Cândido da Agra, C., Cardoso, C., de Maillard, J., O’Reily, C., Ponsaers, P ., Shapland, J. (eds.) (2015). Criminology, Security and Justice: Methodological and Epistemological issues , GERN Research Paper Series n°3, Antwerpen/Apeldoorn/Portland: Maklu, pp. 186.

Abstract

This is the third volume stemming from the annual doctoral conferences organizedby the GERN (Groupement Européen de Recherches sur les Normativités). The GERN is a multidisciplinary consortium of universities, research institutes and scientific researchers in the domain of criminology, deviance and social control. This scientific network of academics organize different types of scientific events (one-day conference-’interlabos’, conferences, research seminars and workinggroups), aiming to foster the culture of collaboration that characterize this group.

Full text

Criminology, Security and Justice
maklu

maklu

GERN Research Paper Series – nr 3
Criminology, Security and Justice
Methodological and epistemological issues
Edited by
Cândido da Agra
Carla Cardoso
Jacques de Maillard
Conor O’Reily
Paul Ponsaers
Joanna Shapland
Maklu
Antwerpen | Apeldoorn | Portland
maklu

Cândido da Agra, Carla Cardoso, Jacques de Maillard,
Conor O’Reily, Paul Ponsaers, Joanna Shapland (eds.)
Criminology, Security and Justice. Methodological and epistemological issues
Antwerpen | Apeldoorn | Portland
Maklu
2015
186 pag. – 24 x 17 cm
ISBN 978-90-466-0778-7
D 2015/1997/45
NUR 822
© 2015 Maklu-Uitgevers nv and the authors
All rights reserved. No part of this publication may be reproduced, stored in a retrieval
system, or transmitted in any form or by any means, electronic, mechanical, photocopy-ing,
recording, or otherwise without the prior permission of the publisher.
Maklu-Publishers
Somersstraat 13/15, 2018 Antwerpen, Belgium, info@maklu.be
Koninginnelaan 96, 7315 EB Apeldoorn, The Netherlands, info@maklu.nl
www.maklu.eu
USA & Canada
International Specialized Book Services
920 NE 58th Ave., Suite 300, Portland, OR 97213-3786,
orders@isbs.com, www.isbs.com
maklu

Maklu 5
Table of contents
Criminology, Security and Justice 7
Methodological and epistemological issues
Cândido da Agra, Carla Cardoso, Jacques de Maillard, Conor O’Reily,
Paul Ponsaers, Joanna Shapland
Should I stay or should I go? 15
Reflecting about experienced methodological issues
Camille Claeys, Els Dumortier & Sofie De Kimpe
Bail decision-making in England and Wales and Canada 41
The value of a comparative approach
Diana C. Grech
Studying the contextual cues associated with fear of crime through eye
tracking techniques 57
An exploratory research
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
Dangerous and endangered 83
The construction of male-homosexual sex work as a social problem in
Germany
Marlen S. Löffler
Studying police use of force 105
Definitional challenges and methodological considerations
Jannie Noppe
Methodological challenges of researching riots in small geographical
communities 133
Deborah Platts-Fowler
The Portuguese version of the YLS/CMI 151
Preliminary data
Alberto Pimente- & Jorge Quintas
Suburban areas behind walls 169
Handling territorial identities in a French prison
Manon Veaudor
maklu

maklu

Maklu 7
Criminology, Security and Justice
Methodological and epistemological issues
Cândido da Agra, Carla Cardoso (School of Criminology, Faculty of Law,
University of Porto, Portugal)
Jacques de Maillard (Université de Versaille Saint-
Quentin-en-Yvelines, France / CESDIP and
Institut Universitaire de France)
Conor O’Reily (Durham University, UK)
Paul Ponsaers (Ghent University, Belgium)
Joanna Shapland (University of Sheffield, UK)
This is the third volume stemming from the annual doctoral conferences organized
by the GERN (Groupement Européen de Recherches sur les Normativités). The
GERN is a multidisciplinary consortium of universities, research institutes and
scientific researchers in the domain of criminology, deviance and social control.
This scientific network of academics organize different types of scientific events
(one-day conference-’interlabos’, conferences, research seminars and working
groups), aiming to foster the culture of collaboration that characterize this group.
Each year the GERN organize a doctoral summer school, given the opportunity to
the PhD students from these universities and research institutes to present and
discuss their ongoing projects and research results as well as meet young and
senior researchers.
After the first conference organized by the Ghent University in 2012 and the
second organized by the White Rose Consortium of Leeds, Sheffield and York
in 2013, this third conference was held in September 2014at the School of
Criminology of the Faculty of Law (University of Porto). The selected theme for
this Summer School was ‘Criminology, Security and Justice: methodological and
epistemological issues’, searching for a fruitful debate about the methodological
and epistemological aspects relevant for the development of PhD thesis.
Scientific research is, in its essence, critical thinking. What is critical thinking?
It is a kind of thinking that differs from magic reasoning, common sense, and
speculation. Scientific evidence contrasts with belief, immediate and apparent
knowledge, illusion and opinion. This is valid for every knowledge domain that
claims to be scientific. It is thus true for the science of crime, Criminology.
maklu

Cândido da Agra, Carla Cardoso, Jacques de Maillard, Conor O’Reily, Paul Ponsaers & Joanna Shapland
8 Maklu
Our discipline is especially permeable to infiltration by non-critical or acritical types
of knowledge. The approach that has been termed ‘Critical Criminology’, which
was dominant in many countries in the 60’s and 70’s, was itself deeply influenced
by many sources of non-critical thinking. This is the great contradiction of the
Criminology that termed itself Critical. This paradox was identified by the Belgian
criminologist Lon Van Outrive in the 70’s. This Professor from the University of
Leuven wondered: how is it possible that those that cultivate Critical Criminology
did not use epistemological methods?
The GERN is a research network on normativity, crime, and justice, which
pursues critical thinking. Let us follow the advice of Lode Van Outrive: let’s be
critical while using epistemological methods, by reflecting about our way of doing
criminological research. It was under this purpose that the organizers of the 2014
GERN Summer School suggested the theme ‘Epistemology and Methodology of
Criminology’. This same theme inspires the 2015 ESC conference ‘Criminology as
unitas multiplex: theoretical, epistemological and methodological developments’
in Porto.
Which are the big questions posed by Epistemology and Philosophy of Science?
They are, in short, the following: How do we distinguish science from other forms
of thinking? What is science? How is it done? How different areas of knowledge
are born, and how do they evolve?
The answer to these questions relies in two traditions: the French historicist
tradition (e.g., G. Bachelard, G. Canguilhem) and the Anglo-Saxon logicist
tradition (e.g., K. Popper, T. Khun). The great philosopher of the 20th century,
Michel Foucault, occupies space apart from these. In fact, he inaugurates, in the
sixties of the 20th century, a new approach to the analysis of knowledge, sciences
and practices. As it is widely known, this philosopher dedicated a great part of his
research to normativity, the formal means of control, penal law, and law in general.
More recently, analytical philosophy has been showing interest in the matters of
crime and justice.
It may be said that all these reflections are useless. It may be said that what is
really important is to collect data and analyse it with sophisticated software,
which show us results in an immediate way. Or that what is interesting is to do
ethnographic research, to conduct interviews, to elaborate life histories. And reach
the point of ‘theoretical saturation’. Who asks ‘what is this complex data analysis
software, which are the theoretical assumptions of these statistical models’? Who
asks ‘what is theoretical saturation and how do we reach it with objectivity’? What
are quantitative methods? What is quantitative Criminology? What are qualitative
methods? Is there one or many qualitative Criminologies? How do we conjugate
quantitative and qualitative approaches? Everyone knows there is no Science
maklu

Criminology, Security and Justice
Maklu 9
without comparison. To compare is a cognitive operation essential for scientific
thinking. What is the comparative method? Which are its rules?
In sum, the aim of this Summer School was to debate critically the methodological
and epistemological issues that the students are facing during the PhD research,
and ultimately provide the students with tools that facilitate to make choices in a
critical informed manner, in particular the way to operationalize the theoretical
framework to better develop their projects.
The Summer School also included a combination of courses, given by Cândido
da Agra (University of Porto), Jacques de Maillard (Université de Versailles
Saint-Quentin-en-Yvelines/CESDIP and Institut Universitaire de France), Paul
Ponsaers (Ghent University) and Joanna Shapland (Sheffield University) and
individual presentation of papers previously prepared by the students. A panel of
senior academics joint this group, Conor O’Reilly (Durham University), Amadeu
Recasens (University of Porto), Jorge Quintas (University of Porto) and Pedro
Sousa (University of Porto) to also commenting on the papers. The best papers
were anonymously peer reviewed and are published in this Maklu series.
The papers
The papers fall naturally into the major theme of the Summer School. The doctoral
students and researchers concentrate on critical discussions of the existing literature
but also about the methodology. They present a reflexion that reveal critical distance
from the selected methodologies. Some papers are quantitative (statistical analysis,
experimental methods) in nature, other qualitative (ethnography, interviews) and
other used mixed methods applied to different domains of Criminology. The
editorial board are very proud to present this volume.
Should I stay or should I go? Reflecting about experienced methodological
issues
Camille Claeys, Els Dumortier and Sofie De Kimpe (Vrije University, Belgium)
This paper derives from the fieldwork experiences of a female PhD student, doing
ethnographic research in the field of police interrogations of juvenile delinquents,
and the discussions with her supervisors. As there has only been a small-scale
empirical research in Belgium on how police interrogates juvenile offenders
in daily practice, the researcher and her supervisors want to discover through
ethnography (more specifically observations and complementary interviews with
the police interrogators) how the interrogations are conducted.
After her first steps into the field (a Belgian Police Youth Unit) and drawing
on her first empirical data, the researcher and her supervisors reflect on their
relationship and on the approach the researcher used to access the field. The
maklu

Cândido da Agra, Carla Cardoso, Jacques de Maillard, Conor O’Reily, Paul Ponsaers & Joanna Shapland
10 Maklu
researcher also reflects on her ‘personal access’ to the field, her personal ‘baggage’
and several methodological issues such as: the role as an observer and avoiding
over-identification. As reflecting about these issues (arising from both the research
aims and the research context) affects the way research is approached and data is
interpreted – and because reflection about this topic is sometimes overlooked in
literature – the researcher and her supervisors believe this reflection is significant
from a theoretical as well as a practical point of view. Methods are a crucial part
of the scientific enterprise and reflexivity offers an opportunity to discuss and
overthink them. Consequently, this paper underlines the relevance of reflecting
about methodology and is intended as a contribution to the discussions about
qualitative research methods in a police setting.
Bail decision-making in England and Wales and Canada. The value of a
comparative approach
Diana C. Grech (University of Leeds, UK)
The use of bail plays a central role in the rise of prison remand populations,
influences defendants and criminal justice institutions, and raises human rights
concerns, yet relatively little is known about the bail decision-making process. For
the purposes of the paper, this is considered to be the process by which court
practitioners decide whether defendants should be released on bail into the
community or detained in custody pending trial. It also involves the decision as
to which, if any, conditions should be imposed upon those released. The objective
of this paper is to demonstrate the value of a comparative approach in developing
a better understanding of the bail decision-making process. It is anticipated that
such an approach will shed light on the relationship between bail practices and
prison remand population trends. The value of a comparative approach will be
illustrated through a preliminary examination of bail decision-making in England
and Wales and Canada.
Studying the contextual cues associated with fear of crime through eye
tracking techniques. An exploratory research
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso (University of
Porto, Portugal)
Fear of crime is a hot topic in the criminological field and a problem in its own
right. Researchers have studied the individual and contextual variables involved
in fear of crime. This exploratory study uses eye tracking techniques to explore
the features of environment that have an impact on fear of crime. A set of urban
pictures were obtained, rated by an external sample and selected to be visualized.
These were then presented to a sample of 10 individuals that observed the pictures
and judged whether the place was insecure or not. Heat map analyses revealed
that the features that participants paid more attention to were the prospect of
the environment and signs of disorder (mainly graffiti). However, comparisons
between pictures did not reveal statically significant differences for latency,
maklu

Criminology, Security and Justice
Maklu 11
number of fixations and total duration of fixations. The results and limitations of
this research will be discussed and a new improved eye tracking protocol will be
outlined.
Dangerous and endangered. The construction of male-homosexual sex work
as a social problem in Germany
Marlen S. Löffler (Tuebingen University, Germany)
Within the realm of sociology of social problems, male homosexual sex work in
Germany can be understood as a social problem. This article serves to investigate
the rise of male homosexual sex work as a social problem in Germany. It presents
some initial results about the problem career of male homosexual sex work on
the macro-sociological level in the time period from 1980 to 1985. It shows that
male homosexual sex work was discussed along the lines of criminalisation and
victimisation. The argumentation will focus on the presentation of two articles,
which on the one hand, pose male homosexual sex workers as dangerous people
and, on the other hand, as endangered people.
Studying police use of force. Definitional challenges & methodological
considerations
Jannie Noppe (Ghent University, Belgium)
In this article it is scrutinised previous (empirical) research on legitimate use
of force by the police with a focus on methodology. First it is described some
definitional challenges each police scholar in this domain is confronted with.
Secondly, it is considered the strengths and weaknesses of different research
methods in relation to the study of police use of force. It is examined the following
five methods: (1) studying official records, (2) survey-research, (3) observational
studies, (4) interview methods and (5) mixed methods. Although the paper did
not focus on the research results, it was also found an important gap in research
on police use of force. The author determined that the police officer’s perspective
on and experiences with use-of-force incidents are mostly neglected. Based on
that finding, at the end of the article, the author will introduce their own research
project on how Belgian police officers think about the legitimate use of force
and how they handle use-of-force incidents. While referring to the results of the
methodological review the author reflect upon the best-suited research design for
the study.
Methodological challenges of researching riots in small geographical
communities
Deborah Platts-Fowler (University of Leeds, UK)
The paper draws on an empirical study of the 2011 English Riots. In the aftermath
of these riots, the media and political discourse spotlighted ‘contagion’ as the main
explanation for the outbreak of disorder across 66 different locations, implying
that context was irrelevant. This research used the case study method to explore
maklu

Cândido da Agra, Carla Cardoso, Jacques de Maillard, Conor O’Reily, Paul Ponsaers & Joanna Shapland
12 Maklu
the role of local contextual factors, especially local policing, partnership relations
and community engagement, in promoting and preventing unrest. The paper
outlines how the contentious nature of ‘riots’ as a subject-matter for empirical
research and the focus on small geographical areas presented particular challenges
that affected the research design, the research experience, and possibly the impact
of the findings. Methodological challenges included the unwillingness of people
to participate in the research; pressures for them to re-write history in particular
ways; risks to the anonymity of individuals and neighbourhoods involved in the
research, with the potential for social harm.
The Portuguese version of the YLS/CMI. Preliminary data
Alberto Pimentel (General Directorate of Reintegration and Prisons – Ministry of
Justice, and University of Porto, Portugal) & Jorge Quintas (University of Porto,
Portugal)
This paper reports data from the normative study of the Portuguese version of the
Youth Level of Service/Case Management Inventory (YLS/CMI), risk assessment
tool designed to assess risk for general recidivism, identify criminogenic needs
and guide case management of young offenders. Normative data of Portuguese
YLS/CMI were established from a sample of 2363 young offenders from juvenile
justice system. We also analyzed the psychometric properties. The results point to
excellent indicators of intra and inter- rater reliability, a good internal consistency
and discriminative power between sub-samples. The preliminary study on
predictive validity (AUC = .77) supports the use of the Portuguese adaptation of
the YLS / CMI in predicting recidivism.
Suburban areas behind walls. Handling territorial identities in a French prison
Manon Veaudor (CESDIP, France)
This study focuses on interconnections between processes of identification in
regard with localities and prison management. The geographical and ethnic
elements are key in the present definition of the localities and territories analysed
in this study – poor urban areas especially. Here, identity is conceived at the
crossroad between the subjectivity of actors (their own definition of who they
are) and social classification. One issue is to analyse how spatial identities are
defined and negotiated between actors. Studying the prison system, a relational
approach to identity is precisely helpful to embrace both prisoners’ and guards’
representations on the field, and the ways they interfere.
From a bottom-up perspective, it is investigated how prison staff and prisoners’
identification with each other shape and is shaped by certain operational modes
within the prison itself. Drawing on some of the findings of ethnographic fieldwork
done by the author, it is outlined a preliminary hypothesis about differentiated
treatment of identities in prison. It will focus on the perceptions of Parisian
poor urban areas at work inside a long-term prison. In a largest perspective, this
maklu

Criminology, Security and Justice
Maklu 13
research falls within academic trends focusing on penal institutions and their ties
with poor areas.
Acknowledgements
The editorial board would like to acknowledge the major contribution made by
all the reviewers of the chapters for this volume, as well as those who helped
to organize the conference in Porto: Pedro Almeida, Gilda Santos and other
colleagues from the School of Criminology (Faculty of Law, U.Porto). The support
of the conference come from GERN, the University of Porto, and the Faculty of
Law (U.Porto).
maklu

maklu

Maklu 15
Should I stay or should I go?
Reflecting about experienced methodological issues
This paper derives from the fieldwork experiences of Camille Claeys, a female PhD student,
doing ethnographic research in the field of police interrogations of juvenile delinquents,
and the discussions with her supervisors Els Dumortier and Sofie De Kimpe1. As there
has only been a small-scale empirical research in Belgium on how police interrogates
juvenile offenders in daily practice, we want to discover through ethnography (more
specifically observations and complementary interviews with the police interrogators)
how the interrogations are conducted.
After her first steps into the field (a Belgian Police Youth Unit) and drawing on her first
empirical data, we reflect on our relationship and on the approach we used to access
the field. Camille Claeys also reflects on her ‘personal access’ to the field, her personal
‘baggage’ and several methodological issues such as the role as an observer and avoiding
over-identification. As reflecting about these issues (arising from both the research aims
and the research context) affects the way we approach our research and interpret our
data – and because reflection about this topic is sometimes overlooked in literature – we
believe this reflection is significant from a theoretical as well as a practical point of view.
Methods are a crucial part of the scientific enterprise and reflexivity offers an opportunity
to discuss and overthink them. Consequently, this paper underlines the relevance of
reflecting about methodology and is intended as a contribution to the discussions about
qualitative research methods in a police setting.
Camille Claeys, Els Dumortier & Sofie De Kimpe2
Introduction: the need for methodological reflections
In a qualitative research project, the methodology used determines what the
field researcher sees, experiences and learns. Without details about the process
and qualitative methods, it is difficult to understand the outcome of the research
project. In other words, if ‘data’, ‘findings’, ‘facts’ are products of the methods
used, they cannot be considered independently of the interactions and relations
with others that comprise these methods: what the ethnographer finds out is
inherently connected with how he/she finds it out (Gubrium & Holstein, 1997)3.
1 In Belgium, we use the term ‘promotor’.
2 Vrije University, Belgium.
3 As a result, these methods should not be ignored; rather, they should comprise an important part
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
16 Maklu
In this paper, a first ethnographic experience in a police force is explored through
a reflexive lens. ‘Reflexivity involves making the research process itself a focus of
inquiry, laying open pre-conceptions and becoming aware of situational dynamics’ in
which the researcher and research-subjects are jointly involved in knowledge
production (Hsiung, 2010, Reflexivity, para. 3)4. The aim of this reflection is thus
to identify some factors that can have an impact on the outcome of the research
and to discuss the methodological issues arising from both the research aims and
the research context. These issues need careful consideration since ethnography
presents researchers with a range of risks and ethical dilemmas relating to the
level of social immersion or involvement (Yates, 2004). Some authors suggest to
expand the meaning of reflexivity, so that it can be considered as a way of ensuring
not just rigorous research practice but also ethical research (Guillemin & Gillam,
2004; Davies, 2008)5.
After explaining briefly the research project and how the field was selected,
a ‘reflexive lens’ is employed to say something about the researcher-supervisor
relationship. Afterwards, the approach that was used during the entrance in
the field is described. This approach contains, among others, the strategies of
gaining access to the field and later on, obtaining information. The researcher
also dwells on her personage in this phase of entering the field. Finally, the main
methodological dilemmas that we encountered during the fieldwork are recited.
These are the observer’s paradox and the concept of ‘avoiding over-identification’.
The description of these aspects – empirically illustrated with extracts (such as
citations) from the field notes6 – is based on the researcher’s personal experience,
showing that fieldwork methodology is often interpreted in a personal and original
way.
The paper contains both common reflections of the supervisors and the PhD
researcher; but also personal reflections and interpretations of the PhD researcher
who conducts the fieldwork: Camille Claeys. To keep the difference between
both kind of reflections and interpretations, we have opted to use the first person
of written field-notes. It thus becomes critical for the ethnographer to document on his/her own
activities, circumstances and emotional responses as these factors shape the process of observing
and recording other’s lives (Emmerson, Fretz & Shaw, 2011).
4 http://www.utsc.utoronto.ca/~pchsiung/LAL/, accessed on the 3rd of august 2014.
5 Being reflexive in an ethical sense means acknowledging and being sensitized to the microethical
dimensions of research practice and in doing so, being alert to and prepared for ways of dealing
with the ethical tensions that arise (Guillemin & Gillam, 2004).
6 The fieldnotes were mainly collected by observing and reflecting on the observations. They were first
quickly written down in a notebook and then later organized as a chronological journal with dates
of entries. The observations and personal reflections were made in the same text, but separated
through a different color. The first field notes are characterized by quite broad and indiscriminate
descriptions, becoming narrower and more focused on specific topics as the inquiry proceeds.
Until now, only the first step of analysis was made: developing manually categories for labelling
chunks of data and then overthink these on a low theoretical level (Davies, 2008).
maklu

Should I stay or should I go?
Maklu 17
‘I’ when the first author’s thoughts are involved and ‘we’ when our common
reflections, interpretations and/or statements are involved.
The research project, selection of the research units and
methods
An important aspect is the fact that the research is part of a FWO7 project that was
accorded to the supervisors. This is a methodological question worth mentioning
because it brings along that the initial input and certain decisions (such as the
selection of the field) were already defined before the PhD-researcher became part
of the research team.
The aim of this research project is to discover how juvenile suspects are being
interrogated by police officers in Belgium. Police interrogations are the starting
point of criminal procedures against juvenile offenders. Research conducted
in the US suggests that juveniles would be particularly vulnerable to various
manipulative and potentially coercive techniques police uses to get suspects to
incriminate themselves (Feld, 2013; Leo & Drizin, 2004). However, empirical
research on police interrogations of juvenile offenders remains scarce. In Belgium
this seems even truer, for there has been very little empirical research on how
police interrogates juvenile offenders in practice. As a consequence, there is no
scientific knowledge on how these interrogations are conducted in (Belgian) daily
police practice. Given the American research results and given the lack of scientific
knowledge in Belgium, the project aims at shedding light on the topic through
ethnography.
The initial design of the research changed in the light of the circumstances and
the researcher’s input, for instance concerning the kind of juveniles who are being
interrogated. The researcher chose to focus solely on the interrogation of juveniles
suspected of having infringed the criminal law and not those suspected of having
committed a status offence8.
The selection of the Belgian police forces (research units) will be briefly explained,
as this choice was relevant for the aspect of ‘gaining access’ (discussed further).
The selection was based on four criteria. The first criterion is the presence of
specialized Youth Units in which the investigation of juvenile delinquency is
the core activity. The second criterion is the urbanity of the police force. As the
specialized unit is situated in a major, multicultural city where juvenile delinquency
7 A Dutch speaking Belgian fundamental research foundation.
8 Like truancy, running away, drinking under the age of 16, etc. The Police Units who were selected
are specialized in interrogation of ‘real’ offenders, suspected from infringing the criminal law.
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
18 Maklu
is a problem, this maximized the possibilities for observing interrogations with
different types of juvenile offenders. Previous research of De Schrijver and De
Kimpe (2009) showed that police agents working in specialized youth units hold a
‘specialized working view’ on the approach of youth delinquency. The supervisors
of this research therefore hoped to encounter police officers with experience in
questions of youth delinquency, which led to the third selection criterion. The
fourth selection criterion was the informal authorization, which the supervisors
received (from the chief constable of the police force) even before the research
started. This constituted a useful tool in gaining access to the field (see infra).
Ethnography generates intellectual insight by exploring the nature of particular
social phenomena in the observable world, rather than setting out to test hypotheses
about them (Liebling, 2001). We therefore believe it is the ideal approach to
examine the interaction between juvenile suspects and police interrogators.
A first method, used during this research project, is doing long period observations
in situ in two Police Youth Units of two major Belgian cities. Observation has
been characterized as ‘the fundamental base of all research methods’ in the social and
behavioral sciences and as ‘the mainstay of the ethnographic enterprise’ (Denzin &
Lincoln, 2008, 161). It has a long and legitimate lineage in research on policing
and is therefore a self-evident method (Fielding, 2006). Another method is
conducting complementary interviews with police interrogators to obtain a better
understanding of the observed practice, because ‘researchers should not privilege any
ways of looking at the world or looking at a particular technique but should instead
continue to question, question and question’ (Denzin & Lincoln, 2008, 118).
Although these research methods (observing, describing and the wish to
understand social issues) are shared with the field of anthropology and sociology,
our theoretical orientation is especially criminological and judicial. We believe
the complexity of this research demands an investigation that combines both the
analysis of legal-judicial sources as well as qualitative methods to observe and
understand the practice. The researcher has a legal background, which is useful for
the judicial aspect of the research. Because of this background however, she had to
acclimate to the methods and disciplinary perspectives of the field of Criminology.
After having conducted a preparatory review of scientific and professional literature
about police interrogation, one of the police forces was visited for a period of
three months, followed by a withdrawal period of three months. This first ‘field-
experience’ in a police force, more specifically a Police Youth Unit, is related in this
paper through reflexivity and the discussion of methodological aspects.
maklu

Should I stay or should I go?
Maklu 19
The researcher-supervisor relationship
As this relationship has an impact on the way doctoral research is conducted and
as other researchers are also concerned with it, we believe it is valuable to mention
some aspects of the way in which our relationship influenced the fieldwork.
In the research period that is recounted in this article, the relationship between
the researcher and her supervisors was not always a ‘flat road’, be it because
of different or colliding viewpoints or communication problems which are,
as literature suggests (see for instance Taylor, 2005; Mainhard, van der Rijst &
van Tartwijk, 2009), a commonly reported difficulty in this kind of relationship.
Miscommunication and disagreement on certain points led to ‘conflict’, which
engendered feelings of anger, vexation and miscomprehension on both sides (see
further). But this ‘conflict’ also induced – for the researcher – the positive exercise
of evaluating distance and positioning and – for the supervisors – reconsidering
their way of supervising.
Premeditated and thoughtful: approaches and strategies to
enter the field
Gaining access to the field: a continual negotiation?
In literature, gaining access is often subdivided in stages: ‘pre-entry’, ‘during
fieldwork’, etc. In this paper, our ‘gaining access process’ is also categorized in
stages or steps.
The first step: ‘formal access’ or ‘pre-entry’
As the police are generally considered a closed setting where entry is not easily
approved (Pruvost, 2007), the first step or stage to gain access was the formal
authorization, after the previously received informal authorization (see above). We
therefore asked and obtained the written permission of the public prosecutor, the
chief constable, as well as the local head of the unit, which are in fact the highest
level of permission. From the moment this formal request was obtained, almost
no resistance was felt from the police force. Surprisingly, after having received
these formal ‘pre-entry authorizations’, the ‘gates of the field’ opened widely and
negotiating access for the next levels appeared quite easy. After a few meetings
with the chiefs of the Police Youth Unit itself, where verbal approval was given
and practical arrangements were made, I was quickly introduced to the lower
ranks: the police inspectors doing the daily interrogations with juveniles. In our
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
20 Maklu
opinion, some underlying reasons for this (almost) effortless entry in the field
are the following. One reason is the fact that we are genuinely interested in their
work and ‘their side of the story’. The tendency of police to separate the police
from ‘non-police’ leads them to a ‘us-versus-them’ thinking which is not only
applied to offenders, but also to the ungrateful public, the uncooperative victims,
the media, etc. (Chan, Doran & Marel, 2010). When the media exposes police
brutality, racism, corruption and blunders in crime investigation, they undermine
the public confidence in the police (Garland, 2001; Manning, 2003). Consequently,
police services engage in ‘image work’ and attempt to project a positive police
image (Mawby, 2010). By showing interest in their job, in their views and opinions
and thus in the ‘us’ story, researchers can count on the goodwill of the Belgian
Police to record ‘their’ story (Devroe et al., 2012). The meetings allowed us to
give explanations about the research aim, to answer their questions and to make
agreements. Moreover, having opted to only interview police officers and not the
young suspects involved, we probably reinforced police’s impression that we are
interested in their side of the story and their side of the story ‘alone’. Besides, the
relationship that the police has with academia and research is not always easy9 but
researchers produce knowledge with methodological rigor based on evidence or
theoretical reasoning which, accordingly, is set apart from opinion on policing as
circulated by the media (Loader & Sparks, 2010). If this knowledge is practically
useful for the police itself, they are even more open to research (Alphert et al.,
2013; Engel & Whalen, 2010). The concerned Police Youth Unit seems to value
academic studies. The discussions with the chief of the Police Youth Unit during
the first informal meeting underlined this. ‘We are curious to know what we can
do better and hopeful to be confirmed that we are doing quite well.’10 Other researchers
have stated: ‘getting access to police organizations is rather easy in Belgium, thanks to
the broad network of scholars in the field’. Many of the gatekeepers, such as chief
constables, have an academic criminology degree and are interested in and familiar
with academic work (Beyens et al., 2013, 17). This is an additional explanation for
their willingness to collaborate, besides the fact that they are also hoping benefits
of the research to the organization itself. Another reason for the easy entry is,
in our opinion, the way we introduced the research to the local unit. We had the
impression that once the ‘first gates’ were opened, the next ones would follow
easily as we would emphasize our permission ‘from above’. I heard the department
chief (of the Youth Unit) say several times to fellow police inspectors: ‘she even got
the permission of the public prosecutor’. The hierarchical structure of the Belgian
police sure was to the advantage of our research and as other researchers have
ascertained: once the official obstacles disappear, the researcher gets unhoped-for
acceptance, politeness and cooperation from the police (Chauvenet & Orlic, 1985).
9 Academics who do police research are often regarded as being overly critical or as eternal pessimists
able to find a dark cloud around any silver lining (Bradley & Nixon, 2009; Cordner & White, 2010).
10 14th of May 2013.
maklu

