seated; get the menu; order; eat; get the cheque; pay; and exit. The concept of
script has been often applied to organizational behaviour (Gioia and Manz, 1985;
Gioia and Poole, 1984; Poole, Gray and Gioia, 1990), and to investigating the
sequence of events leading to phobias and anxiety in psychology (Wenzel and
Golden, 2003; Wenzel and Holt, 2000; Wenzel and Holt, 2003). Scripts have
been applied in criminology to understand property crimes such as robbery,
vandalism, auto theft (Cornish, 1994a; b; 1998), cheque forgery (Lacoste and
Tremblay, 2003) and stolen-vehicle resale (Morselli and Roy, 2008; Tremblay,
Talon and Hurley, 2001). More recently, criminologists have employed scripts in
order to investigate interpersonal forms of crime such as crimes against passen-
gers and employees in public transport (Smith and Cornish, 2006), child sexual
abuse (Leclerc, Wortley and Smallbone, 2011) and serial stranger sex offences
(Beauregard, Proulx, Rossmo, Leclerc and Allaire, 2007), demonstrating the
applicability of crime scripts to crimes that may otherwise seem irrational and
Script analysis for situational crime prevention 223
Although script analysis may have its own limitations in relation to mapping
very complex crimes such as the transnational illegal market in endangered species
(see Moreto and Clarke, this volume), its utility to understand crimes that could
be labelled as ‘organized’ has also been shown by several scholars. Scripts have
been used to design potential preventative methods to disrupt the crime-commission
process of suicide bombing in terrorism (Clarke and Newman, 2006), infiltration
of Italian organized crime groups in the public construction industry (Savona,
2010), human trafficking for sexual exploitation in Italy (Savona, Giommoni and
Mancuso, this volume), child sex trafficking (Brayley, Cockbain and Laycock,
2011), and methamphetamine manufacturing in clandestine laboratories (Chiu,
Leclerc and Townsley, 2011). A theoretical script framework that incorporates
criminal lifestyles and networks of offenders was also proposed for better examin-
ing and preventing organized crime (Hancock and Laycock, 2010).
The crime script concept makes the process of crime commission much easier
to identify and understand. As indicated by Cornish (1994a; b), crime commission
is an event that contains a number of different steps from start to finish. The typi-
cal sequence of script scenes used to break down the crime-commission process
of offenders is detailed in Table 12.1. As an illustration, a hypothetical example of
a street robbery script of an adult stranger is shown. Drawing from rational choice
perspective and earlier work on the crime event model (Clarke and Cornish, 1985;
Cornish and Clarke, 1986; 2008), Cornish (1994a) advanced the script concept
in order to: 1) provide a framework to systematically investigate and identify all
of the stages of the crime-commission process of a specific crime, decisions and
actions that must be taken at each stage and the resources required for success-
ful completion of the crime, and 2) assist criminologists in identifying additional
intervention points for prevention. As shown in Table 12.1, applied to crime, the
script simply represents the complete sequence of actions adopted prior to, during
and following the commission of a particular crime.

Table 12.1 Script scenes used in crime script analysis

Script scenes Script actions

Preparation Get a weapon, select co-offenders

Entry Enter neighborhood, go to public park
Precondition Loiter unobtrusively, look around for potential victims
Instrumental precondition Identify and select a potential victim
Instrumental initiation Approach and verbally intimidate a potential victim
Instrumental actualization Assault
Doing Take money and other valuables
Post-condition Threaten the victim to not report victimization
Exit Walk away
224 Benoit Leclerc
Moving beyond offender scripts
Now it is time to move on. By moving on, I suggest to consider other actors
involved during crime events. Indeed, other actors such as victims (including
past and potential victims) and guardians against crime are largely ignored
when looking at the literature on crime script analysis. Rather than thinking
about scripts only from the offender perspective, one must think about scripts
from the point of view of victims and guardians. At the very foundation of the
crime triangle is the routine activity approach, which argues that crime is more
likely to occur when a motivated offender and a potential victim (or target) con-
verge in time and space in the absence of a capable guardian (Cohen and Felson,
1979). Consistent with this reality and for the sake of prevention, we should
perform script analysis from the perspective of the victim and the guardian too.
Script analysis does not make any distinction between the offender, the victim,
the guardian or any other person who might be involved or intervene during
crime events. In fact, script analysis is designed to map out and understand the
behavioural sequences of any forms of human behaviour leading to an outcome
in a particular situation (Schank and Abelson, 1977).

Victim scripts
This section examines script analysis from the victim perspective. As indicated
by Felson and Clarke (2010), we currently know very little about which routine
precautions taken by ordinary citizens work and which do not. In a context
where governments cannot control crime without the help from citizens, there
is an urgent need to collect information about victims’ experiences with pre-
vention and then to verify the effectiveness of routine precautions (Felson and
Clarke, 2010).
Smith (2009) developed a theoretical model to examine the decision-making
process from the perspective of the potential victim in order to better understand
decisions to change location as a consequence of fear of being victimized. This
model includes the activity scripts of the potential victim to capture what the
potential victim intends to do in a location or what the potential victim did there
if victimized in the past. The decisions made by the potential victim are driven
by concerns about personal security and safety. In addition, as we would assume
from a rational choice perspective (Cornish and Clarke, 2008), potential victims
will try to make the best decision possible to avoid victimization, a decision that
would be constrained by past experiences and lack of knowledge about crime
(Smith, 2009). The model was designed to provide a guide to policymakers and
problem solvers on how to deal with fear of crime and to sensitize professionals
who could be in charge of modifying the physical and social settings in order to
reduce fear of crime (Smith, 2009). Similarly, scripts from the victim perspective
could be examined in order to identify the actions that worked and those that did
not work during crime events. In a preliminary study, Smith (2010) developed
scripts for attempted and completed robberies of taxi drivers. She found that all
Script analysis for situational crime prevention 225
the completed robberies involved a taxi driver who picked up customers on the
street. She also showed that taxi drivers adopted a number of strategies to reduce
their risk of victimization. These strategies included working safer shifts, screen-
ing by geographic destination or pick-up areas, and screening for the type and
number of customers.
The purpose of producing offender scripts (commonly referred to as crime
script analysis) is to identify all stages of the crime-commission process in order
to obtain a fuller range of intervention points. In this case, the ultimate objective
is to disrupt the script before the completion of the crime. However, as shown in
Table 12.2, the goal of the potential victim (or past victim) is to avoid victimiza-
tion not to commit crime, and the purpose of script analysis from the perspective of
the potential victim would be to identify the behavioural sequence leading to non
victimization, that is, the self-precautionary process that potential victims should
adopt to avoid being victimized. Accordingly, the ultimate objective would not
be to disrupt the script but rather to facilitate its execution. This exercise is in line
with the need to learn about which routine precautions work and assist potential
victims in protecting themselves (Felson and Clarke, 2010).
Particular groups of potential victims could be targeted for this purpose, such
as clerks working in convenience stores, petrol stations or banks; women work-
ing in pubs late at night; and employees working in public transport (see Smith,
2010; Smith and Cornish, 2006) such as buses, trains, planes and even boats
travelling on international waters. The identification of such scripts could be
achieved by interviewing the victims about their experience of victimization
and by examining police records and their statements following crime. As the
goal is to avoid victimization, victims could also be interviewed about their
experiences where they actually managed to avoid victimization if possible. For
potential victims, routine precautions and reactive strategies taken in particular
locations could be the focus of script analysis (Smith, 2009). This may include
what to do in particular circumstances such as when the offender is not alone or
when a guardian is present but unlikely to intervene. These would all feed up
into the script analysis in order to maximize the chances of assisting potential
victims to avoid victimization.

Table 12.2 The application of script analysis to the main actors of crime events

Dimension Offender Victim Guardian

Objective pursued by Commit crime Avoid victimization Prevent crime

actor during event
Focus of script Crime-commission Self-precautionary Intervention process
analysis process process
Prevention goal Disrupt script Facilitate script Facilitate script
226 Benoit Leclerc
Table 12.3 A hypothetical victim script

Script scenes Script actions

Preparation Carry mobile phone, capstun canister, put money in more

than one place (e.g. pockets, socks)
Entry Walk on the street
Precondition Look around for other people, walk where street lighting
and/or presence of other people is noticeable
Instrumental precondition Locate potential offenders yelling, do not pay attention
Instrumental initiation Walk faster and change route
Instrumental actualization Knock on door of a residence (or enter commercial
establishment) wherever possible
Doing Enter residence, use mobile phone to call a friend or
relative, ask for help
Post-condition Wait for assistance to arrive
Exit Leave setting

As an illustration of a victim script, and building on the offender script presented

above, Table 12.3 shows a hypothetical example of a potential victim taking pre-
cautions and successfully avoiding street robbery. Similarly to the offender, the
potential victim has to go through a number of stages in order to complete the
script. In the preparation phase, the potential victim can carry a mobile phone
and a capstun canister for protection. Then as precautions, the potential victim
may look around for other people and walk where street lighting is considerable.
In the presence of potential offenders, the potential victim may decide to not pay
attention to them, walk faster, take another route and find a refuge as soon and
wherever possible, which would allow time to call or ask for assistance.

Guardian scripts
Mirroring the previous section, this section examines script analysis from the per-
spective of the guardian. On the basis of current published work in this area, no
one has performed script analysis from the perspective of the guardian to prevent
crime, although one study comes close. Dwyer, Graesser, Hopkinson and Lupfer
(1990) applied scripts to the investigation of law enforcement officers’ decisions
regarding the use of deadly force. The authors used crime scenarios based on
actual incidents in order to determine which situational properties were present
during shooting scenarios. These scenarios were administered to police officers.
For each scenario, police officers had to report whether they would not draw their
weapon, draw their weapon, draw their weapon and aim or shoot the suspect.
Dwyer et al. (1990) did not identify a script as such that set out from beginning
to end the sequence in which police officers shoot suspects, but they applied a
Script analysis for situational crime prevention 227
script framework to investigate and identify the scenarios in which shooting was
more likely.
The concept of ‘capable guardian’ specifically refers to the presence of any
person who has the capacity to interrupt crime commission either directly or indi-
rectly. As noted by Felson (1995), a person can prevent crime simply by his/her
presence only, by being visible and/or in close proximity to the offender. Felson
(2002: 22) indicates that:

a guardian is not usually someone who brandishes a gun or threatens an

offender with quick punishment, but rather someone whose mere presence
serves as a gentle reminder that someone is looking. With a guardian present,
the offender avoids attempting to carry out an offence in the first place.

A guardian is most likely to be someone on the spot when the crime occurs, that
is, a wife, relative or passer-by.
The work completed by Reynald (2010) on the decision-making process of
capable guardians leading to intervention is informative for the performance of
script analysis on guardianship. Using qualitative data, Reynald (2010) showed
that the guardian’s capacity to detect criminal activities and to intervene if nec-
essary increases in cases in which the guardian decides to engage actively in
monitoring the potential crime scene. The decision to monitor is influenced by
factors such as the guardian’s sense of responsibility. After detecting ongoing
criminal activities, the guardian must decide whether or not to intervene, a deci-
sion that is influenced by factors such as his or her physical competence to do
so, the availability of tools for protection if necessary and the severity of the
incident. According to Reynald, this model illustrates how the decisions made
during the process determine how capable of preventing crime a guardian is. It
also highlights the step-by-step nature of the decision-making process related to
intervention in a way that is consistent with script analysis.
As the desired goal of guardians from a situational prevention perspective
is crime control and prevention, the purpose of crime script analysis from the
guardian’s point of view would be to identify all possible stages leading to a
successful intervention during crime, whether the intervention is direct or indi-
rect (see Leclerc and Reynald, in preparation). As with victim scripts, and unlike
offender scripts, the ultimate objective would be to facilitate script completion
(see Table 12.2). Therefore at each step of the script, prevention measures could
be developed to facilitate a successful intervention during crime. The identifi-
cation of such scripts could be achieved by interviewing potential guardians in
different settings (see Reynald, 2010). Similarly, police officers could be inter-
viewed to identify scripts that successfully led to the arrest of an offender. At each
stage of the script, measures could then be developed to facilitate police work in
the most effective and safest ways. Scripts could further be used by police forces
for better understanding and managing hostage situations. In addition, security
guards could be interviewed to identify the scripts that are more likely to lead to
crime prevention and those that were unsuccessful in preventing crime in the past
228 Benoit Leclerc
Table 12.4 A hypothetical guardian script

Script scenes Script actions

Preparation Carry mobile phone and tools for protection

Entry Already in setting
Precondition Capacity to intervene
Instrumental precondition Locate potential offenders and potential victim
walking-by, take responsibility, decide to intervene
Instrumental initiation Alert other potential guardians, approach potential victim
Instrumental actualization Offer assistance to potential victim
Doing Bring potential victim in a safe place
Post-condition Report the incident to police
Exit Leave the potential victim in safe hands (e.g. friend or
relative of victim, police)

(e.g. shoplifting, bank robbery). In all these situations, in line with Smith’s work
(2010), attempted and completed scripts could be identified for comparative pur-
poses but also to detect salient stages – the ones that should be emphasized – in
the intervention process of guardians.
To illustrate a guardian script, and building on the offender and victim script
presented above, Table 12.4 shows a hypothetical example of a potential guard-
ian intervening before the occurrence of a street robbery. In the preparation
phase, the potential guardian could carry a mobile phone and tools for protection
(e.g. capstun canister). To intervene directly, the potential guardian should be
physically capable of doing so if necessary. Then, in order to intervene, locating
potential offenders and the potential victim is essential as is taking responsibility
and deciding to intervene. The next stage could involve alerting other potential
guardians and approaching the potential victim, which would be followed by
offering assistance to the potential victim, bringing the potential victim into a
safe area, reporting the incident to the police, and finally, leaving the potential
victim in safe hands.

