Policing the absence of the victim: An ethnography of raids in sex trafficking operations

Abstract

For this chapter, ethnographic methods (participant observations) have been applied to analyse police raids in areas of commercial sex work as a standardised operating procedure with the purpose of identifying victims of human trafficking. The aim of this chapter is twofold: first, it shows how local policing practices revolve around different and rather contrasting forms of subjectivities – the absent victim and the problematic sex worker –, and second, it indicates how police officers perform their role as saviour and upholder of the social order in their daily practices.

Full text

Policing the Sex Industry
Protection, Paternalism and Politics
Edited by Teela Sanders
and Mary Laing

First published 2018
by Routledge
2 Park Square, Milton Park, Abingdon, Oxon OX14 4RN
and by Routledge
711 Third Avenue, New York, NY 10017
Routledge is an imprint of the Taylor & Francis Group, an informa business
© 2018 selection and editorial matter, Teela Sanders and Mary Laing;
individual chapters, the contributors
The right of Teela Sanders and Mary Laing to be identied as the authors of
the editorial material, and of the authors for their individual chapters, has been
asserted in accordance with sections 77 and 78 of the Copyright, Designs and
Patents Act 1988.
All rights reserved. No part of this book may be reprinted or reproduced or
utilised in any form or by any electronic, mechanical, or other means, now
known or hereafter invented, including photocopying and recording, or in any
information storage or retrieval system, without permission in writing from
the publishers.
Trademark notice: Product or corporate names may be trademarks or
registered trademarks, and are used only for identication and explanation
without intent to infringe.
British Library Cataloguing-in-Publication Data
A catalogue record for this book is available from the British Library
Library of Congress Cataloging-in-Publication Data
A catalog record for this book has been requested
ISBN: 978-1-138-71662-9 (hbk)
ISBN: 978-1-315-19689-3 (ebk)
Typeset in Times New Roman
by Apex CoVantage, LLC

Contents
List of contributors ix
Acknowledgements xiii
Introduction: policing the sex industry: tackling exploitation,
facilitating safety? 1
TEELA SANDERS AND MARY LAING
PART I
Protection through policing: plurality and pragmatism 17
1 Policing sex work in Britain: a patchwork approach 19
ALEX FEIS-BRYCE
2 Trans sex workers in the UK: security, services and safety 39
MARY LAING, DEL CAMPBELL, MATTHEW JONES
AND ANGELIKA STROHMAYER
3 Beyond hate: policing sex work, protection and hate crime 53
ROSIE CAMPBELL
4 Decriminalisation, policing and sex work in New Zealand 73
LYNZI ARMSTRONG
5 ‘Notinourname’:ndingsfromWalessupporting
the decriminalisation of sex work 89
TRACEY SAGAR AND DEBBIE JONES

viii Contents
PART II
Policing Operations, Enforcement and Austerity 107
6 Policing the absence of the victim: an ethnography
ofraidsinsextrafckingoperations 109
JULIA LESER
7 Trafcking,pimping,sexworkandthepolice:street
prostitutes’ perceptions in Las Vegas 127
ANDREW L. SPIVAK
8 Condoms as evidence, condoms as a crowbar 143
SYNNØVE JAHNSEN
9 Gentricationandthecriminalizationofsexwork:
exploringthesanitizationofsexworkinKingsCross
with the use of ASBOs and CBOs 159
LUCY NEVILLE AND ERIN SANDERS-MCDONAGH
Index 175

Julia Leser
Introduction
It is another sunny morning in July in a small German city somewhere in the mid-
dle of the country. Two vice squad police ofcers, having entered the apartment of
Sumi and Lulu, are now sitting at their living room table, copying personal details
of Sumi’s passport into a data sheet. A third ofcer is standing in front of Lulu’s
room, knocking impatiently. ‘Sorry, I’ll be out in ve minutes’, Lulu says. She
is still working with a client. In contrast, Sumi, a Thai woman in her late fties,
seems to have had a rather quiet morning right until the police stopped by. Now
she is joking with the ofcers and telling them about her favourite papaya recipes.
The ofcers seem to be hesitant to look at Sumi in her sexy nightgown. As they
nish copying Sumi’s passport and stand up, Sumi slaps the ofcers’ bottoms.
Both of them blush a little, giggling their good-byes on their way out.
Obviously, being entertained by Sumi has not been the vice squad’s main
objective for visiting the sex workers’ apartment today. Rather, Sumi and Lulu’s
place was just one of their destinations on the raid the vice squad is conducting
on a day in July 2015. The vice squad performs these raids regularly upon vari-
ous sites in which prostitution takes place, searching – of course – for victims of
human trafcking.
We all have heard of trafcking. We do seem to know what a victim of trafck-
ing is and what she looks like: the characteristics of the innocent, mostly young
and migrant women are adding up to constitute the ideal victim (Christie 1986).
Further, we seem to know that trafcking is a crime conducted upon young girls
who are forced into prostitution – so do the police, and that is why this particular
vice squad is looking for victims of trafcking in brothels and sex workers’ apart-
ments on this summer morning.
In this chapter, the victim is understood as a relational ‘object of knowledge’
(Bacchi 2012) that emerges from practices and that can be analysed in regard to
certain theoretical premises. First, knowledge is always already situated and local-
ised (Hacking 1999). Its differing framings and contexts affect the way we come
to understand certain situations and act upon this emerging knowledge. In the case
of human trafcking, it has become clear that anti-trafcking campaigns and ini-
tiatives tend to frame phenomena of irregular migration and sex work within the
context of organised crime, violence and exploitation of women, resulting in the
6 Policing the absence
of the victim
An ethnography of raids in sex
trafcking operations
Julia Leser

