Security and Limits to Democracy : A Critical Study of the Danish Ghetto Package

(1)

Security and Limits to Democracy:

A Critical Study of the Danish Ghetto Package

Emelie Thorburn

Thesis, 30 ECTS (hp)

Political Science - Crisis Management and Security Master’s Programme in Politics and War

Spring 2020 Word count: 14 934

(2)

Abstract

This thesis departs from two arguments articulated by Jef Huysmans. First, security, as it is increasingly dispersed and part of every-day decision-making, undermines democracy in previously unprecedented ways. In order to under-stand security, scholars must look beyond insecurities framed as existential threats. They must equally adopt a fractional approach to democracy. Limits to democracy are increasingly found beyond democratic institutional frames of reference. In an empirical endeavour, this thesis builds on theoretical concepts by Huysmans and others in examining the Danish ghetto package as a cluster of securitising techniques. I make the argument that as the ghetto package entails more dispersed securitising techniques, it enacts limits to democracy.

(3)

Acknowledgements

(4)

Introduction

In his annual New Year’s speech in January 2018, the then Prime Minister of Den-mark Lars Løkke Rasmussen introduced what he referred to as the ‘Ghetto package’ (Ghettopakken). The explicit aim of the package was to prevent crime and violence

from spreading beyond ghetto areas, and hinder the continuation of “parallel soci-eties” (Regeringen 2018a). In March 2018, the Danish government published the

package in full, titled “A Denmark without parallel societies - No ghettos in 2030” (Regeringen 2018b, ´Et Danmark uden Parallelsamfund - Ingen ghettoer i 2030,

translated by author). The ghetto package displays interesting insights when read from a critical security perspective. This thesis sets out to conduct a critical

in-quiry of the Danish ghetto package using concepts from Jef Huysmans’ Security

Unbound (2014). What primarily guides Huysmans conduct is the notion that

se-curity inherently undermines democracy. He outlines a shift in sese-curity practice, oriented towards dispersed multiple, rather than unified singular, security threats. By magnifying certain securitising techniques of surveillance and suspicion,

Huys-mans displays how security practices limit democratic categories of privacy, me-diation and political subjectivity. With democratic categories, Huysmans departs

from a widely used balance-metaphor between security and liberty. Security does more than simply “nibble at the edges of democracy” (Huysmans 2014:3). Rather,

security fundamentally alters conditions for pursuing democracy as we know it, all while, an existential crisis or a state of emergency is absent.

(7)

1.1 Research Problem

Building on Huysmans work, this thesis departs from two arguments articulated in his concluding remarks. First, security, as it is increasingly dispersed and part of every-day decision-making, undermine democracy in previously unprecedented

ways. In order to understand security, one must look beyond insecurities framed as existential threats (Huysmans 2014:182). Security is not the concern of security

professionals alone, but include multiple societal actors of state agencies, municipal-ities, private companies, volunteer groups, youth professional, individual citizens,

and many more (Huysmans 2014:186). To scholars interested in dispersed secu-rity, securitising techniques enacted by surveillance offers particular insight. To

understand democracy in this dispersed security context, one must equally adopt a fractional approach to democracy. Limits to democracy can be found beyond

demo-cratic institutions (Huysmans 2014:188). Huysmans’ two arguments are theoretical, adding to the extensive theoretical field of critical security studies (Bigo 2000; Bigo 2002; Ruppert 2011). Between more dispersed techniques of securitisation, and less

institutional limits to democracy, there is hence room for more in-depth empirical engagement. The argument made here is that the Danish ghetto package makes up

an insightful case for such empirical engagement.

1.2 Research Aim and Questions

This thesis leads by a dual research aim. First, the aim is internal, orienting towards

the field of critical security studies. Building on the theoretical work of Jef Huys-mans (2014), Evelyn Ruppert (2011) and others, this thesis approaches the Danish

ghetto package as a cluster of securitising techniques, to which surveillance practice is key. Of particular interest here are the limitations these securitising techniques

may enact on democracy. The aim is to contribute empirical thickness to an exten-sive theoretical field. Second, this aim is external. When viewed through a critical lens, nothing about the Danish ghetto package is inherent or inevitable. This thesis

hence leads by a critical motive, seeking to challenge security practices, along with the seemingly objective accounts of reality they rely on, from becoming naturalised.

(8)

is politically situated (Guillaume 2012:29). Rather than settling with adding to a

scientific field of critical security studies, this thesis seeks to shed light on some securitising techniques detectable in the Danish ghetto package, by examining their

political dimension. The thesis is guided by the following research questions:

(a) In what ways are securitising techniques enacted in the Danish ghetto

pack-age?

(b) In what ways do these techniques enact limits to democratic practice?

1.3 Defining Democratic Practice

Building predominantly on Huysmans’ (2014) theoretical concepts, this thesis equally adopts a fractional approach to democracy rather than an institutional one

(Huys-mans 2014:28). By this definition, security practices are not measured against a democratic ideal type. Rather, democratic practices refer to categories that dis-play fragmented elements of democratic traditions. This fractional approach to

democracy implies that “democracy is not a fixed institutional or conceptual point – a model – but consists of various things that are connected to enactments of

democracy [. . . ]” (Huysmans 2014:28). As he examines techniques of securitisation, Huysmans touches upon democratic categories of privacy, limits to technocratic

governance, citizen autonomy and the capacity for reflection (Huysmans 2014:28). Using a fractional approach, this thesis has no intention to measure policies’

prox-imity to democratic ideals. It rather utilises democratic categories to display what is at stake in security practice.

1.4 The Danish Ghetto Package

The Danish ghetto package was laid out in full in March 2018, in a policy docu-ment called “A Denmark without parallel societies - No ghettos in 2030”

(Regerin-gen 2018b). The package aligned with a Danish multisectoral approach to societal issues (Johansen 2019), including areas of housing, police, school, day-care, job market, social services, and more. The policy document begins by noting that the

(9)

ethnic composition in Denmark has changed in the last 40 years, and while many

immigrants participate actively in society, many do not (2018b:4). It finds that parallel societies among people with non-Western decent have emerged (Regeringen

2018b:4). By data collection, indicators of parallel societies are tied to residential areas. The package therefore lists a number of criteria for classifying residential areas as either hard ghettos, ghettos, or vulnerable areas (Regeringen 2018b:11).