Relational schemas and interpersonal scripts: the

convergence of the offender, the victim and the guardian
One of the most fundamental aspects of the routine activity approach developed
by Cohen and Felson (1979), often overlooked, is the interaction between the
offender, the victim and the guardian. The offender and the victim must converge
in time and space for crime to occur. Crime is more likely to be committed in the
absence of a capable guardian. In other words, the offender script and the victim
script (and the guardian script if any on the spot where crime occurs) will col-
lide. This idea of interaction between scripts from different actors is difficult to
Script analysis for situational crime prevention 229
examine. Fortunately, the literature in cognitive sciences has covered this dimen-
sion of scripts before.
People have cognitive structures to guide them in their relationships with
others. Based on models of cognitive processing, Baldwin (1992; 1995) dis-
cussed the concept of relational schemas, which involve an interpersonal form of
script (for an application of this approach, see Fehr, Baldwin, Collins, Patterson
and Benditt, 1999). According to Baldwin (1992), people develop declarative
and procedural knowledge about their interactions with others through expe-
riencing similar patterns of interaction. This knowledge is organized under
cognitive structures called relational schemas. A relational schema captures the
essence of interactions between people in a specific situation and consists of
an interpersonal script with an associated self-schema (self representation in
this interaction) and an associated other schema. Specifically, the interpersonal
script is a stereotyped relational pattern found in a particular context. It includes
a summary statement about what behaviours tend to be followed by what
responses and is used to interpret social situations and the behaviour of others.
As one may expect, such a script includes a complex sequence of behaviours in
which one person’s choice of behaviours is contingent on the behaviours of the
other person. The script can be used to plan appropriate behaviour required to
interact with others.
To illustrate an interpersonal script, Baldwin (1992: 468) uses a script example
in which a teenage boy is borrowing the car keys of his mother:

His goal is to borrow the keys, and he expects that his mother’s goal is to
make sure the car and her son are returned safely. He therefore knows that
if she seems reluctant, the required behavior is to reassure his mother that
he will act responsibly, so he verbalizes phrases that have been successful in
the past, such as “I’ll be home by 11” and “I’ll drive carefully.” His expecta-
tion is that his mother will respond to these behaviors by perceiving him as
responsible, feeling secure that he will be cautious, and ultimately giving
him the keys. If not, he may engage different routines, such as emphasiz-
ing his urgent need for transportation, whining about the unfairness of her
behavior, and so on.

Leclerc, Smallbone and Wortley (this volume) used an interpersonal script

framework to investigate the sequencing of interpersonal events and patterns of
behavioural interdependence in child sexual abuse. Specifically, they examined
the offender–victim interchange with a focus on the stage where the offender
seeks the cooperation of the child for sexual activity. This stage was broken down
into four steps: 1) Offender action to gain the cooperation of the victim in sexual
activity, 2) Victim response, 3) Offender response to victim and 4) Sexual behav-
iours. Leclerc and his colleagues showed that what seems to matter in explaining
the severity of abuse is not so much what the offender does in the first place
to gain the cooperation of the victim but how the victim reacts in this context.
230 Benoit Leclerc
For instance if the victim is compliant, the intrusiveness of sexual behaviours
performed by the offender and the victim increases. This example shows that if
the victim adopts a particular behaviour, the script of the offender may change
accordingly and lead to another outcome. This finding is consistent with qualita-
tive studies completed on the offender–victim interchange where the outcome of
robbery or homicide has been shown to depend on the victim’s behaviour (e.g.
Indermaur, 1996; Luckenbill, 1977; 1980).
Baldwin (1992) adds that many scripts may include numerous roles that could
have important consequences on the dynamic of the interchange between actors.
Some situations may in fact involve three-person scripts that could place the per-
son in a conflict about the most appropriate or desirable course of action (Baldwin,
1992). This situation can easily be translated to crime where an offender con-
verges with a potential victim in the presence of a guardian. In this case, the
offender would have to choose between the most appropriate and desirable course
of action, that is, to leave the setting without offending against the potential victim
(appropriate action), or commit the crime (desirable action).
Applying interpersonal scripts to crime has the potential to illuminate the
interaction between the offender and the victim (and the guardian if any).
For instance, such an investigation could uncover how the offender reacts
to particular prevention measures in particular situations (e.g., victim using
self-protection, guardian intervention). It could inform us of the strategies
that the potential victim should and should not employ to avoid victimiza-
tion in particular circumstances (e.g., moving from one location to another,
see Smith, 2009). It could reveal at what stage of the script, and under which
circumstances, the guardian is more likely to prevent crime without jeopard-
izing the safety of the victim. This idea of interaction between offender and
victim scripts was previously referred to as script clashes by Ekblom (2007).
During the interchange, as mentioned by Ekblom (2007), each actor may
have to improvise from their script which will generate a different outcome.
According to Ekblom (2007), understanding these situations is vital because
it may offer directions as to which prevention measures to emphasize in order
to favour and protect the victim.

The flexibility of scripts

As shown in the previous section on interpersonal scripts, scripts are not as rigid
as one might initially believe. However, they are usually seen as linear processes
displayed automatically and unconsciously by a person in a particular situation.
Schank and Abelson (1977) define a script as a ‘predetermined and stereotyped
sequence of actions that defines a well-known situation’ (41). However, as
discussed by Schank and Abelson themselves, this is only the simplest way to
define scripts. Indeed, scripts are likely to be flexible for the simple reason that
the immediate environment in which crime occurs shapes how the crime event
unfolds. Note that the following discussion is applicable to any scripts including
victim or guardian scripts.
Script analysis for situational crime prevention 231
Two important concepts central to comprehending the execution of scripts
emerge from the discussion by Schank and Abelson (1977) on the flexibility of
scripts that is interferences and distractions. Interferences are states or actions
that prevent the execution of the script. Schank and Abelson (1977) identify two
types of interference: obstacles and errors. Obstacles are missing conditions that
are necessary for an action to be executed. Errors occur when actions are com-
pleted with an unexpected outcome. The former can be overcome by performing
a new action and the latter by repeating the first intended action. Distractions are
unexpected states or actions that lead the actor towards new goals and conse-
quently out of the original script. In the presence of interferences or distractions,
a detour path is established within the script until the original script is either
re-entered at the point where it was blocked initially to ensure its successful
completion, or simply abandoned. Therefore, people can learn detours from a
variety of situations and refine their scripts to facilitate the accomplishment of
specific goals. An example of a script error is when a person drives his car to go
to work but misses an exit. He then has to use (and perhaps learn) a detour path,
drive back and take the correct exit. A robbery is an event that could interfere
with the execution of the restaurant script. We would expect a person sitting and
eating at the restaurant to panic and try to escape during a robbery. The restaurant
script would then involve a goal that is not part of the original script, namely try-
ing to escape. A script is a predetermined sequence of actions but the context in
which it is used may lead to various and unexpected paths in order to be executed
In line with the concepts of interferences and distractions developed by Schank
and Abelson (1977), Cornish (1994a; b) discussed the concept of permutations in
crime script analysis. Permutations represent the various combinations of alter-
native scenes in the script and refer to the flexibility of the crime-commission
process (Cornish, 1994a, b). Permutations are second-order options that allow the
offender to move forward successfully along the scenes in the crime-commission
process. As crime scripts are particularly likely to be rich in detours, alternative
scenes must be examined too (Ekblom and Tilley, 2000; Tremblay, Talon and
Hurley, 2001). Offenders are likely to come face-to-face with obstacles, such as
alarm systems, door locks, a capable guardian or a reluctant victim. To overcome
these obstacles, offenders may take an unexpected route and use another method
to commit the crime. For instance, an offender may rely on violence to overcome
a guardian or force a victim to perform an action. From this it follows that the
more options the offender has in each step of the process, the more flexible his
script and the better the chances to complete the crime. For example, as pointed
out by Morselli and Roy (2008) (see also Cornish, 1994a), a 1 × 2 × 1 × 2 script
has four steps with very minimal flexibility (one or two options for each scene).
In that case, the maximum number of combinations to commit the crime is only
4 (the product for all options in all scenes is 4). By contrast, a 4 × 4 × 5 × 4 script
offers many more possibilities for the offender to commit his crime. The script
still has only four scenes but with many more options (4 or 5 for each scene). The
maximum number of combinations for this script is 320. This latter script is much
232 Benoit Leclerc
more likely to lead the offender to the completion of the crime when obstacles
arise than is the former script (320 possible combinations versus 4 combinations).
To my knowledge, only two studies have provided a thorough examination of
potential permutations and combinations during the crime-commission process
and their impact on outcomes. In the first study, Lacoste and Tremblay (2003)
investigated cheque fraud scripts as designed by co-offender groups. By analys-
ing the sequence of actions and combinations, Lacoste and Tremblay were able
to identify different groups of co-offenders who have chosen to modify, tinker
with or elaborate upon their initial script. Groups of co-offenders who innovated
and used different script combinations managed to obtain higher returns from
their fraud activities. In the second study, Morselli and Roy (2008) extended the
flexibility notion to the impact of removing key brokers in stolen-vehicle exporta-
tion scripts. After identifying how brokers were distributed across these scripts,
Morselli and Roy found that their removal may greatly reduce the number of
alternative routes that render possible the execution of stolen-vehicle exportation
activities. By neutralizing the flexibility of crime scripts or some key combina-
tions within the script, the likelihood of crime or at least the benefits derived from
crime activities may be reduced.
Another way to look at script permutations and combinations is to investi-
gate each stage of the crime-commission process by using decision trees. Homel,
Macintyre and Wortley (this volume) used decision trees to examine the sequence
of decisions made by burglars for selecting a vulnerable target dwelling. Burglary
scenarios to explore the cues that were most salient to burglars in selecting tar-
get dwellings were administered to a sample of burglars in Australia by using a
computer program. Burglars were allowed to respond to cues for each scenario
sequentially using a 3-point scale (‘deterrent’, ‘attractive’, or ‘neutral’). They
were able to access as much information about a scenario as they needed to make
a decision about whether or not they would commit the crime. An enormous
advantage of this design is that it incorporates the different alternatives for each of
the cues, and the combinations of cue alternatives across burglary scenarios. This,
in turn, offers a more realistic view of the decision-making process before crime.
Then a decision-tree can be graphically developed to visualize the sequence of
decisions made by offenders before committing crime. Although complex, this
approach could be adopted to study any crime and its permutations at any stage of
the script.