110 Julia Leser
subjectication of primarily Eastern European women as ‘victims of trafcking’
(Berman 2003). Many authors have described and analysed the ‘ideal victim’ dis-
course that produces these ‘victims’ as young, innocent and naively acting women
(Blanchette et al. 2013, Doezema 2000; Jakšić 2008; Pates and Schmidt 2007).
In media representations, these discourses are condensed into specic narrative
patterns that are repeated and reproduced continuously. Snajdr (2013) has termed
these patterns ‘master narratives’ of human trafcking that draw upon specic and
analysable techniques of moralisation and emotionalisation. Second, knowledge
production is ubiquitous and central to state practices: gathering knowledge about
its subjects and representing it in policies, laws, maps and documents is a sine qua
non for the state’s exercise of power and control over its subjects, as it has been
frequently underscored by Foucault. In this case, the state has to know the victim
of trafcking in order to govern it accordingly. Third, as we know from science
and technology studies, knowledge is never exclusively objective, rational nor
universal – but quite the contrary: the pathway of producing knowledge is deter-
mined by (competing) practices, interactions, interferences, errors, irrationalities,
discontinuities, inconsistencies and failures. In this process, subjects and objects
are ‘by-products’ or ‘provisional results’ (Latour 2007: 94) and objectivities are
always plural.1 Fourth, and to introduce and apply a term proposed by Mariana
Valverde, ‘bodies of knowledge’ are always dynamic compositions of different
elements (e.g. expert and everyday knowledge) and therefore ‘hybrid’ (Moore and
Valverde 2000), rendering the knowledge production process – once again – plural
and heterogeneous (Valverde et al. 2005). In sum, the victim can be understood as
a relational object of knowledge, which means that the so-called victim of human
trafcking (in this case) is an emerging composition of various and partly compet-
ing institutional logics, bodies of knowledge and their relations. What a victim is
(and what is not) is thus negotiated and enacted in knowledge practices.
In state institutions (such as the police), these practices actualise in the form of
categorisation and classication of women in sex work.2 Through standardised
operating procedures, but always depending on the police ofcers’ framing, pur-
pose and motivations, the domain of the victim is constituted and dened. Further,
the practices of policing reect upon the self-conception of the intervening state
of law: especially within the area of human trafcking, those regulating practices
and techniques intertwine with questions of migration control, moral conceptuali-
sations of sex and sex work, the task of securing social order, as well as sanction-
ing violations to fundamental rights. As recent ethnographic studies of the state
have shown, the state is essentially reproduced in state ofcials’ practices (Sharma
1 With regard to the state’s knowledge production practices, Miller & Rose illustriously articulated:
‘Governing is not the ‘realization’ of a programmer’s dream. The ‘real’ always insists in the form of
resistance to programming; and the programmer’s world is one of constant experiment, invention,
failure, critique and adjustment’ (Miller & Rose 2008, 39).
2 The fact that states need to know their subjects in order to function and therefore have to classify
their populations has been described by James Scott with the concept of ‘legibility’: ‘Legibility is a
condition of manipulation. Any substantial state intervention in society [...] requires the invention of
units that are visible. [...] Whatever the units being manipulated, they must be organized in a manner
that permits them to be identied, observed, recorded, counted, aggregated, and monitored’ (Scott
1998, 183).

Policing the absence of the victim 111
and Gupta 2009; Hansen and Stepputat 2001; Schlichte and Migdal 2005). Thus,
an ethnographic study of police raids can allow for a more complex account on
how policing practices contribute to enacting state-ness and co-producing certain
subjectivities – always within the scope of order and disorder.
For my analysis, I have used ethnographic methods (participant observations)
in order to retrace policing practices in the eld of human trafcking: how do the
police come to know what a victim of trafcking is? During raids, how do they
perceive specic situations and what do they make of it? What instances raise
their suspicion? How do they legitimise their practices? I conducted participant
observations within a vice squad in a German city for six months in 2015 and
observed their raids in sex trafcking operations within that period of time, com-
plementing my eld protocols with 25 qualitative semi-structured interviews with
various policing agents, administrative ofcials and social workers who work in
specialised counselling centres for ‘victims of human trafcking’.3 But I began
my research with a striking paradox: the German city I chose for my inquiries
seemed to have had, quite in contrast to it being a relatively small city, a massive
problem with human trafcking – the city accommodated three specialised coun-
selling centres for human trafcking victims, a specialised police unit (whose
head of department was invited to participate in talk shows on TV on the matter of
trafcking), as well as an institutionalised round table with various state ofcials
that had been implemented in order to ght the crime of trafcking. In short, the
situation was highly problematised, but on taking a closer look into the ofcial
crime statistics, I noticed that the numbers did not match the alleged high scale
proportions of the issue. In fact, the last case of trafcking that led to the convic-
tion of the accused in front of a court of justice dated back to 2013 – two years
before the time I started my research. For six months in 2015, I accompanied the
vice squad ofcers on their raids upon sites of prostitution, searching for victims
of trafcking. But as I did, no victim was ever found, and I started to become
aware that I was witnessing the peculiar and highly paradoxical practice of polic-
ing the absence of the victim.
The legal and organisational framework for policing
trafckinginGermany
In order to frame the main results of my ethnographic inquiries, I shall start with
a few remarks about the German legal and organisational arenas that embed the
state’s handling of the issue of trafcking. The German legislation concerning
human trafcking has been readjusted in 2005 in accordance to the European Pro-
tocol to Prevent, Suppress and Punish Trafcking in Persons, especially Women
and Children (UNODCCP 2000). Until this legislative reform became ratied,
the German law sanctioned human trafcking solely in the form of coercion of
human beings into prostitution. In 2005, for the rst time, the trafcking of human
3 The data I draw on in this chapter have been collected as part of the DFG/ANR-funded research
project ‘Institutionalising Human Trafcking’ that has been conducted in France and Germany from
2014 until 2017 under the direction of Mathilde Darley and Rebecca Pates. The main ambition of
this project is to question how the category of the ‘victim of human trafcking’ is being applied and
translated into local institutional practices.