The designation of ghetto area is based on five criteria. It regards the residents’ ethnicity, level of education, connection to the labour market and the education

sys-tem, income and crime (Regeringen 2018c:6). The ghetto package is made up of a number of policies and legislations. These include rules for settlement in ghetto

ar-eas, renovation and demolition of buildings, policing zones, mandatory day-care and mandatory schooling in Danish language and Danish culture and values

(Regerin-gen 2018b). By tying the policies to residential areas, not all Danish citizens are subjected to the ghetto package. Instead, a majority of the ghetto-labelled citizens

live in suburbs to Denmark’s largest cities.

1.5 Disposition

Five chapters make up the forthcoming critical engagement with the ghetto package.

The literature review intends to give the reader a contextual background to the field of critical security studies. It highlights work written on securitising techniques and their political dimension. After that follows an outline of the chosen theoretical

concepts. The methodology section then touches upon ontology and epistemology of critical security studies, and accounts briefly for methodological decisions made

in conducting this research. Then follows the empirical analysis. It is divided thematically, with the first section devoted to securitising techniques and the second

section devoted to democratic practices. The thesis ends with concluding remarks and some suggestions for further research.

(10)

Chapter 2

Previous Research

The forthcoming review seeks to offer the reader a brief contextual outline to the research subject. It begins by introducing the reader to the critical security studies field at large. It then touches upon the changing security landscape, the notion of

risk and security as securitising techniques. Here, Huysmans (2014) diffusing inse-curities is presented in more detail. His perspective is given contextuality in a brief

outline of the interdisciplinary field of security studies and political sociology. This section highlights research on policing and surveillance constitutive to Huysmans’

work. The review concludes by again account for the research aim and this thesis’ contribution to the field at large.

2.1 Critical Security Studies

As a sub-field to International Relations, the traditional approach to security studies that evolved in the 20th century researched the international system and relations

between states (Herz 1950; Waltz 1979). Scholars diverted attention to military capabilities, the projection of force, and the protection of the state. Traditional

security scholars primarily maintained the state as the referent object of security. In the latter half of the 20th century, academic approaches departing from the tra-ditional view evolved by critically examining and deconstructing some of its key

components. Together, these approaches constitute the field of critical security studies. It builds upon Robert Cox’s (1981) distinction between problem-solving

(11)

theory and critical theory (Williams and Krause 1997:xvi). The former relates to

social and power structures as “the given framework for action” (Cox 1981:208, cited in Williams and Krause 1997:xvi). The latter “calls them into question” by

exam-ining their origin and their potential to change (Cox 1981:208, cited in Williams and Krause 1997:xvi). Critical here is not merely criticising, but rather refers to the practice of challenging knowledge that is taken for granted. The critical field

emphasise the impact of socio-political processes (C.A.S.E 2006:446). Christopher Browning and Matt McDonald (2013:236) sum up the critical approach in three

interrelated themes; a general critique of the epistemology and ontology of tradi-tional scholars; the political dimensions to security; and the ethics of security. The

first theme regards knowledge production. From a critical point of view, insecurity is a constructed concept (Bilgin 2013:95). Contrary to a traditional view, critical

scholars argue that security precedes threats (Huysmans 2014:3). Rather than en-tailing the response to an external reality, security is constructive in creating what

is perceived as reality. “Security is a practice not of responding to enemies and fear but of creating them” (Huysmans 2014:3). As such, there is no pre-defined meaning to security (Bilgin 2013:94). When writing academically on security, scholars take

part in security knowledge production, contributing to the constructive notions of security. The second theme recognises security studies’ political nature. By

exam-ining the political dimensions to security, critical scholars engage in implications and limitations brought about by security, emphasizing its political power. The

po-litical dimension to security is two-fold. First, in being constructed, security is the outcome of political processes. What one perceives as security or insecurity derives

from one’s political outlook and the perceptions of threat that aligns with it (Bilgin 2013:95). Secondly, security enables and restrains political outcomes. The way a

threat is constructed determines the potential political means to contain it (Balzacq 2010:13). The third theme captures the normative dilemma that follows from recog-nising the first two (Huysmans 2002). If scholars partake in constructing notions

of security, and this process renders political implications, what are the normative considerations of critical security studies? The critical perspective hence has an

(12)

must maintain a reflexive stance. This recognition of social theory as partial and

political is what fundamentally sets the critical perspective apart from traditional political and social science (C.A.S.E 2006:445). Summed up Browning and

McDon-ald (2013:238), engaging with what security does politically “encourages reflection on the role of representations of security [. . . ]”.

Huysmans (2014:14) resonates a critical stance when differentiating a ”security read-ing of politics”, from a ”political readread-ing of security”. The former is treatread-ing the

field of security as separated from that of political decision-making. Security is-sues are introduced to the political sphere by security actors, competing with other

issues for attention (Huysmans 2014:13). A security issue hence exists prior to po-litical decision-making. Security studies then primarily concerns ways of handling

security issues (Huysmans 2014:16). To this approach, the political dimension is in the process of determining recourses and priorities for security issues (Huysmans

2014:15). Huysmans however argues that a security reading of politics misses some fundamental aspects of the security field. In his view, rather than exist prior to politics, security is intertwined with politics (Huysmans 2014:14). Politics decide

what can be considered a security issue or the referent object of security. To secu-rity studies, the key question is not “how to best deal with a threat but whether

to approach a particular phenomenon as a security threat or not [. . . ]” (Huysmans 2014:17). Insecurities are politically constructed rather than given. By reading

them politically, critical scholars set out to challenge insecurities and the way they enact politics.