Should I stay or should I go?
Maklu 21
The fact that one of the supervisors has a background in police research and in
the education of police-officers and that she is ‘well known’ in the Flemish ‘police
world’ facilitated, in our opinion, the access as well.
In addition, the way we detailed the research probably influenced the attitude of the
‘gatekeepers’ (actors controlling key resources and avenues of opportunity (Atkinson
& Hammersley, 1994)) about what is being researched on. One of the sensitizing
concepts of the research (i.e. concepts suggesting directions along which to look,
giving a general sense of guidance in approaching empirical instances), namely
‘children’s rights’, was not mentioned as such when we presented the research to
the different gatekeepers. Two reasons legitimated this conscious ‘omission’. The
first one was to avoid that the researcher’s conceptualization and language was
going to be permanently introduced in the discourse of police officers. After all,
one of the aims of this research was and is precisely to find out what the views of
police interrogators on the ‘child’ and ‘children’s rights’ are. Do they know and use
this language? What are their views on the role these rights can/should play during
police interrogations? By introducing ourselves this concept as an important and
sensitizing concept, this might have led to the situation where police officers
answer in a socially or judicial acceptable way, instead of genuinely answering and
reflecting on their views on ‘children’s rights’. The second reason of this conscious
omission was the supervisor’s fear that if ‘children’s rights’ are not in line with
the gatekeepers’ hidden agendas, ideologies and cultures, this could compromise
gaining access and collecting reliable data (Lee, 1993). Children’s rights and the
relationship between juvenile delinquents and police officers are sensitive topics
in Belgium (Kinderrechtencommissariaat, 2012). Referring to ‘children’s rights’
might give police officers the impression that we are ‘not on their side’ after all and
that we are going to ‘control’ them and their respect (or not) for ‘children’s rights’.
This however is not the aim of our research.
With these concerns in mind, communication about the research in the
permission requests was strategically adjusted. Afterwards, this seemed a good
strategy as the researcher often heard the police inspectors say ‘suspects already
have enough rights’ or ‘we always feel controlled, when will somebody listen to our side
of the story?’ However, not stressing all the aspects of the research object made
the researcher feel guilty for not being totally open with the research subjects. As
other researchers have been forced to do the same in the past (Pruvost, 2007) and
as some authors (Goode, 1996) assert that certain kinds of deception are necessary
to gather certain data in certain settings, she felt partly relieved.
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
22 Maklu
The second step: Access during the fieldwork or ‘personal access’11
Access during the fieldwork is sometimes (for instance in business and management
research) called ‘personal access’ because it entails that the researcher knows
relevant individuals in the organization (Laurilla, 1997). Fielding (2006) sees
trust as a guarantor of access, and therefore states that ‘real access’ is more than
a preliminary negotiation. My very first field notes say something about the first
day of access:
‘Today my research officially starts at the Unit of the local police. It is 9:30, I enter
the building and feel tension in my whole body. This is the place where I’m going
to do 6 months of research, they cannot think I’m a moron! Josh Hutcherson12, the
chief of the local Unit, welcomes me in a very friendly way. We immediately start
a two-hour conversation during which he explains to me how the Unit works (…).
It is about 16:00 p.m. and the police team is leaving for a New Year’s Eve dinner
so I say goodbye. I have a headache but I am very pleased, everything went so well
and everybody is being nice to me. I take leave from Josh and he tells me to come
early tomorrow morning, so that I can attend their morning-meeting13.’
In order to gain access to the lower ranked police inspectors who interrogate
juvenile suspects (the relevant individuals), I used a personal ‘strategy’. The first
way to gain the trust of the police inspectors was a ‘soft’ entrance into the field: no
immediate observations of the real research object (the juvenile interrogations) but
a kind of preliminary ‘traineeship’ during which a certain time (sixteen full days)
was spent with the inspectors, observing them, coupled with talking to them about
what they are doing, thinking and saying (Delamont, 2004). Trust was certainly not
immediately there, as a conversation with a police officer on my first day proves:
– The police officer: ‘what is your research about?’
– Me: ‘I’m curious to know how you approach those youngsters.’
– The police officer: ‘Ah so you are going to check whether we are friendly and give
them a cigarette or what?’14
When I explained to them I was interested in everything they were doing,
alongside the interrogations, they demonstrated a particular goodwill to show me
all the facets of their job. I accompanied them during patrol work and during the
control of house arrests of juveniles, went along with them to pick-up juveniles in
youth detention centers or youth institutions and kept them company during dull
11 Here we use the first person, because this part of the article recounts the personal experience of the
researcher.
12 This is a fictive name to protect the anonymity of the respondent.
13 6th of January 2014 (first day in the field).
14 6th of January 2014 (first day in the field).
maklu

Should I stay or should I go?
Maklu 23
hours of waiting in the police station for quests of intervention. I felt ‘like a child
in a new culture, learning via worldly encounters’ (Van Maanen, 2011, 220), trying
to understand their ‘cop culture’. I tried to grasp the practical logic of the whole
context in which the police officers were functioning by developing an intimate
familiarity with their activities.
In this first phase, I also avoided taking field notes in front of the police officers.
The registering of events by writing them down occurred during ‘subtle toilet
visits’, a strategy also used by previous researchers. This was, of course, putting
pressure on the capacity to remember all the relevant events in a setting with many
social actions occurred at the same time (Demaree, Verwee & Enhus, 2013) but I
had the feeling that note-taking would be seen as too inquisitive and unnatural in
this stage of ‘getting to know’ the police-officers.
This first way of interaction during the traineeship was useful to see how police
officers understand the world, to examine what they say and what they do, a
presumed causal relation heavily debated in the police research field (Waddington,
1999). But above all, it was fruitful to build a relationship with the police officers.
In this stage of being accepted and gaining trust, I was sometimes asked to help
in the police work. The Police Unit expected me to be an extra set of eyes, ears and
hands (Spano, 2006). I was willing to give advice or information when asked for,
to help with a face recognition of a suspect, to hold handcuffs, the walkie-talkie or
doors and to look out for suspects when we were patrolling. This was appreciated
and thus helpful in the process of gaining trust and building relationships with
the research participants.
Maintaining the role of a researcher was however not always easy (this is discussed
further in the chapter ‘avoiding over identification’). As ‘trust and acceptance entail
reciprocal relationships, i.e. receiving and giving information of all kinds’ (Demaree,
Verwee & Enhus, 2013, 117), I had to answer (personal) questions and had to
participate on conversations about non-professional topics. When I was asked
about my love life, I mentioned the existence of my partner and engagement, an
opportunity to avoid being seen as a potential or accessible flirt. Other questions
concerned for instance the place where I live, what I was going to do after the
working hours, etc. sometimes, I felt obliged to respond to the nosey questions
and even to give more information than asked, in order to foster the image of
an ‘open and honest’ researcher. The following field note excerpt illustrates an
embarrassing situation in which I felt obliged to give information and then
afterwards wondered if I had not given too much.
‘Around 17 p.m. police officer Olivier asks me if I am joining them for dinner
tonight (I am sitting at a desk in the police station where we are all waiting for an
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
24 Maklu
intervention request). Me: ‘No, I’m already eating with somebody else’. Olivier: ‘Oh
really? Where?’. Me: ‘In De Fabriek (a fictive name of a known restaurant in the
area)’. I then realize that if he asks ‘with who?’ (I experienced police officers are
very curious people) I can’t take the risk to lie because the restaurant is not far away
from the station. Consequently I’ll have to tell it’s with the lawyer who assisted
the interrogation in the morning. So I decide to anticipate the question and tell
him ‘I’m going to eat with the lawyer of this morning. It is a total coincidence that
precisely that lawyer was present during the interrogation, as I have a rendez-vous with
him this evening. I have not seen him for two years. I decided to invite him for dinner
to thank him for his help for my master thesis back then. That man sacrificed two hours
of his time for my interview and I had promised him that if I’d be in Antwerp again I’d
contact him.’ I suddenly feel I’m blushing to the roots of my hair because of all this
information about my private life. But I’m feeling complied to tell all this because
of my fear that they (inspector Paul has joined us and is actively listening) will be
suspicious about my rendez-vous with ‘the other side’ (lawyers are not particularly
appreciated) after an interrogation.15‘
Dealing with the curiosity of the participants was not always easy, especially when
it concerned the research itself. Two conversations (which took place in the second
week of my entrance (when I started to take notes in front of them) illustrate this:
– Police officer: ‘What are you all writing in your little notebook? Well I mean, you
don’t need to show it to me but I see you writing down so much …’
– Me: ‘Well I write down relevant statements, such as Jef’s statement about being
unproductive, because they’re important to garnish my PhD’16.
A male inspector starts a conversation with me about the first interrogation I just
witnessed. I tell him it is my ‘test’ observation that will serve as a reflection basis
for further interrogations. He sees the written field notes I am holding and asks:
‘can I take a look?’ and immediately grabs the document. My first reflex is to say
‘uh, I’d rather not’ but then I think it is more important to keep their trust. So I
say unwillingly ‘go ahead’ and let him take the document. I try to explain to him
that police officers reading my notes does not serve the purpose of the research
because then they will eventually change their behavior. ‘Don’t worry’ he says. ‘We
won’t.’17
When the so-called traineeship was over and I had found my feet, I started to
direct my attention towards the interrogations of the juveniles. The police officers
were cooperating by telling me when they had planned interrogations and even
adapting their time-schedules so that I could be present. They were also eager
15 30th of January 2014.
16 23rd of January 2014 (eleventh day in the field).
17 30th of January 2014 (fourteenth day in the field).
maklu

Should I stay or should I go?
Maklu 25
to respond to my interview requests, although they did not have much time and
some of them were reluctant in the beginning. A field note excerpt illustrates this:
‘I ask inspector Jimmy18 if he can do an interview with me on Friday the 14th about
the interrogation he just did. ‘What interview?’ says Jimmy (knowing very well what
I’m talking about, as I mentioned this already previously and distributed a written
document to all the officers in order to inform them about my scopes). ‘Well, after
each interrogation I’m doing an interview with the interrogator, remember …?’ Jimmy
doesn’t answer. He doesn’t seem to be particularly enthusiastic. I propose him
to do the interview on Saturday (when there is another interrogation) about the
two interrogations at once. Jimmy can live with that. At that moment, I realize an
interview after each interrogation is maybe too much for the interrogators, so I
decide to handle this more strategically in the future19.’
I had managed to develop a working relationship and thus believe that my soft
entrance in the field had been a good thing. The process of gaining trust went along
with getting acquainted in all kind of ways. Initially, the police officers addressed
me with ‘u’ (formal ‘you’ in Dutch) and gave me a hand to greet me. Later on, they
addressed me with ‘jij’ (informal ‘you’ in Dutch) and kissed me hello/goodbye.
The person of the researcher
As the process of access and the dynamics of acceptance are influenced by
the characteristics of the researcher, I consider my ‘personage’ played an
important role in the phase of gaining ‘personal’ access. This personage involves
characteristics such as appearance and communication skills but also gender,
age, personality, class, ethnicity, etc. Since the researcher must also study herself
alongside the others (Chughtai & Meyers, 2014), and because self-representation
and ‘impression management’ are essential aspects of fieldwork, the reflexive lens
is used to discuss some of these influencing characteristics. More specifically, the
effects of my gender (role) and age, appearance and behavior as well as personality
are examined in the context of establishing relationships with the police officers
of the Police Youth Unit.
Gender and age: when the researcher is a young woman
Police settings are generally considered masculine, even macho settings (Miller,
Forest & Jurik, 2003; Reiner, 2000). The concerned Police Youth Unit was
predominantly male (with four women opposed to fifteen men) but the four
women inspectors were clearly not being pushed around. They were addressed
18 This is a fictive name to protect the anonymity of the respondent.
19 12th of March 2014.
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
26 Maklu
with respect and treated gently, although they were sometimes teased in a playful
way. It is not possible to determine whether the fact that there were not many
women in the setting contributed to the male police officers being keen to accept
me, but I felt my gender was an asset rather than a limitation to the fieldwork.
‘Gender closes some avenues of inquiry but clearly opens up others …’ (Martin, 1980,
cited in Punch, 1986, 24). For instance, I sometimes had the feeling that ‘being a
woman seemed to ‘innately’ mean’ that I was ‘trustworthy and understanding’ (Marks,
2004, 881). Here lies, of course, the danger ‘to employ gender as an explanatory
catch-all’ (Mc Kegany & Bloor, 1991, 196).
The concept of ‘gender’ refers to the social and cultural construction of ‘biological’
sexes. As it forms part of our identity, it affects our relationship with others but
is also produced within the interactions with other people: it is assigned to the
researcher by the researched subjects (Fortier, 2001). In this sense it is not a
given, it is always (re-)negotiated and performed in different social situations and
in different locations (Glover & Kaplan, 2000). In this way my gender role (body
language, outfit, …) rather than my gender as such was important in the fieldwork
(Whitehead & Price, 1986).
The number of female ethnographers of policing seems to be quite large, and
several of them have considered the issue of gender and its impact on their field
research (e.g., Loftus, 2009; Marks, 2005; Pruvost, 2007). Whether especially
women are attracted by this research area and whether there is a trend in female
criminologists are questions that will not be explored in this paper.
As I am young (twenty-five years old), I took advantage of my lack of experience
in order to ask naïve questions, and elicit explanations of banal matters, thus
accumulating information concerning the research participants’ accounts and
viewpoints (Corbetta, 2003). As a young woman with few preconceived ideas about
the organization, I had the feeling to be seen as quite inoffensive. One of the
police officers told me my help in an investigation was welcome because ‘you have
an immaculate view as an extern member of the organization’20 Previous researchers
asserted that ‘a young student may be perceived as non-threatening and even elicit a
considerable measure of sympathy from respondents’ (Punch, 1986, 24). Often, I heard
the police-officers say my name diminutively: ‘Ah Camilleke (little Camille), you’re
sitting here’21, ‘Are you ready Camilleke?’22 Nonetheless, I didn’t have the feeling to
be treated as a ‘fragile creature’ they had to care for, unlike other female police
ethnographers experienced (Lumsden & Winters, 2014). Maybe the relatively
average young age in the Police Youth Unit (only two police officers are past 50
20 22nd of January 2014.
21 23d of Janurary 2014.
22 17th of March 2014.
maklu

Should I stay or should I go?
Maklu 27
years old) gives an explanation for the absence of protective and paternalistic
behavior.
Appearance and behavior: the ‘field-outfit’
Concerning the characteristic of appearance, I will confine myself to a short
discussion of my ‘field-outfit’: the dress code, jewelry, make-up and (body) language
I used as well as the manner of conducting myself.
I had not been aware, but I probably copied some elements of the other police
women’s style (like wearing my leather jacket a lot), maybe due to a fear of being
too feminine, a fear of eliciting eventual jealousy or a competitive attitude among
those women.
So I adjusted my ‘style’ or ‘image’ to the other female police officers: ‘feminine but
not accentuated’ (Pruvost, 2007, 143). This was not an important metamorphosis
as I have naturally a common and discrete dress style, quite the opposite of ‘girly’
(with girly I mean very into clothes, hairstyle, nails, makeup, etc.). In my everyday
life I wear little make-up and seldom striking clothes or prominent jewelry.
Consequently, it did not bring along much effort to adapt. At the police station
of the Youth Unit, I was always dressed casually, sometimes sportive (with jeans,
T-shirt, etc.), with (ear)rings as only jewels. I avoided clothes with a décolleté or
inviting skirts, being convinced that an unobtrusive and neutral dress code was the
best (also towards the young suspects being interrogated).
Concerning the (body) language, I had a refined attitude during the serious work
(observations and interviews) but showed playful, rugged behavior (for example
mugging, imitating their accent, …) when the police officers were teasing me.
One day, when I was entering the interrogation room (which is a separate room)
with inspector Jimmy to interview him, we heard the other inspectors (women
included) make a characteristic deep long ‘ouh’ sound to suggest we were going to
do nasty things in that room. I imitated their laughs, giving it a stupid echo in a
playful but defensive way23.
I kept myself from using inappropriate (‘vulgar’) humor to avoid being ungracious.
During my academic course, I had the opportunity to develop good communication
skills: I learnt to express myself in a clear, pondered way, which probably helped to
earn the respect of the research subjects.
23 6th of March 2014.
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
28 Maklu
Personality of the researcher
During the fieldwork, I was often being ‘tested’ through jokes and questions: ‘And
Camille, are you single? ‘Cause we are now on the Singel!’ (during a ride)24. The police
officers would ask me evaluative questions about their job (for instance what I thought
of their behavior during the house arrests control, of the way they interrogated someone,
etc.) but also about the ‘Salduz law’25 and the changes it brought along in the police
practice. During a ride with two police officers they asked me what I thought about
‘this Salduz law’. I felt I had to be bold and frank in giving my opinion but similarly
to other researchers (for ex. Marks, 2004), I felt uncomfortable about being asked these
questions because it entailed that I had to take a position or provide a value judgment:
yes, I thought this Salduz law was a good thing for the rights of suspects but yes, I had to
admit it didn’t facilitate their job, so I answered: ‘Well, the very first time I heard about
it I was only a student and I thought it was a contra productive idea that a lawyer could
assist the interrogations. But after having informed myself better and having read a lot
about false confessions my image concerning this has changed, even if I still think this
must not be easy for you’26.
I also had to react with humor to their teasing jokes or questions, in order to
avoid being walked over and to be seen as a misfit. One evening, while I was
eating with three police officers, they tried to convince me to tell them how much
I was earning as a researcher. I took pleasure in telling them ‘you’ll never know’ and
keeping this information for myself27. I further tried to be kind and helpful and to
show interest in what my research subjects did or said. These patterns of attitude
that in fact make up personality − and so identity − undoubtedly influenced
the process of carrying out the research. Although personality remains fairly
consistent throughout our life, it permits us to project different roles or to disguise
our identities. According to this view that personality − a part of identity − is a
free-flowing and malleable construct (Harvey, 2013), I sense that my personality
in the field was also constructed by the research subjects themselves. Through
social interaction, I constructed a certain ‘field identity’, obtained a certain insider
status. In the local team, some of the police-officers have a nickname. I once asked,
while they were joking around, if they had (secretly) given me a nickname as well.
One of them replied: ‘no, but ‘Fine Fleur de Bruxelles’ would fit you well’28. I then
realized that I had been acting in gentle, polite ways − which had been effective to
24 A roadway surrounding the city.
25 Belgium did not have a specific legal framework that organized and guaranteed the assistance of
a lawyer during the (first) police interrogation. The Salduz judgment of the European Court of
Human Rights was conversed into Belgian law: Belgium passed the « Salduz bill», which went into
effect on January 1st, 2012.
26 16th of January 2014 (eighth day in the field).
27 15th of January 2014 (seventh day in the field).
28 23 of May 2014.
maklu

Should I stay or should I go?
Maklu 29
collect some data − but wondered if I had been too delicate and different from the
common − more rough − police attitude to collect other data.
Methodological dilemmas
‘Being reflexive in doing research is part of being honest and ethically mature in research
practice’ and as such requires researchers to ‘stop being ‘shamans’ of objectivity’
(Ruby, 1980, 154). Recognizing methodological issues in research is part of being
critically reflexive. The observer’s paradox and the concept of going native − which
are the main methodological dilemmas that arose during this first field experience
− will be recounted in the following paragraph.
Observer’s paradox
Being present during the police interrogations was necessary to conduct the
research, as there are only very few (audiovisual) recordings of juvenile (suspect)
interrogations in Belgium. However, a frequent criticism of ethnographic research
is that the presence of the researcher influences the behavior of those being studied,
creating ‘observer effects’. The observer’s presence is thus a source of bias, making
it impossible for ethnographers to ever really document social phenomena in any
accurate, let alone objective way (Wilson, 1977). We want to know what people do
and say in their daily routine but from the moment we observe them, those people
eventually change in their dynamics (Labov, 1972; Schwartz, 1993). During my
observations, I often asked myself in which way I was transforming the attitude
of the observed professionals. Were they not less violent, more ethical, more
conscientious, more efficient, etc. while I was looking, as I would always be an
external member (Ericson, 1980)?
During a car patrol that I was doing with two young male police officers to control
suspects in the public space, I witnessed one of the officers humiliating a suspect
by commenting the breast of his girlfriend (seen in pictures on his cellphone).
The concerned police officer had been joking around with his colleague during
the whole ride, also kidding with me, and I had been laughing the whole time with
them. I felt bad when they humiliated the suspect and kept wondering afterwards
if the fact that I first laughed about their jokes encouraged the police officer to
adopt a macho attitude and humiliate the suspect. Had my characteristics as
an observer (young woman) been a trigger for an altered attitude (exaggerated
humiliating remarks) of the officer?29
29 Also, as they had unlawfully asked the young suspect to look into his phone, I wondered whether I
had to do or say something about it. At that moment, as I had not seen verbal or physical violence I
didn’t think further about what I should do if I were to witness much more controversial behavior.
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
30 Maklu
On the other hand, according to Schwartz (1993) and Pruvost (2007), police officers
cannot control themselves in such a way that they would radically change their
behavior. They can temporarily fool the researcher, but will not do this over time
(Schwartz, 1993) because performing continually in front of the ethnographer
is very difficult (Atkinson, 2014). Several authors posit that the process of
establishing rapport and gaining trust can reduce observer effects, since research
subjects gradually become accustomed to the presence of observers and begin to
act naturally and spontaneously (Spano, 2005).
One day, a conversation with police officer Jef made me believe the participants
were not changing their behavior that much or at least, not what they were saying:
– Jef: ‘Camille, you should not write down in your report what is said here ‘across the
lines’.’
– Me: ‘What do you mean Jef?’
– Jef: ‘Well, when we cut up rough or comment the system, then you shouldn’t pay too
much attention about it.’
– Me: ‘Well, I do know you are only humans and that your statements are made
within a context …’
– Jef: ‘It wouldn’t be human for us any more if we couldn’t say those things you know.’
Helping the police officers is a way of establishing rapport, which I also utilized.
When I was asked ‘Camille will you accompany the lady (the lawyer) to the exit?’30 I
willingly did. During the fieldwork, I had given advice to the police officers when I was
asked to and had been an extra set of ears, eyes and hands (see supra). This was highly
appreciated by the officers, especially when they were working alone on a case.
They started to ask me very often my opinion or advice about things. I remember
the interrogation of a juvenile suspect during which the police officer suddenly
turned towards me to ask: ‘do you know this place he’s talking about’31? I believe
the altering of the normal course of events by helping the officers was minor
because I had established a good working relationship with them and was even
permitted to witness more questionable police behavior (Van Maanen, 1978) but I
was, nonetheless, aware of the influence of my presence. Some researchers have
stated that the capacity of the ethnographer to shape the discourses and practices
being observed can be considered a strength of the method because the observed
performances – however staged or influenced by the observer − often reveal
profound truths about social and cultural phenomena and thus provide rich data
(Monahan & Fisher, 2010).
In this case, I decided to ask another police officer informative questions about this ‘phone-
checking’ but kept my judgement for myself, for the ‘sake of the research’.
30 11th of March 2014.
31 19th of February 2014.
maklu

Should I stay or should I go?
Maklu 31
Avoiding over-identification or ‘going native’
‘In order to gain approval, the outsider will connect and level with the members of the
social context, which is a human interaction reflex’ (Demaree, Verwee & Enhus, 2013,
116). I also experienced this human reflex during my entrance in the field. In fact,
I felt a very strong pull toward increased involvement and found it difficult to
remain an objective and vigilant researcher. After two weeks and a half, I heard a
chief inspector say: ‘but no Camille you don’t bother, you’re now a fixed value here’32
and several police officers told me: ‘come join us after your PhD’. This was in fact a
clear cut signal that I was no longer seen as a threat and was treated equally, as
a member of the police setting (Marks, 2004) but it made it difficult to avoid the
pitfall of getting too close and to avoid being totally ‘resocialised’ (Spano, 2005).
Was being seen as ‘one of them’ to be taken as a compliment or did it have to ring
alarm bells?
The police officers used me as a resource (asking advice) and often treated me
as a colleague while conducting their work. As Horn (1997) asserts, researcher
participants frequently do more than is necessary to ease the way for researchers,
receiving little benefit from the research and taking the risk of confronting them
over their behavior, which produces ‘gratitude towards member settings’ and
‘concern with maintaining rapport’ as Gurney (1985) calls it. As they made a lot of
endeavour to make my work easy (e.g., allowing me to observe everything, giving
me access to the computer system, accepting to be interviewed and recorded,
adapting their time-schedules), a natural consequence was a sympathetic outlook
(Spano, 2005) towards them. Over time, in this process of socialization, it became
difficult for me to avoid ‘overapport’. Moreover, I received often the subtle message
‘be on our side’. For instance when the police officers told me: ‘you really have
to include in your report how difficult Salduz makes (and will make) things for us’33
or, on another day, when they were displeased about the comment of a lawyer
concerning the length of the interrogation: ‘Who does she think she is?! Note that
down Camille!’34.
Like literature suggests, it is very difficult to know whether someone goes
native or not, as the boundaries are very blurry and the phenomenon as such is
subjected to different viewpoints (see for instance Thomas, 1993; Angrosino &
Mays de Perez, 2000; Goffman, 2011; Harrington, 2002; Miller, 2009). This can
be illustrated by a field-experience that led to a different point of view between me
and my supervisors. One day, the police officers asked me to join them on their
teambuilding: a full day excursion. I had to refuse on request of the supervisors
32 12th of February 2014 (eighteenth day in the field).
33 23d of May 2014.
34 27th of March 2014.
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
32 Maklu
because of the blurring lines between the role of a ‘researcher’ and the role of a
‘colleague’. My supervisors feared that, at the end, when ‘as a researcher’ I have
to scientifically and critically assess police’s daily practice, police would no longer
appreciate my ‘colleague-friendly’ positioning during fieldwork because it might
appear ‘false’ afterwards. This, in return, might hinder future research with and
on police from behalf of our research team. However, I felt the police officers were
incomprehensive and disappointed about my refusal and I had the feeling the
relationship with the respondents was not always fair and balanced because they
were investing time in me and I was not investing myself completely in return
by being affectively present (during the non-professional activities for instance)
(Liebling, 2001). I decided to go to the barbecue that the police officers were
organizing − to close their teambuilding − in the evening. My supervisors were
disappointed and upset when they learnt about this because I had ignored their
guidelines and, in their point of view, their trust. Based on our conversations − in
which I was giving feedback about the fieldwork − they had also developed the
viewpoint that I behaved in a too sympathetic way towards the ‘police colleagues’
that I developed only little critical and scientific reflections and that I did not link my
observations and experiences to scientific literature. All this fed their impression
of me ‘going native’. This reproach was painful and made me upset as I had the
impression to be misunderstood. Of course, I was also wounded in my pride. The
reaction of my supervisors led to a more critical inquiry of myself in the field,
though. I recognized the personal inhibition we display ‘to break our rapport with
critical questions’ (Robben, 1995) and became aware of the kind of researcher I want
to be with my participants: more critical and careful. My supervisors decided later
that they would read my field notes, because those were also containing (critical)
reflections. It seems to us that this can help in jointly decision making (when
facing such dilemmas) because of an increased mutual understanding.
According to Atkinson (2014) ‘sometimes going native can be productive’, because
‘(…) the fieldworker’s closeness to others’ daily lives and activities heightens sensitivity to
social life as a process (Emerson et al., 1995, 3-4). Participating to the teambuilding
might have allowed me to experience their world in its full complexity, to observe
them act in a different (non-professional) context where they would maybe have
shared extra (inside) information. On the other hand, my supervisors were more
concerned about the danger of sympathizing too much with the research subjects
as some authors have asserted that observers who participate in police work may be
more susceptible to the biasing effect of going native (McCall, 1978). People who
do research within the police often become more positive towards them, because
they start to see the police as workers who have to face emergency situations, a
deficient judicial system, pressure and the need to behave conform the unwritten
collegial code (Punch, 1983).
maklu

Should I stay or should I go?
Maklu 33
In the opinion of Monahan and Fisher (2010, 363) ‘some of the greatest strengths
of ethnographic research lie in cultivating close ties with others and collaboratively
shaping discourses and practices in the field.’ But how do you find a tenable balance
between stepping into the world of those you want to observe and − at the same
time − avoiding to become ‘too involved’? To our opinion the researcher’s ability
to ‘re-acadamize’ what she has observed is crucial. If this ability is lacking, alarm
bells should ring. However, as Liebling (2001) suggests, keeping distance might
be more important in the (next) stage of data handling. While in the field, Garriott
(2013) believes (in the context of ‘dirty’ police work or evil actions in which you
can become a ‘complice’) that being aware of your involvement, and to critically
question what people say there are doing, what people actually are doing, and what
people say about why people are doing what they are doing is important.
Conclusion
In this paper, first field experiences in a Police Youth Unit were shared, after
briefly detailing our research project. We used a reflexive lens to increase the
self-awareness of the adopted approach to enter the field and in order to enable
contextualization of the research interactions (Arendell, 1997). Coming at the end
of this article, certain recommendations and reflections for further ethnographic
research in a police setting appear appropriate, both on the level of gaining access,
as on the level of the ‘unavoidable’ methodological dilemmas in ethnographic
research.
First of all, one of the trampolines in gaining access in the field clearly was the
cooperation of the local authorities, i.e. the informal authorizations we received
prior to the fieldwork. The formal authorizations later on in the process were a
second important step. Thirdly, the informal meetings with the chiefs of the local
Police Youth Unit allowed us further access. A fourth and final reason for the
smooth entry is probably the chief’s hope for benefits of the research to the Youth
Unit itself. In general, it must be said, access to police organizations is rather easy
in Belgium because of the traditional links between academic scholars and police
officers in higher rankings.
However, after this stage of ‘pre-entry’ access, the ‘real’ access still has to take place.
Here the researcher’s personal abilities play an important role. The researcher
became aware of the strategy she (unconsciously) used: a ‘soft’ entrance, during
which the trust of the police officers − a crucial aspect – is gradually gained in
a subtle way. During this step of ‘personal access’, the researcher’s personal
characteristics such as gender and age inevitably influence the acceptance of the
researcher by the research participants.
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
34 Maklu
In this stage of gaining access, the researcher needs to be strategic and to find
a balance between becoming ‘one of them’ but at the same time, remaining
a scientific researcher. This balancing involves skillful coping with certain
unavoidable methodological dilemmas, such as the ‘observer effects’. Should we try
to neutralize these effects as much as possible or should we, in contrast, constantly
reflect on them, take them into account, in the hope to obtain even richer data
or scientific insights? In this research project, as our presence during the police
interrogations is very noticeable (due to the few people present in the small setting
of the interrogation rooms), we opted for the second path of constantly reflecting
on ‘observer’s effects’ and taking them into account as a way of giving deeper
insight in police’s views and actions. Police officers stating that they do not like
the observer to write down everything because of their need to say some things to
remain ‘human’, seems like a good illustration of how the observer’s effect can also
give a deeper insight in certain police behavior.
However, being clearly ‘present’ and becoming close to police interrogators in
order to make rapport, might lead the researcher to the pitfall of ‘going native’.
Taking distance might then seem to be a ‘good solution’. Nonetheless, it also
involves a clear danger of not being enough affectively present. This can hamper
the collection of robust data but also the empathic capacity to actually understand
in depth ‘what is going on’. These dilemmas can feel like constantly being on a
tightrope between rapport versus distance, participant versus observer, friendly
colleague versus critical researcher. In the field, these dilemmas can lead to very
concrete questions like ‘should I stay (a bit longer to make rapport) or should I go
(now and critically re-academize all things I have observed)’?
As every fieldwork situation is different, we think every researcher should
individually set his/her own boundaries and mediate in order to find a tenable
balance that permits the validity of the outcome. This sustainable balance has to be
created through personal choices, according to the uniqueness of every research
and the singularity of every encounter between the researcher and participants
(Garriott, 2013). However, on the basis of our own experiences, we also advocate
that these choices should be thoroughly discussed between the researcher and
the supervisor(s) in order to avoid misunderstandings. Especially in the case
of inexperienced researchers, clear communication in the initial stages of the
fieldwork is essential. In this case for instance, we could have discussed better
what to do in case of police asking to join them in extra muros activities and to
which extent going native in criminological research is for example different from
going native in anthropological research. Or is there no difference?
These dilemmas and reflections taken into account, we conclude that developing
fair and balanced relationships with the research participants and finding a
scientific approach that resonates well with ideals of engagement, participation
maklu