112 Julia Leser
beings for labour exploitation was punishable by law as well, as the denition of
trafcking had been broadened and moulded into two distinct legal paragraphs:
§232 Criminal Code for human trafcking for the purpose of sexual exploita-
tion, and §233 Criminal Code for human trafcking for the purpose of labour
exploitation. In this chapter, I concentrate on the practices of identifying victims
of sexual exploitation, which have been far more often identied as such than vic-
tims of labour exploitation. While legal measures against labour exploitation have
been taken, the number of cases for labour exploitation recorded with the national
crime statistics is far lower than the number of cases regarding sexual exploita-
tion. In 2013, the Federal Criminal Ofce (Bundeskriminalamt, BKA) nalised
425 cases of human trafcking for the purpose of sexual exploitation, while it
nalised only 53 cases of human trafcking for the purpose of labour exploitation.
This does not necessarily reect the actual situation concerning forced labour,
exploitation and trafcking. In fact, trade unions generally estimate that the num-
ber of persons who have experienced trafcking into labour exploitation outside
the sex industry by far outnumber those who have experienced trafcking within
prostitution.4 Still, institutional agents like the police strongly focus on victims of
trafcking into prostitution within their framing and their practices.
In Germany, there are no national guidelines for the identication of victims
of human trafcking (GRETA 2015). To formally identify a victim, law enforce-
ment agencies themselves are responsible. In some of the German states, law
enforcement agencies and counselling centres for victims of human trafcking
have arranged formalised cooperation agreements that help clarify the routines
and procedures of identifying victims. In order to do that, local and state police
ofcers, representatives of health and social service agencies, and support service
providers or counselling centres have installed round table meetings at communal
and state levels, where issues such as law implementation, victim identication
and responsibilities are discussed on a regular basis.5
In Germany, different police forces are engaged in the enforcement of anti-
trafcking laws. The Federal Criminal Police Ofce (Bundeskriminalamt, BKA),
and the Criminal Police Ofces of the 16 states (Landeskriminalämter, LKA),
investigate trafcking cases within their departments for organised crime. The
BKA engages in the investigation and prosecution of transnational organised
crime networks, often cooperating with EUROPOL and INTERPOL as well as
with other countries’ police agencies. It also offers training for local police of-
cers on human trafcking and victim identication. The LKA serves as a superor-
dinate agency for the criminal investigation departments that leads investigations
within cities and communities. If a local police force is engaged with a trafcking
case that goes beyond their range of operations, i.e. that is connected to victims
or perpetrators in other communities, states or even countries, the LKA (and, in
international cases or those spanning various German states, BKA) gets involved.
Generally though, the processing of human trafcking cases starts either with
counselling centres to whom sex workers report as trafcked victims or within
4 In our research project ‘Institutionalising Human Trafcking’ we have analysed several reasons for
this in Pates et al. (2016).
5 Regular does mean frequent. In most instances, round table meetings occur once or twice a year.

Policing the absence of the victim 113
local criminal investigation departments. In interviewing police ofcers, they
emphasise that the local authorities work within clearly marked jurisdictions,
geographically as well as organisationally.6 Human trafcking here is not located
within the sub-department of organised crime (as it is in BKA and LKA), but in
most cities within the sub-department of the vice squad. In the city that I have
been researching, for example, the vice squad is responsible for all kinds of sexual
offences, child pornography, surveillance of registered sex offenders and all kinds
of offences related to prostitution. Human trafcking, for the vice squad here,
belongs to the latter. As police departments are always limited in their resources,
there is only one ofcer in this vice squad who manages the task of investigat-
ing all kinds of offences related to prostitution. The organisation of local police
ofces seems strictly hierarchical, with all its routines in order and all responsi-
bilities neatly distributed. But those structures will change and adapt due to crimi-
nogenic trends and discourses, shifts in criminal policy and legislation, as well as
innovation in investigation technologies, as the interviewees indicated.
Routines, standards, and ‘doing the state’
So, what do vice squads do in order to ght human trafcking? The very rst thing
that will come to mind (if one has watched too many crime thrillers and/or read
a certain amount of detective stories) is the practice of raiding brothels. In media
representations, ‘[r]aids personify the master narrative of trafcking by casting
state agents in the role of saviours rescuing victims from criminals’ (Hill 2016:
44). Hill argues that the media usually covers a raid as a ‘spectacular performance
of state power that purports to show evidence of trafcking and the need for pro-
active policing’ (Hill 2016: 45). Raids, in this perspective, might be associated
with a large number of police ofcers in full gear storming into a badly lit brothel,
saving young girls from the hands of those ruthless men who exploit them and
thus making the world a much better place. To foreclose, the reality is by far less
spectacular. If we want to look into police practices as a means of ‘doing the
state’, we have to acknowledge that police work is done in the rst place through
the lling out of forms and that ‘‘catching bad guys’ takes up a small fraction of
police forces’ time’ (Valverde 2010: 76). Police ofcers are among other things
‘street-level bureaucrats’ (Lipsky 1980) and police forces are one of the bureau-
cratic organisations that make up the state in its organisational form (Bierschenk
2016: 155). The state, as we know from Migdal and Schlichte,
is a eld of power marked by the use and threat of violence and shaped by
1) the image of a coherent, controlling organization in a territory, which is a
representation of the people bounded by that territory, and 2) the actual prac-
tices involving those stafng its multiple parts and those they engage in their
roles as state ofcials.
(Migdal and Schlichte 2005: 15)
6 City states (Hamburg, Berlin, Bremen) are the exception. Here, an LKA unit is responsible for any
case of human trafcking.

114 Julia Leser
The state is thus not a homogeneous entity but a conglomerate of various state
agents, their practices and knowledges, which do not have to match actual ideas of
the state. The same is true for the police: we can distinguish between certain ideas
about the police (such as the popular representations of police raids stated above)
and – more or less – daily police practices. For Bierschenk, police practices are
always ‘guided by common sense, localised, verbally derived and relayed back-
ground knowledge as well as by their accumulated professional experience’ (Bier-
schenk 2016: 166). And:
Police ofcers are not only law enforcement ofcials, but also protectors of
the social order, educators of society, relationship operators and knowledge
workers. Likewise, the description of the police as a (state) bureaucratic orga-
nization only applies partially. Rather, the daily functioning of the police and
daily police work are characterized by a tension between formal norms and
informal practices, a tension which applies to all state organizations.
(Bierschenk 2016: 170–171)
What distinguishes the police from other state organisations is the monopoly of
the use of the force of law, or, differently put, their ‘legal lawlessness’ (Brodeur
2010: 9). Brodeur denes the police as follows:
Policing agents are part of several connected organizations authorized to use
in more or less controlled ways diverse means, generally prohibited by stat-
ute or regulation to the rest of the population, in order to enforce various
types of rules and customs that promote a dened order in society, considered
in its whole or in some of its parts.
(Brodeur 2010: 130)
Policing thus includes all kinds of police practices that do not ultimately rely
on the use of force, but are illegal to everybody except the police, such as deten-
tions, arrests, seizures, conscations, surveillance, interrogations and so on. In
this sequence, a police raid is dened as an unannounced entering of certain prem-
ises that includes the search for illegal activities or criminal suspects as well as
the checking of the identities of all encountered persons. By law, it is forbidden to
enter an apartment and do an ID check. But according to police law, this ID check
is legitimate if a person sojourns in a particular place in which the police assume
ongoing criminal activity. The interviewees stated:
We perform raids and we enter apartments because we assume that in these
apartments crimes are being committed. That is the basic principle.
(Interview vice squad 10, May 11, 2015)
The police law allows us to enter apartments. Even if the residents do not
want us to. Against the wishes of the residents we can at any time enter and
control certain places if there is the assumption of a concrete threat. This
includes places in which the activity of prostitution is pursued.
(Interview vice squad 11, May 11, 2015)