2.2 The Changing Security Landscape

Critical security scholars, like traditional peers, have devoted much attention to the changing security landscape. By the end of the Cold War, the state-centered

secu-rity perspective was considered inadequate to capture new insecurities of intrastate conflict, environmental changes, new migration patterns, infectious diseases and

terrorism (Metelits 2016:20). Widening the security sphere (Buzan, Wæver and de Wilde 1998) or adopt a human security perspective (Owen 2004) are some attempts to theoretically capture this new security landscape. Critical security studies have

(13)

instead looked at aggregated singular threats being replaced by multiple dispersed

risks (Bigo 2000; Bigo and Tsoukala 2008: Huysmans 2000; Huysmans 2014). The concept of risk society was first introduced by Ulrich Beck (1992) as an inherent

characteristic to modernity. New technology and infrastructure, such as nuclear power, natural resource extraction and global migration movements, brought po-tential unseen global dangers. From this perspective, risk differs from threat in

that it is incalculable, often unpredictable and unforeseeable (Krahmann 2011:355). By risk calculation however, the unpredictable is made substantial. Risk is an

em-bodied form of calculating the unknown future (Burgess 2016:3). The risk concept later evolved after receiving criticism. To Elke Krahmann (2011:353), Beck “fails

to examine in greater detail the mechanisms which contribute to the management and perpetuation of the world risk society”. Engin Isin (2004) equally point to a

political drive behind society’s engagement with risk, rather than the changing re-ality of modernity. Building on this more constructive notion, risk equally refers

to techniques of governing the future, in present times. These techniques work by making the future knowable. Only by then can the future be acted upon in to-day’s decisions (Huysmans 2014:102). From this perspective, risk renders severe

implications on security practice. Risk, in being the unknown and uncertain, has replaced definable threats in the new security landscape. When examining NATO’s

evolvement in the post-Cold War era, Huysmans (2014:78) finds that “the security priority was no longer to manage a global enemy but to govern risks, which referred

to a multiplicity of often not yet known insecurities”. With this new multiplicity and uncertainty, insecurities have been dispersed. This insight has guided critical

scholars, generating research devoted to dispersed insecurities and football hooli-gans (Tsoukala 2009); migration (Huysmans 2000) and counter-terrorism post 9/11

(Bigo and Tsoukala 2008; de Goede 2012).

2.3 Securitisation

To a critical approach, insecurity issues are secondary to the methods or practices

they enact. Huysmans (2014:18) refers to these as “techniques of securitising” – or ”methods of rendering insecurities”. Security then is not a static state of affairs

(14)

but enacted. This perspective “looks at how security practice renders events into

security problems – or, in other words, how it securitises” (Huysmans 2014:19). The Copenhagen school was early in adopting this approach to security politics. They

examined security issues from the lens of securitising by a speech act. Ole Wæver (1995:54) defines securitisation as brought forward by language, by elite actors name a certain development a security issue and thereby claim the right to use exceptional

means. There exists no security issue prior to the act of naming it. In later writing, Barry Buzan, Ole Wæver and Jaap de Wilde (1998:26) emphasized exceptionality

as a key component to securitisation. The Copenhagen school approach has been widely used in examining issues of gender, migration and environmental changes

becoming securitised (Williams 2010:212). Along with its popularity, it has equally

been subjected to critical debate. One key theme to this debate has been the

emphasis on exceptionality (Floyd 2015; Abrahamsen 2005; For other debates, see Balzacq 2010 on audience; Booth 2005 on emancipation; Wilkinson 2007 on

non-Western applicability). To Rita Abrahamsen (2005), the centrality of emergency action limits the perspective of securitisation and its ability to adequately capture processes where issues become security related. Insecurities exist on a continuum,

stretching from normal to worrisome to risk and existential threat. She finds that the process of securitisation is gradual (Abrahamsen 2005:59). Huysmans (2011:375)

equally argues that the sharp distinction between everyday practice and emergency action limits the securitisation framework in identifying processes that fit neither

category. Instead, he examines the security landscape through the lens of diffusing insecurities. Diffusing insecurities is a non-intensive technique of unbinding security

(Huysmans 2014:72). As risk and uncertainty displace existential threat, limits to security are replaced with continuities and boundaries (Huysmans 2014:83). These

continuities extend the meaning of security, renegotiating clear distinctions between security and non-security (Huysmans 2014:80). At once, issues are reconnected by association. In reconnecting insecurities they do not become more urgent, nor can

they be organised by importance or relevance. On the contrary, threats and risks are horizontally interlinked. The associated insecurities make up a diverse patchwork,

(15)

2.4 Security and Political Sociology

Huysmans’ notion of securitising techniques as enacted in everyday practice builds on the interdisciplinary field of security studies and political sociology. Together, this field has been referred to as the Paris school (C.A.S.E 2006:446). When

study-ing security, Paris school scholars mainly cover internal security in a European context (C.A.S.E 2006:449). They display the political construction of security by

analysing policing and surveillance practices (Haggerty and Ericson 1997;2000), mi-gration (Huysmans 2000) and security experts (Bigo 2000) through a political and

sociological lens. On securitising techniques, the research subject sets the Paris school approach apart from the Copenhagen school. Didier Bigo (2002) finds the

Copenhagen approach to engage with elite actors and their speech acts. He ar-gues that by focusing on political discourse, securitisation theory underestimates

the role of bureaucratic officials in conducting securitising practice (Bigo 2002:74). Securitisation, when read from a political sociology perspective, is not a single act of one authority. Rather, they are the everyday decisions of many bureaucratic and

private professionals (Bigo 2002:74). In order to display processes of securitisation, scholars must divert their attention towards the everyday work of bureaucracies

(Bigo 2012:279).

To detect securitising practices in everyday decision-making, the sociological ap-proach has turned to studies of policing and surveillance. In this field, Kevin

Hag-gerty and Richard Ericson’s (1997) ethnographic work on policing is key. Beyond traditional police work, they identify categories of policing on the rise. Policing has

increasingly become knowledge work by gathering, sorting and analysing data. In a risk society, knowledge perceivably heightens the ability to predict, prevent and mit-igate risks (Haggerty and Ericson 1997:6). Policing is hence the work of producing

knowledge by collecting data and conducting risk analyses. Thereto, police increas-ingly function as a central unit among information gathering agencies. In a diffused

security context, data gathering is mainly conducted by surveillance (Huysmans 2014:97). Surveillance in this sense refers to a routinised reporting of fragments

(16)

hence stretches beyond policing work. Scholars have explored surveillance as

re-lated to crime, terrorism and information warfare (Ball and Webster 2003) and as a tool for social sorting (Lyon 2003). In the security studies field, Louise Amoore

and Marieke de Goede (2008) have displayed how surveillance techniques become pre-emptive security measures in the war on terror. However, surveillance practice is not a new phenomenon, nor solely an outcome of risk society (Haggerty and

Er-icson 1997:13). The increase in surveillance practice, along with rapid changes in modern technology, however calls for rethinking surveillance. In particular where

surveillance practices enact politics.