I am not arguing against the use of offender scripts, or recommending to abandon
offender scripts. Rather what I am arguing for in this chapter is to be innovative
and complete script analysis from the perspective of victims (potential and past
victims) and guardians against crime. Interpersonal scripts and the flexible nature
of crime scripts are also worthy of attention.
The application of script analysis to victims and guardians against crime
can not only assist in understanding the behavioural sequences leading to
Script analysis for situational crime prevention 233
non-victimization or intervention during crime (therefore assist in getting a
better grasp of crime events) but also expend the potential of situational pre-
vention. Instead of trying to disrupt the script by mapping measures onto the
crime-commission process as we do with offender scripts, we could facilitate
the execution of victim and guardian scripts by mapping measures onto these
processes too. With a simple theoretical twist, the classification of 25 situa-
tional prevention techniques (Cornish and Clarke, 2003) could be employed as
a template to guide the development of such measures. For instance, rather than
designing measures to increase the risks and efforts associated with the com-
mission of crimes as we would do to disrupt offender scripts, we could design
measures to decrease the risks and efforts taken by victims and guardians in
order to combat crime.
Wortley (2001) introduced years ago a classification of techniques for
controlling situational precipitators of crime. His argument was that the imme-
diate environment can not only offer opportunities to offend but also induce or
encourage people to commit crimes that they would not have contemplated in
a different situation. In this sense, as pointed out by Wortley (2001), rational
choice was only able to provide half of the explanation for the role of imme-
diate environments before his work led to the introduction of the most recent
classification of techniques (Cornish and Clarke, 2003). Similarly, I argue
that offender scripts represent only half the possibilities for the role of script
analysis. On one side of the coin, we have offenders from which crime script
analysis is currently performed by researchers. On the other side of the coin,
we have victims and guardians against crime from which script analysis could
be performed. These developments could critically improve our capacity to
think about and prevent crime events.

Augoustinos, M., Walker, I. and Donaghue, N. (2006). Social cognition: An integrated
introduction (2nd edn). London: Sage Publications.
Baldwin, M.W. (1992). Relational schemas and the processing of social information.
Psychological Bulletin, 112, 461–484.
Baldwin, M.W. (1995). Relational schemas and cognition in close relationships. Journal of
Social and Personal Relationships, 12, 547–552.
Beauregard, E., Proulx, J., Rossmo, D.K., Leclerc, B. and Allaire, J.-F. (2007). Script
analyses of the hunting process of serial sex offenders. Criminal Justice and Behavior,
34, 1069–1084.
Brayley, H., Cockbain, E. and Laycock, G. (2011). The value of crime scripting:
Deconstructing internal child sex trafficking. Policing, 5, 132–143.
Chiu, Y-N., Leclerc, B. and Townsley, M. (2011). Crime script analysis of drug
manufacturing in clandestine laboratories: Implications for strategic intervention.
British Journal of Criminology, 51, 355–374.
Clarke, R.V. and Cornish, D.B. (1985). Modeling offenders’ decisions: A framework for
research and policy. In M. Tonry and N. Morris (eds) Crime and justice: An annual
review of research, Vol. 6 (pp. 147–185). Chicago, IL: University of Chicago Press.
234 Benoit Leclerc
Clarke, R.V. and Eck, J. (2005). Crime analysis for problem solvers in 60 small steps.
Washington, DC: US Department of Justice Office of Community Oriented Policing
Services. Available at:
Clarke, R.V. and Newman, G.R. (2006). Outsmarting the Terrorist. Westport, CT: Praeger
Security International.
Cohen, L.E. and Felson, M. (1979). Social change and crime rate trends: A routine activity
approach. American Sociological Review, 44, 588–608.
Cornish, D.B. (1994a). The procedural analysis of offending and its relevance for situational
prevention. In R.V. Clarke (ed.) Crime Prevention Studies, Vol. 3 (pp. 151–196).
Monsey, NY: Criminal Justice Press.
Cornish, D.B. (1994b). Crimes as scripts. In D. Zahm and P. Cromwell (eds), Proceedings
of the International Seminar on Environmental Criminology and Crime Analysis.
Tallahassee, FL: Florida Statistical Analysis Center, Florida Criminal Justice Executive
Institute and Florida Department of Law Enforcement.
Cornish, D.B. (1998). Regulating lifestyles: A rational choice perspective. Paper presented
at the 7th International Seminar on Environmental Criminology and Crime Analysis,
Cornish, D.B. and Clarke, R.V. (1986). The reasoning criminal: Rational choice
approaches on offending. New York, NY: Springer-Verlag.
Cornish, D.B. and Clarke, R.V. (2003). Opportunities, precipitators and criminal decisions:
A reply to Wortley’s critique of situational crime prevention. In M. Smith and
D.B. Cornish (eds) Crime Prevention Studies. Theory for practice in situational crime
prevention, Vol. 16 (pp. 41–96). Monsey, NY: Criminal Justice Press.
Cornish, D.B. and Clarke, R.V. (2008). The rational choice approach. In R. Wortley
and L. Mazerolle (eds) Environmental criminology and crime analysis (pp. 21–47).
Cullompton, Devon: Willan Publishing.
Dwyer, W.O., Graesser, A.C., Hopkinson, P.L. and Lupfer, M.B. (1990). Application of
script theory to police officers’ use of deadly force. Journal of Police Science and
Administration, 17, 295–301.
Ekblom, P. (2007). Thinking thief: Crime frameworks for design against crime. Presented
at the Design Against Crime Research Centre, UK, London.
Ekblom, P. and Tilley, N. (2000). Criminology, situational crime prevention and the
resourceful offender. British Journal of Criminology, 40, 376–398.
Fehr, B., Baldwin, M., Collins, L., Patterson, S. and Benditt, R. (1999). Anger in close
relationships: An interpersonal script analysis. Personality and Social Psychology
Bulletin, 25, 299–312.
Felson, M. (1995). Those who discourage crime. In J.E. Eck and D.Weisburd (eds) Crime
Prevention Studies: Vol. 4, Crime and place (pp. 63–66). Monsey, NY: Criminal
Justice Press.
Felson, M. (2002). Crime and everyday life (3rd edn). Thousand Oaks, CA: Sage
Felson, M. and Clarke, R.V. (2010). Routine precautions, criminology, and crime
prevention. In H. Barlow and S. Decker (eds) Crime and public policy (2nd edn)
(pp. 106–120). Philadelphia, PA: Temple University Press.
Fiske, S.T. and Taylor, S.E. (1991). Social cognition (2nd edn). New York, NY:
Gioia, D.A. and Manz, C.C. (1985). Linking cognition and behaviour: A script processing
interpretation of vicarious learning. Academy of Management Review, 10, 527–539.
Script analysis for situational crime prevention 235
Gioia, D.A. and Poole, P.P. (1984). Scripts in organizational behaviour. Academy of
Management Review, 9, 449–459.
Hancock, G., and Laycock, G. (2010). Organized crime and crime scripts: Prospects for
disruption. In K. Bullock, R.V. Clarke and N. Tilley (eds) Situational prevention of
organized crime (pp. 172–192). Cullompton, Devon: Willan Publishing.
Indermaur, D. (1996). Reducing the opportunities for violence in robbery and property
crime: The perspectives of offenders and victims. In R. Homel (ed) Crime Prevention
Studies: Vol. 5. The politics and practice of situational crime prevention (pp. 133–157).
Monsey, NY: Criminal Justice Press.
Lacoste, J. and Tremblay, P. (2003). Crime and innovation: A script analysis of patterns
in check forgery. In M.J. Smith and D.B. Cornish (eds) Crime Prevention Studies:
Vol. 16. Theory for practice in situational crime prevention (pp. 41–96). Monsey, NY:
Criminal Justice Press.
Leclerc, B. and Reynald, D. (in preparation). When scripts and guardianship unite: A script
model to facilitate intervention by capable guardians.
Leclerc, B., Wortley, R. and Smallbone, S. (2011). Getting into the script of adult child
sex offenders and mapping out situational prevention measures. Journal of Research in
Crime and Delinquency, 48, 209–237.
Luckenbill, D. (1977). Criminal homicide as a situated transaction. Social Problems, 25,
Luckenbill, D. (1980). Patterns of force in robbery. Deviant Behavior, 1, 361–378.
Morselli, C. and Roy, J. (2008). Brokerage qualifications in ringing operations.
Criminology, 46, 71–98.
Poole, P.P., Gray, B. and Gioia, D.A. (1990). Organizational script development through
interactive accommodation. Group & Organization Studies, 15, 212–232.
Reynald, D. (2010). Guardians on guardianship: Factors affecting the willingness to
supervise, the ability to detect potential offenders, and the willingness to intervene.
Journal of Research in Crime and Delinquency, 47, 358–390.
Savona, E.U. (2010). Infiltration by Italian organized crime (Mafia, N’drangheta and
Camorra) of the public construction industry. In K. Bullok, R.V. Clarke and N. Tilley
(eds) Situational prevention of organized crimes (pp. 130–150). Cullompton, Devon:
Willan Publishing.
Schank, R. and Abelson, R.P. (1977). Scripts, plans, goals and understanding: An inquiry
into human knowledge. Hillsdale, NJ: Erlbaum.
Smith, M.J. (2009). A six-step model of potential victims’ decisions to change location.
Security Journal, 22, 230–249.
Smith, M.J. (2010). Constructing crime scripts from victim and near-victim narratives.
Paper presented at the Annual Meeting of the American Society of Criminology, United
States, San Francisco, CA.
Smith, M.J. and Cornish, D.B. (2006). Secure and tranquil travel: Preventing crime and
disorder on public transport. London: University College London, Jill Dando Institute
of Crime Science.
Tremblay, P., Talon, B. and Hurley, D. (2001). Body switching and related adaptations in
the resale of stolen vehicles: Script elaborations and aggregate crime learning curves.
British Journal of Criminology, 41, 561–579.
Wenzel, A. and Golden, S.B. (2003). Situation-specific scripts for threat in blood fearful
and nonfearful individuals. Journal of Psychopathology and Behavioral Assessment,
25, 213–218.
236 Benoit Leclerc
Wenzel, A. and Holt, C.S. (2000). Situation-specific scripts for threat in two specific
phobias. Journal of Psychopathology and Behavioral Assessment, 22, 1–21.
Wenzel, A. and Holt, C.S. (2003). Situation-specific scripts for threat in socially anxious
and nonanxious individuals. Journal of Social and Clinical Psychology, 22, 144–167.
Wortley, R. (2001). A classification of techniques for controlling situational precipitators
of crime. Security Journal, 14, 63–82.
13 Rational choice and offender
decision making
Lessons from the cognitive sciences
Richard Wortley

As outlined in the introductory chapter, the rational choice perspective (RCP)

has become the primary conceptual underpinning for situational crime prevention
(SCP) and as such has attracted both devoted supporters and trenchant critics. The
starting point for this chapter is the contention that much of the criticism directed
at RCP – and for that matter, much of the support it receives – is based on a mis-
conception of the function that the model is designed to fulfil. The distinction is
drawn between RCP as an organising framework for SCP policy and practice on
the one hand, and RCP as a theoretical model of offender decision making on the
other. The former function is what Cornish and Clarke intended for RCP, and
in that role it has been outstandingly successful. The latter function was never
intended by Cornish and Clarke and nor does RCP credibly fulfil that role.
Cornish and Clarke have been clear from the beginning that their intention was
to develop models of offending that informed prevention, not to provide a detailed
or literal account of how offenders make decisions. They have never claimed their
approach to be a theory, choosing the term rational choice perspective advisedly.
In the first exposition of RCP (Clarke and Cornish, 1985) they could scarcely be
more explicit:

This perspective provides a basis for devising models of criminal behavior

that (1) offer frameworks within which to locate existing research, (2) sug-
gest directions for new research, (3) facilitate analysis of existing policy, and
(4) help to identify potentially fruitful policy initiatives. Such models need
not offer comprehensive explanations; they may be limited and incomplete,
yet still be “good enough” to achieve these important policy and research
(1985, p. 147)

And again:

we have argued (Clarke and Cornish 1983) that simple and parsimonious
accounts of criminal behavior such as those provided by dispositional or situ-
ational theories – can have considerable heuristic value. They do not have
to be “complete” explanations of criminal conduct, but only ones “good
238 Richard Wortley
enough” to accommodate existing research and to suggest new directions for
empirical enquiry or crime control policy. As soon as they no longer serve
these ends they should be modified or discarded.
(1985, p. 149)

And again:

Section II outlines the main requirements of decision models, temporarily

“good enough” to explain the processes of criminal involvement (initial
involvement, continuance, and desistance) and the occurrence of criminal
events. In essence, these models are flowchart diagrams that identify the
main decision points and set out the groups of factors bearing upon the deci-
sions made.
(1985, pp. 149–150)

And again:

This research is still at a relatively early stage, and as yet there is only a
comparatively small body of criminological data relevant to decision making
upon which to draw. Any attempt to develop decision models of crime must
at this stage be tentative. Thus our aim is only to provide models that are at
present “good enough” to accommodate existing knowledge and to guide
research and policy initiatives.
(1985, pp. 163–164)

And again:

The decision models illustrated above should be seen as temporary, incom-

plete, and subject to continual revision as fresh research becomes available.
(1985, pp. 173–174)

And again:

the models have been developed primarily for the limited purposes of
improving crime control policies and developing policy-relevant research.
Such models have only to be “good enough”; they may not necessarily be
the most appropriate or satisfactory for more comprehensive explanations
of criminal behavior – though it seems likely that a decision approach might
provide a useful starting point even for academic purposes.
(1985, p. 178)

I think they make their point. Moreover, they were well aware even then of the
literature specifically highlighting the limitations of rational choice as a decision-
making model. They understood clearly that offender rationality was far from
perfect, compromised as it is by perceptual errors, distorted reasoning, emotional
Rational choice and offender decision making 239
states and the swirling confusion that characterises the typical crime event. Again
their response is pragmatic:

The facts that people do not always make the most “rational” decisions, that
they may pay undue attention to less important information, that they employ
shortcuts in the processing of information, and that group decisions may be
different from individual ones are all clearly relevant to an understanding of
criminal decision making. But cognitive psychology is still at an early stage
in its development and the topics studied so far are not necessarily those
that best illuminate criminal decision making. For example, there has been
perhaps too much concentration upon bias and error in information process-
ing (see Nisbett and Ross 1980), whereas, in fact, the judgmental heuristics
involved usually enable individuals to cope economically and swiftly with
very complex tasks (Bruner, Goodnow, and Austin 1956) – a process Simon
(1983) has termed “bounded rationality.”
(Clarke and Cornish, 1985, p. 160)

Throughout their 1985 paper, Cornish and Clarke invite further development of
their decision models in the light of future research. Their proposed models are
‘tentative’ and ‘accommodate existing research’ that is ‘still at a relatively early
stage’; the models provide a ‘useful starting point’ but ‘should be modified or dis-
carded’ when they no longer serve their purpose. The fact is that despite advances
in decision-making theory and research in the intervening years, current descrip-
tions of RCP are essentially unchanged from those of 1985.
In this chapter I critique RCP as a model of offender decision making. I
contend that the concession by Clarke and Cornish that offender rationality is
bounded – while undoubtedly true – has had the unintended effect of stifling
further theoretical exploration of offender decision making. In the quote imme-
diately above, Clarke and Cornish at the same time acknowledge and dismiss the
lack of rationality that may occur in offender decisions. The concession of non-
rationality, once made, is set to the side; with the focus on policy and practice, the
default position has become to proceed nevertheless ‘as if’ offenders are behaving
rationally. As a consequence, the bounded nature of offender decision making
has been subject to little systematic investigation in its own right. I argue that the
ways in which offenders deviate from rational choice may be as theoretically and
practically important as are the ways in which they conform.

Back to basics
Almost all decisions we make in life involve a degree of uncertainty – we can
never be totally sure of the outcome of our choice. Rational choice approaches
seek to model decision making under uncertainty. Rational decision making in
its purest form is expressed in the expected utility model. The expected utility
model assumes that individuals will make choices that maximise their gains and
minimise their losses (von Neumann and Morgenstern, 1944). Expected utility is
240 Richard Wortley
calculated by multiplying two dimensions of risk – the magnitude of the payoff
and the probability that the payoff will occur. Empirically, decision making is
typically studied by treating decisions as gambles; subjects may be set the task of
choosing between gambles that involve different specified payoffs and probabili-
ties. For example, consider being offered the choice between a) a .5 probability of
winning $100, and b) a .75 probability of winning $80. The respective expected
utilities are as follows:

a) EU = .5 × $100 = $50
b) EU = .75 × $80 = $60

Objectively, bet b represents the best option and should be selected by subjects if
they are acting perfectly rationally.
Of course, in real life situations, decisions rarely involve objectively known
payoffs and probabilities. A prospective burglar, for example, does not know pre-
cisely the value of the goods in a particular house, nor the exact likelihood that
he will be caught. An extension of the expected utility model is the subjective
expected utility model (Savage, 1954). The subjective utility model assumes that
individuals make personal judgements about the value of an outcome and the
likelihood that it will occur. With these judgements made, they will then behave
in ways that maximise subjective expected utility. Rationality is assumed when
decisions fulfil all of the following axioms:

Comparability: alternatives must be comparable; e.g. when comparing house

A and house B a burglar must be able to say that A is a more desirable tar-
get than B, that B is a more desirable target than A, or that A and B are
Transitivity: choices must be able to be ordered; e.g. if a burglar prefers house
A to house B, and house B to house C, he must also prefer house A to house C.
Dominance: an option that is be judged to be better than an alternative on at
least one dimension, and to be at least as good on all other dimensions, must
be preferred; e.g. if a burglar believes house A is less risky to burgle than
house B, and the houses contain equivalent goods, then he must choose house
A over house B.
Independence: if an outcome is unaffected by the choice, then this outcome
should not play a role in the decision making; e.g. if a burglar prefers house
A over house B, then this preference should remain stable irrespective of
whether it rains the following day.
Invariance: choices should be independent of how the alternatives are pre-
sented; e.g. if a burglar judges he has a 90 per cent chance of not being caught
for a burglary, he should arrive at the same decision as he would if he judged
he had a 10 per cent chance of being caught for a burglary.

Decades of research on decision making have shown unequivocally that,

when presented with decision tasks involving uncertain outcomes, people will
Rational choice and offender decision making 241
frequently make choices that do not maximise expected utility; that is to say, their
decisions are not rational. Moreover, many of the deviations from rationality that
individuals make are not random or idiosyncratic but are systematic, indicating
species-typical patterns of thinking. Edwards (1992) reported a survey of leading
decision-making theorists who unanimously agreed that expected utility models
do not account for the way human beings make decisions. The expected util-
ity approaches to decision making are now recognised as normative models of
rationality; they show how judgements should be made and provide benchmarks
against which errors in reasoning can be judged. Approaches that try to capture
how people actually think are referred to as descriptive models.
Simon’s (1955; 1983) concept of bounded rationality, cited by Clarke and
Cornish (1985) as underpinning their adaptation of rational choice, is an example
of a descriptive decision-making model. All descriptive models recognise and
seek to accommodate the ways in which everyday choices deviate from optimal
utility maximisation. Simon believed that human decision making was neither
perfectly rational nor wholly irrational. Individuals strive to benefit themselves
to the best of their ability. Rationality is bounded not just by the fundamental
limits of human information processing, but also by environmental structures that
impose constraints on the quality and amount of information that is available to
the decision maker and on the time in which decisions must be made.
Simon described human decision making as satisficing – satisfactory and suffi-
cient. He argued that decision making typically occurred as a search process guided
by the decision maker’s level of aspiration, which may change from one situation
to the next. In the expected utility model all options are provided to individuals
who then have the opportunity to compare possible alternatives before arriving at a
decision; in satisficing decision making, the decision maker seeks out alternatives
which are examined one at a time until an option is encountered that meets the
decision maker’s minimum requirements at that time. Searching for alternatives
and adjusting our standards according to what is on offer is how most choices are
made in the real world. Consider a burglar searching for a suitable target. He does
not weigh up the costs and benefits associated with every house in a neighbour-
hood before deciding which one to burgle. Rather he keeps looking until he finds a
house that offers the prospect of adequate rewards at an acceptable risk. His deci-
sion may be sub-optimal, but it is good enough for his current purposes.
There are other descriptive decision-making models, notably Kahneman and
Tversky’s (1979) prospect theory (see also Kahneman, 2011), and some of the
research supporting these variants of bounded rationality is discussed in the sec-
tions that follow. In these sections I unpack the concept of bounded rationality. I
examine three factors that impinge upon the capacity of offenders to make rational
choices, namely, cognitive ‘errors’, emotions and automatic cognitive processing.

Cognitive ‘errors’
Descriptive models of decision making recognise that the human brain is not a
perfect computer but, rather, that there are biologically imposed limitations on
242 Richard Wortley
what it can do. Descriptive models examine the subjective perception of outcomes
and probabilities, the cognitive shortcuts and rules of thumb (heuristics) that are
employed to reach a decision, and the resultant biases in reasoning that may occur.
While these deviations from rationality may be regarded as cognitive errors, many
theorists think that they represent adaptive solutions, designed by evolution, that
allow humans to deal efficiently and effectively with the complexity of the physi-
cal and social world that they encounter.
Researchers have identified over a hundred types of systematic deviations
from expected utility predictions in the form of perceptual errors, heuristics and
biases. There are overlaps among these processes and not all are especially rel-
evant to offender decision making. Below I present some of the principles that
underpin the cognitive errors in human decision making, drawing particularly
on Kahneman and Tversky’s (1979) prospect theory, and I make some specula-
tive suggestions about how they might relate to criminal behaviour – necessarily
speculative because the implications of bounded rationality for criminal behav-
iour have rarely been empirically examined.

The greater the gain or loss, the less effect an increase in that gain or
loss has
Expected utility theory assumes that increases in costs and benefits follow a
straight-linear trajectory; every unit increase in a gain or loss will produce the
same unit increase in weighting given to the gain or loss in the decision-making
process. At the heart of Kahneman and Tversky’s (1979) prospect theory is the
finding that subjective gains and losses follow a decay function; the subjective
rate of increase diminishes as the objective size of the gains and losses gets larger.
Put simply, the difference between $0 and $10 is assessed to be greater than the
difference between $100 and $110, which in turn is assessed to be greater than
the difference between $1,000 and $1,010. We might expect, therefore, that a
burglar searching a house where the pickings are slim will place more value on
the discovery of an additional $50 (and expend more time and effort to locate it)
compared with a burglar in a house that offers rich pickings. Similarly, the princi-
ple helps explain why increases in prison sentences do not produce commensurate
increases in deterrence, and perhaps also suggests that there may be diminishing
returns for some situational crime prevention measures.

Losses are felt more acutely than gains

Kahneman and Tversky (1979) further argued that the decay function for subjec-
tive values is greater for gains than it is for losses, meaning that people feel losses
more than they do gains of equivalent (or even greater) value. This phenomenon is
known as loss aversion. Most people would not risk $100 in a 50/50 chance to win
$200, even though the expected utility is $50 (.5 × $200 – .5 × $100) (Benzarti
and Thaler, 1995); the prospect of losing $100 is judged to have greater value
than the prospect of winning $200. Samuelson and Zeckhauser (1988) argued that
Rational choice and offender decision making 243
loss aversion is the basis for status quo bias, that is, people’s tendency to stick
with current or previous choices even when more attractive options are offered.
Perhaps (happily for us) status quo bias encourages many offenders to persevere
with tried-and-true modus operandi and targets rather than to branch out into
potentially more lucrative pursuits. Camerer (2000) also proposed that loss aver-
sion underlies the tendency for some self-employed workers (which we might
extend to include offenders) to quit after achieving their daily income targets. A
street robber who quits after he reaches a pre-set target of, say, $1,000 is behav-
ing irrationally since he will quit earlier on days when the pickings are easy than
on days when the pickings are slim. This is an example of loss aversion since the
focus of the burglar is on not making his target (losing) rather than on increasing
his haul above his target.

People give too much weight to low probability events and insufficient
weight to high probability events
Yet a further principle of prospect theory is that individuals give distorted weight-
ings to probability estimates. They will give disproportionately high weighting
to events that are unlikely, resulting in relatively little discrimination between
extremely unlikely events and moderately unlikely events. They will correspond-
ingly give disproportionately low weighting to events that are likely, meaning that
probabilities that are just short of certainty will be treated as being less probable
than they really are. Thus, while people will buy lottery tickets at infinitesimal
odds, they will be unwilling to exchange the certainty of winning $1 million for a
99 per cent chance of winning $5 million (Allais, 1979). Note that this effect is not
the same as making inaccurate estimations of probability that an event will occur;
it is about the weighting given to estimates. Research indicates that offenders
also underestimate their chances of arrest (Anderson, 2002; Copes and Vieraitis,
2007; Kleck, Sever, Li and Gertz, 2005). This is an example of optimism bias,
that is, the tendency for individuals to believe that they are at a lower risk of
negative outcomes, and more likely to experience positive outcomes, when com-
pared with others (Kahneman, 2011; McAdams and Ulen, 2009). The weighting
phenomenon suggests that the underestimation of risk will be further distorted by
inaccurate weights applied to it. An offender is likely to regard anything short of
certain arrest to be a relatively good bet (that is, he will give insufficient weight
to a high probability of arrest).