Policing the absence of the victim 115
The raid is thus a practice that is – in its capacity to enter residential property
and check IDs of citizens – forbidden to everybody but police ofcials. Accord-
ing to police law a raid is legitimate if the police assume the existence of crimi-
nal activities in the to-be-controlled places. The police law further suggests that
places in which prostitution is (assumed to be) practised are generally suspicious.7
All vice squads have particular standard operating procedures when it comes to
enforcing anti-trafcking laws. The vice squad police ofcers have been working
within the eld of prostitution-related crime for a long time and have gathered a
certain pool of experiences. In the course of their careers, they have attended special
BKA (Federal Criminal Ofce) training courses, they have experienced a number
of trafcking trials from which they learned what factors are relevant in a case to
achieve the conviction of the accused, they know how to process les and collect
evidence, and they know whom to consult when certain issues emerge. But they
also – depending on their personal interests and motivations – have read scientic
literature concerning the topic of trafcking, they have paid attention to the news,
they might have watched documentaries or movies on television, so that they have
amassed a vast pool of everyday experience and knowledge about those ‘victims
of human trafcking’. According to Valverde, administrative knowledge is always
an ‘in-between, hybrid epistemological category’ (Valverde 2009: 20). The ofcers
have, over the span of their career, developed particular victim epistemologies as
well as certain routines and standard operating procedures that help them in nding
these ‘victims’. When I asked one of the ofcers in Kassel how he identies a vic-
tim, he just answered while pointing a nger to his forehead: ‘It is all in my head’.
As all legislative prose does, the German anti-trafcking laws allow for a great
deal of interpretation. In my interviews and observations I have learned that law
enforcement agents identify the key aspects of the criminal law on trafcking
selectively and translate the language of the Criminal Code according to their own
needs, responsibilities and preferred modes of practice. This process of translation
takes place within their nexus of professionalised and everyday knowledge. One
of the ofcers explained this translational process as follows:
Whereby do I recognise the state of helplessness, for example: was the door
[to the woman’s workplace] shut or open? Was he [the assumed trafcker]
outside shopping or drinking a couple of beers with his buddies, well, then
the woman could have used this opportunity to run down on the street and
seek help. Or do I have to look at other signs of helplessness, maybe the
woman’s inability to talk properly, her state of intelligence [. . .] But like it
is written in the Criminal Code: if another person’s, i.e. the prostitute’s, state
of helplessness is exploited, i.e. she was forced to come to a foreign country,
that means she is typically Bulgarian or Romanian. She is helpless because
she doesn’t speak German, she doesn’t know our culture, and is only allowed
to stay within the walls she was put in. So if this person has been coerced into
taking up or continuing work in prostitution, that means that those women
are recruited in Bulgaria under the pretence that they can work in Germany
7 This is stated in the Hessian State Police Law §18 Abs. 2 Satz 1b HSOG (German: Hessisches
Gesetz über die öffentliche Sicherheit und Ordnung).

116 Julia Leser
as a waitress. And when they arrive, they are told that they have to work in
prostitution. And they don’t want to do that. Maybe they have been working
as prostitutes before they came here, but now they don’t want to anymore,
and then we are talking about coercion. And that means they are not allowed
to keep their money but have to give it to the trafcker [. . .] or maybe she
shows signs of injury [. . .] But the legal denition is difcult; often, the
lines of human trafcking are blurred. It is our job to carve out those lines
using interrogation techniques. And we always have to have a possible trial in
mind: what will be the outcome? Which things does the judge want to hear?
At what point does a conviction of the accused become possible?
(Interview vice squad 10, May 11, 2015)
This quote shows that this knowledge about the alleged victim here is a
hybrid corpus of the ofcer’s own experiences, his expert knowledge and
partly corresponding to the ‘ideal victim narrative’ – or ‘master narrative’, as
Snajdr (2013) has termed it – of the helpless, young and Eastern European
woman. He also translates the legal denition in a way that ts the police’s
practices of interrogation, proper ling and securing of evidence that has to be
presented in court later. In the very moment he has to classify the victim, he
already questions the probability if the victim’s particular features will be suf-
cient to prove the case in court and will lead to a conviction. He has witnessed
many trials that were unsuccessful and therefore has developed a routine of
classifying practices in accordance to his anticipation of the criteria applied
during a trial. The practice of the raid is embedded in these logics: I will show
that the raid actually includes a set of practices and bodies of knowledge, per-
formed according to particular and localised logics and objectives, that in turn
have complex consequences and implications and help to understand police
practices of control within the frame of making (in)securities and (dis)order
and as techniques of governing.
Problematisations and paradoxes
As previous research has shown, the main catalysts for law enforcement inves-
tigations in human trafcking are indications by the victims themselves or by
third parties (Herz 2005: 10). Still, many vice squads perform raids as a pro-
active means of investigation. In the city I have been conducting my research,
the ofcers focus on home-based prostitution and visit known apartments on a
regular basis, which usually means every month. In those operations, usually
three to six ofcers visit up to fty apartments on one day. The raid as police
practice is what Farrell and Pfeffer (2014) have called a ‘traditional vice squad
strategy’ that is commonly being utilised by vice squads that proactively look for
victims of human trafcking. Within the police departments Farrell and Pfeffer
have researched in the United States, they noticed that the ofcers usually fail to
identify possible victims during their raids, because the raid as an investigative
strategy is traditionally designed for exposing offences such as illegal residence
or non-existing work permits in the area of prostitution.
The raids I participated in showed similar results. The procedure of the raid
consisted of entering the apartment of the sex workers, asking for their passports,