Bigo, Isin and Ruppert (2019) account for this technological advancement and data as constitutive to politics. They argue that no previous state in history has had

ac-cess to such varied and detailed data on their citizens as Western governing agencies do today (Bigo, Isin and Ruppert 2019:4). Their main interest is however not data

gathering per se, but the political practice that data entails. They approach data “not as a representation (i.e., information collected, stored, and presented without interest) but as an object whose production interests those who exercise power”

(Bigo et.al 2019:6). Data, they argue, is hence generative of politics. Jannick Schou and Morten Hjelholt (2018) equally detect a political link between governance and

data gathering, specifically looking at a Danish context. They build on the concepts of digital citizenship (Isin and Ruppert 2015) and neoliberalisation (Jessop 2002;

Peck 2010) when critically examining how Denmark has pursued a digitalisation of governance. They find that “being a digital citizen intermingles with wider political

calls for personal responsibility, flexibility and optimization”, mirroring neoliberal tropes (Schou and Hjelholt 2018:519). Though emphasising different aspects to data

politics, Schou and Hjelholt equally stress its political dimension. Within a secu-rity context, Huysmans (2014) has engaged with the political dimension to data by specifically looking at ways that securitising surveillance techniques enact limits to

democratic practice. He argues that rather than simply rearrange insecurities in new ways, diffuse securitising techniques has weakened or even undermined democratic

categories of privacy, mediation and political subjectivity (Huysmans 2014:188). In his analysis, he challenges the security-liberty balance metaphor (Bigo 2012:270)

(17)

by stating that security and democracy are inherently incompatible (Huysmans

2014:4). “This inherent tension between democratic politics and security practice defines the political content of security” (Huysmans 2014:5). Democracy is hence

a political stake in security practice. Nevertheless, Huysmans (2016:74) finds that democracy as a driving category has received limited attention from security and surveillance studies. Their focus is rather to understand surveillance. Against this

background, Huysmans (2016:80) calls for studies to “ask if there is more to democ-racy than the institutional eyes see [...]” by looking beyond institutional repertoires

of democratic action. He stresses the need for analysing democratic practice beyond spatial limits of democratic institutions (Huysmans 2016:80).

2.5 A Contribution to Critical Security Studies

From Huysmans call for democratic inquiry to dispersed securitising practices this thesis holds its aim. The review of previous research displayed extensive theoretical

work within the critical security studies field (Bigo 2000; Huysmans 2014; Bigo et.al 2019). Among more subtle and dispersed techniques of securitisation, there is

however room for more in-depth empirical engagement. This thesis approaches the Danish ghetto package as a cluster of securitising techniques, to which surveillance

practice is key. Of particular interest here are the limitations these securitising techniques may enact on democracy beyond institutional boundaries. The aim is to contribute empirical thickness to an extensive theoretical field. Thereto, the

ghetto package has primarily been subjected to sociological examination from the lens of social engineering and social gentrification (Nielsen 2019), as well as to

anthropological ethnography looking at community resistance (Strandholdt Bach 2019). Reading the ghetto package as a cluster of securitising techniques enacting

limits to democratic practice hence offers additional empirical insights beyond what has already been written on this case.

(18)

Chapter 3

Theoretical Framework

In four sections below, key concepts to the analytical framework utilised in the forthcoming empirical analysis, will be outlined. The sections are divided themati-cally, presenting the concepts of diffusing and associating insecurities; a rhizomatic

surveillance structure; purposely reassembling surveillance data and segmentation; and biometric and transactional data. A concluding section reflects upon using the

concepts together rather than separate.

3.1 Diffusing and Associating Insecurities

The concept of diffusing insecurities evolved as a counter-perspective to a theoretical

field premiering existential threats. Insecurity, it was argued, referred to the state of emergency. A shifting security landscape however calls for different theoretical

tools, that better capture the characteristics of security as dispersed and part of everyday life. To Huysmans (2014), this entailed insecurities as diffusing. Diffusing

insecurities is a constant practice rather than a term to capture a static security landscape. Diffusion is continuously enacted by agencies, professionals and

individ-uals. It builds on the notion of risk and unease – of not knowing for certain from where insecurities emerge, or the way they are interlinked. When diffusing insecuri-ties, categorical limits to security spheres are blurred (Huysmans 2014:75). It is no

longer clear where issues of security end and other issues begin. The term entails a transgressional character to security issues as they exceed previously determined

(19)

categories (Huysmans 2014:83). These continuities extend the meaning of

secu-rity, renegotiating clear distinctions between security and non-security (Huysmans 2014:80). The process of diffusing insecurities operates discursively. It works by

policy documents, public speeches, informal text material or information folders ar-ticulating insecurities as spreading. In textual descriptions, insecurities seemingly appear in all spheres of society (Huysmans 2014:183). Diffusing however equally

works by dispersing security practice, by making security a concern of multiple actors (Huysmans 2014:186). As insecurities are unknown and scattered all over

society, private companies, volunteer groups and individual citizens, must equally take part in mitigating them. Diffusing insecurities hence engages actors beyond

the traditional security sphere.