Distant rewards have less effect than proximal rewards

In a process called temporal discounting, the subjective value of rewards reduces
with time (e.g. Green, Fry and Meyerson, 1994). Further, the relationship between
objective and subjective rewards is described by a decay function, leading to a
phenomenon known as preference reversal. If people are asked whether they
would prefer $100 today or $110 in a month’s time, many will select $100 today.
However, if they are asked whether they would prefer $100 in a year’s time or
244 Richard Wortley
$110 in a year and one month’s time, they will switch preference to the $110. The
subjective length of a month reduces as it approaches the time horizon. Temporal
discounting might be viewed as a generic cause of criminal behaviour. Crime
occurs largely because people value immediate rewards more highly than delayed
rewards. This principle reinforces the central logic of situational crime preven-
tion; the most effective strategy to prevent crime is to remove the prospect of
reward at the time the crime is being contemplated.

One of the most persistent criticisms of RCP is that it cannot account for the
decisions made by offenders who commit so-called expressive crimes (assault,
murder, rape) when in a state of emotional arousal (anger, fear, jealousy, sexual
excitation). There are two elements to this criticism. The first is that these are
crimes of passion that do not involve rewards and therefore cannot be subject to a
cost–benefit analysis. The second is that offenders in highly emotional states are
not capable of exercising rational choice.
The first criticism is easily dealt with. While rational choice models are often
said to be based on a view of ‘economic man’ (e.g. Simon, 1955, p. 99), it is
not implied that they are concerned only with decisions that involve the weigh-
ing up of objective costs and benefits. Rational choice theories require only that
individuals have preferences and that they select options that give them greatest
satisfaction. As Cornish and Clarke (2008) explain, ‘The benefits of offending
include satisfying the usual human motives, such as desires for sexual gratifica-
tion, excitement, autonomy, admiration, revenge, control, reduction of tension,
material goods, and so on’ (p. 25).
The second criticism deserves more consideration. The usual response to this
criticism by RCP theorists is to point out that even under high levels of emo-
tional arousal people still retain some rational control over their behaviour. This
is the position eloquently argued by Richard Felson in this volume and elsewhere
(Felson, 2005; Tedeschi and Felson, 1994) with respect to violent behaviour.
Felson challenges the concept of reactive aggression whereby violence is seen to
be an innate and reflexive response to negative affect. The source of aggression
is not a biological urge but an aversive environmental stimulus that the individ-
ual interprets as intentional mistreatment or humiliation. Felson points out that
individuals engaged in ‘reactive’ violence will often adjust the ferocity of their
response to suit the situational circumstances, giving the example of the base-
ball batter who drops his bat before he charges the pitcher. Strong emotions may
overwhelm inhibitions, and individuals may act without much thought to the con-
sequences, but their decision to harm others ‘is a decision nonetheless’ (Felson,
this volume).
I have no issue with Felson’s analysis of the role of anger in violent crime,
and his conclusions can be broadened to cover the role of emotion more gener-
ally in offender decision making. A problem does arise, however, in the way
that RCP theorists and researchers have used the sorts of arguments that Felson
Rational choice and offender decision making 245
presents. The issue is the same as that raised with respect to bounded rationality
more generally. Once the effect of emotion on rationality is conceded it is no
longer discussed; the focus remains on the vestiges of rationality that remain. It
is a bit like the proverbial glass half full/glass half empty scenario – one side of
the issue (reason) is privileged over the other (emotion). Cognitive psychologists
even have a label for this bias: confirmation bias, defined as the tendency to pay
more attention to information that supports one’s existing belief. Individuals who
are emotionally aroused may well retain some level of rationality, but they are
undoubtedly less rational than when they are not emotionally aroused. And the
fact that they are less rational is important and deserves attention.
There is good experimental evidence that individuals tend to engage in
more diverse, more extreme and riskier behaviours when they are emotionally
aroused than when they are not (Ariely and Lowenstein, 2006; Bouffard, 2002;
Leith and Baumeister, 1996; Lerner and Keltner, 2001; Lowenstein, Nagin and
Paternoster, 1997; MacDonald, MacDonald, Zanna and Fong, 2000). To take just
one example, Ariely and Lowenstein (2006) compared the responses of subjects
(male college students) under sexually aroused and non sexually-aroused con-
ditions. Sexual arousal was achieved via masturbation (to sub-ejaculation) to a
series of erotic photographs. Subjects were presented with 20 questions about
how stimulating they would find various activities, many involving fetishes (e.g.
‘Would it be fun to be tied up by your sexual partner?’) and some involving ille-
gal behaviour if acted upon (‘can you imagine being attracted to a 12-year-old
girl?’). They were presented with a further five questions concerning their like-
lihood to engage in immoral (and in some cases illegal) ‘date-rape’ behaviours
(e.g. ‘Would you keep trying to have sex after your date says “no”?’). Finally,
they were presented with eight items concerning their willingness to engage
in unprotected sex (e.g. ‘Would you always use a condom if you didn’t know
the sexual history of a new sexual partner?’). Responses to all items were on a
0–100 scale (0 = no; 100 = yes). Of the 20 questions concerning the attractive-
ness of sexual activities, the arousal condition produced a significant increase
in endorsement for 18 questions. For example, in the case of attraction to a
12-year-old girl, the average score went from 23 (non-aroused) to 46 (aroused).
There was a significant increase in endorsement for all five of likelihood-to-
engage questions. For example, the average score for pursuing sex after a date
said ‘no’ went from 20 (non-aroused) to 45 (aroused). For the eight unprotected
sex questions, aroused subjects indicated the preparedness to take greater risks
in five questions. For example, endorsement for the use of a condom with a
new sex partner fell from 88 per cent (non-aroused) to 69 per cent (aroused). In
summary, under conditions of sexual arousal subjects were interested in a wider
range of sexual activities, they anticipated higher levels of sexual arousal when
engaging in those activities, and they were prepared to be more forceful and to
take more risks in their pursuit of sexual gratification.
There is a burgeoning theoretical literature (most of it post-dating Clarke
and Cornish’s 1985 paper) seeking to accommodate emotion within a cognitive
decision-making framework (Gutnick, Hakimzada, Yoskowitz and Patel, 2006;
246 Richard Wortley
Lerner and Keltner, 2001; Lowenstein and Lerner, 2003; Lowenstein, Weber,
Hsee and Welch, 2001; Slovic, Peters, Finucane and MacGregor, 2005). Theorists
draw the distinction between expected and immediate emotions. Expected emo-
tions refer to how individuals anticipate they will feel if they proceed with a given
behaviour; these anticipated feelings become one of the predicted consequences
of behaviour that feeds into their cost–benefit analysis. Expected emotions are
readily incorporated into normative subjective expected utility models. They
have also been used to help explain the deviations from expected utility exam-
ined in descriptive decision-making models. For example, loss aversion may be
explained as the anticipation of the negative emotion of regret. Immediate emo-
tions refer to how individuals feel at the time the decision is made; these feelings
have the potential to alter the way that the predicted consequences of the behav-
iour are perceived and processed. The role played by immediate emotions has
until recently received less theoretical attention and presents a greater challenge
to rational choice models of decision making.
Lowenstein and Lerner (2003) set out three pathways by which immediate
emotions might affect decision making. First, immediate emotional states may
affect the predicted value of prospective payoffs and probabilities. People will
assess payoff and probability values more positively when they are in a positive
mood, and more negatively when they are in a negative mood. This pathway
suggests, for example, that as an individual becomes more enraged he/she will
increasingly judge that inflicting injury on a victim will produce a more satisfying
outcome and/or will be less risky to carry out. Second, immediate emotional states
may affect the way that information about prospective payoffs and probabilities
is attended to and processed. In this pathway, predicted values do not change but
the individual’s attention to decision-relevant cues shifts. Positive emotional states
are associated with a broadening of attentional focus; negative emotional states are
associated with a narrowing of attentional focus. In a state of anger an individual
will become more present-focused and thus more likely to ignore the longer-term
consequences of engaging in assaultive behaviour, even if those consequences are
judged to be severe. Third, immediate emotions may have a direct influence on
behaviour, bypassing or overwhelming evaluative cognitive processes. A person’s
anger may become so intense that the individual becomes ‘out of control’ and
performs behaviour that is maladaptive and self-defeating. This pathway is most
challenging for rational choice models, suggesting as it does that there may come a
point when rationality is abandoned.