Policing the absence of the victim 117
copying the details of their passport into a form sheet, calling the police headquar-
ters to check if the women are registered sex offenders and, concluding, asking
the women if everything was okay, to which they would reply ‘yes’. Let us look
at one example from the eld protocols8:
In an apartment building, rst oor. The rst ofcer rings the bell to an
apartment that belongs, according to a sign on the door, to ‘Jana’. The
other two ofcers and I hide next to the door. Jana opens the door.
Jana: ‘Occupied!’
Ofcer: ‘Just a moment, we are the police. Can I have a look at your passport?’
Jana: ‘I’m with a client right now, but okay, just a second.’
Everybody enters the apartment.
Jana: ‘Come into the kitchen, quick!’
The rst ofcer starts to copy the details from the passport Jana had
just given to him.
Ofcer: ‘From Belarus. Where are you born?’
Jana: ‘Kobrin. K-O-B-R-I-N. And I live in Berlin.’
Ofcer: ‘Nationality is Belarus.’
Jana: ‘Yes.’
Ofcer: ‘And you live in Berlin, in which street?’
Jana: ‘Hindenburgdamm.’
Ofcer: ‘Hinder . . . .?’
Jana: ‘Hindenburg. I believe that was a chancellor back in the days. (laughs)’
Ofcer: ‘How long are you planning to stay here for work?’
Jana: ‘Not long. This is just a job on the side, but it is legal.’
Ofcer: ‘What is your main job then?’
Jana: ‘I’m a beautician. Just think, what kind of jobs do non-Germans do
here? Hairdresser, beautician, and this here. This is my second job so
to say.’
Ofcer: ‘Can I take a picture of you?’
Jana: ‘Why?’
Ofcer: ‘For our les only.’
Jana: ‘Do I have to?’
Ofcer: ‘No, it’s optional.’
Jana: ‘Then no. You already have the picture in my passport, don’t you?’
Ofcer: ‘Your telephone number?’
Jana dictates her number.
Ofcer: ‘Anything else, is everything alright? Do you have problems with cer-
tain men in this building who want to take your money?’
Jana: ‘Of course, Romanians, Bulgarians . . . (laughs)’
Ofcer: ‘That is what I thought.’
Jana: ‘No, everything is just ne.’
Ofcer: ‘Okay then, thank you. We will not bother you any longer. Bye-bye!’
Everybody leaves the apartment.
(Field protocol raid, February 24, 2015)
8 The names of all participants have been changed in order to respect their anonymity.

118 Julia Leser
So, what happened during these raids was either one of the following
possibilities:
a) The woman examined would have her personal identication documents
in order, which meant that the police would copy these details into their
form sheet, include a photograph of her and a remark of her current where-
abouts. After the raid was nished, the ofcers would enter the gathered
data into a police database exclusively used for the purpose of registering
people working in prostitution (and searchable by name by police ofcers
nation-wide).
b) The woman did not have a valid passport or other acceptable means of iden-
tication, which the ofcers would deem as problematic, because they would
have to detain her, bring her to the station and nd a way to identify her. In
this situation, the woman’s emotional demeanour was decisive to the ofcers’
further procedure: if the woman acted apologetic and remorseful, the ofcers
would leave her with a warning. If, on the other hand, the woman without
proper means of identication would act rather bluntly and showed no signs
of contrition, the ofcers would take her to the station and treat her like an of-
fender to immigration laws. They explained that they interpreted the woman’s
behaviour as disrespectful and therefore had ‘to teach her a lesson’. Here, the
situational assessment includes an affective dynamic (the ofcers’ ‘gut feel-
ings’) that inuences their classication practices (Leser et al. 2017: 25).
c) The woman did not have her passport, but somebody else did. When inter-
viewing the ofcers, they would indicate that a woman not in possession of her
own passport would be a sign that she could possibly be a victim of trafcking.
In the situations I observed during the raids, the woman asked for her passport
would sometimes call for a friend to bring it, and the police ofcers would
have to wait for a few minutes until another woman would enter the apartment
and showed the passport to the ofcers, whose suspicion was in turn not raised,
because the third-party in possession of the raided sex worker’s passport was
a women as well, and not a male ‘trafcker’. The gender dynamic weighs
heavily into the ofcers’ classication practices: reproducing the master nar-
rative of trafcking, in which ‘heinous male predators molest [. . .] female
captives’ (Jones 2010: 1144), the ofcers seem to assume the proper victim to
be female and girlish and the proper trafcker to be male and ruthless. Here,
the master narrative operates as a simplifying device regarding the situational
assessment of the (possible) victim’s features, particularly its gender and age.
It also suggests that male victims and female trafckers are ‘invisible’ (Jones
2010, 2014) – not only in the ‘master narrative’, but also in the state practices.
d) The woman examined had a passport, but the crosscheck with the headquar-
ters’ databases revealed that the woman was a registered offender and wanted
for arrest. In that case, the police would arrest her and take her to the station
in order to initiate the proper detaining procedures.
We can see from these options above that the objective of ‘identifying possible
victims’ has to be understood quite literally: in the centre of the raid as a police
practice (or rather: a set of practices) stands the identication of migrant women