As insecurities are dispersed, they are at once interlinked. They are discursively as well as practically connected in new associative structures. Huysmans refers to this

process as associative securitising (2014:83). In associatively reconnecting insecuri-ties they do not become more urgent, nor can they be organised by importance or relevance. Instead, threats and risks are horizontally outlined side by side. When

associatively reconnected, insecurities do not align in a causal order (Huysmans 2014:84). They rather co-exist, connected by porous links, constituting a disperse

patchwork (2014:84). Issues are outlined as separate, continuously retaining signif-icant diversity and multiplicity, yet together, as components to a larger texture of

unease. Like the process of diffusion, association rearranges traditional categories of security and non-security spheres. Association then “extends the circulation of

insecurity connotations and denotations by increasing the number of units that are included in the text” (Huysmans 2014:85). The process of diffusing insecurities

identified by Huysmans is hence twofold. Previous boundaries between insecurities become porous, dispersing insecurities across various sectors and actors in society. When dispersed, insecurities are at once associatively reconnected. More persons,

sites, transactions and encounters are associatively added to the field of insecuri-ties, creating a social texture of unease (Huysmans 2014:109). Together, they enact

(20)

3.2 A Rhizomatic Surveillance Structure

The rhizome metaphor was first introduced by Gilles Deleuze and F´elix Guattari (1987). To Haggerty and Ericson (2000), it particularly well account for the dis-persed structure of surveillance practices. A rhizome system, contrary to a tree

with branches extending from a root, does not connect at one point. It expands horizontally, evolving into an unstructured network of loosely attached connections.

When cut of somewhere, it will reappear somewhere else (Huysmans 2014:110). Rhi-zomatic surveillance differs from previous dominating notions of surveillance that

premiers a bureaucratic structure. The latter focuses on central information storage and supervision performed by states (Huysmans 2014:111). In a rhizomatic

surveil-lance system, state agencies are merely some out of numerous agents. The data they gather is not stored centrally but remain scattered. Interagency relations and

modes of operation are diverse. They hence differ from a more centralised state con-ception of surveillance as social control (Huysmans 2014:113). A rhizomatic system further continuously extends beyond its previous boundaries, evolving into a

com-plex network of multiple actors. Surveillance is then conducted by state agencies, as well as private security companies, public transport systems, shopping malls, car

parking companies, airport customs and many more. They are not fully linked, nor are they completely detached.

When emerging in a diffused security context, the rhizomatic surveillance structure

operates by enacting suspicion (Huysmans 2014:101). “When turning dangers or contingencies into risks, surveillance makes suspicion an organising principle of

re-lating” (Huysmans 2014:103). As insecurity prevails various spheres of society, all matters are potentially dangerous before cleared. Suspicion is hence not an emo-tion or psychological phenomenon, but a mode of connecting scattered pieces of risk

information and make them meaningful. More importantly however, making sus-picion the mode of relating calls for a particular form of surveillance engaged with

everyday life (Huysmans 2014:104). As risks appear in everyday-life, that is where they can be identified and prevented. When operating in everyday life, surveillance

(21)

more and less suspicious subjects. “Most immediately visible in everyday practice

is that not everyone is equally suspicious” (Huysmans 2014:104). Second, it elevates suspicion to a default position against which to map risks. Everything is potentially

dangerous. Nothing can be trusted (Huysmans 2014:104). Two characteristics are hence significant to the diffused security surveillance structure. It expands rhi-zomatically, necessarily including more actors in data gathering, as more agencies

must concern themselves with security matters. It also leads by a security purpose, making suspicion the organising principle for relating.

3.3 Purposely Reassembling Surveillance Data

Knowing the changing character to diffusing security surveillance structures however does not answer the question of what information surveillance practices will extract.

According to Huysmans (2014:97), “monitoring is a proactive gathering of data and not a passive recording”. Surveillance practices purposely extract information of

individuals and streamline it into standardised data. This standardised reporting generates homogenic and uniform information that can be used as knowledge. It

strips data of history, context, intent or emotion (Huysmans 2014:97). As many agents conduct data gathering, they form a network of information flows. From

these flows, data are reassembled in ‘data doubles’ (Haggerty and Ericson 2000:606). In data doubles, dispersed pieces of information are gathered as a set of connections, summing up information profiles detached from the bodies they were extracted from.

Dispersed data held by many actors are gathered to form the basis of governmental policy. They are intentionally clustered by criteria making them say something

about a phenomenon, a group of citizens or a residential area. When reassembled, disperse data is made meaningful. Collected purposely, homogenized surveillance

data hence cannot be read as “raw data” simply awaiting neutral analysis (Ruppert and Savage 2011:88). Evelyn Ruppert and Mike Savage (2011:88) stress the inability

to analyse data as ‘flat’ descriptions, as if they were objective accounts of a ‘real’ world. Rather, reassembled data mirror existing categories (Lyon 2003:27). They

reflect and repeat prejudicial typing (Lyon 2003:22). They are intentionally gathered in certain ways to display new patterns of correlation (Ruppert 2011:222).

(22)

The practice of purposely reassembling surveillance data makes up a key component

to segmentation. Segmentation primarily refers to classifying practices in private market surveillance, singling out population segments to target or monitor more

closely (Gandy Jr 2012:127). As a securitising technique, segmentation distinguishes areas, population groups or transactions by reassembling surveillance data. When clustering data by certain criteria, segments appear. In policing, segmentation has

been used in crime prevention efforts to identify geographic areas where crime may occur in the future (Lyon 2003:15). From records of crime statistics, geographic

areas are deemed “hot-spots” or “risk zones”. Thereto, segmentation is used to single out groups of citizens (Huysmans 2014:125). From statistical data,

govern-ment agencies classify population groups as potentially dangerous or security risks by using various criteria of income, citizenship, religiosity, shopping patterns,

eth-nicity, criminal record, mental health, family relations, etc. The segments are then subjected to targeted state interventions (Haggerty and Ericson 2000:606). Policies

and practices are put in place to mitigate identified segments, either by altering their composition or by monitoring them closely (Huysmans 2014:125). The securi-tising practice of segmentation is hence dual. It refers to the reassembling process

of clustering data that produce areas or groups of particular interest – the segment. It equally entails governing interventions targeting these segments, often differently

from the way society is governed at large.

3.4 Biometric, Biographical and Transactional Data

When examining surveillance practices, one benefits from making a distinction

be-tween biometric or biographical data, and transactional data (Ruppert 2011). The former is primarily extracted from bodies and/or subjects. The latter is generated

as a by-product from everyday activities (Huysmans 2014:107). Transactions, like bodies, are detached from their contexts and standardised into comparable data.