Should I stay or should I go?
Maklu 35
and collaboration (Monahan & Fisher, 2010) is and remains a difficult craft in
ethnographic work.
References
Alphert, G.P., Rojek, J. & Andrew Hansen, J. (2013). Building bridges between
police researchers and practioners: agents of change in a complex world. U.S.
Department of Justice. National Institute of Justice. Retrieved from https://www.
ncjrs.gov/pdffiles1/nij/grants/244345.pdf
Angrosino, M.V. & Mays de Perez, K.A. (2000). Rethinking observation. From
method to context. In N.K. Denzin & Y.S. Lincoln (Eds.), Handbook of Qualitative
Research (2nd ed.). Thousand Oaks, CA: SAGE Publications, 673-702.
Arendell, T. (1997). Reflections on the researcher-researched relationships: a
woman interviewing men. Qualitative Sociology, 20(3), 341-369.
Atkinson, P. & Hammersley, M. (1994). Ethnography and participant observation.
In N.K. Denzin & Y.S. Lincoln (Eds.), Handbook of Qualitative Research.
Thousand Oaks, CA: SAGE Publications, 248-261.
Atkinson, P. (2014, June). The craft of ethnography. In doing research in
Criminology. Symposium conducted at the Vrije Universiteit Brussel, Brussels,
Belgium.
Beyens, K. (2013). Introduction. Giving voice to the researcher. In K. Beyens,
J. Christiaens, B. Claes, S. de Ridder & H. Tournel (Eds.), The pains of doing
Criminological Research. Brussels: VUBPress, 13-21.
Bradley, D. & Nixon, C. (2009). Ending the dialogue of the deaf: evidence and
policing policies and practices. An Australian case study. Police Practice and
Research, 10(5-6), 423-435.
Chan, J., Doran, S. & Marel, C. (2010). Doing and undoing gender in policing.
Theoretical Criminology, 14, 425-446.
Chauvenet, A. & Orlic, F. (1985). Interroger la police. Sociologie du Travail, 4, 452-
467.
Chughtai, H. & Myers, M. (2014, June). Ethnographic field research: interpreting
one’s entrance into the field as thrownness. Paper presented at the Twenty Second
European Conference on Information Systems, Tel Aviv, IL.
Corbetta, P. (2003). Social Research: theory, methods and techniques. London: SAGE.
Cordner, G. & White, S. (2010). The evolving relationship between police research
and police practice. Police Practice and Research, 11(2), 90-94.
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
36 Maklu
Davies, C.A. (2008). Reflexive Ethnography: a guide to researching selves and others.
London & New York: Routledge.
Delamont, S. (2004). Ethnography and participant observation. In C. Seale, G.
Gobo, G. Gubrium & D. Silverman (Eds.), Qualitative Research Practice. London:
SAGE, 217-299.
Demaree, C., Verwee, I. & Enhus, E. (2013). Observing the observers. Participant
observations in police settings. In K. Beyens, J. Christiaens, B. Claes, S. de
Ridder & H. Tournel (Eds.), The pains of doing criminological research. Brussels:
VUBPress, 105-123.
Denzin, N.K. & Lincoln, Y.S. (2008). Collecting and interpreting qualitative materials.
London: SAGE.
De Schrijver, A. & De Kimpe, S. (2009). De politionele aanpak van minderjarigen
in Vlaanderen. Een congruentie tussen de organisatorische aanpak en de
beleidsmatige visie? Panopticon, 5, 38-51.
Devroe, E., Pauwels, L., Verhage, A. Easton, M. & Cools, M. (Eds.) (2012).
Tegendraadse Criminologie. Liber Amicorum Paul Ponsaers. Antwerpen: Maklu.
Emerson, R.M., Fretz, R.I. & Shaw, L.L. (1995). Writing ethnographic fieldnotes.
Chicago: University of Chicago Press.
Engel, R.S. & Whalen, J.L. (2010). Police-academic partnerships: ending the
dialogue of the deaf, the Cincinnati experience. Police Practice and Research,
11, 105-116.
Ericson, R.V. (1980). Reproducing order: a study of police patrol work. Toronto:
University of Toronto Press.
Feld, B. (2013). Kids, Cops and Confessions: inside the interrogation room. New York:
New York University Press.
Fielding, N. (2006). Fieldwork and police work. In D. Hobbs & R. Wright (Eds.),
The Sage Handbook of Fieldwork. London: SAGE, 277-291.
Fortier, A.M. (2001). Gender, ethnicity and fieldwork: a case study. In C. Seale
(Ed.), Researching Society and Culture. London: SAGE, 48-57.
Garriott, W. (2013). Policing and contemporary governance. The anthropology of police
in practice. New York: Palgrave Macmillan.
Garland, D. (2001). The culture of control. Crime and social order in contemporary
society. Oxford: Oxford University Press.
Goode, E. (1996). The ethics of deception in social research: a case study. Qualitative
Sociology, 19(1), 11-33.
maklu

Should I stay or should I go?
Maklu 37
Goffman, E. (2011). On fieldwork. In P. Atkinson & S. Delamont (Eds.), SAGE
Qualitative Research Methods. London: SAGE, 121-127.
Gubrium, J.F. & Holstein, J.A. (1997). The new language of qualitative method. New
York: Oxford University Press.
Guillemin, M. & Gillam, L. (2004). Ethics, reflexivity, and ‘ethically important
moments’ in research. Qualitative Inquiry, 10, 261-280.
Gurney, J.N. (1985). Not one of the guys: the female researcher in a male-dominated
setting, Qualitative Sociology, 8(1), 42-62.
Glover, D. & Kaplan, C. (2000). Genders. London: Routledge.
Harrington, B. (2002). Obtrusiveness as strategy in ethnographic research,
Qualitative Sociology, 25, 49-61.
Harvey, J. (2013). Footprints in the field: researcher identity in social research,
Methodological Innovations Online, 8(1), 86-98.
Horn, R. (1997). Not ‘One of the Boys’: women researching the police. Journal of
Gender Studies, 6(3), 297-308.
Hsiung, P.C. (2010). Lives and legacies: a guide to qualitative interviewing. Retrieved
from http://www.utsc.utoronto.ca/~pchsiung/LAL/
Kinderrechtencommissariaat (2012). Jongeren en politie: aanbevelingen voor meer
toenadering. Retrieved from http://www.kinderrechtencommissariaat.be/sites/
default/files/bestanden/2011_2012_5_advies_jongeren_en_politie.pdf
Laurilla, J. (1997). Promoting research access and informant rapport in corporate
settings: notes from research on a crisis company. Scandinavian Journal of
Management, 13(4), 407-418.
Labov, W. (1972). Sociolinguistic patterns. Oxford: Blackwell.
Lee, R. (1993). Doing research on sensitive topics. London: SAGE.
Leo, A.R. & Drizin, A.S. (2004). The problem of false confessions in the post DNA
world. North Carolina Law Review, 82, 891-1007.
Liebling, A. (2001). Whose side are we on? Theory, practice and allegiances in
prison research. British Journal of Criminology, 41, 472-484.
Loader, I. & Sparks, R. (2010). What is to be done with public criminology?
Criminology and Public Policy, 9(4), 771-781.
Lumsden, K. & Winters, A. (2014). Reflexivity in criminological research: experiences
with the powerful and the powerless. Basingstoke: Palgrave Macmillan.
maklu

Camille Claeys, Els Dumortier & Sofie De Kimpe
38 Maklu
Mainhard, T., van der Rijst, R. & van Tartwijk, J. (2009). A model for the supervisor-
doctoral student relationship. Higher Education, 58(3), 359-373.
Manning, P. (2003). Policing Contingencies. Chicago: The University of Chicago
Press.
Marks, M. (2004). Researching police transformation, the ethnographic imperative.
British Journal of Criminology, 53, 866-888.
Mawby, R.C. (2010). Police corporate communications, crime reporting and the
shaping of policing news. Policing & Society, 20(1), 124-139.
Mc Kegany, N. & Bloor, M. (1991). Spotting the invisible man: The influence of
male gender on fieldwork relations. British Journal of Sociology, 42(2), 195-210.
Miller, S., Forest, K. & Jurik, N. (2003). Diversity in blue: lesbian and gay police
officers in a masculine occupation. Men and Masculinities, 5, 355-385.
Miller, J.M. (2009). 21st century Criminology: a reference handbook, 1. USA: SAGE.
Mccall, G. (1978). Observing the law: field methods in the study of crime and the
criminal justice system. New York: The Free Press.
Monahan, T. & Fisher, J.A. (2010). Benefits of ‘observer effects’: lessons from the
field. Qualitative Research, 10(3), 357-376.
Pruvost, G. (2007). Enqueter sur les policiers: entre devoir de réserve, héroïsation
et accès au monde privé. Terrain (Revue d’Ethnologie de l’Europe), 48, 131-148.
Punch, M. (1983). De Warmoesstraat: politiewerk in de binnenstad: een etnografische
studie van politiewerk in Amsterdam. London, UK: Macmillan.
Punch, M. (1986). The politics and ethics of fieldwork. London, UK: SAGE.
Reiner, R. (2000). The politics of the police. Oxford: Oxford University Press.
Robben, C.G.M. (1995). The politics of truth and emotions among victims and
perpetrators of violence. In C. Nordstrom & C.G.M. Robben (Eds.), Fieldwork
under fire. Berkeley: University of California Press.
Ruby, D. (1980). Exposing yourself: reflexivity, anthropology and film. Semiotica,
30, 153-179.
Schwartz, O. (1993). Postface: l’empirisme irréductible. In N. Anderson & O.
Schwartz (Eds.), Le Hobo. Sociologie du sans-abri. Paris, Armand Colin, 380-390.
Spano, R. (2005). Potential sources of observer bias in police observational data.
Social Science Research, 34(3), 591-617.
maklu

Should I stay or should I go?
Maklu 39
Spano, R. (2006). Observer behavior as a potential source of reactivity: describing
and quantifying observer effects in a large-scale observational study of police.
Sociological Methods & Research, 34, 521-553.
Taylor, S. & Beasley, N. (2005). A Handbook for doctoral supervisors. London:
Routledge.
Thomas, J. (1993). Doing critical ethnography. USA: SAGE.
Van Maanen, J. (1978). Epilogue: on watching the watchers. In P. Manning & J. Van
Maanen (Eds.), Policing: a view from the street. Santa Monica, CA: Goodyear,
309-349.
Van Maanen, J. (2011). Tales of the field: on writing ethnography. Chicago: University
of Chicago Press.
Waddington, P. (1999). Police (canteen) sub-culture. An appreciation. British
Journal of Criminology, 39(2), 287-309.
Whitehead, T.L. & Price, L. (1986). Summary: sex and the fieldwork experience.
In T.L. Whitehead & M.E. Conaway (Eds.), Self, sex and gender in cross-cultural
fieldwork. Urbana, IL: University of Illinois Press, 289-304.
Wilson, S. (1977). The use of ethnographic techniques in educational research.
Review of Educational Research, 47(2), 245-265.
Yates, J. (2004). Criminological ethnography: risks, dilemmas and their negotiation.
British Journal of Community Justice, 3(1), 19-33.
maklu

maklu

Maklu 41
Bail decision-making in England and
Wales and Canada
The value of a comparative approach
The use of bail plays a central role in the rise of prison remand populations, influences
defendants and criminal justice institutions, and raises human rights concerns, yet
relatively little is known about the bail decision-making process. For the purposes of the
paper, this is considered to be the process by which court practitioners decide whether
defendants should be released on bail into the community or detained in custody pending
trial. It also involves the decision as to which, if any, conditions should be imposed upon
those released. The objective of this paper is to demonstrate the value of a comparative
approach in developing a better understanding of the bail decision-making process. It
is anticipated that such an approach will shed light on the relationship between bail
practices and prison remand population trends. The value of a comparative approach
will be illustrated through a preliminary examination of bail decision-making in England
and Wales1 and Canada.
Although there have been some studies that investigate bail decision-making within
specific jurisdictions, there is a paucity of research that makes comparisons on an
international level.2 This paper will demonstrate how this type of approach will offer
lessons of theoretical and practical significance. Specifically, a comparative analysis will
facilitate a greater understanding of an important part of the court process that has
rarely been discussed in previous research and will lay the foundation for interventions
in the legal systems examined.
The paper will first identify the significance of bail-decision making and explain how a
comparative approach could fill gaps in the research on this subject. Second, the value
of the comparative approach will be illustrated through a preliminary examination of
remand population trends, bail laws and informal practices in England and Canada.
Finally, a methodological design will be proposed for future comparative research.
Diana C. Grech3
1 For ease hereafter, the jurisdiction of ‘England and Wales’ will be referred to exclusively as England.
2 A study by the Open Society Justice Initiative (2014) takes a comparative approach to this topic,
but its focus is on the overuse of pretrial detention broadly, as opposed to bail decision-making
specifically.
3 University of Leeds, UK.
maklu

Diana C. Grech
42 Maklu
The significance of bail decision-making
The bail process has historically been conceived as a summary procedure in which
the liberty of defendants prior to trial is determined in a timely fashion (Webster
et al., 2009). This conceptualisation can perhaps explain why processes that are
traditionally considered to be more central, such as trial or sentencing, are more
commonly the focus of research. In reality, however, the determination of bail has
evolved into an influential part of the court process that is in need of examination.
This is because, first, bail decision-making has a major impact on the number
of individuals detained in remand custody. This portion of the prison population
is made up of defendants who have not received bail and are, instead, detained
in custody awaiting the conclusion of their criminal proceedings. Although
decisions made during the bail process cannot wholly explain increases in the
remand population,4 they provide considerable insight into potential contributing
factors at one stage of the process. Indeed, bail decision-making can be expected
to impact the two main drivers of the remand population: the number of accused
persons detained in custody and the length of time they spend in custodial remand
(Hucklesby, 2009; Webster et al., 2009).
There are also individual and institutional costs associated with bail decision-
making. The bail decision dictates whether defendants will be remanded in
custody or released into the community pending trial. Defendants remanded in
custody can suffer damage to their social, domestic, and financial circumstances
(Player et al., 2009) and may have an increased difficulty defending themselves
during the court process (Trotter, 2010). In addition, this decision determines to
what extent courts and correctional facilities must cope with those defendants who
are remanded in custody. These institutions must deal with the economic costs of
transporting and housing these individuals as well as the complex issue of their
management. For instance, the remand population can be unpredictable in terms
of their length of stay, must often be separated from sentenced offenders, and can
remain idle for long periods of time as they are often unable to access activities or
programs (Webster, 2007).
Finally, the determination of bail has important human rights implications given
that defendants are considered innocent until proven guilty.5 Bail decisions that
restrict the freedom of defendants can raise concerns associated with the right to
liberty and procedural fairness. Specifically, domestic and international human
rights law dictates that unconvicted individuals should only be detained with good
4 Decisions made outside of the bail process can also be expected to influence the size of the remand
population. For example, the length of time accused persons spend in custody following a decision
on bail would also have an impact.
5 This principle would not apply to the portion of the remand population that is awaiting sentencing
following a conviction.
maklu

Bail decision-making in England and Wales and Canada
Maklu 43
cause and in limited circumstances (see, for example, the European Convention
on Human Rights, the Canadian Charter of Rights and Freedoms, and the
International Covenant on Civil and Political Rights). These laws are intended to
protect individuals from being remanded in custody unless there are exceptional
circumstances that suggest their detention is warranted. When restraint in the
use of custodial remand is not exercised, the extent to which human rights
standards are being adhered becomes questionable and the central principle of
the presumption of innocence can become strained (Webster, 2007).
Despite the obvious human rights implications associated with bail decision-
making as well as the impact it has on individuals, institutions, and the prison
population, very little is known about this process. In fact, it has been called the
‘Cinderella’ of the justice system (Hucklesby & Sarre, 2009), receiving minimal
academic and political attention until quite recently. Given the rise in many
jurisdictions’ remand populations, however, researchers from around the world
have started to investigate the bail process in an attempt to understand the factors
that underlie these trends (see, for example, Booth & Townsley, 2009; Hucklesby,
2009; Myers, 2009; Player et al., 2010; Webster et al., 2009). It is important to
expand on this research in order to develop a more thorough understanding of
what has been referred to as the ‘remand problem’ (Doob, 2013). A comparative
approach will offer additional insight as to the factors that contribute to different
bail decision-making patterns and the effect of such decisions across jurisdictions.
Ultimately, this knowledge is anticipated to produce potential explanations for
remand population trends and offer suggestions for policy-makers seeking to alter
their direction.
Comparative law: towards a comprehensive methodology
With increasing global interdependence and the current practice of borrowing
and imitating the laws of other jurisdictions, it is perhaps unsurprising that the
use of comparative law has increased in the last decade (Nelken, 2007; Orucu,
2007). However, as its popularity increases so do its forms and the objectives
associated with its use. As such, the aims of the research will ultimately shape the
way in which comparative law is conceived (Nelken, 2007; Samuel, 2013). Indeed,
there is on-going controversy amongst comparative legal scholars as to whether
comparative law is a method used for practical purposes or a subject matter in
itself. It is anticipated that the more ‘traditional’ form of comparative law, in
which the approach is used primarily as a method, would be the most useful to a
comparative researcher interested in bail decision-making. This is often referred
to as the ‘functional’ method of comparative law (Michaels, 2005; Zweigert & Kotz,
1998) and its focus is discovering which legal system − or in this case, which bail
system − fulfils certain legal demands (Jansen, 2006). Proponents of the traditional
maklu

Diana C. Grech
44 Maklu
approach seek to use comparative law for primarily instrumental purposes and are
interested in comparing legal rules and institutions for practical purposes related
to adjudication and law reform (Orucu, 2004). They often attempt to learn from
other systems how to improve their own and seek to borrow an institution, practice,
technique or idea in order to reach this objective (Nelken, 2010). The functional
method would be useful in determining the impact of various bail decision-making
practices across jurisdictions. Policy-makers aiming to reform their bail systems
could use the knowledge gained in this process to adopt laws, policies, or practices
that have been shown to achieve desired objectives in other legal systems. For
instance, jurisdictions with rising remand rates could implement procedures used
in jurisdictions with decreasing or stable remand rates in order to reduce their
prison remand population.
Although this approach is both useful and widely employed, it is not without
criticism. Indeed, some researchers have noted the dangers in reducing
comparative law to attempts to improve policy without acknowledgement of the
context that underpins certain laws and practices (Ewald, 1995; Nelken 2007;
Orucu, 2007). Nelken (2007) asserts that it is challenging to find ‘solutions’ to
domestic problems given that many of the ‘problems’ are closely intertwined with
otherwise valued features of the society. Consequently, Nelken (2010) warns that
finding explanations or differences that translate quickly into policy arguments
may be difficult. This does not, however, imply that traditional comparative legal
methods are impracticable. Rather, solutions can be obtained with regard for the
context underlying the comparison. For instance, Orucu (2007) has stated that
different values pursued by different legal systems can and should be investigated
and acknowledged when making recommendations. As such, researchers who
are comparing bail decision-making in multiple jurisdictions must remember to
take the context underlying various laws and practices into account when they are
suggesting that another jurisdiction adopt them. It may be that an approach that
achieves a particular objective in one jurisdiction would not have the same impact
in another jurisdiction with a significantly different culture or value system.
While previous studies that have investigated bail decision-making have provided
valuable insight into this topic in individual jurisdictions, the use of the functional
method of comparative law would fill gaps in the research by offering a broader
understanding of the subject. This is because the previous research rarely
investigates jurisdictions relative to each other. Zweigert and Kotz (1998) suggest
that approaches that focus on one jurisdiction only offer legal solutions to practical
problems ‘on their own terms’ and argue that solutions should be freed from the
context of their own system in order for evaluation to take place. For example,
Myers (2009) has shown that an over reliance on sureties6 in Ontario courts can
6 A surety offers to pay a sum of money on behalf of a defendant if he or she does not comply with
maklu

Bail decision-making in England and Wales and Canada
Maklu 45
lengthen bail proceedings and increase the time defendants spend in remand
custody. While this is valuable from a procedural standpoint, it does not answer
broader questions about the appropriate role of a surety or investigate alternatives
to current practices. For instance, a logical solution to the problem identified by
Myers (2009) might be to instruct court practitioners to reduce the use of sureties
in Ontario courts. A solution based in a comparative approach is capable of going
further. Rather than suggesting a decrease in the number of sureties imposed, a
comparative study may suggest that the expectations placed on sureties be altered
to be more in line with another jurisdiction that does not impose sureties to the
same extent. This type of solution, which targets the mentality underlying the
problem, is likely to have a broader, more enduring impact.
While there have been edited collections which examine a number of different
jurisdiction separately, this approach can also be problematic in the sense that no
actual comparison has taken place. Nelken (2010) asserts that this individualistic
approach is often high value in terms of offering guides to local developments, but
argues that what emerges is often of only tangential relevance to cross-national
generalisations. For example, a special issue was published in Current Issues in
Criminal Justice in 2009 (vol. 21, no. 1) that examines bail in the United Kingdom,
Australia and Canada. Multiple authors from all three jurisdictions offered potential
explanations for their respective remand population trends. Although the articles
offered valuable information about the United Kingdom, Australia and Canada
separately, they largely focused on the issues pertaining to each jurisdiction in
isolation. For instance, Hucklesby (2009) suggested that government initiatives
that aim to reduce the remand population often have a limited impact on bail
decision-making patterns in England. This conclusion lends itself to further
analysis on a wider scale. Is this resistance to policy changes exclusive to England
or is it applicable to the other jurisdictions? Is policy an appropriate way to reduce
remand populations in Canada or Australia? These broader questions can only be
answered by removing oneself from the confines of single jurisdiction studies and
examining them in a comparative context.
Ultimately, it is projected that this type of approach can achieve what Samuel
(2013) sees as a fundamental component of comparative law: the acquisition of
new knowledge through the process of comparison. One must be able to draw
conclusions from the comparison that could not otherwise be drawn had the two
objects been analysed separately. It is anticipated that a comparative approach to
understanding bail decision-making will expand on and overcome many of the
obstacles involved with single jurisdiction studies through the acquisition of this
new knowledge.
bail conditions (in Canada) or does not appear for court (in England).
maklu

Diana C. Grech
46 Maklu
The use of bail in England and Canada
In order to ensure that useful knowledge is obtained from a comparative analysis,
appropriate jurisdictions must be selected as sources of comparison. An examination
of remand population trends worldwide indicates that England and Canada are ideal
subjects of a comparative approach. These jurisdictions share similar historical
foundations and adhere to similar international human rights instruments, yet
statistics reveal a marked difference in their use of remand custody.
Figure 1.0 illustrates that, in England, the overall rate of imprisonment fluctuated
from the late 1970s to the early 1990s, when it increased sharply initially and then
more gradually following the turn of the century. The remand rate, however, has
remained relatively stable over this same time period. In contrast, Canada’s overall
adult imprisonment rate has fluctuated, but maintained fairly stable since the late
1970s, while its remand rate has nearly tripled.
Source (England): Home Office, 1981, Table 1.4; Home Office, 1984, Table 1.4; Home Office, 1987,
Table 1.9; Home Office, 1990, Table 1.2; Home Office, 1993, Table 1.2; Ministry of Justice, 2004, Table
8.1; Ministry of Justice, 2010, Table 7.1; Ministry of Justice, 2011, Table 1.1; Ministry of Justice, 2013,
Table 1.1a.
Source (Canada): Statistics Canada, 2013a; Statistics Canada, 2013b.
Figure 1. Components of the Prison Population in England and Wales7 and Canada8
(Rates9 per 100,000 Population)
7 Due to the availability of data, rates from 1971 to 1984 do not include police cells in England. This
period is also constituted by yearly averages, while from 1985 to 2013 is represented by June 30th
counts.
8 Data limitations required the inclusion of the adult prison population only in Canada (and thus
adult population estimates).
9 Rates were calculated with population estimates from Office for National Statistics, National
Records of Scotland, Northern Ireland Statistics and Research Agency in England and Statistic
Canada’s CANSIM table 051-0001 Estimates of Population by age group in Canada.
maklu

Bail decision-making in England and Wales and Canada
Maklu 47
In England, the remand population comprises a relatively small percentage of
the total prison population (13%) and the remand rate has decreased slightly,
from 21.9 per 100,000 population in 2000 to 19.3 per 100,000 population in 2013
(Ministry of Justice, 2004; Ministry of Justice, 2013). The remand population rate
in Canada has increased significantly, from 31.4 per 100,000 population in 2000 to
49.4 per 100,000 population in 2012, and remand prisoners comprise 35% of the
adult prison population (Statistics Canada, 2013a; Statistics Canada, 2013b).
Given that remand rates are, to a large extent, dictated by bail decision-making
practices in each jurisdiction, the disparity in these trends indicates that these
practices are quite different in England and Canada. However, it is currently
unclear what it is about the bail decision-making in each jurisdiction that explains
this disjuncture. A comparative approach that investigates the factors contributing
to the bail decision-making in England and Canada would be useful in explaining
why their remand population trends have taken such different directions.
The law related to bail in England and Canada
If identifying the factors that contribute to bail decision-making is the goal of the
comparative analysis, beginning with an examination of the law related to bail
in England and Canada may prove instructive. This is because these laws can be
expected to frame the decision-making of practitioners in both jurisdictions.
Both English and Canadian legal systems share a similar common-law tradition.
This, in part, explains why the bail laws in each jurisdiction have a similar basis.
The main statute governing the grant of bail in England is the Bail Act 1976
while the bail laws in Canada are primarily rooted in amendments made to the
Criminal Code of Canada introduced by the Bail Reform Act 1972. These Acts were
preceded by studies from both jurisdictions that stressed the importance of human
rights and due process in relation to the determination of bail (see, for example,
Bottomley, 1970; Canadian Committee on Corrections, 1969; Friedland, 1965;
King, 1971). They argued that remand custody was being used in cases where bail
was more appropriate and that it had a detrimental impact on defendants. Many
of the recommendations contained in these reports were reflected in the Bail Act
1976 and the Bail Reform Act 1972, forming the basis of a new ‘enlightened’ era in
the history of bail law (Trotter, 2010).
The guiding philosophies of both Acts are underscored by themes related to the
presumption of innocence. Section 4 of the Bail Act 1976 declares a general right
to bail, usually referred to as ‘the presumption in favour of bail’, to most categories
maklu

Diana C. Grech
48 Maklu
of defendants10 who are brought before the court in England. This principle
suggests that, unless specific exceptions apply, accused persons should be granted
bail. The amendments to the Criminal Code that were introduced by the Bail
Reform Act 1972 also specify a presumption of release in Canada. The legislation
specifies that bail decisions are to be governed by the underlying presumption
that, unless the Crown can justify otherwise, accused persons should be released
into the community pending their next court appearance (Criminal Code, 1985, s.
515(1)).11 The rights enshrined in the Bail Act 1976 and the Bail Reform Act 1972
are intended to protect defendants from being remanded in custody unless there
are exceptional circumstances that suggest their detention is warranted. In other
words, defendants are, in the main, entitled to their liberty.
Both England and Canada have demonstrated their commitment to this principle
by promising to comply with the standards set out in the International Covenant
on Civil and Political Rights (ICCPR). Under Article 9, Paragraph One of the
ICCPR, ‘everyone has the right to liberty and security of person. No one shall be
subjected to arbitrary arrest or detention. No one shall be deprived of his liberty
except on such grounds and in accordance with such procedure as are established
by law.’ In order to uphold this basic human right, both jurisdictions must ensure
that accused persons are only remanded in custody when their detention can be
justified for lawful reasons. Although the individual right to liberty is protected by
the ICCPR in both England and Canada, it is generally the national and regional
systems for human rights that are referred to if a violation is suspected. Specifically,
Article 5(3) of the European Convention on Human Rights12 guarantees a right
to trial within a reasonable time or to be released pending trial in England and
Section 11(e) of the Canadian Charter of Rights and Freedoms specifies a right to
reasonable bail in Canada.
In both Canadian and English bail systems, the emphasis placed on the
presumption of bail clearly demonstrates that defendants are to be released
into the community unless the state can demonstrate that there is a legitimate
reason to remove this right. However, in both jurisdictions this principle has been
eroded by a number of amendments that restrict the right to bail. In England,
these amendments were precipitated by concerns over offences committed on
bail (Hucklesby & Marshall, 2000) that occurred while the climate of the criminal
10 Section 4(2) and Section 4(7) of the Bail Act 1976 specify that the general right to bail does not apply
to convicted persons (unless they are awaiting a report), fugitive offenders, or persons charged with
treason (who must appear before a judge of the High Court or a Secretary of State).
11 Section 515(1) does not apply to those serious offences listed in s. 469 (e.g. treason, piracy, murder),
which must be heard by a Superior Court Judge. In these cases the accused must prove why he or
she should be released.
12 The European Convention has been incorporated into domestic law through the Human Rights
Act 1998.
maklu