Policing the absence of the victim 119
engaged in sex work in form of checking their documents, i.e. their names, age,
and nationality. In this way, they become registered as prostitute migrant women
in police databases later on and as such, they are approached as subjects who
might be in danger (every prostitute migrant women could be a victim of trafck-
ing) and someone who is dangerous, threatening (possible offender and/or dis-
rupter of the social order) at the same time. This process of identication involves
the registration and classication of a distinct target group and contains means
of formalising persons into data sheets, accumulating and processing these data
into databases. Thus, this process rst and foremost produces knowledge about a
target population in order to make this group more manageable via gathering this
group into a database. Second, different conclusions about the identied param-
eters are drawn according to the logics of policing practices to perform a situ-
ational assessment: could this be a possible victim of trafcking? Is she helpless?
Or is she dangerous or disruptive regarding the social order? These parameters do
not only include technical details about a person’s age, nationality and gender, but
also certain affective behavioural features of the controlled person.
While the raid as such constitutes a standardised operating procedure in vice
squads, the ofcers who engage in performing the raid have – to a certain extent –
discretionary powers as street-level bureaucrats (Lipsky 1980) in direct contact
with the controlled population, i.e. they can prioritise certain locations and times
for the raid. In the raids I observed, police ofcers focus strongly on Bulgarian and
Romanian women in sex work as problematic and therefore as target populations
for raids. Thus, they control apartment buildings that these women inhabit more
often than e.g. brothels where Germans work. Bulgarians and Romanians are, as
they stated in interviews, more prone to be victims. When the ofcers talk about
their target groups for nding victims in terms of nationality, they almost exclu-
sively refer to women from Bulgaria and Romania, explaining that these women are
more helpless than Germans and thus more easily exploitable.9 In the interviews,
the police ofcers clearly identify Romanians and Bulgarians as the primary target
group of potential victims, but paradoxically, in the raids these women do not come
to the attention to the ofcers as victims, but as offenders. In two instances they
found Romanian women without passports or other proper means of identication,
but did not perceive the missing passport as a sign of being trafcked, but instead
took both of the women to the station, treating them as possible offenders to immi-
gration laws. In another instance the crosscheck with police headquarters revealed
that the Romanian woman they just checked was an offender wanted for trafcking
in France, leaving the ofcers perplexed because of her gender. A female trafcker,
they commented, was highly atypical.10 In none of the raids I observed, was a
9 In the ofcial statistics regarding formally identied victims of trafcking, Germans are equally
represented as Bulgarians and Romanians are: In 2015, out of 416 identied victims, 98 were
Romanian, 97 were German, and 71 were Bulgarian (BKA 2015). But building up such a statistic
also means to produce certain narratives of who does what to whom, where they are from, and
how to intercept the criminals. Thus, clear narratives are produced concerning women and men
from certain countries with particular migration statuses (Pates et al. 2016, Blanchette et al. 2013).
10 Although ‘the United Nations found that an overwhelmingly disproportionate number of human
trafckers were in fact not males, but females, and that female trafckers may outnumber male
trafckers worldwide’ (Jones 2014, 147).

120 Julia Leser
‘victim of human trafcking’ identied. The ofcer told me afterwards in an inter-
view that they in fact never have been successful in nding a victim during a raid:
The ways of coming to the attention to the police as a victim are different.
Mostly the women just self-report. But within the eight years I have per-
formed raids, I never met a victim who would reveal herself to us.
(Interview vice squad 11, May 11, 2015)
Ye t, th e y pe r s is t in t h ei r re gu la r r ai d s . A nd t he se ra i d s a r e i n s of a r ‘t r a di t io na l’ , a s
Farrell and Pfeffer (2014) are calling it, as they constitute standardised operating pro-
cedures that have existed in the vice investigations’ repertoire for a long time and that
form the ofcers’ actions and expectations of what will happen during their opera-
tions. It could be assumed that the search for evidence of human trafcking is not the
main objective of their raids, and that they rather check on violations of immigration
laws. These practices, rather than developed in order to identify victims of trafcking
within the area of commercialised sex, seem to be quite an unchanged continuation
of traditional practices developed in times when the whole commercial prostitution
sector as such was criminalised and controlled mainly via repressive policing.
We thus have to look further into the logics and legitimisations of these prac-
tices: why are raids still performed when no victim was ever found? Why is there
such a strong focus on Bulgarians and Romanians? If success does not legitimate
the raids, what does?
Producing order and deviant subjects
Let us look at another example from the raids I have been observing: three ofcers
and myself went to raid a local apartment building where sex workers rent an apart-
ment to work in (in this instance it was an apartment building where mostly Bulgarian
women used to work). The police ofcers went into every apartment, rang the bell
and checked the passports of every single sex worker they met. As we went from one
apartment to another, we nally arrived at the last one where the ofcers found them-
selves confronted with a different situation than in the other apartments they checked
so far. They rang the bell and a Bulgarian woman opened the door, letting us in. We
went into the kitchen and found ve Bulgarian men sitting around the kitchen table,
smoking and eating food. I felt that the ofcers were suddenly very tense. They started
questioning the men, what they were doing here and so on. Then the ofcers started
discussing among themselves and after a while they announced that they wanted to
check all the men separately in the room next door. One of the policemen and I waited
in the kitchen, while the other two went into the room next door, calling in the Bulgar-
ian men one after the other. They body-searched them and checked their passports. It
took them quite a while to get done with that. While I waited in the kitchen, I asked
the police ofcer who was waiting with me why they were doing this. He pointed out
to me that he noticed a piece of tinfoil on the kitchen counter when he came in. He
explained that they suspected this piece of tinfoil was used to wrap some kind of drug,
and he suspected that drug to be weed. If those men in the kitchen were in the posses-
sion of drugs like that, they have to check them, he explained. So they did, and after

Policing the absence of the victim 121
about half an hour they were done, and they started searching the kitchen and the rest
of the apartment for more leads and evidence. They found nothing. The piece of tinfoil
turned out to be just a piece of tinfoil that had been used for cooking. They did not
nd any drugs. It turned out that the woman living and working in this apartment had
invited her friends over for dinner and that was it (Field protocol raid, April 14, 2015).
In this example, several things are striking. First, the ofcers make ad hoc legit-
imisations for their practice of raiding (mostly private) apartments. The tinfoil
as a lead to drug use was their legitimisation for a full check of the apartment
including every person who was there at the time. Second, while this instance has
nothing to do with victims of trafcking, it illustrates how the policing practices
revolving around trafcking produce particular (dis)orders, specic subjects and
norms. And third, more specically, it (re)produces the target population, that is
Bulgarians and Romanians, as an inherently suspicious group. So they raid within
a narrow and focused area where they assume to nd these populations in order
to perform control upon them and, if possible, ‘catch the bad guys’ they associ-
ate with this population. That Bulgarians and Romanians stand as a metonym for
disorder and crime becomes more apparent in an interview done with a vice squad
ofcer in another city:
We started raiding in the 1990s. It was cruel. It was so cruel. We found mat-
tresses, in apartments with fty square metres we found up to twelve mat-
tresses on the oor. Fourteen-year old girls in prostitution. And in some parts
you couldn’t stand for more than three seconds on one spot because it was too
sticky. It was the meanest of the mean. [. . .] We faced this problem very early,
all those disgusting buildings. A lot of people cashed in on it, those who had
residential property in those areas to let. We had some foreigners who those
houses belonged to, they really cashed in on it. And I had colleagues who
had been to Plovdiv, Bulgaria, and they told me how bad it was there. They
were really disturbed by it. We just had to go against it. We had to force an
extended presence on the curb.
(Interview vice squad 20, July 28, 2015)
Even when control is not as explicit as in the example above, the presence of
the ofcers and their visibility is central to the raids and strategically utilised for
maintaining and (re)producing order.
Particularly in the eld of prostitution and human trafcking, police of-
cers I interviewed had difculties to not involve their own personal beliefs and
opinions – things they came to know through their own personal interests, which
interlace with their practices and institutional discourse, which in turn results in
this ‘institutional ction’ that highly resembles Snajdr’s ‘master narrative’ (2013),
recalling images of femininity, youth, innocence, vulnerability and poverty. For
example:
The trafckers start to prepare their merchandise in the country of origin.
They selectively target women they want to deliver into Western European
countries. Sometimes they chose women from Bulgarian orphanages who