As such, transactional data invites policy makers, as well as private companies, to “know individuals and whole populations on the basis of what people do rather than

what they say” (Ruppert and Savage 2011:83). Ruppert (2011) specifically looks at how governance is increasingly reliant on transactional data. She distinguishes

(23)

censuses, specifically requesting citizens to fill out biometric data on themselves,

from population metrics, extracting transactional data from agency databases by income, tax, public benefits, and more (Ruppert 2011:222). By population metrics,

what people do in relation to the government is rendered more useful than what they say they do, or even who they say they are (Ruppert 2011:227). In censuses, people are asked to formulate their own subjective identifications by filling out age,

sex, residency, place of birth, etc. They can lie or even refuse to provide informa-tion at all. Populainforma-tion metrics depend less on the active engagements or deliberate

actions of subjects. Governments build policy upon knowledge on what people do or don’t do, rather than what they say they do, or who they say they are.

3.5 The Theoretical Framework

The theoretical concepts introduced in the sections above make up the theoretical framework of this thesis and the forthcoming analysis. Individually, they address

particularities and themes to the empirical material. However, if applied together as a guiding framework, they paint a larger picture of interconnected practices. The

theoretical framework will function as a map in the forthcoming analysis, from which patterns and structures can be detected and made visible. As a guiding framework,

it seeks to display patterns beyond what is instantly visible in the empirical material. It is indented to point to connections beyond already familiar frames of reference.

(24)

Chapter 4

Methodology

4.1 Methodological Implications of Reading Security

Politically

A critical stance in security studies has methodological implications. Rather than uncritically adopting a positivist notion of method as toolkits for research,

Clau-dia Aradau, Jef Huysmans, Andrew Neal and Nadine Voelkner (2014) stress the political dimension to method in research. The production of knowledge, they

ar-gue, is not normatively or politically neutral. A threshold for relevant, valuable or correct knowledge is set in a historical context with political dynamics (Aradau

et.al 2014:14). Methods bring perceptions of ‘truthfulness’ into being (Aradau and Huysmans 2013:598). To Xavier Guillaume (2012:29), it is therefore crucial that critical scholars acknowledge their ontological assumptions as well as their

episte-mological choices as political. Deciding upon research design and methodological choices should hence entail a reflexive mediation rather than follow a pre-existing

code of academic praxis. To this end, I will briefly reflect upon the ontological preconceptions guiding this thesis. At large, the ontological assumptions of this

thesis adhere to an interpretivist school of thought. An interpretivist ontology en-tails that reality is constructed (Yanow 2013:6). As such, there is no true or hidden

dimension to reality that can be captured or discovered by social scientists. Instead, realities, as they are known and given meaning by humans, are meditated through

(25)

interpretation. To Dvora Yanow (2013:6), “it is an approach that sees concepts and

categories as embodying and reflecting the point of view of their creators”. When conducting security analysis, this ontological perspective has two key implications.

The first implication regards research subjects as co-constructing security. This ties into Huysmans (2014:14) differentiation between a security reading of politics and a political reading of security. By the latter, security issues or the referent object of

security, is formed politically. From this perspective, security is not a static state of affairs. Nor is it an end-point to which all security measures strive. Security is

performed, enacted, and pursued by politicians, government officials, security pro-fessionals, private company officials, lobbying groups, individual citizens, and many

more. When setting out to examine security issues, an interpretivist approach is interested in security as co-constructed by these individuals. By which securitising

techniques do they bring insecurities into being? What are the political effects of these insecurities brought into being? The second implication regards the researcher.

As ‘reality’ is mediated through interpretation (Yanow 2013:6), the analysis of the ghetto package is in no way complete or exhaustive. Rather, it is one way of read-ing it. This thesis will emphasise some areas of particular interest in relation to a

number of theoretical concepts. Understanding the ghetto package will hence be mediated by several research choices, as well as an interpretive reading guided by

the theoretical framework. By adhering to an interpretivist ontology, both the ma-terial and the reading of it entail partaking in co-constructing security. Following

this reflection on ontology, I will briefly address some of my epistemological choices.

4.2 Conducting a Critical Single-Case Study

In conducting a case study, Gary Thomas (2011) differentiate between the subject

and the object of a case. He makes a distinction between the subject, as the event or policy of interest, and the object, as the theory or analytical framework from which

the subject is analytically explored. “The subject of the study is thus an instance of some phenomenon, and the latter—the phenomenon— comprises the analytical

frame” (Thomas 2011:512). Leaving out the object, research becomes descriptive rather than explanatory (Thomas 2011:513). In line with Thomas’ typology, the

(26)

Danish ghetto package is the subject of this thesis. It is not deemed representative of

a wider population of policies or legislative packages. Instead, it is selected because it particularly well displays some limitations on democratic practices by securitising

techniques. As such, it is selected on being a key case, or in Thomas’ words, “because it is an interesting or unusual or revealing example through which the lineaments of the object can be refracted” (2011:514). The object of this thesis is hence Huysmans’

theory on securitising techniques and limits to democracy. It is redundant whether the Danish ghetto package is a typical or exceptional case of securitising techniques

(Thomas 2011:514). Its utility lies in its capacity to function as an illuminating example of theoretical concepts brought forward by Huysmans and others. Selecting

the Danish ghetto package as the case for this research however entails a number of decisions. The first regards purpose, and the aim of this research to be theory

guided rather than illustrative or descriptive (Thomas 2011:516). If categorised by typology, this thesis adheres to a theory seeking approach. By that, the aim is not

to falsify the theory, but to add to the field by exploring and applying theoretical concepts (Kuhn 1972:91, in May 2010:34). Thereto, it has no comparative scope. It is a single-case study, examining one policy plan. From a critical lens, the way

securitising practices are enacted is in one way or another always idiosyncratic (Neal 2012:42). They evolve within particular contextual settings, where they are

continuously rearranged and negotiated. To engage empirically with securitising practices empirically hence offers valuable insights to the theoretical field of critical

security studies.