Bail decision-making in England and Wales and Canada
Maklu 49
justice system was shifting to a reduced emphasis on the rights of defendants and
a greater focus on the protection of the public and the rights of victims (Hucklesby,
2009). For instance, the Criminal Justice and Public Order Act 1994 introduced a
change in legislation that substantially restricts the right to bail for defendants
accused of serious offences (e.g. murder, manslaughter, and rape) who have
previous convictions for similar offences. In addition, the Criminal Justice Act
2003 reversed the presumption of bail for defendants alleged to have committed
indictable or either way offences while on bail and instructed that defendants who
test positive for drug use need not be granted bail if the offence is drug related and
they refuse to undergo assessment and/or treatment for drug dependency.
Similar changes to the bail legislation indicate that a comparable attitude was
forming in Canada. In this jurisdiction, there has been a growing trend towards the
use of ‘reverse onus’ provisions that place the responsibility on defendants, rather
than the Crown, to justify why they should be released on bail. Provisions like this
are ‘slowly eroding years of legislative and jurisprudential change sparked by the
Bail Reform Act in the 1970s’ (Trotter, 2010, 1-4). The stream of provisions began
in the mid-1970s, when the Criminal Amendment Act 1975 reversed the onus
in cases when an indictable offence was alleged to have been committed while
the accused was at large awaiting trial for another indictable offence, when the
accused was charged with an indictable offence and was not a resident of Canada,
and when the accused was charged with drug trafficking (importing or exporting).
Several similar provisions were implemented before this trend culminated in
2008 when a package of reverse onus provisions related to firearm offences came
into force under the Tackling Violent Crime Act. These amendments more than
doubled the number of reverse onus provisions in the Criminal Code. The growing
use of such provisions clearly represents a gradual shift in the attitudes towards
bail similar to those in England.
Although the preceding comparison is not exhaustive, it does demonstrate that
both the foundation and evolution of the bail laws in England and Canada share
striking similarities. Both jurisdictions appear to have moved from an approach to
bail that prioritises the rights of defendants to an approach that restricts these rights
in the name of public safety. Given this shift, one might assume that the remand
rates in both jurisdictions would have been rising steadily over the last several
decades. However, the statistics clearly demonstrate that only Canada has followed
this trend. It is evident that despite having similar starting points in relation to
bail, the two jurisdictions are experiencing extremely different outcomes. This
suggests that there are other factors influencing the bail decision-making process
irrespective of each jurisdiction’s bail laws. Indeed, it may be informal practices,
rather than formal rules and procedures that account for differences in bail
decision-making in these jurisdictions.
maklu

Diana C. Grech
50 Maklu
The importance of informal practices
There is some research to support the suggestion that an explanation for the
disjuncture in remand population trends in England and Canada is rooted in
informal practices by bail decision-makers. For instance, a recent study by the
Open Society Justice Initiative (2014) found that a multitude of informal practices
adopted by bail decision-makers facilitates the overuse of pre-trial detention. These
practices include, but are not limited to, the flouting of legislative time limits
on detention, adherence to populist pressures that promote restrictive decision-
making, the abuse of discretionary power to facilitate the overuse of custody,
and a lack of coordination between criminal justice officials resulting in delays.
Although it is clear that these issues vary between jurisdictions, these findings
demonstrate that informal practices, broadly, can have a significant impact on
remand population trends.
Researchers from England and Canada, specifically, have also pointed to the
importance of informal practices in efforts to explain remand population trends
(see, for example, Hucklesby, 2009; Myers, 2009; Webster et al., 2009). There is
some − albeit limited − research to suggest that there are substantial differences
in informal practices in English and Canadian courts. Studies that have examined
courts in which bail decisions are made have shown disparate uses of processes
with similar functions. For example, The Canadian Civil Liberties Association
(2014) observed that sureties are frequently required to secure defendants’
releases on bail in certain provinces in Canada while Hucklesby (2011) found that
they are rarely required in England. Defendants are required to stay in custody
until appropriate sureties are available to come to court and, in cases where a
surety cannot be obtained, defendants may be denied bail as a result. Further,
bail conditions have been found to be attached to the release of defendants much
more frequently across provinces in Canada (Canadian Civil Liberties Association,
2014; John Howard Society, 2013; Myers, 2009) than they have in England
(Dhami, 2004; Hucklesby, 2002). In circumstances in which multiple, restrictive
conditions are imposed, defendants may breach these conditions and experience
difficulties obtaining another release on bail when they are brought back to court.
Finally, a ‘culture of adjournments’ has developed in Canadian courts in which
bail proceedings are frequently put over to another day before a bail decision is
made (Webster et al., 2009). This practice extends the time that defendants spend
in remand custody awaiting a determination of bail. It is unclear to what extent
adjournments are used in England as no recent studies have examined this issue.
The informal practices that are employed in each jurisdiction appear to be in
line with broader attitudes associated with bail in each jurisdiction. In England,
the government has become increasingly concerned about overcrowding in the
general prison population. Although there has been legislative restrictions placed
maklu

Bail decision-making in England and Wales and Canada
Maklu 51
on the presumption to bail, there have also been an influx of policy initiatives
which send the message to court practitioners that they should make less use of
custodial remand (Hucklesby, 2009). These initiatives have persisted despite the
Conservative government coming into power in 2010.13 In contrast, the prevailing
climate in Canada is marked by a culture of risk aversion. In this culture court
practitioners are reluctant to make decisions related to bail out of fear they will
be held accountable for potential bail breaches or public dissatisfaction (Doob,
2012; Myers, 2009; Webster et al., 2009). Unlike the bail legislation in England
and Canada, informal practices seem to correspond with their respective remand
population trends. In Canada, where the remand rate has been increasing, practices
within the court and mentalities surrounding bail seem to favour a restrictive use
of bail. In England, where the remand rate has been decreasing, it would seem
that there is – relatively speaking − more leniency in relation to practices and
mentalities associated with bail.
It is clear that a more thorough examination of bail decision-making practices
in England and Canada is necessary in order to develop a greater understanding
of their remand population trends. This preliminary comparative analysis has
demonstrated that close attention should be paid to informal practices, rather
than formal rules and procedures, in attempts to explain the use of remand in
these jurisdictions. The use of a comparative approach has been instrumental
in guiding future research in this direction as it has allowed the examination to
extend beyond the confines of a single jurisdiction, permitting the identification
of factors that contribute to bail decision-making across different legal systems.
A methodological design can be developed and implemented as a result of these
findings.
The way forward
It is evident that a comprehensive comparative approach must move beyond the
examination of bail laws and directly compare informal practices in England and
Canada. Indeed, Nelken (2010) advises that comparative researchers should pay
attention to differences between the ‘law in books’ − what the rules say about what
is supposed to happen and the ‘law in action’ − how the law is or is not used in
practice. The law in action may include frontline professionals’ decision-making
or conduct, often motivated by third party directives (e.g. police guidelines,
prosecutorial instructions) rather than primary legal rules. According to Roberts
(2007), these unwritten operational policies occupy the shadowlands of informal
13 The Legal Aid, Sentencing and Punishment of Offenders Act 2012 instructs magistrates to consider
the likelihood of imprisonment in their bail decisions, presumably with the aim of reducing
remands in custody for those who would otherwise not receive a prison sentence.
maklu

Diana C. Grech
52 Maklu
agreements, institutionalized routines, shared professional understandings, and
taken for granted cultural assumptions.
The ‘law in action’ is particularly important to the examination of bail decision-
making in England and Canada given the aforementioned similarities in their bail
laws. It is likely that, in addition to external factors, differences in their use of
remand custody are rooted in a disjuncture in the interpretation and the application
of the law as opposed to the legislation itself. An appropriate way to examine these
factors would be through ‘structured observation’ of bail proceedings. This involves
the systematic observation of behaviour in terms of a schedule of categories that
has been devised prior to the commencement of data collection (Bryman, 2004).
Information about the defendant and the case, the position of court practitioners,
and the information they provide in open court could be recorded in this schedule
for each course observed. This type of method would present an idea of the working
culture in each court as well as identify patterns in bail decision-making.
Court observations should not, however, be the only method employed in a
comparative analysis of bail decision-making. Rather, they should be combined
with interviews in order to provide context for the behaviour observed and reveal
the underlying values and attitudes of court practitioners. Interviews should
be conducted with judicial officials, defence solicitors, prosecutors,14 and legal
advisors15 as these are the primary court ‘decision-makers’ in both jurisdictions.
The interviews should be conducted in a semi-structured format as it encourages
the interviewees to expand on their answers, allowing the researcher to gain insight
as to what they see to be relevant and important (Bryman, 2004). Questions might
focus on the relevance of law and policy to the decision-making process and ask
about the reasons underlying certain practices observed in court. It is anticipated
that the information obtained in the interviews will provide valuable information
about the legal culture in each jurisdiction. Orucu (2007) argues that since legal
culture is a configuration of values, concepts, practices and institutions through
which individuals interpret and apply legal norms, it is rooted in the general
culture of a particular society. It is critical to develop an understanding of this
culture in order to develop a thorough understanding of the context underlying the
bail decision-making process.
This mixed methods approach, which includes quantitative and qualitative court
observations as well as qualitative interviews with court practitioners, will ensure
that broad trends are identified while at the same time the findings are placed in
14 It should be noted that, while it is not impossible, it can be difficult to obtain interviews with Crown
counsel in Canada because they can be reluctant to participate in research. However, efforts should
still be made to gain their input.
15 Legal advisors could only be interviewed in England as this type of position does not exist in
Canada.
maklu

Bail decision-making in England and Wales and Canada
Maklu 53
the appropriate context. Ultimately this is expected to contribute to a thorough
understanding of the use of informal practices in each jurisdiction and the impact
these practices have on bail decision-making patterns and the use of remand.
Conclusion
It is clear that a comparative approach will contribute to a greater understanding
of the bail decision-making process across multiple jurisdictions. The value of this
approach has been exemplified through a comparative analysis of bail decision-
making in England and Canada, ultimately demonstrating that an explanation for
the disjuncture in their use of remand is rooted in informal practices related to
bail decision-making as opposed to in the laws related to bail. An examination of
the law in action as exhibited by court observations and legal culture as expressed
in interviews with court practitioners is expected to shed further light on the
relationship between these informal practices and remand population trends.
In addition to offering a better understanding of a poorly understood court
process, the comparative approach can also provide valuable information to policy-
makers who are hoping to make changes to the bail systems in both jurisdictions.
By contrasting informal practices that are taken for granted to be the ‘norm’ in
one culture with another culture in which the practice is much different, it may
enable practitioners to explore possibilities they would not have anticipated. When
examining the attitudes and practices in two jurisdictions, we avoid taking for
granted the subjectivity of values embedded in a specific legal culture that may
have been assumed to be universal truths. It is only by developing a greater
understanding of one’s own system in relation to another that a comprehensive
understanding of the impact of the bail decision-making process can be obtained.
References
Bottomley, A.K. (1970). Prison before trial, occasional paper in social administration,
no. 39. London: Bell and Son.
Booth, T. & Townsley, L. (2009). The process is the punishment: the case of bail in
New South Wales. Current Issues in Criminal Justice, 21(1), 41-58.
Bryman, A. (2004). Social research methods.(2nd ed). Oxford: Oxford University
Press.
Canadian Civil Liberties Association and Education Trust. (2014). Set Up to Fail:
Bail and the Revolving Door of Pre-trial Detention. Toronto: Canadian Civil
Liberties Association.
maklu

Diana C. Grech
54 Maklu
Canadian Committee on Corrections. (1969). Report of the Canadian Committee on
Corrections – Toward unity: Criminal justice and corrections. Ottawa: Department
of the Solicitor General.
Dhami, M.K. (2004). Conditional bail decision making in the magistrates’ court.
The Howard Journal, 43(1), 27-46.
Doob, A.N. (2012). Bail and remand: one problem or two? Fourth National
Symposium, Re-Inventing Criminal Justice, 12 January, Victoria, B.C.
Doob, A.N. (2013). The Bail Process in Ontario: An Overview. Centre for Criminology
and Sociolegal Studies, University of Toronto, 5 June 2013, 17.
Ewald, W. (1995). Comparative jurisprudence (1): what was it like to try rat?
Pennsylvania Law Review, 143, 1889.
Friedland, M.L. (1965). Detention before trial: A study of criminal cases tried in the
Toronto magistrates’ courts. Toronto: University of Toronto Press.
Home Office. (1981). Prison Statistics England and Wales 1980. London: HMSO.
Home Office. (1984). Prison Statistics England and Wales. 1983. London: HMSO.
Home Office. (1987). Prison Statistics England and Wales 1986. Cm 210. London:
HMSO.
Home Office. (1990). Prison Statistics England and Wales 1989. Cm 1221. London:
HMSO.
Home Office. (1993). Prison Statistics England and Wales 1992. Cm 2581. London:
HMSO.
Hucklesby, A. (2002). Bail in criminal cases. In M. McConville & G. Wilson (Eds).
Handbook of the Criminal Justice Process. Oxford: Oxford University Press, 115-
136.
Hucklesby, A. (2009). Keeping the lid on the prison remand population: the
experience in England and Wales. Current Issues in Criminal Justice. 21(1), 3-23.
Hucklesby, A. (2011). Bail support schemes for adults. Bristol: The Policy Press.
Hucklesby, A. & Sarre, R. (2009). Bail in Australia, the United Kingdom and
Canada: Introduction. Current Issues in Criminal Justice, 21(1), 1-2.
Hucklesby, A. & Marshall, E. (2000). Tackling offending on bail. The Howard
Journal, 39(2), 150-170.
Jansen, N. (2006). Comparative law and comparative knowledge. In M. Reimann.
& R. Zimmermann (Eds), The Oxford Handbook of Comparative Law. Oxford:
Oxford University Press, 305-338.
maklu

Bail decision-making in England and Wales and Canada
Maklu 55
John Howard Society of Ontario. (2013). Reasonable Bail? Toronto: John Howard
Society of Ontario.
King, M. (1971). Bail or custody? London: Cobden Trust.
Leverick, F. & Duff, P. (2002). Court culture and adjournments in criminal cases:
a tale of four courts. The Criminal Law Review, 39, 39-52.
Michaels, R. (2005). The functional method of comparative law. Duke Law School
Faculty Scholarship Series, 26, 1-47.
Ministry of Justice. (2004). Offender management caseload statistics 2003. London:
Ministry of Justice.
Ministry of Justice. (2010). Offender Management Caseload Statistics 2009. London:
Ministry of Justice.
Ministry of Justice. (2011). Offender management statistics quarterly bulletin April to
June 2011, England and Wales. London: Ministry of Justice.
Ministry of Justice. (2013). Offender Management Statistics Quarterly Bulletin
January to March 2013. London: Ministry of Justice.
Myers, N. (2009). Shifting risk: bail and the use of sureties. Current Issues in
Criminal Justice, 21(1), 127-147.
Nelken, D. (2007). Comparative law and comparative legal studies. In E. Orucu
& D. Nelken (Eds.), Comparative Law: A Handbook. Portland: Hart Publishing,
3-42.
Nelken, D. (2010). Comparative criminal justice: making sense of difference. London:
SAGE.
Open Society Justice Initiative. (2014). Presumption of guilt: The global overuse of pre-
trial detention. New York: Open Society Foundations.
Orucu, E. (2004). The enigma of comparative law: variations on a theme for the 21st
century. The Hague: Martinus Nijhoff.
Orucu, E. (2007). Developing Comparative Law. In E. Orucu. & D. Nelken (Eds.).
Comparative Law: A Handbook. Portland: Hart Publishing, 43-66.
Player, E., Roberts, J., Jacobson, J., Hough, M. & Robottom, J. (2010). Remanded
in custody: an analysis of recent trends in England and Wales. The Howard
Journal, 49(3), 231-251.
Roberts, P. (2007). Comparative Law for International Criminal Justice. In E.
Orucu, & D. Nelken (Eds.), Comparative Law: A Handbook. Portland: Hart
Publishing, 339-370.
maklu

Diana C. Grech
56 Maklu
Samuel, G. (2013). Comparative law and its methodology. In D. Watkins. & M.
Burton (Eds.), Research Methods in Law. Oxon: Routledge, 100-118.
Statistics Canada. (2013a). Table 251-0006 – Adult correctional services, average
counts of offenders in federal programs, annual, 1978/1979 to 2011/2012,
CANSIM [Database]. Retrieved from: http://www5.statcan.gc.ca/cansim/
a26?lang=eng&retrLang=eng&id=2510006&tabMode=
dataTable&srchLan=-1&p1=-1&p2=9
Statistics Canada. (2013b.) Table 251-0005 – Adult correctional services, average counts
of offenders in provincial and territorial programs, annual, 1978/1979 to 2011/2012,
CANSIM [Database]. Retrieved from: http://www5.statcan.gc.ca/cansim/
a26?lang=eng&retrLang=eng&id
=2510005&tabMode=dataTable&srchLan=-1&p1=-1&p2=9
Trotter, G.T. (2010). The law of bail in Canada. Scarborough, Ontario: Carswell.
Webster, C.M. (2007). Remanding the problem: An evaluation of the Ottawa bail court.
Ottawa: Ministry of the Attorney General.
Webster, C.M., Doob, A. & Myers, N. (2009). The parable of Ms. Baker:
understanding pre-trial detention in Canada. Current Issues in Criminal Justice,
21(1), 79-102.
Zweigert, K. & Kotz, H. (1998). Introduction to comparative law (3rd ed.) Oxford:
Oxford University Press.
maklu

Maklu 57
Studying the contextual cues associated with fear
of crime through eye tracking techniques
An exploratory research
Fear of crime is a hot topic in the criminological field and a problem in its own right.
Researchers have studied the individual and contextual variables involved in fear of
crime. This exploratory study uses eye tracking techniques to explore the features of
environment that have an impact on fear of crime. A set of urban pictures were obtained,
rated by an external sample and selected to be visualized. These were then presented to
a sample of 10 individuals that observed the pictures and judged whether the place was
insecure or not. Heat map analyses revealed that the features that participants paid more
attention to were the prospect of the environment and signs of disorder (mainly graffiti).
However, comparisons between pictures did not reveal statically significant differences for
latency, number of fixations and total duration of fixations. The results and limitations of
this research will be discussed and a new improved eye tracking protocol will be outlined.
Inês Guedes1, Pedro Fernandes2,
Cândido da Agra1 & Carla Cardoso1
Introduction
Crime and fear of crime have an uneven distribution over space and time. They
tend to concentrate in specific areas and situations (Nasar & Fisher, 1993). These
areas and situations in which fear of crime is concentrated have been termed
‘hot spots of fear’. The study of these ‘hot spots’ at a micro-level are important to
develop situational strategies regarding the reduction of fear of crime. In fact, fear
of crime is more prevalent than actual victimization (Hunter, 1978) and affects
community health. The lack of lighting or simply darkness, social and physical
incivilities and lack of prospect have been the main contextual cues associated
with higher levels of fear of crime (Wilson & Kelling, 1982; Fisher & Nasar, 1995;
Painter, 1996). Studies addressing the relation between fear of crime and spatial
features usually employ surveys to measure the reactions to different urban spaces.
This paper presents the results of an exploratory study which had the main goal
of exploring the contextual features associated with fear of crime. Eye tracking
1 School of Criminology, Faculty of Law, University of Porto.
2 Department of Physics and Astronomy, Faculty of Science, University of Porto.
maklu

Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
58 Maklu
is the process of analyzing the spatial locations to which people are looking in
order to understand human behavior. Under normal circumstances (Hoffman &
Subramaniam, 1995) it reflects the selective attention that individuals deploy to
stimuli, since the direction of gaze is generally considered to be coupled with the
orienting of attention.
In the present study, we define the concept of fear of crime and describe the
individual and contextual variables that have been associated with it. Although the
studies that have applied eye tracking techniques to safety questions are scarce,
some authors gave important contributions to this field (Davoudian & Raynham,
2012; Kim et al., 2014). These studies, along with the ones that have explored the
selective attention to emotional pictures will be outlined. Then, the methodology
and the results of this study are presented. Finally, the results are discussed at the
light of the literature on fear of crime.
Theoretical background
Definition of fear of crime
There are two broad categories in what concerns the definition of fear of crime
(Vanderviver, 2011). A more restricted approach argues that fear of crime is an
emotional dimension which makes part of a larger concept – insecurity feelings
(e.g., Warr, 2000; Ferraro & LaGrange, 1987). The other view is wider, reflecting a
multidimensional approach, in which fear of crime includes emotional, cognitional
and behavioral elements (e.g., Gabriel & Greve, 2003; Madriz, 1997; Skogan, 1999).
The emotional dimension of fear of crime if often defined as a negative reaction to
crime or symbols associated with it (Ferraro & LaGrange, 1987). When people fear
crime, there may be a series of complex changes in bodily functioning that alerts
them to potential danger (Guedes et al., 2013). Accordingly, Warr (2000, 453-454)
considers that fear of crime is an emotion, a feeling of alarm or dread caused by an
awareness or expectation of danger (…) associated with certain physiological changes,
including increased heart rate, rapid breathing, sweating, decreased salivation and
increased galvanic skin response. In this perspective, fear of crime is not a perception
of the environment but an emotional reaction to the perceived environment.
Beyond this emotional component, insecurity feelings also encompasses cognitive
and behavioral dimensions. The behavioral component if reflected by what people
do when they fear or perceive risk. These behavioral manifestations can be divided
in ‘avoidance’ (e.g., avoiding dangerous places and people, walking accompagnied
after dark) and ‘mobilization’ (e.g., outdoor lights, door locks, window bars). In fact,
Liska, Sanchirico and Reed (1988) argue that these security behaviors are extended
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 59
not only towards avoiding some places and situations related to crime but also
towards protecting oneself from places and situations related to crime. Lastly, the
cognitive dimension is related to an assessment of reality or the dangerousness
of victimization (Mesch, 2000). Skogan (1999) defines the risk perception as
the perception made by individuals about the likelihood of being victimized. In
the same line, Ferraro and LaGrange (1987) constructed a taxonomy where they
differentiate fear of crime and risk perception. According to the authors, while
fear of crime is an emotional reaction to crime, risk perception is a judgment of
individual’s security. Although fear of crime and risk perception may be related,
some authors (e.g., Mesch, 2000; Guedes, 2012) argue that these variables are
differently affected by individual and social characteristics of subjects.
The view of Gabriel and Greve (2003) complements the definitions referred above.
The authors suggest that the state of fear only appears with the concurrence of three
conditions: the individual’s cognitive perception of being threatened (cognitive
dimension), a corresponding affective experience (affective dimension) and an
appropriate motive or action tendency. In this conceptualization, and contrary to
other authors (e.g., Skogan, 1999; Liska et al., 1988), the behavioral expression
covers not only the intentionally planned and controlled behavior, but also the
visible behavior that is not intentionally planned, and also physiological reactions
(Gabriel & Greve, 2003).
But is fear of crime a general experience (such as anxiety) or a specific reaction
to a stimulus? Farrall et al. (2009) discuss the difference between experiential
fears and expressive fears. Experiential fears are fleeting experiences that are
transitory, infrequent and short-lived. They appear in reaction to external stimuli
when individuals think that threat of crime may be possible. On the other hand,
expressive fears are constructed through the knowledge and perceptions of
people’s community and broader socio-cultural concerns. These expressive fears
are related to mental processes (ibidem).
In the present study fear of crime is equivalent to the expression safety or security.
Individuals were asked to assess the safety of locations in the images presented
and then to evaluate if the images were or not insecure. Hence, we are measuring
an ‘experiential fear’ rather than an ‘expressive fear’.
Individual determinants of fear of crime
Researchers have been focusing on individual and contextual variables that are
related to fear of crime (see Hale, 1996; Guedes et al., 2013 for a review). Regarding
individual variables, it is largely accepted that women fear more, perceive higher
risk and adopt more security behaviors (Reid & Konrad, 2004; Machado & Agra,
maklu

60 Maklu
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
2002). Although elderly individuals often present higher levels of fear of crime,
other studies showed that younger people fear more (e.g., Chadee & Ditton, 2003;
Ziegler & Mitchell, 2003; Gomme, 1988). On the other hand, studies consistently
show that the least educated are the ones who present higher levels of fear of crime
(e.g., Smith & Hill, 1991; Guedes, 2012). If age, sex and education have been often
associated with fear of crime, the same is not true for personality dimensions,
where the literature is scarce. The few studies that correlated personality
dimensions with fear of crime conclude that neuroticism (a trait related with
anxiety) is positively correlated with fear of crime (Klama & Egan, 2011; Guedes,
2012). Lastly, researchers have studied the relationship between past victimization
experiences and fear of crime, concluding that the results are mixed (Hale, 1996).
Van der Wurff and colleagues (1989) constructed a model based on the assumption
that fear of crime is associated with four social psychological components:
attractivity, evil intent, power and criminalisable Space. The first component
(attractivity) refers to the extent to which people see themselves or their possessions
as an attractive target for criminal activities. Evil intent component concerns the
wrongdoer’s role in the phenomenon and it is represented by the extent to which
a person attributes criminal intentions to another individual or particular group.
Power component refers to the degree of self-assurance and feeling of control
that an individual has relating to possible threat by another. The last component
(criminalisable space) concerns the situation in which a crime may take place. In
conclusion the emphasis is on characteristics of place, time and the presence of
others. Testing this model with a sample of 440 individuals, van der Wurff et al.,
(1989) found that the social psychological model explained 24% of the variance of
fear of crime.
Contextual determinants of fear of crime
Since the hotspots of crime and hot spots of fear of crime do not always overlap
(Nelson et al., 2001) it is important to understand what are the environmental
features that may increase fear of crime. In fact, fear of crime (as the crime itself)
has an unequal spatial and temporal distribution (Fisher & Nasar, 1995; Nasar &
Fisher, 1993). Certain countries, cities, neighborhoods and specific places appear
to have higher levels of crime and fear of crime. Hotspots of fear of crime are
places where individuals feel fear of being victimized but where crimes may not
be so frequently (Fisher & Nasar, 1995). So, what are the environmental cues that
have an impact on fear of crime?
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 61
Incivilities or disorders
Several authors point out incivilities − social and physical conditions that are viewed as
troublesome and potentially threatening (Taylor, 1999, 65) as having a negative impact
on fear of crime (Hunter, 1978; Wilson & Kelling, 1982; Doran & Burgess, 2012;
Guedes et al., 2013). Incivilities may be physical – graffiti, litter, abandoned houses
and cars, vandalism – and social – prostitution, drug trafficking, insults, gangs and
people arguing on the streets (Sampson, 2009; Perkins & Taylor, 1996; Doran &
Lees, 2005). From a social perspective, incivilities denote possible victimization and,
on the other hand, physical incivilities convey messages about social conditions
rising fear of crime (Nasar & Fisher, 1993). Hunter (1978) argued that since fear
of crime in urban areas is higher than victimization itself, researchers should
look to the situational factors that generate it. Concretely, urban environments
with signs of disorder such as destroyed buildings, litter on the streets and others
aspects increase the perception about a loss of a civil society. Also, due to the fact
that these symbols are more experienced than crime itself, they have a higher
ability to increase fear and insecurity of individuals: ‘these physical incivilities, like
their inter-personal counterparts, are more frequently experienced more ubiquitous in
daily routines than crime, and therefore are more experientially significant in generating
fear and insecurity among urban residents’ (Hunter, 1978, 7). Wilson and Kelling
(1982) in their Broken Windows theory adopt a temporal and sequential perspective
regarding the relationship between disorder and crime in a community. They
argue that if a window in a building is broken and not repaired in a short period
of time, residents will infer that informal social control is low and that the other
residents do not care about what is going on in their neighborhood. Consequently,
the perception that no one cares will active a set of responses that result in the
attraction of serious predatory crime (Barker & Crawford, 2006). At the same time,
the levels of fear of crime though the residents will increase (Doran & Lees, 2005).
In turn, Sampson (2009) argues that it is the perception of disorder and not the
disorder itself that has the main relevance. Moreover, the disorder depends on
the social context where it occurs. Adbullah et al. (2015) conducted a research in
Malaysia to assess the indirect relationship between perceived disorder and fear of
crime through collective efficacy. Firstly, they found that high perception of disorder
was negatively associated with collective efficacy. Then, that low fear of crime was
associated with high collective efficacy. When analyzing the relationship between
fear of crime and disorder, the authors provide empirical support for the indirect
effect of disorder on fear of crime through collective efficacy (ibidem). Another
important contribution was the study of Hinkle (2015) which used a specific
operationalization of fear as an emotional dimension different from the perceived
safety and risk of victimization. Concretely, the author explored the implications
of using different proxies of fear of crime for testing the broken windows theory.
Hinkle (2015) found that the levels of perceived social and physical disorder
maklu