122 Julia Leser
don’t have any relatives left, so nobody will look for them if they go miss-
ing. Of course they take an advantage of the women’s economic distress.
The women are generally very poor, most of them have children and struggle
in order to support their own families. They don’t have a proper education.
Some of them are illiterate and cannot write or read. All of that makes them
prone to become a victim. Another important thing is the women’s self-image
and their role as women in Bulgaria which is dened in totally different ways
as it is in Germany. They don’t really perceive themselves as victims because
they don’t know better. The women obey the men, they follow them any-
where and full their duties towards their husbands.
(Interview vice squad 11, May 11, 2015)
In the police ofcer’s explanation, the Bulgarian woman is especially vulnerable
because of the conservative gender roles and beliefs in her country of origin. Those
specic problematisations used by the ofcers are producing normative orders in
terms of who is being subjectied as a victim and why (in theory). But most inter-
estingly, the notion of knowledge used for identifying victims (in theory) is not
specialised expert knowledge, but common knowledge enmeshed with standard
narratives that serves as a clear horizon of expectations (either their own expec-
tations or the expectations they are confronted with when interacting with other
institutions), which causes irritation to the ofcers as they face the more complex
realities of their clientele.11 In their daily practices, these expectations are not met:
the controlled population they encounter does not t the ofcers imagined criteria
of someone being vulnerable and in need of help (a ‘true’ victim). Rather, they are
deemed inherently suspicious – and thus rendering the ‘victim’ a truly absent gure.
Counterbalancing the absence of the victim
The implementation of anti-trafcking policies and legislations depends on the suc-
cessful production of a victim in daily policing practices. As we have seen so far, the
police seem to fail to ‘produce’ victims during their regular raids for a variety of rea-
sons, while concurrently these raids seem to serve a different purpose than initially
assumed. Still, the failure of the police in producing the assumed high numbers in
ofcial crime statistics accordingly is counterbalanced by certain strategies: (i) the
vice squads developed alternative modes of data acquisition. The data they gather on
their raids in known brothels and sites of home-based prostitution – names, nation-
alities, birth dates, phone numbers, photographs, whereabouts – were fed into police
databases for internal uses, i.e. accessible by other vice police ofcers across the
state; (ii) police ofcers articulate numbers other than the ofcial statistics which
11 Ethnographic research on trafcking in human beings indicates, for example, that smuggling
and labour migrations often is blurred with trafcking, that there are many cases of exploited
labour into jobs other than sex work, and that it is often unclear where to draw the line between
economic exploitation and trafcking. It suggests that victims are coerced by a range of factors
including poverty, kinship obligations, fear of violence, debt, or the desire for the trappings of
modernity (Jacobsen & Skilbrei 2010, Engle Merry 2015, Skilbrei & Tveit 2008, Zhang 2009).
On taking a closer look, it is unclear who the victims are, how big the problem actually is, and
what kinds of exploitation are involved.

Policing the absence of the victim 123
are almost always estimates or add-ups from the prostitution databases; and (iii),
the abstraction from certain numbers, like estimates of city wide working sex work-
ers or the count of Bulgarian prostitutes in a one-day raid, as possible victims of
trafcking. In this way, ‘victim of trafcking’ becomes a certain semantic category
which almost always refers to ‘migrant prostitute’, and thus an indication that is the
foundation for most of the numbers and knowledge being produced on trafcking
victims. For example, one of the city’s counselling centres states that they counsel
up to 40 women a year who are mainly Bulgarian and Romanian sex workers, and
though they are not ofcially identied as victims, they are possible victims of traf-
cking. Similarly, the police estimate that there might be up to 400 possible victims
of trafcking in the city, which is their estimate of overall sex workers present. What
happens here is a transformation of very complex, messy, singular stories into more
stereotype overall narratives through the techniques of counting: of classifying and
turning them into numbers that can be summarised, allowing for allotting very dif-
ferent things in the same categories and creating ‘trafcking victims’ in the plural.
It can be assumed that the police contain and arrange broader political interests, and
through their production of knowledge (in form of databases and statistics), they
legitimise their performance of control and securing order – thereby enacting state-
ness and co-producing certain subjectivities.
Conclusion
As we have seen, the raid is not a singular practice, but a conglomerate of a variety
of governmental techniques aimed at the policing and control of migrant women
engaged in sex work. It includes techniques of identication, registration and clas-
sication of ‘problematic’ subjects, and draws on particular logics of perform-
ing control, maintaining order, of data acquisition and knowledge production in
order to make those subjects of control more manageable. In these practices, the
‘victim of trafcking’ is enacted as a ‘relational object of knowledge’ along differ-
ent dimensions: temporal relations concerning the expectation and anticipation of
future events (like proving the victim’s features in front of a court of justice), situ-
ational relations of assessing a person’s status along affective and gender dynam-
ics, and performative relations regarding the role-conception of the police ofcers,
of their duties and responsibilities. Meanwhile, the ‘master narrative’ of trafck-
ing is selectively invoked in the knowledge practices of the ofcers as a simplify-
ing device, which leads to particular paradoxes in their daily performance. Those
women who should be typical victims, the Bulgarians and Romanians, do come to
the attention of the state not as victims, but more likely as offenders to immigra-
tion laws, as disruptors of the social order, as suspicious and deviant. The imagi-
nary gure that police ofcers come to know as a ‘true victim’, deserving of being
rescued, remains absent. But while these policing practices revolve around those
rather contrasting forms of subjectivities – the absent victim and the problematic
sex worker (possible victim, possible offender and/or disruptor of the social order) –
the police and the state as such are being reproduced in a certain way: in being
burdened with the task of upholding the social order, they identify (not the victim,
but) what is deemed normal, deviant, order-less or morally questionable, in short:
what sort of problem requires a response by the state. In practice, this becomes