When conducting data collection and analysis, I methodologically adhere to a crit-ical approach that favours a more relaxed stance in research by improvisation and

bricolage (Aradau et.al 2014:7). Rather than begin the research process by establish-ing predetermined ways of research, Aradau et.al (2014:7) suggest that research is conducted experimentally, by assembling concepts, methods and empirical data. To

this approach, ‘messiness’ or complexity cannot be overcome by structured method-ological tools. Rather, complexity is inherent to reality. This approach resonates

particularly well with the framework and scope of this thesis. As insecurities are dispersed, what institutions, agents, policy documents, governmental practices or

(27)

articles that may be interesting to the analysis will not be identified beforehand.

The process of collecting data is rather one of ‘going back and forth’ between empir-ics and theory. This approach is however different from that of aimlessly searching

or ‘anything goes’ (Aradau et.al 2014:9). Huysmans’ (2014) and Ruppert’s (2011) theoretical concepts of diffusing and associating insecurities, rhizomatic surveillance, segmentation and transactional data function as a map in data collection. As the

theoretical framework guides data collection, some institutions, some agencies, some policies, and some documents, are rendered more useful than others. I adopt Tim

May’s (2011:27) stance on social theory as informative in making research decisions and sensitising empirical data. “Theories enable us to break free from everyday

thinking to consider issues beyond our normal frames of reference” (May 2011:29). Going back and forth with theory, new insights and connections appear in the

em-pirical material. I will consciously keep the messiness in my material until the very end. By then, more structured categories have crystalized, connecting the empirical

data to the theoretical concepts in interesting and illuminating ways.

4.3 Text, Techniques and Practice – The Selection of

Material

In line with a single-case study approach, data collection is however set by tem-poral and institutional boundaries (Thomas 2011:516). All empirical material is

published between 2015 and 2019. The ghetto package was introduced and imple-mented in 2018, however, the Danish centre-right government led by Lars Løkke

Rasmussen was elected in 2015 and replaced by a social democratic government in 2019. Implementation has however continued with the current government. The time frame is hence that of the centre-right government in power. As the case

sub-ject is a governmental policy, state agencies will be the primary resources for data. Departing from a dispersed notion of securitisation however, additional actors

be-yond state agencies will be examined as well. Huysmans (2014) find insecurities to be increasingly dispersed, spreading beyond previously bounded security spheres.

Surveillance techniques equally spread rhizomatically, to new areas of society as well as new agents of surveillance.These insights call for an initial broader search

(28)

for empirical material, looking at the Danish police, the Danish intelligence service,

municipalities, state departments, as well as private security companies, lobbying networks and consultancy firms. This approach aligns with a more fractured

in-terpretation of securitising practices (Huysmans 2014:127). In proceeding with the analytical process however, the material will be narrowed down to texts that relate to the ghetto package initiative. In conducting the empirical analysis, the empirical

material will include the Prime Minister’s New Year’s speech, reports on ghettos and/or parallel society issues, the ghetto package policy document and ghetto

pack-age related policies for governmental pack-agencies.

The empirical data is solely made up of text. For the purpose of this thesis, the text in itself is secondary to the practices that are described and commanded by the

texts (for a discursive approach on the concept ‘Ghetto’ in Denmark, see Bakkær Simonsen 2016). These practices include increased penalty zones, database

interac-tion, ordering house demolition in selected city areas, making day-care mandatory, and more. By approaching text as practices, they are not viewed as displays of hidden meaning. Rather, texts are treated as descriptions of intended

governmen-tal practice. Bigo (2000:194) highlights some limitations with relying primarily on state level and text in approaching securitising practices. The practical work, he

argues, does not necessarily reflect policy or discourse. Securitising is rather a field effect, reflecting agreements and struggles of the many security professionals and

ex-perts engaged in conducting everyday security work (Bigo 2000:194). When looking specifically at the context of Denmark, texts may however offer valuable insights to

securitising practices. In David Sausdal’s (2019) 900 hours ethnographic fieldwork in two Danish police departments, police officers vocally disagreed with the new

policy for policing terrorism. Sausdal (2019:3), in line with Bigo (2000), identified a different point of view in managerial policy from that of individual officers. Police officers did not agree with, or endorse, the new ways of policing terrorism

(Saus-dal 2019:17). They were rather annoyed by the terror preventive tasks that kept them from carrying on with their “actual work” (Sausdal 2019:12). Despite their

disagreement or annoyance, Sausdal (2019:12) however accounts repeatedly for po-lice officers preforming their assigned tasks. Thereto, he finds that “Danish officers

(29)

had become more heavily armed, they had been legally granted extra surveillance

and stop-and-search discretions, and their ethnic profiling had become cruder ow-ing to racialized stereotypes” (Sausdal 2019:8). In policow-ing terrorism in Denmark,

security professionals (police officers in this case) disagreed with the policy yet exe-cuted assigned tasks accordingly. The new policing policy also rendered changes in equipment and techniques. Mette-Louise Johansen (2019) equally build on

empiri-cal fieldwork when examining a multiagency strategy to counter violent extremism in Denmark’s second largest city Aarhus, the so-called “Danish model”. She finds

that by following new instructions, concerns of radicalisation were increasingly filed by professionals in the public youth sector (Johansen 2019:481). With this new

ap-proach, youth professionals departed from their previous information chain where they report concerns on children’s welfare to social services. Instead, concerns were

reported to the police (Johansen 2019:485). According to Johansen, the new Danish model changed the direction of information exchange (2019:485). Rather than

be-ing anecdotal encounters of Danish public officials, the empirical findbe-ings of Sausdal and Johansen display the effect of policy on practical work in the securitising field in a Danish context. In the context of interest to this thesis, text is hence found

to say something about practice. Ethnographic fieldwork offers more detail and complexity to the securitising field, especially as professionals adapt and resist

pol-icy. Nevertheless, for mapping securitising techniques and some implications they render on democratic practice, I consider texts useful and potentially insightful. As

(30)

Chapter 5

Empirical Analysis

The analysis is divided into two parts. The first part examines a number of se-curitising techniques in the Danish ghetto package. The structure of this part is guided by the theoretical concepts, examining diffusing and associating insecurities,

rhizomatic surveillance, reassembling surveillance and segmentation and biometric and transactional data. It is important to emphasise here that these are some

out of many securitising techniques accounted for by Huysmans (2014) and others. The analysis is hence not exhaustive. The second part builds on the first part by

examining ways these securitising techniques enact limits to democracy. It looks specifically at democratic control of surveillance and political subjectivity.