62 Maklu
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
increase fear and perceptions of safety and risk. However, interestingly, when
taking into account the ‘emotional fear’ the results suggested that while perceived
social disorder increased fear, perceptions of physical disorder had no impact. The
author suggests that the perceptions might be attached to the view that there are
dangerous people in the area rather than the physical decay.
Lack of prospect, blocked viewpoints and blocked escape
Other features of environment that have an impact on fear of crime are the lack
of prospect, blocked viewpoints where potential offenders can hide and areas with
blocked escape (Nasar & Jones, 1997). The term ‘prospect’ concerns the ability of
individuals to see the openness of its immediate environment (having an ‘open
vision’) both looking at it and walking through it (Fisher & Nasar, 1995). The lack
of perspective (blocked prospect or lack of prospect) is associated with Goffman’s
(1971 cit. Warr, 1990) concept of ‘lurk lines’ which are those beyond the vision
line of individuals. Many studies concluded that lack of prospect increases the
levels of fear of crime (Schroeder & Anderson, 1984; Petherick, 2000; Herzog &
Chernick, 2000). Also the difficulty that individuals anticipate in escaping when
confronted with a potential offender triggers higher levels of fear of crime (Nasar
& Jones, 1997). From an offender perspective, hiding places such as dense shrubs
and corners are desirables since they allow surprising the victims (Nasar & Fisher,
1993). According to a typology adapted by Fisher and Nasar (1993) individuals feel
more insecure when in the same area there are many hiding places for potential
offenders and blocked prospect (see also Herzog & Chernick, 2000; Petherick,
2000).
Lack of lighting and novelty
Guedes at al. (2013) conducted an exploratory study (published in the previous
GERN book) in order to understand the impact of different environmental
conditions to feelings of unsafety. These conditions were day/night and
rehabilitated/non-rehabilitated neighborhoods. In that study a set of pictures
of urban spaces representing those conditions was presented to a sample of
individuals. Reactions such as positive and negative emotions, unsafety feelings,
arousal, valence and others were measured through a survey. It was consistently
observed that in the night condition subjects reported that it would be more
difficult to see other people, being seen by other people, anticipate behaviors of
other individuals, manage unexpected situations, being helped in case of need
or threat and find a place of refuge in case of need or threat (ibidem). Individuals
also reported higher levels of insecurity in night condition comparing to daylight.
When inquired about the importance of environmental features to the level of
unsafety in the night condition, individuals included the lighting (99%), presence
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 63
or absence of people (89%) and corners or alleys (86%). These findings supported
the idea that the level of lighting is an important environmental feature to feelings
of insecurity (Guedes et al., 2013).
Fotios et al. (2014) in their review gave important contributions to the
understanding of the relation between road lighting and pedestrian reassurance
after dark. Reassurance was defined as the confidence when using a road. Higher
reassurance implies higher perceived safety and lower fear of crime. The analysis
of several studies using pictures, evaluations of real locations and evaluations from
memories showed that lightning enhances reassurance. Moreover, the review
concluded that higher illuminance − which implies higher brightness − increases
reassurance.
In fact, lack of lighting or poor lighting is one of the greatest cues associated
with increased fear (Doran & Lees, 2005; Nasar & Jones, 1997; Machado, 2004).
Painter (1994) argues that darkness increases feelings of vulnerability due to
the reduction of area vision field. Streets which are dimly lit create a number of
places where visibility is unbearable and, moreover, rises the difficulty of escape
when facing a potential offender (Guedes et al., 2013). Hanyu (1997) studied the
relationship between a set of affective responses of environmental experience and
visual properties. In its study, twenty-eight students rated stimuli of night-time
scenes. The author found a positive relation between security and good visibility.
Moreover, physical/behavioral activity was related positively to assessment of
safety. Subjects adopt more physical and behavioral activities in well-lit areas since
surveillance by other people reduce fear at night (Hanyu, 1997). Accordingly, a
low level of reassurance can lead to constrained behavior such as avoid walking
(Fotios et al., 2015). In the same line, Bishop and Rohrmann (2003) found that
the perception of insecurity is higher for night scenarios. Their goal was to assess
the validity of computer-generated environmental simulations. They constructed a
computer model of a real urban park with a set of details such as trees, buildings,
vehicles and sounds. Four groups of university students saw the animations and
were taken for a walk in the real environment both during the day and at night.
Although the results showed that the computer simulations did not generate the
same answers comparing to the real environment, it was observed that differences
between day and night conditions were mostly the same in both environments.
Also, Painter (1996) studied the effect of increased lighting in perceived safety in
three streets. The author detected that fear of physical attack decreased when the
lighting was enhanced. Moreover, they found that women had a higher likelihood
of noticing an increase in lightning comparing to men (idem). Similarly, Boomsa
and Steg (2014) found that women perceived a potential situation less safe
compared to men. Their study was aimed at analyzing the effects of physical
and individual factors on acceptability of reduced lighting levels and safety. The
maklu

64 Maklu
authors concluded that the acceptability of reduced street lighting is associated to
perceived safety which depends on entrapment, lighting and gender.
It is important to understand how the amount of light effects the perception of
safety at night. In order to address that problem, Boyce et al. (2000) carried out
a study using field surveys of 24 car parks in urban and suburban areas. They
observed that as illuminance increased, the difference between ratings of perceived
safety at day and night-times tended to decrease.
Since fear of crime is higher after darkness, researchers argue that if ones reduce
the darkness, fear of crime will be also reduced (Atkins et al., 1991).
Lastly, the novelty or mystery is an important feature of the environment that can
be associated with fear of crime. There is a known tendency for individuals to
perceive their own houses and neighborhoods more safety than other areas of
their cities (Skogan, 1999). On the contrary, unfamiliar places may increase the
levels of fear of crime due to lack of previous experience and the appearance of
new danger signs (Warr, 1990).
Applying eye tracking techniques to the study of fear of crime
Fear of crime has been scarcely studied using eye tracking techniques. One known
exception is the study of Davoudian and Raynham (2012) which intended to identify
and characterize the main visual tasks of pedestrians at night and the key aspects
of the visual environment that were important to pedestrians. In order to achieve
these goals, the authors was asked members of the general public to walk along
three residential routes chosen for the study during day and night-times. Route
one was a residential street with two antisocial behavior incidents, an illuminance
level S33 and an obstacle level in pavements classified as moderate. Route two was
a residential street with an illuminance level S4, three crime incidents recorded
in the past year and an obstacle level classified as high. Lastly, route three was a
collector road of residential streets where all the residents had to pass along to get
to their respective streets. The illuminance level was S1 and the route had one anti-
social behavior and three crime incidents in the past year. The obstacle level for the
route three was considered low. In total, fifteen subjects participated in the night
time experiments and five subjects took part in the day time. Individuals were
required to walk along the three routes with a head mounted eye tracking device.
Results showed that individuals looked at the pavement between 40% and 50%
of the time. Subjects looked more to the pavement on route two since there were
3 The levels of illuminance in UK is based on BS 5489-1 (see Davoudian & Raynham, 2012 for more
details). ‘S’ is the class of light that covers the lighting of residential areas and ranges between S1
and S6. S1 is the brightest level.
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 65
more obstacles in it. Also on the route two individuals looked less to the footpath
at night comparing to day time. The authors advanced with the explanation that
subjects reported feeling insecure on route two at night. Hence, this would lead
participants spending longer times making sure of their environment and looking
less to the pavement. As for the route three, individuals tended to spend longer
looking at the footpath. These result were explained by the fact that subjects were
accompanied on this route and there being lack of hiding place for potential
offenders. Accordingly, no participants reported feeling insecure on this route.
Hence, the argument would be that since people did not feel insecure in this route,
they did not have to pay much attention to threats in the environment. Despite
these important results, Davoudian and Raynham (2012) study shows the problem
already pointed out by Duchowski (2012 cit. Davoudian & Raynham, 2012) of the
dissociation of visual attention from the point of regard. In one hand, it is possible
to fix one’s gaze at a specific point and the attention be focused on a nearby region.
On the other hand, there is the possibility that individuals fixate in an object while
not paying attention to any part of their visual field including the object. Concerning
this problem, Fotios et al. (2015, 135) also assert that ‘fixations do not necessarily
indicate attention is being directed towards the fixated objet, or that the fixated objet is
important to the task’. To overcome this issue, the authors used a complementary
cognitive task (reaction time to an auditory stimulus) running concurrently with
the task of walking in their study. Their study had the main goal of identifying
pedestrians’ critical visual tasks when walking after dark. Concretely, participants
walked a short route whilst wearing an eye-tracking equipment. Whenever an
auditory stimulus appeared, they would press a button. Each section of the route
was composed by different characteristics in which participants walked twice (day
and night-times). Researchers found that after dark individuals fixate less in other
people and more in the path comparing to daylight. The main reason suggested
for these results was participant’s need to detect obstacles during night time while
walking (Fotios et al., 2015).
Kim and colleagues (2014) analyzed people’s fear responses to different outdoor
night-time environments through eye tracking techniques. Specifically, the
author’s goal was to study eye movements in response to actual environments
and simulated photographs of nightscapes (image-based environments). Two
hypotheses were the base of this study. Firstly, they expected that people might feel
different levels of fear depending on the arrangement of trees, lighting, shrubs,
and buildings in their physical environment and secondly, perceived fear would
be slightly different in actual environments than in image-based environments.
To operationalize these hypotheses the authors constructed a study where 23
university students were tracked while viewing six night-time environments. They
were then required to rate the level of fear experienced in each environment. After
this procedure, the same sample rated the fear of six photographs of the same six
nightscapes. Kim et al. (2014) found that there were differences in levels of perceived
maklu

66 Maklu
fear between the actual environments and the image-based environments. Overall,
participants rated the image-based environment as less frightening than the actual
environments. However, the differences were only statistically significant for
three of the six environments. Regarding eye movements results, it was found
that the most fixated environment was different between actual and image-based
conditions. However, using heat maps generated from eye-tracker data, authors
found that areas of focus were very similar between the actual environment and
image-based environment (idem). Interestingly, in all the environments presented
to the sample, the areas that individuals tend to focus were the center and the end
of the path. So that, the study suggests that eye-tracking research using images can
be as effective as bringing subjects to real environments.
In sum, the studies above contributed to the study of walking tasks and unsafety
questions after dark. However, more studies are necessary in order to understand
the features in (un)safe environments that most capture the attention of
individuals. In the eye tracking literature, much is known about selective attention
to emotional pictures. For instance, Wadlinger and Isaacowitz (2008) attempted
to analyze the effects of valence (or pleasantness) in selective attention. Valence is
defined as a dimension of affective experience which can be classified as positive
or negative (Charland, 2005). They found that subjects presented a heightened
attentional breadth in emotional positive images comparing negative images. Also
Nummenmaa et al. (2006) found that individuals first looked at emotional pictures
and fixated more on emotional than neutral pictures. Moreover, individuals
looked more to emotional pictures with negative and positive content comparing
to neutral pictures. Quigley et al. (2012) compared the effects of trait and state
anxiety on selective visual attention to emotional pictures. The authors found
that state anxiety (but not the trait anxiety) was related to increased attention to
threatening images. Also, they concluded that individuals first looked at emotional
pictures comparing to neutral ones. Wieser et al. (2008) tested the hypervigilance-
avoidance hypothesis through eye tracking techniques. The hypothesis defends
that anxious individuals initially attend to threatening stimuli and then avoid them.
They conclude that individuals with higher levels of ‘fear of negative evaluation’,
a feature related to social anxiety, looked first to emotional faces comparing to
neutral ones more often. However, 1 to 1.5 seconds later they tended to avoid those
faces and looked more to neutral ones.
Goals and hypothesis of the present study
The present study aims at using eye tracking techniques to explore the contextual
cues associated with fear of crime. For this, we intend to explore the selective
attention to urban images and, concretely, to environmental cues contained in
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 67
these. Since this is an exploratory study, the hypotheses in which this research is
based are general.
Hypothesis 1: Pictures previously rated with higher levels of fear of crime will
capture more the attention of individuals.
Hypothesis 2: Individuals will look more to the physical features that have been
associated with fear of crime.
Methods
Participants
The results of eye tracker study are from 10 voluntary subjects (5 females). All the
participants are students of the Faculty of Law of University of Porto.
Stimuli
Database of urban space images
Stimuli of this experience were obtained through a number of steps. Firstly, a data
of urban images was constructed. Pictures were obtained by researchers following
a set of criteria (see Guedes et al., 2013 for details). To complement, a number of
images of city of Porto were chosen from Google Maps®. Researchers gathered
more than 200 pictures that represented urban spaces in different conditions:
daylight and night, spaces with incivilities (such as graffiti, broken windows,
and litter) or corners and rehabilitated places. Secondly, after a set of reunions
between researchers, 50 pictures to be rated by a sample were selected. In order to
do so, two online questionnaires were constructed using Survey Monkey®, each of
them constituted by 25 pictures. The online questionnaire was constituted by four
questions that assessed valence, arousal, safety, and victimization risk perception.
The valence is related to the pleasantness of a stimulus, while arousal dimension
concerns the intensity of emotion provoked by a stimulus (Warriner, Kuperman &
Brysbaert, 2013). Valence and arousal varied between one and nine. Regarding the
safety dimension, the question was ‘how safe would you feel in a place represented
by the picture’ and ranged between one (very unsafety) to five (very safety). For the
victimization risk perception, participants were asked to measure the likelihood
of being victimized in a place represented by picture. The questionnaires were
administered to a large sample of 1300 students from different universities of the
North of Portugal. Hence, the validity of ratings were guaranteed.
maklu

68 Maklu
Pictures analysis and selection to eye tracking experience
After the data collection, statistical analysis was done through IBM SPSS®, in
order to obtain mean and median values of arousal, valence, and safety. The mean
values of safety, arousal and valence of the selected pictures are presented in Table
1. Three pictures with high levels of safety were selected (pictures 1, 2 and 5) and
three pictures with low levels of safety were chosen (pictures 3, 4 and 6). In sum,
from the total of 50 pictures, six were chosen for the present study.
Table 1 Mean (
X
) and Standard Deviation (SD) of perceived safety, arousal and
valence in the different pictures selected for the present study.
Valence Arousal Safety
SD SD SD
Picture 1 7.41 1.46 3.85 2.33 4.09 .83
Picture 2 7.78 1.32 5.33 2.29 3.60 .76
Picture 3 2.17 1.25 4.98 2.27 1.67 .71
Picture 4 2.31 1.46 5.46 .73 1.68 .73
Picture 5 5.06 1.70 3.31 1.66 3.62 .78
Picture 6 3.01 1.55 4.73 2.07 1.98 .75
Since the surveys were administered online, and the participants were not
interviewed, the description of the selected pictures was made by the researchers
of this project. Picture 1 depicts a clean avenue in daytime with careful vegetation
between roads, good visibility and high prospect. The values of security and
valence were rated very high. Picture 2 represents a carefully maintained garden
in the center of the city with open spaces to walk. The garden has many trees and
bushes disabling a very good perception of the surrounding space. However, it was
rated as moderately safety. In turn, picture 3 describes an entrance to a dimly light
tunnel. The stairs and the walls are aged and vandalized. Since it is not possible to
observe the end of the tunnel, or where it leads, the space of this picture has a low
prospect. Picture 4 represents an abandoned place with untidy vegetation, many
places to hide and graffiti on the walls. Moreover, the place has a lack of luminosity
and visibility. The level of security in this place is 1.68 which is very low, as well as
the valence. On the other hand, picture 5 represents a very large street with high
buildings in a daylight context. It is possible to see a few people walking by and
two graffiti on the wall. This place has a high prospect and visibility. The security
mean is moderately high. The last picture (6) depicts a street in night context with
shadows, corners and a building under construction. These features do not allow
having a whole perspective of the street, so the prospect is low. Furthermore, the
sense of security in this picture was rated as very low.
X
X
X
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 69
Picture 1 Picture 2
Picture 3 Picture 4
Picture 5 Picture 6
Image 1. Pictures selected to the eye tracking experience
Apparatus
Stimuli were presented in a 19-in (48.3 cm) HP monitor (1440 by 900 pixels of
resolution) with a Dell Intel® Core ™ i5 computer. Participant’s eye movements
were recorded with a Mirametrix Tracker S2. The sample rate of the eye tracker
was 60 Hz. The presentation of stimuli was programmed in PsychoPy v1.80, an
open-source package for running experiments. To draw and select the areas of
interest (AoI) of pictures Paint.net was used.
maklu

70 Maklu
Stimulus displays and procedure
Each stimulus display consisted of an individual picture representing an urban
context with high or low levels of fear of crime. The pictures were presented in
the center of the screen. After each picture the question ‘Is this place insecure?’
appeared at the center of the screen. The main goal of this question was to direct
the attention of individuals to the cues related to insecurity. Following this,
individuals were required to press a button to answer ‘yes’ or ‘no’ according their
evaluation. The size of each individual picture was 1720 x 720 pixels. In order to
control order effects, pictures were randomized. Before each trial a fixation point
(a black plus sign) appeared on the center of the screen during 3000ms. After the
fixation point, each individual picture was projected during 6000ms. The duration
of individual pictures was initially 4000ms but was modified after pretests with
the aim of subjects looked more attentively to the features of images. Volunteers
were tested individually. Upon arriving to the laboratory, they were forwarded to
the data collection room. Firstly, it was explained the aims of the study project,
and they were invited to read the consent form and sign it if they agreed with the
experience terms. After that, they were seated in a comfortable chair located 85 cm
from the screen (distance from face to the screen). A computer mouse was handed
to them for responding to the questions. After these instructions the eye tracker
was calibrated. The calibration was accepted if the estimated error was lesser than
40 pixels. Then, the researcher would start the experimental block. Participant
were required to complete a questionnaire for personal characterization. After
filling the questionnaire, participants sealed it and put it in a box to guarantee
anonymity and confidentiality.
Measures
The eye-tracking metrics taken in account were: a) number of total fixations in
each picture; b) total time visualization in each picture (ms), and c) reaction time
(ms) from the moment when picture appears to the first ocular movement.
Results
Comparison between level of safety from the external sample and the evaluati-
on of safety from the eye tracking sample
The first analysis is related to the question presented after each image appeared
in the screen. This question had the main goal of creating a mindset for the
individual while they were observing the images. So that, instead of looking freely
to the images, individuals would look at them with a mindset on the safety of each
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 71
picture. The proportion of ‘yes’ and ‘no’ answers given by the 10 individuals that
compose the sample are shown in the table 2. Additionally, in the same table, are
the results for mean safety given by the sample that answered the online surveys
(already presented in the table 1). Although we have in mind that they are not fully
comparable (due to the differences on the way they were measured) we believe that
it is interesting to compare the ratings.
Table 2. Proportion of negative and positive answers to the question
‘Is this place insecure?’ for each picture presented to the sample of 10 individuals.
In addition, the table presents mean values of safety evaluated by an external sample,
for comparison.
‘Is this place insecure?’ Safety (1-5)
Yes (%) No (%) mean
Picture 1 0 100 4.09
Picture 2 20 80 3.60
Picture 3 100 0 1.67
Picture 4 70 30 1.68
Picture 5 10 90 3.62
Picture 6 80 20 1.98
Picture 3 was the picture with the highest proportion of ‘yes’ answers to the
question ‘is this place insecure?’. Concretely, the total sample perceived the place
as insecure. Regarding the level of security reported by an external sample it is
possible to observe that picture 3 was rated as the most unsafe place. Most of
the respondents (80%) agreed that picture 6 represented an insecure place. On
the contrary, individuals tended to answer that picture 2 is secure. While picture
6 has a low mean safety rated by other external sample, picture 2 was rated as
moderately safety. All of the respondents agreed that picture 1 was not an insecure
place. Accordingly, the level of safety in this picture was the highest comparing to
the others (4.09). For most of the individuals in the sample, picture 5 represents
a secure place. The level of safety (3.62) rated by an external sample was also
moderately high. Lastly, concerning picture 4 it is possible to analyze that 70% of
the sample answer that the place was insecure. Also, the level of safety rated by the
external sample is low (1.68).
Pictures heatmaps (eye tracking)
Images 2, 3, 4, 5, 6 and 7 represent the heatmaps, that is, the specific stimuli of
pictures that had higher gaze duration. Regarding the image 2 it is possible to
maklu

72 Maklu
observe that individuals looked longer time to the bottom of the street (red color).
They also tended to look at the vegetation (yellow and green colors).
Image 2. Heatmap of picture 1
Concerning the image 3, the stimulus where individuals focused more time was
the center of the garden (where the main crosswalk ends). As it is possible to
observe, the rest of the image was not fixated by individuals.
Image 3. Heatmap of the picture 2
Image 4 shows that individuals focus more time on the black tunnel (red color).
However, other elements of the place were looked at by the sample, namely the
graffiti on the walls, the graffiti on the top of the tunnel, and the stairs.
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 73
Image 4. Heatmap of picture 3
In image 5 there were several elements that captured the attention of individuals.
As it is observable, the participants looked more time to the top of the stairs, where
there is a graffiti painted on the wall (red patch). Other features such as graffiti in
the left column of the image and vegetation were looked at by the participants,
although not for a long time (green patch).
Image 5. Heatmap of picture 4
In image 6, individuals paid more attention to the bottom of the road and the
crosswalk. Also, they also seemed to look at the person walking in the image.
maklu

74 Maklu
Less time was centered in the buildings, although both of them were looked at by
participants.
Image 6. Heatmap of picture 5
In the last picture (image 7) there were a set of features that individuals paid
attention to. Firstly, they looked longer at the building under construction on the
right side of image (red patch). Secondly, they also observed the window bars of the
house represented in this picture. Moreover, participants focused their attention to
the shadows and the corner of the left size of image.
Image 7. Heatmap of picture 6
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 75
Quantitative measures: latency, total fixations of each image and total durati-
ons of fixations in each image
Concerning latency, number of fixations and total durations of fixations in each
picture, it was possible to observe (table 3) that there were no significant statistical
differences between pictures (p >.05 for all comparisons).
Table 3. Results (mean and standard deviation) of latency, total fixations
and duration of fixations in each pictures. Non-parametric tests to analyze the
differences between variables were done. The p-value is the result of Friedman test.
Latency (ms)
X±SD
Total fixations
X±SD
Total duration of
fixations (ms)
X±SD
Picture 1 672.1± 833.4 6.7 ± 3.4 4143.5 ± 1515.4
Picture 2 995.6± 616.4 6.5 ± 3.8 4256.7 ± 660.7
Picture 3 823,1 ± 472.4 7.2 ± 3.6 4500.2 ± 669.2
Picture 4 423.9 ± 212.4 7.3 ± 1.9 4903.6 ± 495.2
Picture 5 995.0 ±814.2 6.9 ± 2.8 4474.2 ± 619.9
Picture 6 1401.2 ±1521.6 5.6 ± 2.9 4183.9 ± 1483.9
p-value .198 .273 .646
Discussion
This exploratory study had the main goal of analyzing selective attention to
different contextual features of images depicting urban spaces. In this paper we
intended to analyze which were the features of urban spaces previously rated for
insecurity, arousal and valence levels that captured more attention of the sample.
To address these goals, an experience using eye tracker was planned. A set of trials
were presented to a sample of 10 individuals and a few measures were collected
(latency, number of fixations and total number of fixations).
Firstly, we compared the answers to insecurity assessment in the present sample
(N=10) with the mean levels of safety previous rated by an external sample
(N=1300). Although this was not a main goal of the present study, we thought it
would be interesting to compare the ratings of safety. We are aware of the problem
of validity and reliability when making this comparison: on one hand, the sample
of the eye tracking study was very small (n=10) comparing to the size of the external
sample (more than 1000 individuals). On the other hand, to assess the safety of
pictures in order to choose them for the eye tracking study we used a scale of 5
maklu

76 Maklu
points. In turn, we did not ask the participants of the eye tracking study to evaluate
the pictures in the same scale. Rather, we just asked them to evaluate if the place
represented by the pictures were or not were insecure. The goal of this question
was to create a mindset in participants so they thought about the insecurity while
looking at pictures. Supposedly, they would look at pictures analyzing the cues
or features that were more important to assess (in)security. Taking into account
the referred precautions, it was possible to observe an agreement in almost every
pictures between the two samples.
The second set of results is related to the attention paid to the contextual features of
the urban pictures. Usually, studies that explore the contextual cues associated with
safety employ methods such as quantitative surveys. Our study has the advantage
of using eye tracking methodology to explore the features mostly looked at. Also,
these images contrast concerning time of the day (daylight/night), presence
or absence of disorder (such as abandoned places, graffiti, litter) and also the
prospect or visibility of images. The analysis of heat maps revealed the importance
of prospect of urban spaces, since the bottom of the streets was the feature that
captured more subject’s attention. So, when walking in an urban environment it
is suggested that participants make an evaluation of (in)security relying in how
much of that space they can prospect. One may pose the question of whether
these results would be different if individuals were in the real environments
represented by the pictures. The study of Kim et al. (2014) seems relevant for this
discussion. Although they found that reported fear of crime was slightly different
between real environments and image-based environments, the analysis of heat
maps revealed that individuals focused in the same areas when comparing the real
and the image-based contexts. Another problem of eye tracking studies is the fact
that fixations may not necessarily indicate that attention is being directed towards
the fixated objet, or that the fixated objet is important to the task. As Fotios et al.
(2015) suggests, it would be important, in next studies, to create a complementary
cognitive task in order to overcome that problem.
Fisher and Nasar (1993) had already sustained the relevance of places with
prospect (open view) since those places offer an observation point to see, react and
to defend if necessary. On the contrary, places with low prospect or with ‘lurk lines’
(Warr, 1990) constitute a part of the space in which individuals cannot see. These
environments trigger higher levels of insecurity (Fisher & Nasar, 1992; Petherick,
2000). In our study, images with low prospect or less visibility were the ones rated
with lower security. Other elements of places most looked at by individuals were
the graffiti on the walls. This was a consistent result in every image that depicted
this physical disorder. According to the literature, physical disorders convey
messages about social conditions, can lead to serious crime (Wilson & Kelling,
1982) and trigger fear of crime (Hunter, 1978). Window bars and buildings under
construction were also environmental features looked by the sample. The window
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 77
bars could transmit to individuals that the space is not secure and needs extra
precaution. It would be interesting to compare the same place at daylight in order
to understand if the features most looked at. This leads to a limitation of this work:
the low amount of trails. In fact, since this is an ongoing and exploratory study,
in the next stage we intend to increase the amount of trials adding more images
of urban contexts. For instance, it would be important to have the same places
differing in daylight and night time and also to compare places with and without
the presence of people.
Lastly, the quantitative measures were analyzed in order to understand if there
were differences between latency (reaction time from the moment when the
picture appears to the moment when individuals make the first ocular movement),
number of fixations and the total time of fixations in each pictures. Non-parametric
tests did not indicate any significant difference between pictures. In following
studies, as Kim et al. (2014) suggest, it would be important to measure pupil size
since it may be a valid measure of fear of crime.
At the same time this study is being developed, we are also applying qualitative
methods in order to explore the meanings attached to places that people perceive
as secure and insecure. Specifically, semi-structured interviews are being made to
a sample of individuals that differ in sex and age. So far, both men and women of
all ages agreed that ‘lack of luminosity’ is the most fearful feature. When we try to
understand what are the meanings attributed to the darkness, individuals explain
that this feature diminishes visibility, people movement and, in consequence,
increase the opportunity for victimization. Other environmental cue that people
associate very frequently with fear is ‘narrow streets’. These make people feel
trapped in a situation of attack. Participants also referred ‘abandoned places’ as
unsafe, associating those environments as being frequented by possible offenders.
Lastly, deprived neighborhoods of the outskirts of the city are very often mentioned
by our sample. Although these are preliminary results, they seem to be very
important and a valuable complement of eye tracking techniques.
It is important to remind that this eye tracking study was exploratory. Currently, we
have constructed a new experimental protocol which is improved in many ways.
Firstly, the number of trials was increased. Instead of only 6 pictures, 30 pictures
were selected to this new experience. Concretely, 10 were positive, 10 were neutral
and 10 were negative in terms of fear of crime assessment. We also added a survey
with individual measures such as sex, age, victimization, fear of crime (emotional
component), victimization risk perception, behaviors for security reasons,
personality dimensions, self-control and trait emotions. Hence, it will be possible
to analyze the influence of individual differences in the selective attention to
pictures presented. Moreover, all the participants will evaluate the 30 images after
the eye tracking experience for valence, arousal, fear of crime and risk perception
maklu

78 Maklu
of victimization. This procedure will allow to relate in a more direct manner the
relationship between fear of crime and the cues observed by participants.
In conclusion, this exploratory study intended to use eye tracking techniques
to contribute to the relationship between fear of crime and environmental cues
associated with it. In spite of the limitations of this investigation, which will
be overcome through a new study, we believe that the heat map results were
important to explore the cues that individuals supposedly paid more attention
during the observation of images. In order to have a more complete understanding
of this phenomenon, qualitative work is being complemented with eye tracking
techniques.
Acknowledgements
The first author of this paper is recipient of a PhD Grant (SFRH/BD/76921/2011)
of Portuguese Science Foundation (Fundação para a Ciência e a Tecnologia – FCT).
References
Abdullah, A., Marzbali, M., Bahauddin, A. & Tilaki, M. (2015). Broken Windows
and Collective Efficacy: do they affect fear of crime? Sage Open, January-March,
1-11.
Atkins, S., Husain, S. & Storey, A. (1991). The influence of street lighting on crime and
fear of crime. Crime Prevention Unit Paper 28. London: Home Office.
Barker, A. & Crawford, A. (2009). Fear of crime and insecury in Europe. Workpackage
4 Report, CRIMEPREV, Paris.
Bishop, I. & Rohrmann, B. (2003). Subjective responses to simulated and real
environments. Landscape and Urban Planning, 65(4), 261-277.
Boomsma, C. & Steg, L. (2014). Feeling safe in the dark: examining the effect of
entrapment, lighting levels and gender on feelings of safety and lighting policy
acceptability. Environment and Behavior, 46(2), 193-212.
Boyce, P.R., Eklund, N.H, Hamilton, B.J. & Bruno, L.D. (2000). Perceptions of
safety at night in different lighting conditions. Lighting Research and Technology,
32(2), 79-91.
Chadee, D. & Ditton, J. (2003). Are older people most afraid of crime? Revisiting
Ferraro and LaGrange in Trinidad. British Journal of Criminology, 43(2), 417-433.
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 79
Charland, L. (2005). Indeterminacy of valence. In L. Feldman-Barret, P. Niedenthal
& P. Winkielman (Eds.), Emotion and Consciousness. New York: Guilford Press.
Davoudian, M. & Raynham, P. (2012). What do pedestrians look at night? Lightning
Research and Technology, 44(4), 438-448.
Doran, B. & Burgess, M. (2012). Putting fear on the map: Investigating perceptions of
crime using geographic information systems. New York: Springer.
Doran, B. & Lees, B. (2005). Investigating the spatiotemporal links between
disorder, crime and the fear of crime. The Professional Geographer, 57(1), 1-12.
Farrall, S., Jackson, J. & Gray, E. (2009). Social order and the fear of crime in
contemporary times. Oxford New York: Oxford University Press.
Ferraro, K. & LaGrange, R. (1987). The measurement of fear of crime. Sociological
Inquiry, 57(1), 70-97.
Fisher, B. & Nasar, J. (1995). Fear spots in relation to microlevel physical cues –
exploring the overlooked. Journal of Research in Crime and Delinquency, 32(2),
214-239.
Fotios, S. (2014). Road lighting and pedestrian reassurance after dark. A review of
the evidence. Proceedings of CIE 2014 Lighting Quality & Energy Efficiency. CIE
2014 Lighting Quality & Energy Efficiency, 23-26 April 2014, Kuala Lumpur,
Malaysia, 843-852.
Fotios, S., Uttley, J., Cheal, C. & Hara, N. (2015). Using eye-tracking to identify
pedestrians’ critical visual tasks, Part 1. Dual task approach. Lighting Research
and Technology, 47, 133-148.
Gabriel, U. & Greve, W. (2003). The psychology of fear of crime: conceptual and
methodological perspectives. British Journal of Criminology, 43, 600-614.
Gomme, I.M. (1988). The role of experience in the production of fear of crime: a
test of a causal model. Canadian Journal of Criminology, 30, 67-76.
Guedes, I. (2012). Sentimento de insegurança, personalidade e emoções disposicionais:
que relações? Tese de Mestrado. Faculdade de Direito da Universidade do Porto.
Guedes, I., Cardoso, C. & Agra, d. C. (2013). Emotional and insecurity reactions
to different urban contexts. In P. Ponsaers, A. Crawford, J. Maillard, J.
Shapland, A. Verhage (Eds.), Crime, violence, justice and social order: monitoring
contemporary security issues. GERN Research Papers. Antwerpen, Apeldoorn &
Portland: Maklu.
Hale, C. (1996). Fear of crime: a review of the literature. International Review of
Victimology, 4, 79-150.
maklu