124 Julia Leser
a matter of emphasis, of negotiation, contestation and sometimes even discrete
decision. It is a process that incorporates a variety of hybrid bodies of knowledge,
institutional logics, interactions and interrelations. And although these practices
are unmistakably local, situated and embodied in individual state ofcials, they
assemble a complex composition of how the state is done.
References
Bacchi, C. (2012) ‘Why study problematizations? Making politics visible’, Open Journal
of Political Science, 2, 1–8.
Berman, J. (2003) ‘(Un)popular strangers and crises (un)bounded: Discourses of sex-
trafcking, the European political community and the panicked state of the modern
state’, European Journal of International Relations, 9, 37–86.
Bierschenk, T. (2016) ‘Police and state’, in Bradford, B., Jauregui, B., Loader, I., and Stein-
berg, J. (Eds.) The SAGE handbook of global policing. London: Sage, pp. 155–178.
Blanchette, T. G., Silva, A. P., and Bento, A. R. (2013) ‘The myth of Maria and the imagin-
ing of sexual trafcking in Brazil’, Dialectical Anthropology, 37, 195–227.
Brodeur, J. P. (2010) The policing web. Oxford: Oxford University Press.
Bundeskriminalamt [BKA]. (2015). Bundeslagebild Menschenhandel. Wiesbaden: BKA.
Christie, N. (1986) ‘The ideal victim’, in Fattah, E. (Ed.) From crime policy to victim
policy. Basingstoke: Palgrave Macmillan UK, pp. 17–30.
Doezema, J. (2000) ‘Loose women or lost women? The re-emergence of the myth of white
slavery in contemporary discourses of “trafcking” in women’, Gender Issues, 18,
23–50.
Farrell, A., and Pfeffer, R. (2014) ‘Policing human trafcking: Cultural blinders and orga-
nizational barriers’, The Annals of the American Academy of Political and Social Sci-
ence, 653 (1), 46–64.
Group of Experts on Action against Trafcking in Human Beings [GRETA] (2015) Report
concerning the implementation of the Council of Europe Convention on Action against
Trafcking in Human Beings by Germany. Available at: https://rm.coe.int/1680631c3b
(Accessed 1 October 2017).
Hacking, I. (1999) The social construction of what? Cambridge, MA: Harvard University
Press.
Hansen, T. B., and Stepputat, F. (2001) States of imagination: Ethnographic explorations
of the postcolonial state. Michigan: Duke University Press.
Herz, A. L. (2005) Menschenhandel: Eine empirische Untersuchung zur Strafverfolgung-
spraxis. Freiburg i. Brsg, Germany: Max Planck Institut.
Hill, A. (2016) ‘How to stage a raid: Police, media and the master narrative of trafcking’,
Anti-Trafcking Review, 7, 39–55.
Jakšić, M. (2008) ‘Figures de la victime de la traite des êtres humains: de la victime idéale
à la victime coupable’, Cahiers internationaux de sociologie, 124, 127–146.
Jones, S. V. (2010) ‘The invisible man: The conscious neglect of men and boys in the war
on human trafcking’, Utah Law Review, 4, 1143–1188.
Jones, S. V. (2014) ‘The invisible women: Have conceptions about femininity led to the
global dominance of the female human trafcker?’, Albany Government Law Review, 7
(1), 143–165.
Latour, B. (2007) ‘A textbook case revisited: Knowledge as mode of existence’, in Hackett,
E. J., Lynch, M., Wajcman, J., and Amsterdamska, O. (Eds.) The handbook of science
and technology studies. Cambridge, MA: MIT Press, pp. 83–112.

Policing the absence of the victim 125
Leser, J., Pates, R., and Dölemeyer, A. (2017) ‘The emotional leviathan: How street-level
bureaucrats govern human trafcking victims’, Digithum, 19, 19–36.
Lipsky, M. (1980) Street-level bureaucracy: Dilemmas of the individual in public services.
New York: Russell Sage Foundation.
Migdal, J. S., and Schlichte, K. (2005) ‘Rethinking the state’, in Schlichte, K. (Ed.) The
dynamics of states: The formation and crises of state domination. Aldershot: Ashgate,
pp. 1–40.
Moore, D., and Valverde, M. (2000) ‘Maidens at risk: Date rape drugs and the formation of
hybrid risk knowledges’, Economy and Society, 29 (4), 514–531.
Pates, R., and Schmidt, D. (2007) ‘Wahrheiten über Opfer: Menschenhandelsdiskurse
im Vergleich: HU Berlin: Zentrum für transdisziplinäre Geschlechterstudien: Bulletin
Texte’, Der involvierte Blick Zwangsprostitution und ihre Repräsentation, 35, 90–105.
Sharma, A., and Gupta, A. (Eds.) (2009) The anthropology of the state: A reader. Malden;
Oxford; Victoria: Blackwell Publishing.
Snajdr, E. (2013) ‘Beneath the master narrative: Human trafcking, myths of sexual slav-
ery and ethnographic realities’, Dialectical Anthropology, 37, 229–256.
United Nations Ofce for Drug Control and Crime Prevention [UNODCCP] (2000) Pro-
tocol to prevent, suppress and punish trafcking in persons, especially women and
children. Available at: http://www.unodc.org/documents/treaties/UNTOC/Publications/
TOC%20Convention/TOCebook-e.pdf (Accessed 1 October 2017).
Valverde, M. (2009) Law’s dream of a common knowledge. Princeton: Princeton Univer-
sity Press.
Valverde, M. (2010) The force of law. Ontario: Groundwood Books Ltd.
Valverde, M., Levi, R., and Moore, D. (2005) ‘Legal knowledges of risks’, in Law Com-
mission of Canada (Ed.) Law and risk. Vancouver, Toronto: UBC Press, pp. 86–120.