Automatic cognitive processing

Cognitive psychologists distinguish between two types of cognitive process
– controlled processes and automatic processes (Bargh, 1984; Chartrand and
Bargh, 1996; Kahneman, 2003; Moors and De Houwer, 2006). Controlled pro-
cesses are those cognitions of which we have conscious awareness, that we
intentionally initiate, that we deliberately control, that are relatively slow,
and that require cognitive effort. Learning to drive a car involves controlled
Rational choice and offender decision making 247
processes. In the early stages of learning to drive we must attend to and con-
sciously initiate every element of the task – scanning the road, judging the
distance to the car in front, turning the steering wheel, applying the brake,
and so on. Driving consumes all of our attention and effort. Automatic pro-
cesses, in contrast, occur below the level of conscious awareness, are activated
without intention, operate in the absence of conscious control, are fast, and
consume little in the way of cognitive resources. Many controlled processes
become automatic processes with practice. Most practised drivers will have
the experience of arriving at their destination and realising that they have little
recollection of the journey. It is as if they have been driving on ‘auto-pilot’.
Since automatic processes require minimal cognitive effort, they allow humans
to maximise information processing resources and to undertake multiple tasks
simultaneously. It is universally acknowledged by cognitive psychologists that
most cognition occurs automatically and that individuals have access to just a
tiny fraction of the neuronal activity that occurs within their skulls.
One explanation for many automatic processes involves the concept of schema
(Fiske and Taylor, 1991; see also Leclerc, this volume). Schemas are content-
based memory constructs that contain learned information about a specific topic
or domain. All of the accumulated knowledge and assumptions that an individual
has acquired about the world are collected and organised into schemas. The pur-
pose of schemas is to permit individuals to manage the complexity of their world
and to provide them with a basis to deal efficiently with new situations. Drawing
on exiting schemas the individual can rapidly interpret familiar stimuli, fill in any
gaps, and make predictions and judgements without the need to engage in delib-
erative decision making.
Both common sense and the empirical evidence tell us that a great deal of
criminal behaviour occurs as the result of an automatic process. Few researchers
seriously believe that offenders carefully draw up a ledger of costs and benefits
before engaging in a crime. Research on the decision making of burglars indicates
that experienced offenders – compared with inexperienced burglars and non-
offenders – make rapid-fire judgements about likely targets based on a few salient
cues (Garcia-Retamero and Dhami, 2009; Nee and Meenaghan, 2006. See also
Homel, Macintyre and Wortley, this volume). Through their experience they have
developed a ‘suitable target for burglary’ schema that is activated by the sight of
features of a potential target that are consistent with the schema. Once inside the
targeted premises, offenders operate on the basis of an event schema, or script.1
Crimes that go according to script can be performed as ‘mindless’ behaviour. Nee
and Meenaghan (2006) reported that three-quarters of their sample of experienced
burglars actually used terms such as ‘automatic’, ‘routine’, ‘second nature’ and
‘instinctive’ when describing their burglary strategies.
A closely related phenomenon is that of priming. Priming refers to the increased
sensitivity to a stimulus as a result of previous exposure to it. Primes facilitate
the retrieval of stored information from implicit memory, that is, memory of
which there is no conscious awareness. In a classic demonstration of priming
effects, Marcel (1983) required subjects to identify whether a word presented
248 Richard Wortley
to them was a real word (e.g. ‘doctor’) or a non-word (e.g. ‘glayner’). Subjects
were quicker at identifying a real word if it was preceded by a related real word.
For example, ‘doctor’ was identified more quickly if it was preceded by ‘nurse’
rather than by an unrelated word; ‘nurse’ acted as a prime to facilitate retrieval
of ‘doctor’. Priming has been used to explain the so-called weapons effect. In
the original study demonstrating this effect, Berkowitz and LePage (1967) found
that the sight of a gun increased the tendency for subjects to behave more aggres-
sively (to deliver electric shocks to other subjects). The gun primed subjects for
aggression. The weapons effect has been replicated many times using a wide
variety of primes. Carlson, Marcus-Newhall and Miller (1990) conducted a meta-
analysis on 23 studies that examined the effects of aggression cues that included
guns, knives, the names of people who had been previously paired with violent
images, Ku Klux Klan clothing, aggressive verbalisations, vengeance-themed
bumper-stickers, aggressive films and boxing films. The mean effect size for the
association between aggression cues and the expression of aggressive attitudes
or behaviour was .38 (medium). The effects were stronger when the target of the
aggressive response was of low status or an outgroup member.
So where does the concept of automatic processing leave RCP? To what extent
does rationality necessarily imply a consciously controlled process? There are
divided opinions on this issue. Some definitions of rational choice explicitly
require that decisions involve a conscious process, ‘as opposed to being motivated
by habit, tradition, or social appropriateness’ (MacDonald, 2003, p. 552). Other
definitions are more liberal; Herrnstein (1990) allows that even animals can be
acting rationally if their behaviours maximise utility. Clarke and Cornish (1985)
have it both ways. On the one hand, they were interested in ‘the conscious thought
processes that give purpose to and justify conduct, and the underlying cognitive
mechanisms by which information about the world is selected, attended to, and
processed’ (p. 147). At the same time, they recognised that the choices to engage
in criminal acts might be made rapidly and efficiently on the basis of ‘standing
decisions’ or ‘knowledge structures’ – effectively, schemas.
The general consensus would seem to be that while reasoning is by defini-
tion a deliberate and effortful process, automatic processes do not necessarily
preclude the production of (boundedly) rational outcomes (Glöckner and Betch,
2008; Kahneman, 2003; Slovic et al., 2005). However, controlled and automatic
processes have different strengths and weaknesses and can produce conflicting
outcomes. According to Kahneman (2003), because controlled processes involve
greater attention to the data they are relatively flexible and rule-governed. In com-
parison, automatic processes are governed by habit, are more likely to be affected
by emotion, and are more difficult to change. The quality of decisions made by
automatic processes is only as good as the schemas on which the decisions have
been made. For example, schemas that contain stereotypical assumptions about
particular groups of people support discriminatory decisions and practices with
respect to those groups.
Rational choice and offender decision making 249
It may seem odd to include a critique of RCP in a volume paying homage to the
publication of the Reasoning Criminal. Most criticisms of RCP come with hostile
intent, designed to denigrate the situational perspective on crime more broadly. I
have no such agenda. I have written this chapter not as a trenchant critic of SCP
– for there are enough of them – but as someone who has devoted much of his
academic career to the field. My aim is to strengthen the situational perspective,
not to weaken it.
The scope of the critique is clearly circumscribed. RCP was designed by
Clarke and Cornish as a heuristic to guide policy and practice. It has been an out-
standing success in this role. The critique focuses narrowly on the limitations of
RCP as a comprehensive and accurate theoretical account of how offenders make
decisions. Clarke and Cornish did not intend this role for it and nor does it fulfil
this role. However it seems that many researchers – both critics and adherents –
believe that RCT is meant as a literal description of offender decision making.
The problem, in fact, is not so much that bounded rationality is necessarily
an inadequate model of crime, but rather that the full implications of the model
have been broadly ignored by researchers in the field. They have focused on the
‘rationality’ part of the model and forgotten the ‘bounded’ part. In doing so they
have thrown away some of the most interesting and potentially useful bits. The
fact is, there has been almost no theoretical development of RCP since it was
first presented in 1985. Over this period, in the cognitive sciences there has been
an explosion of interest in decision-making research, almost none of which has
found its way into environmental criminology.
Attacks on RCP from outside the field are often attacks on SCP (e.g. Hayward,
2007). The logic runs as such: SCP is based on RCP; offenders are not rational;
therefore SCP is flawed. Critics of SCP can draw no comfort from the critique
offered in this chapter. Most of the factors that impinge upon offender rational-
ity – the ‘bounded’ bit of the model – are themselves situationally dependent.
Cognitive errors are the systematic misinterpretation of situational data (often
based on how those data are presented); the source of emotional arousal is almost
always located in the immediate environment; and the automatic processes that
produce the decision to commit a crime are typically primed by situational cues.
The issue is not that situational factors have been rendered unimportant by the
current critique, but that their effects are shown to be more complex than a simple
model of rationality would suggest.
The theory that underpins SCP must satisfy the needs of two audiences. In the
first instance it needs to provide clear and persuasive guidance for practitioners,
and as I have noted, this has been the primary focus of Clarke and Cornish. For
this audience, arguably ‘good enough’ theory is, well, good enough. The second
audience comprises the researchers and academics whose job is to enrich and
carry forward the situational perspective. By the very nature of their role, aca-
demics should not be satisfied with ‘good enough’ theory; they should strive for
nothing less than ‘good’ theory.2
250 Richard Wortley
Clarke and Cornish were rightly concerned about the dangers of developing
criminological theory for its own sake. However, there are three pragmatic rea-
sons to pay greater attention to the theoretical bases of SCP. First, good theory
provides a smaller target for critics of situational crime prevention. In its cur-
rent neglected and unelaborated state, RCP offers itself as an easy target for
those who want to undermine the situational perspective. Second, drawing on
recent developments from the cognitive sciences can help to better integrate the
SCP perspective with other disciplines such as psychology and economics. Pease
and Laycock (2012) have made the point that SCP research is poorly cited by
academics working within a similar theoretical framework in other disciplines.
Third, we may find that the current model of RCP isn’t ‘good enough’ for prac-
tice after all – that there are new crime prevention strategies to be discovered or
that the deployment of current strategies can be carried in a more nuanced and
targeted manner.

1 While Cornish (1994) borrowed the concept of script from cognitive psychology, the
application of the concept in environmental criminology has been very different from
the way it was originally conceptualised. In cognitive psychology scripts describe molar
activities that require no deliberative thought processes. Cornish, on the other hand,
deconstructs the script to identify decision points during the crime commission process.
Once again, Cornish’s goal is to develop a model for prevention, not to describe the
cognitive processes of offenders.
2 In the words of Kurt Lewin (1952, p. 169), ‘There is nothing more practical than a good

Allais, M. (1979). The foundations of a positive theory of choice involving risk and criticism
of the postulates and axioms of the American school. In M. Allais and O. Hagen (eds)
Expected Utility Hypothesis and the Allais Paradox (pp. 27–145). Dordrecht: Reidel.
Anderson, D.A. (2002). The deterrence hypothesis and picking pockets at the pickpocket’s
hanging. American Law and Economics Review, 4, 295–313.
Ariely, D. and Lowenstein, G. (2006). The heat of the moment: The effect of sexual arousal
on sexual decision making. Journal of Behavioral Decision Making, 19, 87–98.
Bargh, J.A. (1984). Automatic and cognitive processing of social information. In
R.S. Wyer, Jr. and T.K. Scull (eds) Handbook of Social Cognition (Vol. 3) (pp. 1–43).
Hillsdale, NJ: Erlbaum.
Benzarti, S. and Thaler, R. (1995). Myopic loss aversion and the equity premium puzzle.
The Quarterly Journal of Economics, 110, 73–92.
Berkowitz, L. and LePage, A. (1967). Weapons as aggression-eliciting stimuli. Journal of
Personality and Social Psychology, 7, 202–207.
Bouffard, J.A. (2002). The influence of emotion on rational decision making in sexual
aggression. Journal of Criminal Justice, 30, 121–134.
Bruner, J.S., Goodnow, J.J., and Austin, G.A. (1956). A Study of Thinking. New York:
Rational choice and offender decision making 251
Camerer, C.F. (2000). Prospect theory in the wild: Evidence from the field. In D. Kahneman
and A. Tversky, A. (eds) Choices, Values and Frames (pp. 288–300). Cambridge, UK:
Cambridge University Press.
Carlson, M., Marcus-Newhall, A. and Miller, N. (1990). Effects of situational aggressive
cues: A quantitative review. Journal of Personality and Social Psychology, 58, 622–633.
Chartrand, T.L. and Bargh, J.A. (1996). Automatic activation of impression formation and
memorization goals: Nonconscious goal priming reproduces effects of explicit task
instruction. Journal of Personality and Social Psychology, 71, 464–478.
Clarke, R.V. and Cornish, D.B. (1983). Crime Control in Britain: A Review of Policy
Research. Albany, NY: State University of New York Press.
Clarke, R.V. and Cornish, D.B. (1985). Modeling offenders’ decisions: A framework for
research and policy. In M. Tonry and N. Morris (eds) Crime and Justice: An annual
review of research, Vol. 6 (pp. 147–185). Chicago, IL: University of Chicago Press.
Copes, H. and Vieraitis, L. (2007). Identity Theft: Assessing Offenders’ Strategies and
Perceptions of Risk. Washington, DC: US Department of Justice.
Cornish, D. (1994). The procedural analysis of offending and its relevance for situational
prevention. In R. Clarke (ed.) Crime Prevention Studies: Vol. 3 (pp. 151–196). Monsey,
NY: Criminal Justice Press.
Cornish, D.B. and Clarke, R.V. (2008). The rational choice perspective. In R. Wortley
and L. Mazerolle (eds) Environmental Criminology and Crime Analyses (pp. 21–47).
Cullompton, Devon: Willan Publishing.
Edwards, W. (1992). Towards the demise of economic man and woman: Bottom lines from
Santa Cruz. In W. Edwards (ed.) Utility Theories: Measurements and Applications
(pp. 253–270). Dordrecht: Kluwer.
Felson, R.B. (2005). Violence as instrumental behavior. In K. Kelloway, J. Barling and
J. Hurrell (eds) Handbook of Workplace Violence (pp. 7–28). Thousand Oaks, CA:
Sage Publications.
Fiske, S.T. and Taylor, S.E. (1991). Social Cognition (2nd edn). New York: McGraw-Hill.
Garcia-Retamero, R. and Dhami, M.K. (2009). Take-the-best in expert-novice decision
strategies for residential burglary. Psychomonic Bulletin and Review, 16, 163–169.
Glöckner, A. and Betsch, T. (2008). Modeling option and strategy choices with connectionist
networks: Towards an integrative model of automatic and deliberate decision making.
Preprints of the Max Planck Institute for Research on Collective Goods, No. 2008, 2.
Available at:
Green, L., Fry, A. F. and Myerson, J. (1994). Discounting of delayed rewards: A life-span
comparison. Psychological Science, 5, 33–36.
Gutnik, L.A., Hakimzada, A.F., Yoskowitz, N.A. and Patel, V.L. (2006). The role
of emotion in decision-making: A cognitive neuroeconomic approach towards
understanding sexual risk behavior. Journal of Biomedical Informatics, 39, 720–736.
Hayward, K. (2007). Situational crime prevention and its discontents: Rational choice
theory versus the ‘culture of now’. Social Policy and Administration, 41, 232–250.
Herrnstein, R.J. (1990). Rational choice theory: Necessary but not sufficient. Americam
Psychologist, 45, 356–367.
Kahneman, D. (2003). Maps of bounded rationality: Psychology for behavioral economics.
The American Economic Review, 93, 1449–1475.
Kahneman, D. (2011). Thinking, Fast and Slow. London: Allen Lane.
Kahneman, D. and Tversky, A. (1979). Prospect theory: An analysis of decisions under
risk. Econometrica, 47, 263–291.
252 Richard Wortley
Kleck, G., Sever, B., Li, S. and Gertz, M. (2005). The missing link in general deterrence
research, Criminology, 43, 623–660.
Leith, K.P. and Baumeister, R.F. (1996). Why do bad moods increase self-defeating
behavior? Emotion, risk tasking, and self-regulation. Journal of Personality and Social
Psychology, 71, 1250–1267.
Lerner, J.S. and Keltner, D. (2001). Fear, anger, and risk. Journal of Personality and Social
Psychology, 81, 146–159.
Lewin, K. (1952). Field Theory in Social Science: Selected Theoretical Papers by Kurt
Lewin. London: Tavistock.
Lowenstein, G. and Lerner, J.S. (2003). The role of affect in decision making. In
R.J. Davidson, K.R. Scherer and H.H. Goldsmith (eds) Handbook of Affective Sciences
(pp. 619–642). Oxford: Oxford University Press.
Lowenstein, G., Nagin, D. and Paternoster, R. (1997). The effect of sexual arousal on
expectations of sexual forcefulness. Journal of Research in Crime and Delinquency,
34, 443–473.
Loewenstein, G.F., Weber, E.U., Hsee, C.K. and Welch, N. (2001). Risk as feelings.
Psychological Bulletin, 12, 267–286.
McAdams, R.H. and Ulen, T.S. (2009). Behavioral criminal law and economics. In
N. Garoupa (ed.) Criminal Law and Economics (pp. 403–436). Northampton, MA:
Edward Elgar.
MacDonald, P.K. (2003). Useful fiction or miracle maker: The competing epistemological
foundations of rational choice theory. The American Political Science Review, 97(4),
MacDonald, T.K., MacDonald, G., Zanna, M.P. and Fong, G.T. (2000). Alcohol, sexual
arousal, and intentions to use condoms in young men: Applying alcohol myopia theory
to risky sexual behavior. Health Psychology, 19, 290–298.
Marcel, A.J. (1983). Conscious and unconscious perception: Experiments on visual
masking and word recognition. Cognitive Psychology, 15, 197–237.
Moors, A. and De Houwer, J. (2006). Automaticity: A theoretical and conceptual analysis.
Psychological Bulletin, 132, 297–326.
Nee, C. and Meenaghan, A. (2006). Expert decision-making in burglars. British Journal of
Criminology, 46, 935–949.
Nisbett, R. and Ross, L. (1980). Human Inference: Strategies and Shortcomings of Social
Judgment. Englewood Cliffs, NJ: Prentice-Hall.
Pease, K. and Laycock, G. (2012). Ron and the Schipol fly. In N. Tilley and G. Farrell (eds)
The Reasoning Criminologist: Essays in honour of Ronald V. Clarke (pp. 172–183).
London: Routledge.
Samuelson, W. and Zeckhauser, R. (1988). Status quo bias in decision making. Journal of
Risk and Uncertainty, 1, 7–59.
Savage, L. (1954). The Foundations of Statistics. New York, NY: Wiley.
Simon, H.A. (1955). A behavioral model of rational choice. Quarterly Journal of
Economics, 69, 99–118.
Simon, H.A. (1983). Reasoning in Human Affairs. Oxford: Blackwell.
Slovic, P., Peters, E., Finucane, M.L. and MacGregor, D.G. (2005) Affect, risk, and
decision making. Health Psychology, 24(4) (Suppl.), S35–S40.
Tedeschi, J.T. and Felson, R.B. (1994). Violence, Aggression, and Coercive Actions.
Washington, DC: American Psychological Association.
von Neumann, J. and Morgenstern, O. (1944). Theory of Games and Economic Behavior
(1st edn). Princeton, NJ: Princeton University Press.