80 Maklu
Hanyu, K. (1997). Visual properties and affective appraisals in residential areas
after dark. Journal of Environmental Psychology, 17, 301-315.
Herzog, T. & Chernick, K. (2000). Tranquility and danger in urban and natural
settings. Journal of Environmental Psychology, 20, 29-39.
Hinkle, J. (2015). Emotional fear of crime vs. perceived safety and risk: implications
for measuring ‘fear’ and testing the broken windows thesis. American Journal
of Criminal Justice, 40(1), 147-168.
Hoffman, J. & Subramaniam, B. (1995). The role of visual attention in saccadic eye
movements. Perception & Psychophysics, 57(6), 787-795.
Hunter, A. (1978). Symbols of incivility: social disorder and fear of crime in urban
neighborhoods. Paper presented to the Annual Meeting of the American
Criminological Society, Dallas.
Kim, M., Kang, Y., Hong, S. & Abel, T. (2014). Difference of perceived fear between
actual environment and image-based environment using eye tracking method,
331-340. Wissen Hayek, U., Friker, P. & Buhmann, E. (Eds.): Peer Reviewed
Proceedings of Digital Landscape Architecture 2014 at ETH Zurich. Herbert
Wichmann Verlag, VDE VERGLAG GMBH, Berlin/Offenbach.
Klama, E. & Egan, V. (2011). Personality and mental health influences on fear of
crime and attitudes to punishment. Personality and Individual Differences, 51,
613-617.
Liska, A.E., Sanchirico, A. & Reed, M.D. (1988). Fear of Crime and Constrained
Behavior: Specifying and Estimating a Reciprocal Effects Model. Social Forces,
66, 827-837.
Machado, C. (2004). Crime e insegurança. Discursos do medo, imagens do outro.
Lisboa: Editorial Notícias.
Machado, C. & Agra, C. d. (2002). Insegurança e medo do crime: da ruptura da
sociabilidade à reprodução da ordem social. Revista Portuguesa da Ciência
Criminal, 12, 79-101.
Madriz, E. (1997). Nothing bad happens to good girls: fear of crime in women’s lives.
Berkeley: University of California Press.
Mesch, G.S. (2000). Perceptions of risk, lifestyle activities, and fear of crime.
Deviant Behavior, 21(1), 47-62.
Nasar, J. (1988). The evaluative image of the city. Thousand Oaks: SAGE Publications.
Nasar, J. & Jones, K. (1997). Landscapes of fear and stress. Environment and
Behavior, 29(3), 291-323.
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Studying the contextual cues associated with fear of crime through eye tracking techniques
Maklu 81
Nasar, J. & Fisher, B. (1993). Hot spots of fear of crime: A multiple-method
investigation. Journal of Environmental Psychology, 13, 187-206.
Nelson, A., Bromley, R. & Thomas, C. (2001). Identifying micro-spatial and
temporal patterns of violent crime and disorder in the British city centre.
Applied Geography, 21, 249-279.
Nummenmaa, L., Hyona, J. & Calvo, G. (2006). Eye movement assessment of
selective attentional capture by emotional pictures. Emotion, 6, 257-268.
Painter, K. (1994). The impact of street lighting on crime, fear and pedestrian use.
Security Journal, 5, 116-124.
Painter, K. (1996). Street lighting, crime and fear of crime: a summary of research.
In T. Bennett (Ed.), Preventing Crime and Disorder. Cambridge, UK: Institute of
Criminology.
Perkins, D. & Taylor, R. (1996). Ecological assessments of community disorder:
their relationship to fear of crime and theoretical implications. American
Journal of Community Psychology, 24(1), 63-107.
Petherick, N. (2000). Environmental design and fear: the prospect-refuge model
and the university college of the Cariboo Campus. Western Geography, 10, 89-
112.
Quigley, L., Nelson, AL., Carriere, J., Smilek, D. & Purdon, C. (2012). The effects of
trait and state anxiety on attention to emotional images: an eye tracking study.
Cognitive Emotion, 26(8), 1390-411.
Reid, L.W. & Konrad, M. (2004). The gender gap in fear of crime: assessing the
interactive effects of gender and perceived risk on fear of crime. Sociological
Spectrum, 24, 399-425.
Sampson, R. (2009). Disparity and diversity in the contemporary city: social (dis)
order revisited. The British Journal of Sociology, 60(1), 1-31.
Schroeder, H. & Anderson, L. (1984). Perception of personal safety in urban
recreation sites. Journal of Leisure Research, 16(2), 178-194.
Skogan, W. (1999). Measuring what matters: crime, disorder, and fear. In R.
Lanworthy (Ed.), Measuring what matters: Proceedings from the Policing Research
Institute Meetings. Research Report. Washington, DC: U.S. Department of
Justice, National Institute of Justice and Office of Community Oriented
Policing Services, 37-53.
Smith, L.N. & Hill, G.D. (1991). Perceptions of crime seriousness and fear of
crime. Sociological Focus, 24(4), 315-327.
maklu

82 Maklu
Taylor. R. (1999). The incivilities thesis: theory, measurement and policy. In
Langworthy, R. (Ed.), Measuring what matters: Proceedings from the Policing
Research Institute meetings. Washington, DC: National Institute of Justice, 65-
88.
Van der Wurff, A., Staalduinen, L. & Stringer, P. (2001). Fear of crime in residential
environments: testing a social psychological model. The Journal of Social
Psychology, 129(2), 141-160.
Vandeviver, C. (2011). Fear of crime in the EU-15 and Hungray: Assessing the
impact of vulnerability, victimization and incivilities model on the fear of crime
in a European cross-national context. Master Thesis, Hogeschool-Universiteit
Brussel.
Wadlinger, H. & Isaacowitz, D. (2008). Looking happy: the experimental
manipulation of a positive visual attention bias. Emotion, 8(1), 121-126.
Warr, M. (1990). Dangerous situations: social context and fear of victimization.
Social Forces, 68(3), 891-907.
Warr, M. (2000). Fear of crime in the United States: avenues for research and
policy. Measurement and analysis of crime and justice, 4, 451-489.
Warriner, A., Kuperman, V. & Brysbaert, M. (2013). Norms of valence, arousal and
dominance for 13,915 English lemmas. Journal of Behavior Research Methods,
45(4), 1191-1207.
Wieser, M., Pauli, P., Weyers, P., Alpers, G. & Mühlberger, A. (2009). Fear of
negative evaluation and the hypervigilance-avoidance hypothesis: an eye-
tracking study. Journal of Neural Transmission, 116(6), 717-723.
Wilson, J.Q. & Kelling, G. (1982). Broken windows. Atlantic Monthly, 211, 29-38.
Ziegler, R. & Mitchell, D. (2003). Aging and fear of crime: an experimental approach
to an apparent paradox. Experimental Aging Research, 29, 173-187.
Inês Guedes, Pedro Fernandes, Cândido da Agra & Carla Cardoso
maklu

Maklu 83
Dangerous and endangered
The construction of male-homosexual sex work
as a social problem in Germany
Within the realm of sociology of social problems, male homosexual sex work in Germany
can be understood as a social problem. This article serves to investigate the rise of male
homosexual sex work as a social problem in Germany. It presents some initial results
about the problem career of male homosexual sex work on the macro-sociological level
in the time period from 1980 to 1985. It shows that male homosexual sex work was
discussed along the lines of criminalisation and victimisation. The argumentation will
focus on the presentation of two articles, which on the one hand, pose male homosexual
sex workers as dangerous people and, on the other hand, as endangered people.
Marlen S. Löffler1
In Germany, where selling sex has been legal since 2002 (Bundesministerium
für Familie, Senioren, Frauen und Jugend, 2005), seven drop-in and counselling
centres for male sex workers (DICC) have been established since 1986 (Arbeitskreis
der Strichereinrichtungen in Deutschland, 2012; Grabatsch, 2013; Gusy, Krauß,
Schrott & Heckmann, 1994). The DICCs locate themselves between ‘AIDS-
prevention and youth welfare’ (Arbeitskreis deutschsprachiger Stricherprojekte,
2003, 123) and their goal is to support male sex workers in difficult situations as
well as sensitise them with regards to the transmission of HIV and other sexually
transmitted infections (STIs). These goals confirm that male sex work is associated
with neediness in living conditions and health (Bochow, 2003; Fachkreis für
Stricherarbeit im deutschsprachigen Raum, 2007; Fink & Werner, 2005; Gusy et
al., 1994; Möbius, 1991; Pfister, 2009). The installation of these institutions proves
that the political recognising of male sex work as a topic, requiring social control
(Schetsche, 2008).
Even if there seems to be a continuous political interest in supporting male sex
workers, seeing as how the first of these institutions has existed nearly 30 years,
male homosexual sex work2 is an almost unnoticed subject in the German public,
1 Tuebingen University, Germany.
2 Male-homosexual sex work is understood here as a sexual service between men for financial or
material exchange (Gusy, Krauß, Schrott & Heckmann (1994, 1088). The term ‘male-homosexual’
means the sexual practices between men. Therefore, this term neither testifies to male sex workers’
sexual identity nor the clients’ sexual identity.
maklu

Marlen S. Löffler
84 Maklu
but also in international social science (Scott & Minichiello, 2014; Stallberg, 1990;
Weitzer, 2005). In contrast to publications about female sex work, studies about
male sex work are limited (Dennis, 2008). Most national and international research
focuses on male sex workers, their safer sex strategies, and on a differentiation of
male sex worker in categories (Bimbi, 2007; Morrison & Whitehead, 2007; Pfister,
2009).
Aside from these topics, rare studies exist focussing the conditions or other
protagonists in the sex industry, like clients (Bimbi & Koken, 2014; Crofts, 2014;
Logan, 2014; Scott, John, Callander & Minichiello, 2014; Tyler, 2014) and the
institutions of social control working with male sex workers, like the DICCs are
(Lautmann, 2002). Nevertheless, some research has been done about the history of
male sex work in societies (Friedman, 2014; Kaye, 2014) and about the perspective
on male sex workers in cinematic films (Sheaffer, 2014).
With their studies, Scott (2003) and Bimbi (2007) started to explore the macro level
of this topic by examining the scientific discourse about male homosexual sex
work. While Bimbi (2007) is focusing on the central topics of studies about male
sex workers, Scott (2003) is investigating the socio-scientific discourse about male
homosexual sex work in English-speaking publications from the 19th to the 20th
century. He discovered that male sex workers were seen in the 19th century as self-
identified homosexuals, being effeminate, deviant, and social disordered.
In the middle of the 20th century, this perspective altered in the context of
reforming the concept of sexual identity. For this reason, male sex workers were
not automatically defined as even being homosexuals whatsoever. A distinction
was made, on the one hand, in that the public male sex worker who was depicted
as heterosexual and masculine, but young and innocent (Bimbi, 2007; Scott et
al., 2005). On the other hand, there was the group of the private male prostitutes,
which was described as effeminate and homosexual. Based on the conclusions that
male sex workers were not only homosexually, but also heterosexually oriented,
with the emergence of HIV/AIDS at the beginning of the 1980s, a concern about
a transmission of HIV into the heterosexual population arose (Scott, 2003; Scott
et al., 2005).
Even though these results are productive, it does not become clear how male sex
work was constructed in public by focussing only on the socio-scientific discourse.
Besides that, it is also difficult, to apply Scott’s (2003) results on the discourse in
Germany, because discourses are integrated in the local culture (Schetsche, 2008).
In light of this, this article investigates the question as to how male homosexual
sex work has been represented in German public discourse before the installation
of the DICCs. The theoretical frame of this work, which is what is first introduced,
maklu

Dangerous and endangered
Maklu 85
is the constructionist sociology of social problems. After the explanation of
the methodology and the research question, an overview of the quantitative
significance of male sex work in three German weekly magazines from 1980 to
1994 is presented. Two articles in two weekly magazines are studied in detail about
the construction of male sex workers as dangerous and endangered. These initial
results are discussed at the end of the article.
Theoretical background
Definition of social problems − sociology of social problems
The constructionist sociology of social problems takes the view ‘that social problems
are fundamentally products of a process of collective definition’ (Blumer, 1971, 298;
Spector & Kitsuse, 1977). Seen as such, a social circumstance is defined as a
social problem if different protagonists react to an individual trouble with a
special interest and refer to it as a general public concern. Consequently, after
having raised the topic in society, the institutionalisation of measures to solve
this problematised circumstance, e.g., the institutions of social work, are then
installed (Groenemeyer, 2011; Schetsche, 1996). By the institutionalisation of these
organisations, the already mentioned social circumstance is verified and accepted
as a social problem in society (Groenemeyer, 2010; 2011). According to Schetsche,
in the process of problematisation, it is irrelevant whether the problematised
circumstances are true or false under scientific criteria (Schetsche, 1996). The
deviation from ‘collective normative expectations and values’ (Groenemeyer, 1999,
8) is more important for process. A problematisation of a social circumstance has
its origin in unmet expectations or values of society. Hence, for the analysis of a
social problem, the reference to the values and the moral concept of the mentioned
society is indispensable (Groenemeyer, 1999; 2001).
The aim of studies in the sociology of social problems is not the verification or
falsification of different social problems, because there are no objective criteria
upon which an evaluation of ‘true’ or ‘false’ could be based (Schetsche, 1996).
Schetsche asserts that ‘a social problem […] is rather everything, what is seen and
identified as a social problem by the collective protagonists, the public or the welfare
state’ (ib., 2). Therefore, the analysis of social problems focuses on ‘conditions and
processes, on which social phenomena are problematised and are made as a public topic’
(Groenemeyer, 2001, 13). In this constructionist view, the research interest in
social problems is how the problematic issue is constituted and why the social
phenomenon or behaviour is seen as a problem (Groenemeyer, 2001; Loseke,
2003).
maklu

Marlen S. Löffler
86 Maklu
The cocoon model of social problems according to Schetsche (2008)
Schetsche’s cocoon model is a macro-sociological perspective of the social
construction of social problems and their development within societies, which
also delivers a methodical approach to examine the career of a social problem (ib.).
This model is helpful in demonstrating the different steps in the problematisation
of a social circumstance and the role of different protagonists and media with their
different views and goals in problematising a social circumstance. In Schetsche’s
perspective, social problems are always understood as being the results of
discursive processes. The career of a problem starts with the problematisation of
an already known or new social topic. Afterwards, a verbalisation of a problem
frame by collective protagonists follows, which defines that the social circumstance
is contravening against moral concepts and values of society. The problem frame
also contains statements about affected groups, combating possibilities and
consequential damages. These statements are based on the individual’s previous
knowledge (ib.).
To get a public problem perception, one will need to connect the new social topic
to already existing problem perceptions (Schetsche, 2000; 2008). One will also
have to address the recipients in a moral and emotional way, which consequently
arouses interest and the intentions about the topic. These practices are called
discourse strategies. One of these practices is to use dramatic statistics, but also
the use of selected examples or moralising the social subject (Schetsche, 2008) can
be defined as discourse strategies. Not only the public media, scientists are using
these strategies also (ib.). Nevertheless, the public media has a great significance,
because the continuous public discussion of the topic is a precondition of
problematising a social circumstance successfully (ib.). The successful use of these
strategies by media and other collective protagonists results in a public perception
of the problematised circumstances, which leads to measures on the part of welfare
state authorities, for instance the installation of institutions of social work. With
the installation of institutions, which should solve the specific accredited social
problem, the so-called institutions of social problems work (Holstein & Miller,
1993), and by including legal and financial arrangements, the social circumstance
is verified as a social problem and public awareness of the social problem will
spread further and will be reproduced (Schetsche, 2008).
Applying these considerations to the topic of male homosexual sex work in
Germany, it becomes obvious that, with the establishment of the first DICC for
male homosexual sex workers in the year 1986, male homosexual sex work became
recognised as a social problem.
maklu

Dangerous and endangered
Maklu 87
Methodology
Based on this, the following study’s focus is the public discourse of male sex
work in media. The goal is to investigate, how male homosexual sex work was
constructed before DICCs were applied. In addition, the discourse strategies with
regard to male homosexual sex work will be identified.
This article is embedded in an extensive study that includes articles about male
homosexual sex work in three German weekly magazines, namely Der Spiegel,
Stern, and Die Zeit, which include the keyword ‘Stricher’ (hustler) and were
published between 1980 and 1994. The three chosen weekly magazines can be
seen in this time period as the defining media in Germany and German daily
newspapers have chosen their topics based on these weekly magazines (ib.). This
time period is chosen, because the institutionalisation of the first DICCs was in
1986 (Grabatsch, 2013). On the other hand, HIV/AIDS first came up in this time
period (Wright & Rosenbrock, 2012) and as a result an increasing public interest
of male sex workers then followed (Scott, 2003). Aside from that, the larger study
uses different types of documents to examine how three German DICCs were
installed locally. This will not be discussed at this point.
With the help of Schetsche’s model of the empirical problem analysis (Schetsche,
2008), two articles that were published in 1985 in the weekly magazines Der
Spiegel and Stern are studied. The first problematisation of male sex work and the
discourse strategies used in the articles will be investigated by a content analysis.
Both articles were chosen with regard to their content. In the examined time
period, these are the first articles that discuss male sex work centrally. The other
23 articles that have been published before in these two weekly magazines often
just mention the word ‘Stricher’ and do not discuss male sex work in detail. With
the chosen articles, it is possible to show distinctive constructions about male
homosexual sex workers and discourse strategies in constructing male sex work
as a social problem.
Results
To begin, the quantitative amount of male sex work in the three weekly magazines
Der Spiegel, Stern, and Die Zeit is presented. The low public attention about male
sex work will be shown by comparison with other topics discussed in the weeklies.
Afterwards, an overview of the discussion about male sex workers in the time
period 1980 to 1985 in two weekly magazines and a more detailed investigation of
two articles follows.
maklu

Marlen S. Löffler
88 Maklu
Male homosexual sex work in the weekly magazines Der Spiegel, Stern, and
Die Zeit
Looking at the quantity of reports using the keyword ‘Stricher’ (hustler) in the
three chosen weeklies between 1980 and 1994, 141 articles are found. The weekly
Der Spiegel published 56 articles, Stern 44, and Die Zeit 41. Using the keywords
‘Callboy’ or ‘Prostituierter’ (German for male prostitute), the result is numerically
low. In the three weekly magazines in the same time period, only 7 articles are
found that use the term ‘Prostituierter’ and 13 articles that use the term ‘Callboy’.
By comparison, the articles include the words ‘Stricher’, ‘Prostituierter’ or ‘Callboy’
with articles including the words for female sex worker, ‘Hure’ (whore), ‘Nutte’
(another word for whore), ‘Prostituierte’ (German for female prostitute) or
‘Homosexueller’ (male homosexual), it becomes clear that male sex work is not
a topic of special interest (see Table 1). The amount of articles using the word
‘Prostituierte’ is nearly ten times higher than articles using the word ‘Stricher’.
Table 1. Amount of articles using the different keywords between 1980 and 1994
Keywords in the time period 1980-1994 Der Spiegel Stern Die Zeit
‘Prostituierte’ (German for female prostitute) 459 495 491
‘Homosexueller’ (male homosexual) 156 83 676
‘Hure’ (whore) 268 288 491
‘Nutte’ (another word for whore) 106 90 144
‘Stricher’ (hustler) 56 44 41
‘Callboy’ (callboy) 4 1 8
‘Prostituierter’ (German for male prostitute) 1 - 6
Apart from this, the number of articles that pay attention to male sex work in detail
in this data source is low. Mostly male sex workers are mentioned in literature,
theatre, film, and TV reviews or in portraits of artists. Altogether, 61 of the 141
articles refer to these topics. Coming in second is the mentioning of ‘Stricher’
in the context of HIV/AIDS or other STIs (27 articles). Male sex workers are also
included in 13 articles about criminal offences. Finally, in 12 articles, male sex
workers are portrayed with reference to male homosexual sex work, to sex work in
general or to the work of DICCs. The rest of the articles refer to a political affair
in Germany about a general called Kießling (8 articles), to other political debates
(6 articles), to homeless children or teenagers (5 articles), to child prostitution in
foreign countries and sex tourism (5 articles), as well as portraits of police work
(4 articles).
maklu

Dangerous and endangered
Maklu 89
The highest amount of articles including the keyword ‘Stricher’ are found in the
years 1987 and 1993 with 14 articles, followed by 13 articles in the year 1989.
Ranking third are 12 articles published with the keyword in 1988 and 1992. In the
years 1984 and 1986, we can see that there were 11 articles per year (see Figure 1).
Figure 1. Amount of articles with the keyword ‘Stricher’ (hustler)
per year and weekly magazine from 1980 to 1994.
Regarding the development of the amount of the articles, it becomes clear that
the discussion about male sex work in the three weeklies increased after 1983
(see Figure 1). After that, the number of articles on this topic continues to remain
steady. Despite the fact that the total amount of the articles is low, the rise of the
number of articles is remarkable. This may lie in the fact that HIV/AIDS emerged
in 1981. For Germany, Wright & Rosenbrock (2012) define the time period from
1981 to 1986 as the phase of the state of emergency with respect to HIV/AIDS.
The phase of development of strategies dealing with the infection from 1986 to
1991 followed afterwards. These strategies were, for instance, the implementation
of prevention offers, HIV-tests, and the improvement of health care (ib.). The
numerical increase in the articles about male sex workers at the time of the rise
of HIV/AIDS underlines the results of Scott (2003). In his study on the discourse
of male sex work in social sciences, it becomes clear that male sex work was
understood more and more as a public health problem in conjunction with the
rise of HIV/AIDS (ib.).
Problematising male homosexual sex workers: dangerous and endangered
Though problematising male sex work seems to correspond with HIV/AIDS, it
is not the only way it can be looked at. Regarding all of the 25 articles, even if
male sex work is just mentioned in a few words or a sentence in say 23 of them,
it becomes obvious that the topic is discussed in a manner slotting in as between
maklu

Marlen S. Löffler
90 Maklu
criminalisation and victimisation. The criminalisation contains statements that
male sex workers would be greedy egoists or cheaters who only have their own
benefit in mind and who have a high risk of STI infection. In this thread, the sex
workers scene is equated with the criminal or junkie milieu3. Within the framework
of victimisation, male sex workers are seen as juveniles who likely come from
bad living conditions such as a broken-home background with experience with
physical violence, drugs, and alcohol abuse, no education, and having to remain
in detention centres4.
The following study of two chosen articles provides a detailed investigation of the
construction of male sex work and focuses on the question as to why male sex
work is seen as problematic. These findings will be connected with the results
from Scott’s study (2003) about the socio-scientific discourse of male sex work. It
shows that male sex work is constructed as a health risk for society, but also that
male sex work is constructed as a risk for children or teenagers who offer sexual
services.
The article ‘With ‘safe sex’ I can‘t make any dosh’. The Hamburg hustler-scene and
the Aids disease of the weekly Der Spiegel (‘‘Mit ‘Safer-Sex’ kann ich keine Kohle
machen’,’ 1985) presents a portrait of a 23 year old hustler named Christipher
Dohmin who started conducting sex work at the age of 15. The main theme is
the change of the sex workers scene after the appearance of HIV/AIDS. In this
article, the criminalisation of male homosexual sex workers becomes obvious.
The male sex workers are constructed as an uncontrollable health risk for the
entire (heterosexual) society. Here, male sex workers are portrayed as uninterested
people, who do not understand or know the simplest biological processes:
‘[…] and what if he just got infected with AIDS? The question leaves him cold and
he swears by his alleged robust constitution: ‘To this day’, he says, ‘I never caught
a sexually transmitted disease and that after more than 2,000 jobs.’ He goes to
the doctor, who is also a friend, for regular check-ups but he doesn’t get tested for
AIDS: ‘I’m not going to make myself crazy over that’’ (ib., 13; 23).
‘[…] doch wenn er sich eben jetzt mit Aids infiziert hätte? Die Frage läßt ihn kalt,
er schwört auf seine angeblich eiserne Konstitution: „Bis heute’, sagt er, „bin ich
noch nie geschlechtskrank gewesen, und das nach immerhin mehr als 2000 Jobs.’
3 Limmer (1980); ‘Knapp und teuer,’ 1980; Limmer (1981); Karasek (1981); Sternsdorff (1983);
‘Flambierte Frau, ganz cool’; Umbach (1983); Karasek (1983); Kohl: ‘Das läuft nicht gut’, (1984);
‘Ein Abgrund von Sumpf hat sich aufgetan’, (1984); ‘Wörner – ‚der Lächerlichkeit preisgegeben’,
(1984); ‘Affäre Wörner’, (1984); ‘Fünf Jahre Quarantäne?’, (1984); Noack (1985); Schneider (1985);
‘Mit ‘Safer-Sex’ kann ich keine Kohle machen’, (1985).
4 Timmerberg (1981); Blechschmidt (1984), Röhl (1984, 1985), Brocher (1985); Prase & Thieme
(1985); Schaper (1985); Bier (1985); Lemmen & Winhuisen (1985).
maklu

Dangerous and endangered
Maklu 91
Zwar läßt er sich regelmäßig von einem befreundeten Arzt untersuchen, aber
nicht auf Aids: „Ich mach‘ mich doch deswegen nicht verrückt’‘ (ib.).
The article also shows that the portrayed sex worker is not willing to voluntarily
be tested for HIV. A medical practitioner underlines this statement by saying that
only heterosexual family fathers would test themselves for HIV:
‘This kind of ostrich attitude is typical for the sex professionals of the Reeperbahn
is confirmed by Dr. Gerd Burchard, a doctor in the Berhard-Nocht-Institute in
St. Pauli. For weeks now, approximately 300 blood samples are being tested for
AIDS daily. However, ‘neither hustlers nor other at-risk groups from the red-light
district’ show up to be tested voluntarily (price: 68 German Marks) – but rather
‘Nearly all are frightened fathers getting in touch with us, often using a false
identity’’ (ib., 22).
‘Daß die Vogel-Strauß-Haltung typisch ist für die Sex-Profis von der Reeperbahn,
bestätigt der Ambulanzarzt Dr. Gerd Burchard vom Bernhard-Nocht-Institut
in St. Pauli; seit Wochen werden dort täglich rund 300 Blutproben auf Aids
untersucht. Doch zum freiwilligen Test (Kosten: 68 Mark) erscheinen, so
Burchard, ‚weder Stricher noch andere Risikogruppen vom Kiez’ – vielmehr: ‚Es
sind durchweg verängstigte Familienväter, die sich, meist unter falschem Namen,
hier bei uns melden’.’
There are also some bisexually defined clients amongst the sex workers who do
not care about the infection risk and who are not willing to practice safer sex, as
the portrayed Dohmin says:
‘‘Even now, most don’t want durex protection.’ He is convinced: ‘With ‘safe sex’ I
can‘t make any dosh’’ (ib., p. 21).
‘‘die meisten wollen auch jetzt keinen Gummischutz.’ Für ihn steht fest: ‘Mit
‘Safer Sex‘ kann ich keine Kohle machen’’ (ib.).
Even if there are some clients who are not interested in safe sex and who are
also family fathers, the sex workers are made responsible for the spread of HIV/
AIDS, because of their disinterest and insufficient biological knowledge. The
responsibility of the male sex workers for the spread of the infection goes on by
demonstrating an egoistic attitude. Thus, the sex workers would not have any
moral scruples about practicing unsafe sex with their clients. They act negligently,
and do so almost intentionally:
maklu

Marlen S. Löffler
92 Maklu
‘The thought that he could infect his clients with an initially undiscovered AIDS
infection doesn’t give him any scruples: ‘If I have to die’, he explains detached, ‘for
all I care, the others can die as well. I couldn’t care less.’’ (ib., 14).
‘Daß er nach einer zunächst unentdeckten Aids-Infektion seine Kunden anstecken
könnte, dieser Gedanke bereitet ihm keinerlei Skrupel. ‘Wenn ich sterben muß’,
erklärt er ungerührt, „können von mir aus auch die anderen draufgehen, das
wäre mir egal’’ (ib.).
Still, the article also discusses that male sex workers are getting fewer and fewer
work orders, because of the clients’ fear about HIV/AIDS. This leads to financial
difficulties that produce higher pressure of competition between the sex workers.
Consequently, as the author claims, the male sex workers are forced to practice
unsafe sex work in absence of other alternatives (ib., p. 20). On the other hand,
the picture of an egoistic and hedonistic sex worker is developed in parallel by
reporting that Dohmin is leading an extravagant and wasteful lifestyle:
‘He goes out for a steak and for dessert he treats himself to a joint. The next
morning he takes a taxi to the hotel ‘Vier Jahreszeiten’. There, Mr. Dohmin is
offered a table by the window with a view of the river Alster. With fragrant coffee
and fresh bread rolls, he succumbs to a daydream. He dreams of a Porsche and a
luxurious penthouse in the city. With a clientele from the upper-class of Hamburg,
he could become self-employed. As long as there is no vaccine against AIDS, he’s
gradually realising that this will never happen’ (ib., 24-26).
‘Er geht ein Steak essen, zum Nachtisch leistet er sich einen Joint. Am Vormittag
darauf läßt er sich im Taxi zum Hotel ‘Vier Jahreszeiten’ fahren. Dem Herrn
Dohmin wird ein Fenstertisch mit Alsterblick angeboten. Bei duftendem Kaffee
und frischen Brötchen gerät er ins Phantasieren. Er träumt von einem Porsche
und einem luxuriösen Penthouse in der City. Da könnte er sich, mit einer Klientel
aus den allerfeinsten Hamburger Kreisen, selbständig machen. Daß daraus
nichts wird, solange es keinen Impfstoff gegen Aids gibt, begreift er allmählich‘
(ib.).
At this point, it also becomes obvious that Dohmin has some naïve future plans.
He broke his confectioner education (ib., 27) and he does not want back in a civil
life:
‘If nothing happens on the game anymore’, he considers, ‘I would like to become
a coke dealer’ (ib., 27).
‘Wenn auf dem Strich nichts mehr läuft’, überlegt er, ‘würde ich gern Koks-Dealer
werden’ (ib.).
maklu