Page numbers followed by ‘f’ refer to 246–8, 249; cognitive errors 241–4,
figures, followed by ‘n’ refer to notes and 249; concept of self-control 17, 21;
followed by ‘t’ refer to tables. a descriptive decision-making model
241; emotion 241, 244–6, 249; lack of
9/11 attacks 188, 202 investigation of 239, 249
Brayley, H. 6, 146, 165, 209, 223
Abelson, R.P. 102, 143, 165, 222, 230, 231 bulk cash smuggling 201
absconders from juvenile residential burglary, British study demonstrating
school 1 highly specific nature of 210–11
advertisements 155 burglary, computer-based scenario
African American communities 14, 22 approach to study of decisions on
aggression: defining 12–13; and deviance targets 31–47; advantages of approach
13; instrumental vs. reactive 15–17, 26–7; computerized information
244; research on effects of cues 247; board 34; conclusion 43–5; cue
sources of 16, 244 combinations 33–4, 39, –40, 40t,
Al Qaeda 186, 188, 189, 204n 41t; cues selected 36–7, 36f, 37t,
alcohol 18, 57, 76 38–9, 39t, 43–4; decision tree 41–3,
Allaire, J-F. 48, 103, 104, 122, 143, 222 42t, 232; development of cues 31–3,
Anderson, C.A. 15 32–3t; effects of age and experience
anger 15, 16, 246; motivating stalking 73, 35–6, 35f, 38, 38f; features of an
88, 91 attractive target 43–4; first cue
Ariely, D. 16, 245 selected 38–9, 39t; implications for
armed robbery 5, 20 prevention of burglaries 44; method
arrest: auto thieves perceptions of risk of 31–5, 44–5; number of cues selected
58–61, 65; estimating chances of 243 40; participants 31; procedure 34–5;
auto theft see motor vehicle theft, risks and research on burglar decision-making
rewards of 27–31; results 35–43; strengths and
automatic cognitive processing 246–8, 249 limitations of method employed 44–5
burglary, research on decision making
Baldwin, M.W. 102, 116, 229, 230, 232 5, 27–31; alarms 29; automatic
Beauregard, E. 5, 19, 20, 48, 74, 101, 103, processing and schema of experienced
115, 122, 143, 146, 209, 222 burglars 247; categories of burglar 27;
behavioural economics 16 Cromwell and Olson’s study 28–31;
Belli, R. 186, 187, 188, 189, 203, 204n dogs 29; heroin use and burglary
Bennett, T. 5, 27, 29, 30, 48, 51, 66 27, 31; house location 30–1; inside
blame, assigning 16 information 29; levels of affluence
blending in by auto thieves 62 29; locks and security 30; means of
bounded rationality 50, 74, 93, 239; travel for burglary 28; motivations for
automatic cognitive processing burglary 27; selecting targets 28; signs
254 Index
of occupancy 29–30; visibility 30 purchase 196; purchase, application of
Bushman, B.J. 15, 18 SCP techniques 198–201; residency
192, 195; residency, application of
Camerer, C.F. 243 SCP techniques 197; situational crime
capable guardian concept 227 prevention 191–2; situational crime
caviar 211, 212 prevention techniques, application
cheque forgery, crime scripts 6 of 193–4t, 197–202; ties 195; ties,
Cherbonneau, M. 5, 51, 53, 56, 62, 64, 65, application of SCP techniques 197–8;
66, 165 timeline of events 192–6, 193–4t; tools,
Chermak, S.M. 186, 188, 189, 191, 198 control of 200; tools used 193–4t;
child sex trafficking crime scripts 146 transportation, storage and distribution
child sexual abuse, interpersonal scripts 196; transportation, storage and
and victim reactions in 101–19, distribution, application of SCP
229–30; conclusion 116–17; crime techniques 198–201
scripts and 103; crime scripts and cigarette smuggling in Germany 122, 191,
interpersonal scripts 101–2, 116–17; 204
discussion 112–16; duration of sexual cigarette trafficking in USA, problems
contact 13t, 108, 109, 112, 114; index with 187–9
of intrusiveness in sexual behaviours CITES (Convention on International Trade
performed 107; intrusiveness of in Endangered Species of Wild Fauna
offender sexual behaviours 107, 109, and Flora) 210, 213, 215
110t, 112, 114; intrusiveness of victim Clarke, R. 1, 2, 3, 4, 5, 6, 48, 51, 70, 73,
sexual behaviours 107, 108, 109, 74, 101, 122, 142, 153, 186, 190, 191,
111t; limitations of study 115–16; 192, 197, 198, 204n, 209, 216, 223,
method 105–8; persuasive and non- 224, 225, 233, 237, 238, 239, 244, 248,
persuasive strategies of offender 106, 249
107, 109, 112, 114, 115; prevention cognitive errors 241–4, 249; distant and
programs for victims, designing proximal rewards 243–4; distorted
effective 115; procedural variation weightings to probability estimates
107, 112, 115; recruitment of 243; gains and losses 242; loss aversion
participants 105; research on modus 242–3
operandi, current 103–4, 104f; results cognitive science, borrowing from xix, 44,
109–12; sample 105; self-reported 143, 146, 165, 250
survey 105; stages of offender–victim Cohen, L.E. 120, 121, 224, 228
interchange 101; statistical analysis community policing 198
108; variables and analytic strategy comparability 240
105–8, 106t; victim characteristics confirmation bias 245
108, 109, 112, 114; victim compliance consensual crimes 135
and non-compliance 106–7, 112, Contraband Cigarette Trafficking Act
114–15, 115; victim reaction, offender (CCTA) 199
response and duration of sexual controlled cognitive processing 246–7, 248
contact 112, 113t, 114 Convention on International Trade in
choice model 2, 3 Endangered Species of Wild Fauna and
cigarette smuggling and terrorism Flora (CITES) 210, 213, 215
financing, script approach 186–208; Copes, H. 5, 48, 50, 51, 52, 54, 56, 58, 59,
analysis 192–202; case summary 62, 64, 66, 67, 165, 243
190; crime scripts 190–1, 193–4t; Cornish, D. 1, 2, 3, 4, 6, 48, 51, 70, 73, 74,
data 189–90; facilitating conditions 101, 102, 142, 143, 145, 166, 190, 191,
193–4t; financing mechanisms 196; 192, 209, 216, 221, 223, 231, 233, 237,
financing mechanisms, application 238, 239, 244, 248, 249
of SCP techniques 201–2; fraudulent corruption in public procurement (CPP)
loan 196; method 190–2; problem of 164–85; aims of study 166; crime
cigarette trafficking in USA 187–9; scripts, merging key stages to visualize
prosecutions and convictions 190; 176, 177f, 178; data and analytic
Index 255
strategies 167–8; data sources 166–7; reactions in; cigarette smuggling and
dialogue of tension between trust and terrorism financing, script approach;
fear 180; facilitators of CPP in works corruption in public procurement
contracts 176–9, 178t; identification (CPP); drug dealing in Amsterdam’s
of ‘target/need for’ and of partner Red Light District; human trafficking
(KS1) combined with facilitators for sexual exploitation; Transnational
178t, 179; identification of ‘target/ Illegal Market in Endangered Species
need for’ and of partner (KS1) moves (TIMES) script analysis
170–2, 170f, 177f; key stages of CPP crime scripts, moving beyond offender
in works contracts 168–9; from key scripts 224–8; convergence of offender,
stages to moves 169–76, 169t; meeting, victim and guardian 228–30; guardian
negotiation and conclusion of corrupt scripts 225t, 226–8, 228t; script
deal (KS2) combined with facilitators analysis applied to main actors in a
178t, 179; meeting, negotiation and crime event 225, 225t; underutilization
conclusion of corrupt deal (KS2) of approach 232–3; victim scripts
moves 173–4, 173f, 177f; methods 224–6, 225t, 226t
166–8; participants 167; performance crimes of passion 244
guarantee and execution of corrupt criminal persistence: research on street
deal (KS3) combined with facilitators offenders and 50–1; risks and rewards
178t, 179; performance guarantee of auto theft and implications for 48–9,
and execution of corrupt deal (KS3) 65–7
moves 174–6, 174f, 177f; from rational Cromwell, P.F. 5, 27, 28, 29, 30, 31, 48, 51
choice perspective 164–5; results crowd violence 5
168–79; situational crime prevention Cupach, W.R. 71, 72, 73, 77, 78, 79, 80,
interventions 180–1 90
crime scripts 6; applied to different cut-points 79
criminal phenomena 143, 165, 191, cyber crime 5
209; cheque forgery 6; child sex
trafficking 146; for complex crimes decision trees 41–3, 42t, 232
xx; from concept of scripts in cognitive Decker, S. 5, 22n, 29, 30, 32, 50, 51, 54,
psychology 6, 102, 165, 221, 250n; 58
designing prevention measures with depression, impairing self-control 18–19
aid of 134–5; drug manufacturing descriptive decision-making models 241,
6; a dynamic changing process 242, 246
203–4; employee computer crime desisting from crime 49, 67
143; flexibility of 230–2; general deviance: aggression and 13; defining 13;
scripts 122, 124–5, 125t, 135; human sociology of 4
trafficking and 142–4; illegal waste differential association 53
smuggling 146; and interpersonal displacement 17, 21
scripts xx, 101–2, 103, 116–17, 222, dispute-related offences 14–15, 19, 20
229; for irrational and unpreventable distractions 231
crimes 5, 74, 222; a modified script domestic violence 20, 79, 80, 116
analysis for TIMES 216–17, 217–19; dominance 240
motor vehicle theft 49, 143, 165, drug dealing in Amsterdam’s Red Light
232; organized crime 6, 143, 159, District 120–39, 125t; agreeing on and
165, 209, 222; permutations 231–2; going to trade locale 129–30; agreeing
popularity 143–4; scripts and analysis trade terms 130–1; customer schema
of 221–3; sexual crimes 5, 103, 122, 127; data collection 123–4; dealer
143, 222; steps in 145–6, 223, 223t; adversaries 134; drug addicts 126,
suicide bombing 223; terminology 127, 134, 135; drug tourists 127, 130,
146; terrorism 186; track scripts 131, 134; facilitating steps 125–30;
122, 124–5, 125t, 135–6; vandalism failure 133–4; formal control 120–1;
165 see also child sexual abuse, fraud 132–3; general scripts of drug
interpersonal scripts and victim dealing 124–5, 125t, 135; implications
256 Index
for prevention 134–6; informal control problem in US 187–9; situational crime
133; law enforcement practices and prevention techniques 201–2, 203, 204
costs to society 136; making an forged documents 155–6, 212, 213
exchange 131–2; methods 122–4; Freilich, J.D. 5, 186, 188, 189, 191, 198,
necessary steps 130–3; opportunity and 204n
rationality in choices 121–2, 127, 129, Froming, W.J. 18
132, 134, 135; participants 123; rip-offs funds, lack of research 212–13
132; script analysis 124–34; searching
for a customer 126–8; soliciting gains and losses 242
customers 123, 128–9; track scripts game theory 165
122, 124–5, 125t, 135–6; walking or general scripts 122, 135; of drug dealing
standing 127–8 124–5, 125t, 135
drug manufacturing crime scripts 6 giving up crime, motivations for 49, 67
drugs, burglary to finance use of 27, 31 Grapendaal, M. 120, 121, 122, 123, 128,
132, 136
Eastern Europe sex trafficking to Italy guardian scripts 225t, 226–8, 228t; and
case study 144, 145, 147–50, 149f; convergence with offender and victim
situational crime prevention techniques scripts 228–30
153–8, 158t guilt, neutralizing of 56, 67; auto thieves
ECDB (Extremist Crime Database) 188, 56–8, 66–7
189, 205n
Edwards, W. 241 Hamas 186, 187, 189, 195, 204n
Ekblom, P. 154, 204, 215, 216, 230, 231 harm-doing 12–14; auto thieves denial
elephant ivory 211, 215 of 58; relationship between attitude
emotion 66, 241, 244–6, 249 toward harm and method of harm doing
employee computer crime, crime scripts 14–15, 15t
143 Hayward, K. 3, 4, 5, 249
environmental influences on crime 1–2 heroin use and burglary 27, 31
evaluation of seriousness of a crime, Hezbollah 186, 187, 188, 189, 190, 195,
methods for 107–8 196, 202, 204n
event decisions 48–9, 164, 165, 180 homicide: applicability of rational choice
event schemas 6, 103, 222, 247 perspective to 244; motivations for
expected emotions 246 13–14, 21–2; outcome dependent on
expected utility model 239–41 victim’s behaviour 230; self-control
Extremist Crime Database (ECDB) 188, and 13, 20
189, 205n human trafficking: corruption in
155; crime script and 142–4; data
fear: and anxiety in committing a crime comparability problems 142; defining
61, 66; CPP and dialogue of tension 140–1; Dutch legislation to predict
between trust and 180; of crime, model victims of 152, 156; forged documents
to deal with 224; of retaliation 14, 155–6; inconsistencies in data
21–2, 142, 153 collection 141–2; literature on 141;
Fehr, B. 102, 116, 229 phases of 141; under-reporting of 142;
Felson, R. 4, 13, 14, 15, 16, 18, 19, 20, 21, traffickers 141; victims 141
50, 74, 117, 120, 154, 165, 167, 176, human trafficking for sexual exploitation
181, 197, 224, 225, 227, 228, 244 144–63; aftermath 149, 149f, 151f,
financial institutions, misuse of non- 153; aftermath, prevention 157;
banking 157 data sources 144–5; debt of victims
financial terrorism 186–7, 188, 203; 152, 153; exploitation 147–8,
by domestic extremists in US 186, 149f, 151f, 152–3; exploitation
188–9; ECDB study 189; excluded prevention 156–7; ‘the joint’ 152–3;
from terrorism studies 186; financial law enforcement and collaboration
scheme concept 188; future research with victims 153, 157; methodology
to understand 203; importance of 145–6; Nigerian ‘madams’ 145, 150;
Index 257
offender adaptation 156; recruitment Magnus, N. 212
147, 149f, 150, 151f; recruitment Maguire, M. 27
prevention 154–5; sex trafficking Marcel, A.J. 247
from Eastern Europe to Italy case master scripts xix–xx, 210; of TIMES 210,
study 144, 145, 147–50, 149f; sex 212, 215, 216, 217
trafficking from Nigeria to Italy Matza, D. 56
case study 144, 145, 150–3, 151f; Mead, G.H. 19
situational crime techniques for Meenaghan, A. 5, 26, 27, 28, 29, 30, 44,
preventing 153–8, 158t; stages in 247
process of 146; terminology 146; money laundering 157, 201, 203; National
transportation 147, 149f, 150–2, 151f; Money Laundering Strategy 2002
transportation prevention 155–6; 201–2
victims as exploiters 153; voodoo Morgenstern, O. 165, 239
rituals 145, 153 Morselli, C. 107, 143, 146, 165, 209, 222,
immediate emotions 246 motor vehicle theft, risks and rewards of
immigrant communities in US 198 48–69; belief in protection of juvenile
impulsive crime: implications for status 60; blending in 62; conclusion
situational crime prevention 20–1; vs. 65–7; crime scripts 49, 143, 165, 232;
premeditated crime 19–20 driving to avoid attracting attention 64;
independence 240 fear and anxiety in committing a crime
injury, auto thieves denial of 58 61, 66; financing extravagant lifestyles
instrumental aggression, vs. reactive 50, 53, 54, 66; guilt neutralization
aggression 15–17, 244 techniques 56–8, 66–7; implications
interferences 231 for criminal persistence 48–9, 65–7;
international borders, illegal trade across for joyriding 54–5; license plates 63–4;
142, 187–8, 211, 217–18 maintaining a positive self-image 56;
interpersonal scripts xx, 101–2, 103, methods 51–2; money a primary motive
116–17, 221; example of 229; for 53; need for proper connections
relational schemas and 228–30 see 53–4, 66; offenders views on probation
also child sexual abuse, interpersonal and incarceration 60; participants 52;
scripts and victim reactions in rational choice perspective in context
INTERPOL 210, 213 49–51; research on offender decision
invariance 240 making 5; reward, perceptions of 50–1,
involvement decisions 49, 51, 164 52–8, 65–6, 67; risks, perceptions of
irrational crimes 5, 74, 222 50–1, 58–65, 67; sacrifice of legitimate
employment for 66; sampling strategies
Jacques, S. 14, 52, 124, 126, 130, 131, 52; selection of models to minimise
132, 133, 134 suspicions 62–3; shift to persistent
joyriding 54–5 offending 55, 65; strategies to avoid
juvenile residential school absconders 1 detection 62–5; views on what would
juvenile status, belief in protection of 60 happen if caught 58–61