Dangerous and endangered
Maklu 93
Consequently, an additional danger results from male sex workers. They endanger
society from a health standpoint, but also morally, particularly in matters of
sexuality, deviance, and crime.
To repeat the essential points: Male sex workers are seen here as vectors for the
transmission of HIV/AIDS to the larger public. They are pictured as naïve and
uncaring about the health of other people. Also, financial difficulties that arise in
context to the clients’ fear about HIV/AIDS, but also as the sex workers’ own fault,
are forcing the sex workers to conduct commercially unsafe sex. Furthermore, the
use of drugs is discussed in this article, such that male sex work is also connected
to the drug and criminal milieu.
According to results gained by Scott (2003), we can see that the emergence of male
sex work as a social problem is, in Germany, also attributed to a concern about
public health (Bimbi, 2007). Firstly, at the beginning of the 19th century, male sex
work was constructed as being a moral problem. This view is based on sexual
deviance and the crossing of class boundaries (Scott, 2003). Later, in the middle
of the 20th century, intergenerational sexual contacts and exploitive relations
were problematised. Therefore, in this time, male sex work was constructed as
being more of a welfare problem. Before the appearance of HIV/AIDS, hardly
any connections were made between male sex workers and STIs. This changed in
1960s and 1970s, because male sex workers back then were pictured as being fluid
in their sexual identity.
These findings lead to worries about the health of society on whole: ‘The idea,
that a prostitute could slip ‘unseen’ into the general population created much
anxiety within epidemiological quarters’ (Scott, 2003, 194-195; Scott et al.,
2005). Consequently, male sex work was discussed as a public health problem.
Comparing these findings with the previous results in the anxiety about the
spreading of HIV/AIDS through the larger public is a point of agreement. But
there is a different view about the dangers cause. While Scott sees the emergence
of male homosexual sex work as a public health problem in the undefined sexual
orientation of male sex workers (Scott, 2003; Scott et al., 2005), this article locates
the risk in the irresponsible behaviour of the male sex workers and their often
heterosexually active clients.
The other article, named For example Cologne: teenager prostitution in Germany.
‘That’s better than stealing, right?’ (Lemmen & Winhuisen, 1985) and published
in the weekly Stern, reports on juvenile prostitution. It contains portraits of three
teenagers (two boys and one girl) aged between 14 and 16 who conduct sex work.
Here, juvenile (male) sex workers are discussed as endangered persons, and thus,
in a victimising way. Although this article discusses attempts to help the juveniles,
a clear description of the problematic issue or a clarified threat for the young sex
maklu

Marlen S. Löffler
94 Maklu
workers cannot be identified. Just the information that juveniles are doing sex work
frames the issue as being problematic, because juveniles need special protection.
Some clues are indicated, such as the harmfulness of commercial sexual contacts
between juveniles and male adults:
‘[…] Michael […] thinks our question is stupid: ‘The men are disgusting. Sometimes I
really hate the old bastards but what can you do, they have the money and for that you
have to work for it’. The 14-year old Petra: ‘I lay there stiff as a board, just thinking of
the money and that he will soon be finished’. The question about what some of the men’s
requests are obviously embarrasses Ralf, Michael and Petra alike. They giggle and don’t
want to tell us’ (ib., 32-33; 54).
‘[…] Michael […] findet unsere Frage blöd: „Ekelhaft sind die Männer, manchmal hab’
ich einen richtigen Haß auf diese alten Säcke, aber was willst du denn machen, die
haben schließlich das Geld, und dafür muß man schon was tun.’ Die 14jährige Petra:
„Ich liege da wie ein Brett und denke nur an das Geld und daß der bald fertig wird.’
Wenn wie genauer nachfragen, was die Männer denn so von ihnen wollen, werden Ralf
und Michael und Petra verlegen, sie kichern und mögen es uns nicht sagen’ (ib.).
Also, the authors seem to see a danger for the juveniles in remaining in the field
of prostitution:
‘If he starts an apprenticeship, he wants to stop working at the train station.
Really? We ask and could he manage with considerably less money? ‘Well, now
I can earn a quick thousand a month on the side, so it really would be a change.
Maybe I’ll just stop by for a while on the weekend if I need money – that’s better
than stealing, right?’’ (ib., 39-40).
‘Wenn er jetzt mit der Lehre beginnt, will er am Bahnhof aufhören. Wirklich?
fragen wir, und ob er dann mit deutlich weniger auskommen wird? ‘Na ja, jetzt
verdien ich ja so nebenbei auf die Schnelle einen Tausender im Monat, es wird
schon eine Umstellung sein. Vielleicht komm ich dann mal kurz am Wochenende
wenn ich Geld brauche – das ist doch besser als Klauen, oder?’’ (ib.).
Although HIV/AIDS is discussed shortly in this article, it has a convincing effect
in demonstrating the danger for juveniles:
‘A doctor from the public health department in Cologne: ‘We found antibodies
of the immunodeficiency AIDS pathogen in a ten-year-old boy who goes on the
game’’ (ib., 52).
‘Eine Ärztin des Kölner Gesundheitsamtes: ‘Bei einem 10jährigen Jungen, der
auf den Stricht geht, haben wir Antikörper auf den Erreger der Immunschwäche
Aids gefunden’’ (ib.).
maklu

Dangerous and endangered
Maklu 95
Pedagogical experts are quoted in the article. They report an increase in juvenile sex
workers and an increasing number of juveniles less than 14 years of age doing sex
work (ib., 51). Correspondingly, the juvenile families are pictured as being intact,
and the sex workers would not always suffer from a broken-home background:
‘The street worker: ‘We are registering a change in the child prostitution scene.
Now, there are not only more and even younger children, mostly runaways from
orphanages, but there is an increasing number of adolescents coming from stable
living conditions who do it on the side to add to their pocket money’’ (ib.).
‘Die Streetworker: ‘Wir registrieren eine Veränderung auf dem Kinderstrich
– es gibt inzwischen nicht nur immer mehr und immer jüngere weggelaufene
Heimkinder, da sind auch immer mehr Jugendliche aus geordneten Verhältnissen,
die’s mal nebenbei machen, die ihr Taschengeld aufbessern’’ (ib.).
This information shows that there is an urgency for the topic and that every child
or family could be affected. Complicating matters, the juvenile sex workers are
not easy to recognise for people who do not know the scene and the juveniles go
about doing it in an inconspicuous manner (ib., 29; 7-9; 12-16). For this reason,
it is difficult for the police, social workers, and parents to offer them protection,
because the juveniles do not seem to see any problematic issue in doing commercial
sex with older men. As such, the police, social workers, and parents seem to be
helpless with this situation. Another problem for social workers is seen in the
fact that the youth male sex worker isn’t coming from a milieu where existential
problems play a role:
‘This is not about existential necessities; it’s about shirts, jumpers, a new stereo
system. They want to ‘keep-up’ at the disco with money and with the latest fashion
from the expensive boutiques’ (ib., 50; 36).
‘Da geht es nicht um existentielle Not, da geht es um Hemden, um Pullover, die
neue Stereo-Anlage. Die wollen in der Disco mithalten können, mit Geld und mit
den neuesten Klamotten aus den teuren Boutiquen’ (ib.).
They would conduct sex work in order to be able to afford expensive clothes or
nightlife, so that the social work could not offer them any attractive help (ib., 3;
17-18). Moreover, the juveniles who would have existential issues could not be
helped, because social work was underfunded (ib., 49-50). Social workers and
police criticise that juvenile sex work is not yet recognised by all administrations.
Therefore, they demand politics to recognise this topic and to support the social
work or the police in this subject (ib., 27-28; 52).
maklu

Marlen S. Löffler
96 Maklu
To summarise the main results: Even though there is no specific description of the
problematic issue, the commercial sexual interaction between adults and children
or teenagers is a fact that is problematised in this article. Thus, this article has a
victimising view on male sex workers. The juveniles do not understand the risky
consequences of their behaviour, so they are portrayed as being naïve.
With respect to the results gained by Scott (2003), the construction of male sex
workers as endangered juveniles has remained in the socio-scientific discourse since
the middle of the 20th century. The view on male sex workers as being homosexual,
effeminate, and perverse men existed before its change into a victimisation of the
male sex worker. From that point on, the group was seen as being the young, weak,
innocent, and heterosexual victims of older male homosexual ‘perverts’ (ib., 183).
The task of institutions of social control was to protect the juveniles from these
older male offenders, because the relation between younger and older men was
regarded as being exploitive (ib.). As a consequence, male sex work was assessed
as a welfare issue. These associations are apparent in the analysed article. The
portrayed male juveniles identify themselves as heterosexuals and are pictured
as being inexperienced. The clients are exclusively older men, homosexually
orientated or family fathers, who partly have some bizarre sexual tendencies.
Furthermore, some institutions of social control are active in trying to protect the
juveniles.
Comparison: commonalities and distinctions
The problematisation in these two articles sways between criminalising and
victimising male sex workers. Although these two poles are different, some
common knowledge about male sex workers can be identified in both articles as
well as in the 23 articles published between 1980 and 1985. There, sex work is
not seen as a commendable activity earning money, but rather is labelled as a
surviving strategy after difficult circumstances. For instance, for homeless people,
sex work is more accepted. Sex work is an unpleasant source for income as well.
It is an activity that is endured or only feasible with drug use. Both articles portray
the male sex workers as juvenile, naïve, and carefree. The depictions of the sex
working scene are similar. They mostly work in central stations in bigger cities
and the contact between the sex workers and clients is initiated in a non-verbal
manner. The clients are described as old, unattractive, and male, and as being
either bisexual or homosexual.
Still, this common knowledge does not lead to the same assessments. The young
age of the portrayed sex worker in the article of Der Spiegel is just mentioned and
not rated in any way. While the age of sex workers in the article of Stern gets a
central role in problematising male sex work. Although the juveniles’ motivation
to conducting sex work is not only a survival strategy, but to buy some luxury
maklu

Dangerous and endangered
Maklu 97
goods, they are not associated with hedonism. Instead, they are pictured as being
ignorant, naïve, and not knowing how they are harming themselves. The clients
are seen in the victimising construction of male sex work as offenders. In the
meantime, this aspect is not discussed in the criminalising perspective of male sex
workers. As a countermove, HIV/AIDS is just dealt with shortly in the article of the
magazine Stern while this topic is a central argumentation of problematising male
sex work in the magazine Der Spiegel.
With regard to the different point of views on male sex work, some questions
do arise, namely: At which point does the hedonistic male sex worker who is
dangerous to society turn into the male sex worker, for whom sex work is a survival
strategy? What criteria have to be fulfilled in order for the juvenile sex workers to
transform into a health offender?
In the analysed articles different discourse strategies (Schetsche, 2008), necessary
to lead to a successful problematisation, are identified. First one may state that
offering sexual services is regarded as a deviant behaviour, which was convicted
as a sex crime in Germany until 2001 (Bundesministerium für Familie, Senioren,
Frauen und Jugend, 2005). In this sense, male sex work is identified for this time
period as a norm deviation. Both articles use dramatic examples for particular cases
to describe the topic. The depictions are supported by scenesters and different
experts who also report scandalous statistics about the situation. Der Spiegel
reports about the high amount of HIV tests amongst the simultaneous absence of
risk groups such as that of male sex workers. In the same way, the article about the
juvenile sex workers is using numbers about the amount of juvenile sex workers
and their young age to demonstrate the significance of this subject.
This strategy also affects the magazines’ recipients in an emotional manner. Both
the criminalising and the victimising construction awaken sorrow and fear among
the readers. On the one hand, there is fear about HIV/AIDS. This fear is expanded
by the information that the clients are not only homosexual men, but also family
fathers. On the other hand, there is a sorrow to be felt about the children and
juveniles, because they do not get enough protection. Schetsche (2000) also
explains that, for a successful problematisation, a dichotomisation of guilt must
be present. This criterion can be found in both reports. In the article ‘With ‘safe sex’
I can‘t make any dosh’ The Hamburg hustler-scene and the Aids disease (‘‘Mit ‘Safer-
Sex’ kann ich keine Kohle machen’,’ 1985), the dichotomisation of guilt lays in the
health risk, which comes from the male homosexual sex worker. They are seen as
dangerous persons and the accusations are expanded while the sex workers are
pictured as greedy, hedonistic, and criminal persons who infect health conscious
(heterosexual) family fathers and probably their families with HIV.
maklu

Marlen S. Löffler
98 Maklu
In the second report For example Cologne: Teenager prostitution in Germany. ‘That’s
better than stealing, right?’ (Lemmen & Winhuisen, 1985), guilt is not attributed
to the juvenile sex workers. The children and young persons who prostitute
themselves are seen as endangered persons and as victims. They are victims of
their (childish) ignorance, but also victims of a society in which consumption and
financial independence seems to be amongst the highest values. Consequently,
the older male ‘perverts’ who use the offered services of the juvenile sex workers
are seen as culprits.
It is remarkable that the two weeklies have tendencies in their way of reporting
about male sex workers: The weekly Der Spiegel mainly criminalises the sex workers
while the weekly Stern reports about male sex workers in a victimising manner.
These differential representations are explainable in context of a competition
between media that are understood here as business groups attempting to gain
enough public attention (Schetsche, 2008). As was already shown, by using the
discourse strategies, the attention of the public about a topic can be produced. Aside
from that, the weekly magazines have to fulfil the preferences and expectations
of their recipients in order to remain economic (ib.). Accordingly, these findings
could provide us with a clue on the different expectations about this subject in the
readership. While the readers of Der Spiegel seem to prefer a sceptical and negative
view on male sex work and male homosexuals, the recipients of Stern view (male)
sex workers with compassion and clients with contempt.
Comparing the two articles, it becomes obvious that in both, the problematising
of male (homosexual) sex work is seen in different development stages. The first
article has a nearly exclusive focus on HIV/AIDS. HIV/AIDS popped up in 1981
as a deadly infection. In the analysed time period from 1980 to 1985, almost no
measures fighting against this disease existed. The discussion of both HIV/AIDS
and male homosexual sex work was in its early stages. Only medical practitioners
were identified and questioned as experts at this point. Schetsche (2008) says that
coherence with already recognised knowledge is necessary to recognise a social
circumstance as a social problem. By picturing male sex workers as a vector of
HIV/AIDS, this coherence is made.
HIV/AIDS is a sexually transmitted infection and at that time, only male
homosexuals and drug users were identified as risk groups (Wright & Rosenbrock,
2012, 198). The act for fighting STIs installed from 1953 until 2000 in Germany,
which focused exclusively on female sex workers, can be understood as already
prevailing knowledge about dealing with this subject. Correspondingly, the
problematisation of HIV/AIDS in the context of male homosexual sex work,
which includes another probable risk group, namely homosexual males, seems
to be successful under these conditions. In addition, the portrayed male sex
worker is described as a potential criminal (‘‘Mit ‘Safer-Sex’ kann ich keine Kohle
maklu

Dangerous and endangered
Maklu 99
machen’,’ 1985), which follows the view that male sex work was located in a
criminal environment. Therefore, the growing public interest on male sex work
in the context of HIV/AIDS can also be interpreted as a culmination of the already
prevailing criminalisation of male sex workers who are supposedly homosexual.
The process of problematising male sex workers as being endangered persons
in the second article (Lemmen & Winhuisen, 1985) seems to be more advanced.
In this discourse, more protagonists such as policemen, street workers, medical
practitioners, and currently unofficially installed measures for fighting this
problem remain. They demand adequate measures for fighting this situation, but
the problem is only recognised in particular by politics.
The lead of problematising male sex work as a danger for juveniles in comparison
with the construction of male sex worker as a health risk for society is explainable
in that the chronology of the socio-scientific discourse about male sex work is
taken into consideration. A new perspective on male sex workers was established
in the 1950s and 1960s. While male sex workers were pictured as homosexual
perverts beforehand, the emerging comprehension drew them as heterosexual
and exploitive victims (Scott, 2003). As a result of the distinction of male sex
workers as homo and heterosexual, the connection between STIs and male sex
work was possible (ib.). For this reason, the victimisation of male sex workers and
the understanding of it as a welfare problem can be defined as the foundation of
the already taken measures, like street work. Prior the emergence of HIV/AIDS,
male sex work was not discussed in the context of STIs. Accordingly, in the year
the articles were published, the construction of male sex work as a public health
problem was merely at its beginning.
Discussion
Aside from some of the basic point of views about male sex work, including the
young age or the local structures of male sex work, there are different constructions
of male sex workers: First, male sex workers are pictured as a health risk to all of
society and as responsible for spreading of HIV/AIDS. Second, male juveniles who
are actively conducting sex work suffer from a physical and psychological risk. The
juveniles are subjected to harm by the sexual and exploitive contact with clients.
Although these problematisations proceed parallel, the view on sex workers as a
health risk marks the transformation of the topic from a nearly trivial problem to
a risk for all of society.
In comparison to the results obtained by Scott (2003), it is considerable that the
public discourse in the two analysed weeklies is not too different to that of the
socio-scientific discourse. This fact demonstrates the significance of science
maklu

Marlen S. Löffler
100 Maklu
in problematising social circumstances (Schetsche, 2008). In contrast, until
1985, scientific research about male sex work in Germany was rarely published
(Redhardt, 1954; 1968; Schickedanz, 1979; Schmidt-Relenberg, Kärner & Pieper,
1975). This gives rise to the suspicion that the view on male sex work in the
scientific discourse has been promoted by public representations (Schetsche,
2008). From this point of view, further research could examine what connection
might exist between media and scientific research on male sex work and how
public and scientific views interact.
Apart from rendering male sex work in media to be problematic, the effects of
political decisions and law changes affecting male sex work are topics that are
underrepresented in social science. In Germany, for example, the abolition of
the law, which prohibited sexual relationships between men and boys under the
age of 21 in Germany in 1973, or the act regulating the legal status of prostitutes
installed in 2002, could be seen as starting points for further research about the
public and scientific discourse on male sex work. These changes have immediate
consequences for male sex workers, thus making them a significant subject of
research.
References
‘Ein Abgrund von Sumpf hat sich aufgetan’ (1984, January 23). Der Spiegel, 15-27.
Affäre Wörner: Zweideutig eindeutig (1984, February 20). Der Spiegel, 30-32.
Arbeitskreis der Strichereinrichtungen in Deutschland (AKSD). (2012).
Projektstandorte. Retrieved from http://aksd.eu/projekte.html.
Arbeitskreis deutschsprachiger Stricherprojekte (AKSD). (2003). Leitlinien für
die soziale Arbeit mit Strichern. In M.T. Wright (Ed.), Prostitution, Prävention
und Gesundheitsförderung. Teil 1: Männer (pp. 119-128). Berlin: Deutsche AIDS-
Hilfe e.V.
Bier, P. (1985, November 28). Nachwuchs für den Knast. Stern, 184.
Bimbi, D.S. (2007). Male Prostitution: Pathology, Paradigms and Progress in
Research. Journal of Homosexuality, 53(1/2), 7-35.
Bimbi, D.S., & Koken, J.A. (2014). Public Health Policy and Practice with Male Sex
Workers. In V. Minichiello & J. Scott (Eds.), Male sex work and society. New York:
Harrington Park Press, 198-221.
Blechschmidt, P. (1984, February 23). Keine Ruhe für den Minister. Stern, 224-225.
Blumer, H. (1971). Social problems as a collective behavior. Social Problems, 18,
298-306.
maklu

Dangerous and endangered
Maklu 101
Bochow, M. (2003). Die Lebenswelten von Strichern: Interviews aus der Szene.
In M.T. Wright (Ed.), Prostitution, Prävention und Gesundheitsförderung. Teil 1:
Männer. Berlin: Deutsche AIDS-Hilfe e.V, 25-48.
Brocher, T. (1985, July 11). Die Flucht der Männer. Stern, 46.
Bundesministerium für Familie, Senioren, Frauen und Jugend. (2005). Untersuchung
‘Auswirkungen des Prostitutionsgesetzes’: Abschlussbericht. Berlin. Retrieved from
http://www.bmfsfj.de/doku/Publikationen/prostitutionsgesetz/pdf/gesamt.
pdf
Crofts, T. (2014). Regulation of the male sex industry. In V. Minichiello & J. Scott
(Eds.), Male sex work and society. New York: Harrington Park Press, 178-197.
Dennis, J.P. (2008). Women are Victims, Men Make Choices: The Invisibility of
Men and Boys in the Global Sex Trade. Gender Issues, 25(1), 11-25. doi:10.1007/
s12147-008-9051-y
Fachkreis für Stricherarbeit im deutschsprachigen Raum (AKSD). (2007). Leitlinien
fuer die sozialpädagogische Arbeit mit Strichern. Retrieved from Fachkreis für
Stricherarbeit im deutschsprachigen Raum (AKSD) website: http://www.aksd.
eu/download/leitlinien.pdf
Fink, K. & Werner, W.B. (2005). Stricher: Ein sozialpädagogisches Handbuch zur
mann-männlichen Prostitution. Lengerich [u.a.]: Pabst.
‘Flambierte Frau’, ganz cool (1983, May 9). Der Spiegel, 175.
Friedman, M. (2014). Male sex work from ancient times to the near present. In V.
Minichiello & J. Scott (Eds.), Male sex work and society. New York: Harrington
Park Press, 2-33.
‘Fünf Jahre Quarantäne?’ (1984, December 24). Der Spiegel, 28-32.
Grabatsch, S. (2013). 26 Jahre sozialpädagogische Arbeit mit Strichern in
Deutschland. In Arbeitskreis der Strichereinrichtungen in Deutschland
(AKSD) (Ed.), 20 Jahre AKSD – Ein Bericht. Eine Textsammlung anlässlich des
20-jährigen Bestehens des AKSD, 3-8.
Groenemeyer, A. (1999). Soziale Probleme, soziologische Theorie und moderne
Gesellschaften. In G. Albrecht, A. Groenemeyer & F.W. Stallberg (Eds.),
Handbuch soziale Probleme. Opladen: Westdeutscher Verlag, 13-72.
Groenemeyer, A. (2001). Soziologische Konstruktionen sozialer Probleme
und gesellschaftliche Herausforderungen: Eine Einführung. Soziale
Probleme. zeitschrift für soziale Probleme und soziale Kontrolle, 12(1/2), 5-27.
Groenemeyer, A. (2010). Doing Social Problems – Doing Social Control:
Mikroanalysen der Konstruktion sozialer Probleme in institutionellen
maklu

Marlen S. Löffler
102 Maklu
Kontexten – Ein Forschungsprogramm. In A. Groenemeyer (Ed.), Doing
Social Problems. Mikroanalysen der Konstruktion sozialer Probleme und sozialer
Kontrolle in institutionellen Kontexten (1st ed.). Wiesbaden: VS Verlag für
Sozialwissenschaften, 13-56.
Groenemeyer, A. (2011). Soziale Probleme. In H.-U. Otto & H. Thiersch (Eds.),
Handbuch. Handbuch Soziale Arbeit (4th ed.). München: Reinhardt, 1390-1405.
Gusy, B., Krauß, G., Schrott, G. & Heckmann, W. (1994). Aufsuchende
Sozialarbeit in der AIDS-Prävention: Das ‘Streetwork’-Modell. Schriftenreihe
des Bundesministeriums für Gesundheit: Bd. 21. Baden-Baden: Nomos
Verlagsgesellschaft.
Holstein, J.A. & Miller, G. (1993). Social constructionism and social problems
work. In J.A. Holstein & G. Miller (Eds.), Reconsidering social constructionism.
Debates in social problems theory. Hawthorne, New York: Aldine de Gruyter, 152-
172.
Karasek, H. (1981, April 20). Narziß im Bahnhofsviertel. Der Spiegel, 229-231.
Karasek, H. (1983, June 13). Wie man Hexen flambiert. Der Spiegel, 169-170.
Kaye, K. (2014). Male Sex Work in Modern Times. In V. Minichiello & J. Scott
(Eds.), Male sex work and society (pp. 34-48). New York: Harrington Park Press,
33-48.
Knapp und teuer (1980, October 13). Der Spiegel, 261-264.
Kohl: ‘Das läuft nicht gut’ (1984, January 16). Der Spiegel, 15-24.
Lautmann, R. (2002, October). Arbeiten auf dem Männerstrich – Stationen einer
Karriere. Café Strich-Punkt. Vortrag zur Fünfjahresfeier von «Café Strich-
Punkt» in Stuttgart, Stuttgart. Retrieved from http://www.verein-jugendliche.
de/downloads/vortrag_lautmann.pdf
Lemmen, C. & Winhuisen, M. (1985, July 11). Zum beispiel köln:
teenagerprostitution in Deutschland: ‚Das ist doch besser als Klauen, oder?‘.
Stern, 44-45.
Limmer, W. (1980, August 28). Der Lockvögler. Der Spiegel, 155-156.
Limmer, W. (1981, January 19). Unruhiger Docht. Der Spiegel, 153-155.
Logan, T.D. (2014). Economic Analyses of Male Sex Work. In V. Minichiello & J.
Scott (Eds.), Male sex work and society. New York: Harrington Park Press, 106-
147.
Loseke, D.R. (2003). Thinking about social problems: An introduction to constructionist
perspectives (2nd ed.). Social problems and social issues. New York: Aldine de
Gruyter.
maklu

Dangerous and endangered
Maklu 103
‘Mit ‘Safer-Sex’ kann ich keine Kohle machen’: Die Hamburger Stricher-Szene
und die Krankheit Aids (1985, October 14). Der Spiegel, 140-143.
Möbius, T. (1991). Straßensozialarbeit in der Hamburger Stricherszene. In B.
Bader & E. Lang (Eds.), Stricher-Leben. Hamburg: Galgenberg, 45-53.
Morrison, T.G. & Whitehead, B.W. (2007). ‘It´s a Business Doing Pleasure with
You’: An Interdisciplinary Reader on Male Sex Work. Journal of Homoexuality,
53(1/2), 1-6.
Noack, H.-J. (1985, January 28). ‚Plötzlich stirbst du ein Stück weit‘. Der Spiegel,
178-183.
Pfister, A. (2009). Sexarbeiter und HIV/Aids: Karrierewege und HIV-Schutzverhalten
im mann-männlichen Sexgewerbe. Marburg: Tectum-Verlag.
Prase, F. & Thieme, W. (1985, August 22). Bahnhofsviertel Frankfurt: Ein Revier
zeigt sein Gesicht. Stern, 14-16.
Redhardt, R. (1954). Zur gleichgeschlechtlichen männlichen Prostitution. In H.
Bürger-Prinz & H. Giese (Eds.), Beiträge zur Sexualforschung: Vol. 5. Studien zur
männlichen Homosexualität. Stuttgart, 22-72.
Redhardt, R. (1968). Zur gleichgeschlechtlichen männlichen Prostitution. In H.
Bürger-Prinz & H. Giese (Eds.), Beiträge zur Sexualforschung: Vol. 45. Prostitution
bei weiblichen und männlichen Jugendlichen. Stuttgart, 58-107.
Röhl, W. (1984, March 22). Ein Mann wie Frieda. Stern, 28-30.
Röhl, W. (1985, May 9). Panik-Macher. Stern, 29.
Schaper, M. (1985, October 25). Der Mann, der den Tod liebte. Stern, 133-134.
Schetsche, M. (1996). Der Karriere sozialer Probleme: Soziologische Einführung.
München: R. Oldenbourg.
Schetsche, M. (2000). Wissenssoziologie sozialer Probleme: Grundlegung einer
relativistischen Problemtheorie. Wiesbaden: Westdeutscher Verlag. Retrieved
from http://www.worldcat.org/oclc/45541452
Schetsche, M. (2008). Empirische Analyse sozialer Probleme: Das wissenssoziologische
Programm (1st ed.). Wiesbaden: VS Verlag für Sozialwissenschaften / GWV
Fachverlage, Wiesbaden.
Schickedanz, H.-J. (1979). Homosexuelle Prostitution: E. empir. Unters. über sozial
diskriminiertes Verhalten bei Strichjungen u. Call-Boys. Campus Forschung: Bd. 85.
Frankfurt/Main, New York: Campus-Verlag.
Schmidt-Relenberg, N., Kärner, H. & Pieper, R. (1975). Strichjungen-Gespräche:
Zur Soziologie jugendlicher Homosexuellen-Prostitution (Erstausg). Sammlung
Luchterhand: Vol. 188. Darmstadt, Neuwied: Luchterhand.
maklu