Kahneman, D. 241, 242, 243, 246, 248 National Money Laundering Strategy 2002
Leclerc, B. 5, 6, 19, 20, 48, 74, 101, 103, Nee, C. 5, 26, 27, 28, 29, 30, 44, 59, 247
104, 106, 108, 114, 115, 122 Nelen, H. 120, 132
Leuw, E. 120, 121, 132, 136 Netherlands: legalization of prostitution
license plates 63–4 156; legislation to protect victims of
lifestyles xix; of auto thieves 50, 53, 54, human trafficking 152, 156 see also
66; and impact on choices 50 drug dealing in Amsterdam’s Red Light
loss aversion 242–3 District
Lowenstein, G. 245, 246 Neumann, J. von 165, 239
258 Index
neutralization techniques 56, 67; of auto Opium Act 1976 120
thieves 56–8, 66–7 opportunity theory 2, 121, 127
Newman, G.R. 5, 6, 74, 186, 191, 192, optimism bias 243
198, 204n organized crime: crime scripts 6, 143,
Nigerian sex trafficking to Italy case study 159, 165, 209, 222; model for TIMES
144, 145, 150–3, 151f; debt of victims 213–14; situational crime prevention
152, 153; the’joint’ 152–3; Nigerian applicable to 5
‘madams’ 145, 150; situational crime
prevention techniques 153–8, 158t; paranoia 16–17
victims as exploiters 153; voodoo parrot trade, neo-tropical 211, 217
rituals 145, 153 Patriot Act 2001 201, 202
normative models of rationality 241 permutations, concept of 231–2
person schemas 222
obstacles 221, 231, 232 planning crimes 19–20
offender adaptation 156; to preventive police officers’ shooting scenarios 226–7
interventions 203–4; in TIMES predatory offences 14–15, 19, 20
interventions 215–16 preference reversal 243–4
offender decision making 239–50; and premeditated crime: implications for
attacks on situational crime prevention situational crime prevention 20–1; vs.
249–50; automatic cognitive processing impulsive crime 19–20
246–8, 249; choosing or abstaining Prevent All Cigarette Trafficking (PACT)
from crime 49, 67; cognitive errors Act 2010 199
241–4, 249; comparability 240; priming 247–8
controlled cognitive processing 246–7, prison, views on 60
248; crime-specific nature of 3, 51, prisoners: depression and self-control
166; descriptive decision-making 18–19; paranoia a predictor for violent
models 241, 242, 246; dominance 240; offences 16–17
and emotion 241, 244–6, 249; expected probability estimates, distorted weightings
utility model 239–41; gains and losses 243
242; growing body of literature on probation, views on 60
5; independence 240; on initiating a ‘The procedural analysis of offending and
corrupt career 164–5; invariance 240; its relevance for situational prevention’
loss aversion 242–3; methods for 221
studying xvii; probability estimates, procedural variation 107, 112, 115
distorted weightings 243; rational prospect theory 241, 242, 242–3
decision making 239–41; rewards, prostitution: Dutch legislation on 156;
distant and proximal 243–4; satisficing sanctions against clients 157 see
decision making 241; sociocultural also human trafficking for sexual
context of 50–1; subjective expected exploitation
utility model 240; transitivity 240 see Proulx, J. 48, 101, 103, 104, 122, 143, 222
also burglary, computer-based scenario public transport, crimes against employees
approach to study of decisions on and passengers, crime scripts 5
targets; burglary, research on decision
making; motor vehicle theft, risks Racketeer Influenced and Corrupt
and rewards of; stalking, study of sex Organizations Act (RICO) 190
differences in; violent crime, offenders rape: emotional arousal and committing of
decision making during 244, 245; involving predatory violence
offender scripts: and convergence with 15; planning 20
victim and guardian scripts 228–30; rapid assessment techniques 216
moving beyond 224–8 rational choice perspective: appeal of
offender–victim interchange see child 48; applicability to homicide 244;
sexual abuse, interpersonal scripts and basic tenets 2–3; in context of street
victim reactions in offenders 49–51; crime-specific nature
Olson, J.N. 5, 27, 28, 29, 30, 31, 48 of 3, 51, 166; decisions to choose or
Index 259
abstain from crime 49, 67; design and to probability estimates for 243;
intended function xviii, xix, 237–9; emotional arousal and willingness to
development 1–3; elaborations over take 245
25 years xviii–xix; future development robberies 20, 21; armed 5, 20; guardian
xvii–xix; human trafficking from interfering before a street robbery,
142–3; informing research on a career hypothetical script 228, 228t; involving
in CPP 164–5; for irrational crimes predatory violence 15; outcome
5, 74, 222; limit