Marlen S. Löffler
104 Maklu
Schneider, P. (1985, November 25). ‚Dir müsste man einen Orden geben‘. Der
Spiegel, 238-251.
Scott, J., Callander, D. & Minichiello, V. (2014). Clients of Male Sex Workers. In V.
Minichiello & J. Scott (Eds.), Male sex work and society. New York: Harrington
Park Press, 150-177.
Scott, J. (2003). A Prostitute´s Progress: Male Prostitution in Scientific Discourse.
Social Semiotics, 13(2), 179-199.
Scott, J. & Minichiello, V. (2014). Reframing Male Sex Work. In V. Minichiello & J.
Scott (Eds.), Male sex work and society (pp. XII–XXVII). New York: Harrington
Park Press.
Scott, J., Minichiello, V., Mariño, R., Harvey, G.P., Jamieson, M. & Browne, J.
(2005). Understanding the New Context of the Male Sex Work Industry. Journal
of Interpersonal Violence, 20(3), 320-342.
Sheaffer, R. (2014). Representations of Male Sex Work in Film. In V. Minichiello
& J. Scott (Eds.), Male sex work and society. New York: Harrington Park Press,
50-79.
Spector, M. & Kitsuse, J.I. (1977). Constructing social problems. Cummings series in
contemporary sociology. Menlo Park, CA: Cummings Pub. Co.
Stallberg, F.W. (1990). Das Strichjungenphänomen – Akutalität, Perspektiven,
Befunde. In Katholische Sozialethische Arbeitsstelle e.V. (Ed.), Aktuelle
Orientierungen: Jugendschutz: Vol. 7. Strichjungen. Fakten zur männlichen
Prostitution : Sachverhalte, Hintergründe, Informationen (pp. 17-26). Hamm:
Hoheneck.
Sternsdorff, H.-W. (1983, January 10). Auf dem Weg zum Überwachungsstaat. Der
Spiegel, 46-67.
Timmerberg, H. (July 1981). Die Familie vom Stachus. Stern, 12-14.
Tyler, A. (2014). Advertising Male Sexual Services. In V. Minichiello & J. Scott
(Eds.), Male sex work and society. New York: Harrington Park Press, 82-105.
Umbach, K. (1983, June 6). Klassenkampf als Katzenjammer. Der Spiegel, 174-175.
Weitzer, R. (2005). New directions in research prostitution. Crime, Law & Social
Change, 43, 211-235.
Wörner – ‘der Lächerlichkeit preisgegeben’ (1984, January 30). Der Spiegel, 17-26.
Wright, M.T. & Rosenbrock, R. (2012). Aids – Zur Normalisierung einer
Infektionskrankheit. In G. Albrecht & A. Groenemeyer (Eds.), Handbuch
soziale Probleme (2nd ed.). Wiesbaden: Springer VS, 195-218.
maklu

Maklu 105
Studying police use of force
Definitional challenges and methodological considerations
In this article we scrutinise previous (empirical) research on legitimate use of force by the
police with a focus on methodology. First we describe some definitional challenges each
police scholar in this domain is confronted with. Secondly, we consider the strengths and
weaknesses of different research methods in relation to the study of police use of force. We
examine the following five methods: (1) studying official records, (2) survey-research, (3)
observational studies, (4) interview methods and (5) mixed methods. Although we did
not focus on the research results, we also found an important gap in research on police
use of force. We determined that the police officer’s perspective on and experiences with
use-of-force incidents are mostly neglected. Based on that finding, at the end of the article,
we will introduce our own research project on how Belgian police officers think about the
legitimate use of force and how they handle use-of-force incidents. While referring to the
results of our methodological review we reflect upon the best-suited research design for
our study.
Jannie Noppe1
Introduction
In democratic societies the police are assigned a monopoly on legal violence (West-
ley, 1970), which means they are allowed to use force when they are pursuing a
legal goal that cannot be achieved in another non-violent way (Goossens, 2006).
Following Bittner (1980), police scholars state that this unique characteristic is
the essence of the police function (Birkbeck & Gabaldon, 1996; Garner, Maxwell
& Heraux, 2002; Klockars, 1996). Bittner describes the role of the police by means
of two basic propositions. First of all, he sees the police as ‘a mechanism for the dis-
tribution of non-negotiably coercive force employed in accordance with the dictates of an
intuitive grasp of situational exigencies’ (Bittner, 1980, 46). Secondly, he characteriz-
es the typical circumstances in which the police are called to act as ‘something-that-
ought-not-to-be-happening-and-about-which-someone-had-better-do-something-now’
(Bittner, 1974, 249). According to Brodeur (2010), by means of this dual definition
Bittner developed the core elements of a theory on the police, which are used by
several police scholars. Brodeur (2010) calls it the ‘police-use-of-force’ paradigm.
The key feature of the paradigm is that the use of force is not conceived as force
in action, implying that every routine intervention consists of the actual use of
1 Ghent University, Belgium.
maklu

Jannie Noppe
106 Maklu
force, but refers to a capacity to use force (‘coping with problems in which force
might be used’). Furthermore, Bittner did not presume that the state possesses a
monopoly on violence. He did suppose that police officers have a defeasible and
almost unrestricted capacity to employ force, contrary to all other citizens/profes-
sional groups who can only use force under strictly defined conditions (Bittner,
1980; Brodeur, 2007). On the other hand, Bittner underlined that only by limiting
police force to the minimum needed to deal with a problem, police legitimacy is
maintained (Bittner, 1974; Brodeur, 2007). However, on the street the assessment
of the adequate amount of force needed to de-escalate the situation is left entirely
to the discretion of the police officers, without any concrete criteria that allow to
judge whether a forceful intervention is appropriate (Bittner, 1974, 1980; Brodeur,
2010; Uildriks, 1997).
Westley (1953) was one of the first scholars that examined police use of force.
It took, however, until the mid-sixties/early seventies before the topic drew the
attention of other scholars (Bittner, 1974; Reiss, 1968; Van Maanen, 1974). Since
then, many scholars aimed to find predictors for police use of force (whether legal
or not), and tried to identify in which kind of circumstances police officers do
actually resort to the use of force and which factors are precipitating those actions.
In literature, three approaches are distinguished (Adams, 1999; Friedrich, 1980;
Terrill, 2001). First, studies that follow the individual or psychological perspective,
which means they start from the idea that a police officer’s characteristics,
experiences, views and outlooks can explain the amount of police use of force
(Bazley, Lersch & Mieczkowski, 2007; Kop & Euwema, 2001; Manzoni & Eisner,
2006; Paoline & Terrill, 2007). Those scholars believe that in situations where
officers follow legal and institutional guidelines, the reaction on a typical situation
can be different because of personal traits, experiences, or attitudes. Secondly,
research that follows the situational approach (Alpert, Dunham & Mac Donald,
2004; Crawford & Burns, 2008; Engel, Sobol, & Worden, 2000; Garner et al., 2002)
and thus explains police use of force by looking at factors in the social and physical
context in which use-of-force events take place (for instance characteristics of
the citizen/suspect the police officer is confronted with, the number and type of
persons at the scene, the behaviour of the suspect, presence of weapons, drug
and alcohol use by the suspect). And thirdly, the organizational approach (Alpert
& Mac Donald, 2001; Terrill, Paoline & Manning, 2003), connecting police use of
force with formal (e.g. number of employees, training and supervision policies,
registration policy, departmental philosophy) and informal aspects (e.g. local
police culture) of the police organization. In this last category, one can also include
items external to the organization that shape the work environment, for instance
crime rates and poverty rates (Terrill & Reisig, 2003).
Within the research domain on ‘police use of force’, a large part of the research is
focused on police brutality or excessive use of force (Brandl, Stroshine & Frank,
maklu

Studying police use of force
Maklu 107
2001; Coady, James & Miller, 2000; Geller & Toch, 1996; Reiss, 1968; Skolnick & Fyfe,
1993; Westley, 1970), while less attention has been paid to legitimate, justified and
acceptable use of force (Stenning et al., 2009). Because we are currently engaged
in an ongoing research on the legitimate use of force by Belgian police officers, we
want to map previous research on legitimate police use of force. By doing this, we
aim to get a general overview of what has been done in the past and particularly
want to scrutinise the empirical methods that have been used for studying police
use of force. Although many scholars have reviewed previous research on police use
of force, the focus was mainly on the results (Jobard, 1999; Riksheim & Chermak,
1993; Skogan & Frydl, 2004) or the conceptualization and operationalization of
police use of force (Garner et al., 2002; Garner, Schade, Hepburn & Buchanan,
1995; Klahm IV, Frank & Liederbach, 2013; Terrill, 2014). So with this article we
not only want to establish the baseline for our own research project, by focusing
on the methodology, we also aim to contribute to literature. In their review of
the causes of police behaviour, Skogan and Frydl (2004) also scrutinised empirical
methods for studying police behaviour (including arrest decisions, engaging in
community policing as well as the use of force). They distinguished four methods:
studying police records, direct observations of police in the field, surveys of police
officers and surveys of the public. Unlike Skogan and Frydl (2004), we will not
examine citizen surveys, since we are not interested in citizen’s perceptions of
police behaviour, but we want to know why police officers resort to the use of
force. Without being exhaustive, we will distinguish five categories: (1) studying
official records, (2) survey-research (3) observational studies (4) interview methods
and (5) mixed methods. The majority of research discussed in this article is
conducted in the U.S., in a context that differs largely from other countries. In
addition, each country has different police systems, different laws and customs
with regard to acceptable (and unacceptable) police practices, different levels of
relationships between the police and citizens, which all might influence the use of
force (Stenning et al., 2009). Because of these reasons, it is not possible to simply
transfer results of previous research to other national contexts, such as Belgium.
Therefore, we deliberately choose not to describe previous research findings in
detail here. We will only present some results to illustrate the methodological
strengths and weaknesses of the methods.
However, in the discussion we do want to draw the attention to an important gap
in research on legitimate use of force by the police. In addition to that conclusion,
we will shortly introduce our own research project and reflect upon the most
suitable research design based on the results of the methodological review. But
first we want to address some definitional challenges each police scholar on this
topic is confronted with.
maklu

Jannie Noppe
108 Maklu
Definitional challenges
The first thing that every new researcher in the domain of police use of force
is confronted with is the definitional struggle with the terms ‘coercion’, ‘force’,
and ‘violence’2. Actually, most police scholars treat them as similar (Terrill, 2014).
Nevertheless, we think it is important to make a clear distinction. First of all,
between the terms ‘coercion’ and ‘force’. In Belgium, for instance, the use of
force is considered as one of the means of coercion police officers can appeal to,
besides identity controls, body search and handcuffing (Goossens & Hutsebaut,
2007). So, the term ‘coercion’ is broader than the term ‘force’. However, in
previous research, most police scholars have avoided this difference and only use
an implicit definition, in the form of a list of force types or tactics (Terrill, 2014).
Some of them include other means of coercion, such as handcuffing, others do
not. In other words, for a long time, police scholars did not explicitly conceptualize
what they meant by ‘police use of force’. And if they had a conceptual definition
(e.g. Garner et al., 1995; Terrill & Mastrofski, 2002), the discrepancy between the
conceptual and operational definition led to confusion and blurred the meaning of
their findings with regard to correlates of police use of force (Klahm IV et al., 2013)
This leads us to our second definitional challenge: how to operationalize police
force? When we look at previous research, various operational definitions have
been used in the past (Alpert & Dunham, 2004; Terrill, 2014). Police use of force
has been operationalized as the actual use of physical force (Binder & Scharf,
1980; Cancino, 2001; Engel et al., 2000; Klinger & Brunson, 2009; Uildriks, 1997).
Most of those studies measured force as a simple dichotomy: the absence of force
versus the use of force. However, those studies have been criticized because a
simple dichotomy does not capture the full range of police behaviour (Garner,
Buchanan, Schade & Hepburn, 1996) and in daily practice police officers are using
several levels of force in one and the same situation (Terrill & Reisig, 2003). By
only focusing on physical force nonphysical acts, such as verbal commands and
threats, fall beyond the research scope, while they are also part of the monopoly
on violence and are reflecting the coercive character of police work (Stenning et
al., 2009). More recently, there seem to be a growing consensus amongst police
scholars that forceful acts include physical behaviour as well as verbal threats and
warnings (Terrill, 2014).
Nevertheless, the operationalization of the use-of-force measure still differs
from study to study. In most studies ‘force’ is operationalized by means of a
force continuum, in which levels of force are distinguished based on the extent
of (physical or psychological) damage that is inflicted on the suspect (Terrill &
Mastrofski, 2002). As already mentioned, some studies included verbal threats
2 In Dutch: ‘dwang’, ‘geweld’ en ‘politiegeweld’.
maklu

Studying police use of force
Maklu 109
and unarmed tactics in their measure of force (Garner et al., 1995), while others
included also handcuffing as a level in the force continuum (Terrill & Mastrofski,
2002; Terrill & Reisig, 2003). Some scholars made a clear distinction between
different weapons: the baton, chemical weapons, firearm (Alpert & Mac Donald,
2001; Bazley et al., 2007; Crawford & Burns, 2008; Manzoni & Eisner, 2006). In
addition, a number of studies operationalized the use of force in relation to the
level of suspect’s resistance (Alpert & Dunham, 1997; Alpert & Dunham, 2004;
Terrill, 2001). As shown in table 1, for each actor (police officer and suspect) they
distinguished different levels of force. According to Mc. Laughlin (as cited in
Terrill, 2001) the force continuum, unlike legislation, clarifies which type of force
may be used legitimately in a given situation. Furthermore, it allows researchers to
search for explanations for dominant use of force3 against accommodating force4.
Table 1. An example of a force continuum
Officer Force Suspect’s Resistance
1. Presence 1. Cooperative
2. Strong order 2 Verbal resistance
3. Defensive force 3. Defensive resistance
4. Offensive force 4. Active resistance
5. Intermediate weapon 5. Non-deadly weapon
6. Deadly force 6. Deadly weapon
Source: Alpert, Dunham & Mac Donald, 2004
In practice, however, as Terrill noticed (2014), most scholars adapt their operational
definition of force to the type of data they have at their disposal. For instance, when
the dataset consists of observational data, a broader definition (including verbal
forms of force) can be used in comparison with official data, in which verbal forms
of force are mostly not included. So in the latter, researchers have to narrow their
definition.
Finally, we want to clarify the difference between the terms ‘force’ and ‘violence’.
According to Fassin (2013) this distinction is essentially normative: ‘The police must
know how to use force, but must not exercise violence’ (Fassin, 2013, 126). One speaks
of police violence when the use of force is unjustifiable, referring to the principles
of legality, proportionality and subsidiarity, which means that those acts could lead
to a court conviction. However, Fassin (2013) also warns for blindly following the
3 Operationally defined as ‘relative force greater than the suspect’s resistance’ (Alpert et al., 2004, 476).
4 Operationally defined as ‘a level of force equal to or less than the level of suspect’s resistance’ (Alpert et
al., 2004, 476).
maklu

Jannie Noppe
110 Maklu
judicial interpretation, which limits violence to the excessive use of physical force,
and does not take into account moral violence, by which he means assaults on the
personhood or dignity of the victim. When we look at previous research, to a great
extent, studies followed the judicial definition and were focused on excessive use of
physical force (Brandl et al., 2001; Coady et al., 2000; Skolnick & Fyfe, 1993; Terrill
& Reisig, 2003; Worden, 1996). Although some of these studies are discussed here
because they compare excessive use of force with justifiable use of force, we do not
want to focus on excessive use of force in this article, since this is not the scope of
our research. So to be clear, in this article, the term ‘force’ refers to legitimate use
of force. We included in our review studies that examined verbal as well as physical
levels of force.
Methodological considerations
In the following part, we will examine previous research on police use of force.
We searched for relevant studies through different channels. We went through
databases like Google Scholar and Web of Science with the search term ‘police use
of force’. But we also examined bibliographies of other articles in search for relevant
studies. Important criteria to include studies into our literature review were: (1)
it concerned empirical research (2) published in English or Dutch, (3) in which
scholars tried to identify why police officers do actually resort to the use of force
and which factors are precipitating those actions (4) with the focus on ‘legitimate’
use of force (except for some studies in which legitimate as well as excessive
use of force are discussed). When selecting the articles we especially looked for
publications who paid attention to methodological strengths and weaknesses of
their own or other studies on police use of force. We did not conduct a systematic
review, so we are aware of the fact that the list of studies included in this review is
not exhaustive. Nevertheless, we think we were able to map major strengths and
weaknesses of each method in relation to the study of police use of force.
The majority of the research that is discussed, was conducted in the United States,
with the exception of some studies in Latin-America, the Netherlands, Germany,
England, Australia, and Switzerland.
Studying official data (use-of-force reports and citizen complaints)
A large part of the research on police use of force is based on the study of official
use-of-force reports (Alpert & Dunham, 2004; Alpert et al., 2004; Bazley et al., 2007;
Morabito & Doerner, 1997; Timmer, 2005; Wolf, Mesloh, Henych & Thompson,
2009). In many U.S. police departments officers are obliged to fill in a use-of-
force report each time they use force against a civilian. In the report, they have
maklu

Studying police use of force
Maklu 111
to describe the circumstances of the event and the nature of force that was used
(Adams, 1999; Terrill, 2001).
In that way, official police records provide relatively structured data on a broad
range of use-of-force incidents, over a broad spectrum of police behaviour, and
over a longer period in time (Garner et al., 2002). Based on those reports, as shown
in the summarizing table in annexe, police scholars have tried to get insight into
the prevalence and frequency of different levels of force and have tried to explain
the relationship between the level of force used by the police officer and the
behaviour of the suspect (level of resistance). Some studies (Alpert & Dunham,
2004; Bazley et al., 2007; Wolf et al., 2009), for instance, found that in the vast
majority of cases the police officer maintained a level of force that was very close
to the level of resistance given by the suspect. Most of the extreme deviations were
cases in which the police officer deployed less force than the level of resistance
they encountered from suspects. However, those results must be interpreted with
caution because there may be some bias of the reporting officer, since input from
the other actor(s) involved in the encounter is neglected. Furthermore, less harsh
levels of force are underreported (e.g. verbal commands, threats, handcuffing and
takedowns), because data are only gathered on specific types or high levels of force,
and not the full range of force levels. In addition, police departments may differ in
the kinds of force they require officers to report, which offers some limitations for
generalisation and comparison across agencies (Alpert & Dunham, 2004; Bazley et
al., 2007; Wolf et al., 2009). In other words, this research method is more useful for
studying evolution in the use of force within one jurisdiction than for comparison
among jurisdictions (Garner et al., 2002).
Besides that, when studying official records, one should keep in mind some
general weaknesses of this kind of data. Firstly, the data is subject to selection
bias, since all police-suspect encounters where force occurred yet no arrest was
made are left out (Crawford & Burns, 2008; Garner et al., 2002; Manzoni & Eisner,
2006). Secondly, as with most official data, it concerns quantitative information
reported by one party, that does not (always) include the rationale for the (use-of-
force) decision (Jobard, 1999). And finally, and not unimportant, all depends from
the availability of data.
In Belgium, for instance, it is hard to find reliable statistics on police use of force.
Although the GPI625 prescribes that each incident where force is used, whether
or not police weapons and intervention techniques or tactics are deployed, has to
be officially reported, little is known about the prevalence and frequency of the
legitimate use of force by Belgian police officers. In practice, incidents are not
5 Circular GPI62 of February 14, 2008 concerning the weaponry of the integrated police, structured
on two levels, B.S., February 29, 2008.
maklu

Jannie Noppe
112 Maklu
systematically reported and if reported, different reporting forms are used, with the
result that at this moment no consistent national data is available. As alternative
one could examine citizens’ complaints, that are registered on a national level by
several monitoring committees. Yet they only give an indication of cases in which
excessive force was used6.
Survey research
Conducting (paper-and-pencil) surveys amongst police officers is a second
method that has been used for studying police use of force (Alpert & Mac Donald,
2001; Crawford & Burns, 2008; Garner et al., 1996; Garner et al., 2002; Holmes,
Reynolds, Holmes & Faulkner, 1998; Manzoni & Eisner, 2006; Norris, Birkbeck
& Gabaldon, 2006). In the studies of Garner et al. (1996, 2002), police officers
in different districts were asked to complete a one-page survey each time they
made an arrest during a 2-week period. In this form specific behavioural aspects of
the arrests and information on police mobilization, the nature of the offense, and
officer and suspect characteristics were recorded in a standard manner. Manzoni
and Eisner (2006), on the other hand, collected their data by means of a written
mail survey, sent to the home address of the police officers. Since surveying is a
more direct way of gathering information, in comparison with for instance official
records, survey data is less subject to selection bias, resulting in more complete
data on different kinds of force (Garner et al., 2002). Survey research, for instance,
also provides insight into the prevalence and frequency of less severe levels of
force, such as oral commands and threatening with the use of force. Most studies
in this type, in fact, conclude that the use of weapons and injury-producing tactics
are used only exceptionally and that less severe types of force are occurring more
commonly (Crawford & Burns, 2008; Garner et al., 2002; Manzoni & Eisner, 2006).
However, reliability of these results can be questioned because of reporter bias.
It is very likely that police officers do not report minor physical force or force
believed to be illegitimate (Manzoni & Eisner, 2006), which can lead to a strong
underestimation of police use of force. Another threat for the reliability of survey
data is the fact that police officers are questioned about experiences in the past
which makes them susceptible to post-hoc rationalisation (see infra).
In addition, the results of survey-research can vary dependent on the definition
of force that is used and on the timeframe being referenced. Less events will be
reported when defining police use of force as physical force, contrary to a definition
that includes verbal as well as physical force. With regard to the timeframe, it is
also obvious that questions about lifetime experiences will capture more incidents
than questions about the past year (Adams, 1999). The studies of Garner et al.
6 The external monitoring service Committee P reports, in the period 2009-2012, 39 cases (of the 91
proceedings that were instituted) in which the police officer is found guilty of police violence.
maklu

Studying police use of force
Maklu 113
(1996; 2002) also demonstrated that the influence of situational characteristics can
vary substantially depending on the measure of force that is used. They examined
the prevalence as well as the severity of force and found, for instance, that if the
location was known for criminal activity, if the suspect was intoxicated, or if the
intervention was officer-initiated the prevalence of force increased, but not the
severity. Factors that did increase the severity of force were a higher number
of suspects or the presence of bystanders that were unknown to the suspect.
Besides the influence of the force measure, they also emphasize the importance
of controlling for the level of suspect’s resistance. For a number of situational
characteristics (e.g. nature of the offense, suspect’s race), the effect decreased or
significance disappeared after including suspect’s resistance in the analysis.
Some scholars used a specific kind of survey to examine police use of force. Holmes
et al. (1998), for instance, used a factorial survey to explain variation in officers’
perceptions of the threat level or risk inherent in a police-citizen encounter. In
the factorial survey approach vignettes are used to study ‘the underlying principles
behind human judgements (or evaluations) of social objects’ (Rossi & Anderson, as
cited in Wallander, 2009, 2). Holmes and colleagues (1998) for instance found that
important predictors of the appropriate amount of force that officers believe to
be reasonable are: the level of threat they perceive, the number of warnings they
would issue before using force, the level of resistance, the gender of the suspect,
the offense severity and the number of officers present at the scene.
Observational studies
A third, and frequently used method for studying individual officer decision-
making with regard to the use of force, is the observational method (Skogan
& Frydl, 2004). A great number of studies that used the observational method
for studying police use of force used social systematic observations (Mastrofski,
Parks & Mc Cluskey, 2010), which means that observations are conducted by
multiple researchers and according to explicit procedures (clarified by means of
a questionnaire or a logbook) that permit applying the logic of scientific inference
and replication and make effects of observing and measuring measurable (Reiss,
1971). However, many scholars completed their observational data with data
retrieved by interviews, a survey or official records (see infra). In this paragraph
we only discuss the research that solely used observations.
Corresponding with results mentioned earlier, Bayley and Garofalo (1989) found
that force was only exceptionally occurring, by as well as against police officers.
Only in 37 cases (about 8%) of the 467 police-citizen encounters (classified ‘as
potentially violent’) they observed that some form of physical force was used by the
police officers. Furthermore, they found that officers became involved in forceful
maklu

Jannie Noppe
114 Maklu
incidents7 about once every 81/3 working days. It mainly concerned grabbing,
pulling and pushing or restraining. Besides that, explanatory analyses illustrated
that existence of conflict at arrival, possession of a weapon by a citizen, and use of
obscene or insulting remarks or gestures by a citizen influenced police use of force.
However, since they only had 37 cases in which force was used, no substantive
significant effects could be distinguished. Engel et al. (2000) confirm the influence
of a suspect’s demeanour on the use of force, but could not determine significant
interaction effects for e.g. sex/demeanour, race/ demeanour. And Friedrich (1980)
who reanalysed the observational data of Reiss (1966) concluded that police use
of force varies with a variety of individual, situational and organizational factors.
His final multiple regression analysis, however, showed that police use of force
depends primarily on how the offender behaves and whether or not other citizens
and police are present. The background characteristics of the officer as well as the
citizen seem to make no difference.
Police scholars address several strengths of the observational method. First of all, it
allows a more thorough and detailed examination of how an encounter can result
in use of force by the officer or the suspect, and how both parties interact with each
other (Alpert & Dunham, 2004; Terrill, 2001). Secondly, since observational data is
not gathered for organizational purposes, unlike for instance use-of-force reports,
they are more valid. Observational studies allow the researcher to collect more
complete and less biased information on both the formal and informal setting in
which the police action takes place (Skogan & Frydl, 2004). Furthermore, it gives
the researcher the opportunity to experience first-hand forceful police practices,
so there is no bias of the reporting officer. In this way, field observations allow a
researcher to get insight into unreported instances of police use of force (Worden,
1996) and police behaviour is measured more independently, as opposed to official
records, surveys or interviews (Garner et al., 2002; Terrill, 2001). On the other
hand, there may be observer bias (Morabito & Doerner, 1997). A particular source
of observer bias is the risk of going native, meaning that the observer adopts the
perspective of his research subjects (Mastrofski et al., 2010). Therefore, in case
of social systematic observations, the observers are neutral third parties, who are
trained to remove themselves from the target, as such facilitating the testing and
improvement of the reliability of observations (Terrill, 2001; Terrill & Reisig, 2003).
Another observer error that partially undermines the validity of this research
method is the risk of reactivity. Namely, it is quite conceivable that police officers
will act differently when they are accompanied by a researcher who intends to make
a detailed study of the police officer’s actions (Alpert & Dunham, 2004; Worden,
1996). Another concern with regard to validity is the limited generalizability of
results obtained by observational methods (Alpert & Dunham, 2004; Terrill, 2001).
Since police use of force is a rather infrequent event (Bayley & Garofalo, 1989;
7 Incidents in which either they or members of the public used physical force against each other.
maklu

Studying police use of force
Maklu 115
Friedrich, 1980; Worden, 1996), which is definitely the case for high levels of force,
observation is also a very time-consuming method, which makes it really hard to
obtain a reasonable sample of use-of-force incidents for analysis (Adams, 1999;
Alpert & Dunham, 2004; Morabito & Doerner, 1997; Terrill, 2001). Police scholars
are, therefore, often concentrating on high-risk precincts and high-risk shifts
(Adams, 1999) or are using systematic social observations conducted by multiple
independent observers (Bayley & Garofalo, 1989; Engel et al., 2000; Paoline &
Terrill, 2007; Terrill, 2001; Terrill & Mastrofski, 2002; Terrill et al., 2003).
Finally, although systematic social observations have often been used to study how
police officers deal with the tension between formal standards and standards that
issue from the occupational culture, the method is not well-suited for studying
internal thought processes, motivations or judgments (Mastrofski et al., 2010).
As Mastrofski et al. (2010) suggest, for this kind of research questions, a mixed
approach may be considered. In the observational study of Mastrofski and Parks
(1990), for instance, police officers were asked immediately after an encounter
to describe what they perceived, thought and felt during the encounter. Those
comments were very interesting for the researchers because they gave more
insight into how judgmental issues of problem-solving and personal moral issues
were resolved and how important reference groups (such as the department’s
formal hierarchy, the officer’s peer group and the community) influenced their
decision. This mixed approach allowed the researchers to gather objective as well
as subjective data on police behaviour.
Interview methods (single, in group, focus groups)
As far as we know, interview methods have not often been used as a sole method
to study police use of force. More often interviews are an additional part of a
larger research project with multiple research methods (see infra). However, the
studies that used interview methods are focusing on a different kind of research
questions, in comparison with the research discussed above. More particularly,
they are examining how police officers experience and think about the use of force
and how they justify use-of-force decisions. In other words, they start from the
idea that besides quantitatively analysing the relationship between individual,
situational and organizational characteristics and the use of force by means of
official data or observational data, one should also scrutinise how police officers
perceive those characteristics and which characteristics they use to justify the use
of force.
The study of Birkbeck and Gabaldon (1996), for instance, in which interviews
were conducted with supervisory police officers in Venezuela, illustrated that
tactical rules used by police officers on the street differed significantly from legal-
maklu

Jannie Noppe
116 Maklu
administrative rules set out in the department. They suggest that, in the first place,
the nature and dynamics of citizens’ complaints shape police officers’ prescriptions
and justifications for the use of force. Another example is the international research
project on the use of force by the police (Baker, 2009; Klukkert, Ohlemacher
& Feltes, 2009; Stenning et al., 2009; Waddington et al., 2009), in which police
officers in six different countries8 were asked, during focus groups, to talk about
a particular scenario that describes an attempt to arrest two youths. The study
indicated that police officers’ perceptions of the situation influence the use of force
strongly and serve as an important justification ground afterwards. The researchers
found that police officers, indeed, take into account individual, situational and
organizational factors in a cost-benefit analysis of the decision to act. However,
the degree of rationality depended on the officer’s subjective assessment of the
situation and on general conditions (as legality, proportionality, subsidiarity).
Besides that, by means of interviews, Klinger and Brunson (2009) determined that
police officers are experiencing several perceptual distortions, such as a reduced
visual and cognitive reaction time; a temporary diminished sense of hearing and
sight (tunnel vision), when they are discharging their firearm9.
As research results above show, qualitative data obtained from interviews or focus
groups, are more direct and richer compared to the more abstract quantitative data.
By means of interviews or focus groups, the researcher obtains a richer picture of
the subject of the study, in our case police use of force, through colourful language,
anecdotes and jokes, and in case of focus groups through the conversations and
discussions between the participants. According to Stenning et al. (2009) police
use of force is, like any violent encounter, more than an aggregation of attitudinal
dispositions, but is rather a complex interplay of considerations and interacting
participants. Therefore, focus groups are more suitable (contrary to for instance
attitudinal scales) to get insight into the complexities of a real encounter in which
force might be used. However, we should also keep in mind that focus groups are
not that well-suited for studying highly sensitive topics. For instance, since the
police world is still considered to be a macho-world, we might expect that police
officers are less likely to admit in focus groups (in front of their colleagues) that
acute stress and especially fear play a part in their decision making process. One-
on-one interviewing may be a better method to address these kinds of sensitive
topics.
Another weakness of interviews as well as focus groups is that you only get a
view on the individual’s own accounts. By which we mean that the rationales
8 Australia, Brazil, England, Germany, the Netherlands and Venezuela.
9 Similar results were found in research with experimental designs (Akinola & Mendes, 2012;
Nieuwenhuys & Oudejans, 2010, 2011; Nieuwenhuys, Savelsbergh, & Oudejans, 2012; Taverniers,
Smeets, Van Ruysseveldt, Syroit, & von Grumbkow, 2011). Since those studies are explicitly focused
on shooting decisions, and not on less severe levels of force, we do not discuss them in detail here.