5.1 Securitising Techniques in the Ghetto Package

5.1.1 Diffusing and Associating Insecurities

In his New Years’ speech, the former Prime Minister Lars Løkke Rasmussen states

that he is deeply concerned (Regeringen 2018a:2). He “seriously fears” the new real-ities of parallel societies and the way they spread beyond ghetto areas (Regeringen

2018a:3, translated by author). Among these new realities, he lists unemployment, forced marriage, and gender inequality alongside gang criminality, social control and child abuse. When listed horizontally, it is not clear where sources of serious

fear – insecurities – end, and other issues begin. The Prime Minister’s speech hence works by diffusing insecurities discursively. Security issues are not interlinked, nor

(31)

connected to a single organisational unit. Security then continues beyond its

tradi-tional spheres of criminality and violence (Huysmans 2014:83). These continuities extend the meaning of security, renegotiating the clear distinctions between

secu-rity and non-secusecu-rity (Huysmans 2014:80). As a multisectoral policy, the ghetto package equally diffuses insecurities in practice. According to Huysmans (2014:88), diffusing insecurities enables security practices to travel beyond its former limits.

The multisectoral approach makes implementing the ghetto package a concern for many state actors, there among police, social services, schools, private building

com-panies, criminal correction systems and municipality officials. These actors are not tasked to provide security. However, their everyday decisions are tied to a larger

security incentive that the ghetto package entails. In reporting, counting, testing language skills, offering or declining housing and terminating social services, they

are collectively guided towards this texture of unease outlined by the Danish Prime Minister (2018a). Managing security issues is dispersed over many points of decision

(Huysmans 2014:186).

The package however securitises equally by linking together seemingly non-security

related issues by association. Areas in need of improvement; voting, actively taking part in civil society, school results and language skills, are listed next to criminality,

domestic violence, child abuse and violent extremism (Regeringen 2018b). There is no explicit relation between them, nor is it stated in what way they contribute to

ghettos or parallel societies. They merely co-exist. The package displays a general context of insecurity by associating these issues as sources of unease (Huysmans

2014:85). When listed together, the different areas of improvement are enacted as interrelated with spheres of insecurity. The ghetto package hence displays an

as-sociative relation between residential areas, the residents and their ethnicity, their occupation and their language skills, and insecurities of violence and crime. This relation is not causal, nor cumulative. It is rather one of listing issues horizontally,

“putting them side by side” (Huysmans 2014:85). Issues are outlined as separate, continuously retaining significant diversity and multiplicity, yet together, as

compo-nents to the larger ghetto issue. Its associative character extends the connotations to insecurity as it increases the number of units included (Huysmans 2014:85). When

(32)

read through the theoretical lens of Huysmans, the ghetto package does not

articu-late a threat, but rather diffuses insecurities, linking them by association. In other words, the issues listed in the ghetto package resembles sources of unease rather than

urgent security threats. As embedded in everyday life, that is where they can be prevented. The ghetto package hence calls for a multiagency approach, increasing the number of agencies concerned with the everyday life of ghetto residents.

5.1.2 A Rhizomatic Surveillance Structure

This multiagency approach in the ghetto package can be read as a rhizomatic surveil-lance network (Huysmans 2014:111). When pursuing their everyday work, Danish

agencies collect data on citizens. The criminal correction agency hold data on crim-inal records. The tax agency hold data on employment. The municipality hold data on public housing and tenants. Schools hold data on student performance and

attendance. Holding data on citizens is in no way new to the agencies concerned, however, as of the ghetto package, more actors are tasked with data collection,

in ways that serves new purposes. With the numerous yet seemingly unrelated agencies involved in the ghetto package, the information network they make up is

rhizomatic. No central agency holds the data. Instead, actors engage horizontally upon request. The ghetto package suggests that persons with a criminal record will

be denied residency in ghetto areas (Regeringen 2018b:23). Thereto, persons living in public housing in ghettos who commit criminal acts will be evicted from their

homes (Regeringen 2018b:23). By these suggestions, the ghetto package requires a temporary information exchange between police, municipal offices and public hous-ing agencies. The concerned agencies are tasked to combine their respective data

before making decisions if citizens can rent, or in case they commit a crime. The agencies are not fully linked, nor are they completely detached. Rather, they remain

separated. From this perspective, the ghetto package not only encourages dispersed surveillance data gathering, but rather depend on it. It hence works by

renego-tiating the information structure and increase the number of agencies required to govern ghetto areas. By that, the data gathering structure expands, including more

agencies. In porously linking these agencies, the ghetto package makes surveillance increasingly rhizomatic (Huysmans 2014:113).

(33)

More importantly however, the ghetto package alters surveillance purposes, by

en-act suspicion as the organising principle of relating (Huysmans 2014:104). With the ghetto package, day-care personnel are tasked to register and report young

children’s attendance in daycare and their performance in learning Danish (Børne -og Undervisningsministeriet 2019:2). Rather than a concern of children’s welfare, or enable staff scheduling, reporting attendance is a way to mitigate the risk of

children becoming isolated. When abstracted as data, children’s performance and attendance are stripped of contextuality, there among the motive for parents to

abstain from day-care. Instead, simply by living in a ghetto labelled area, any ab-sence from day-care could potentially be a sign of the risk of isolation. In the ghetto

package, suspicion of isolation makes up the background condition against which agencies and individual personnel are tasked to mitigate risk (Huysmans 2014:104).

It leads by general categories rather than individual actions or particular context (Huysmans 2014:104). Day-care personnel are hence tasked to take part in the

work of preventing the risk of parallel societies and ghettos by gathering data and reporting it to municipalities.

The ghetto package’s rhizomatic surveillance structure and suspicion as an organis-ing principle resonates Johansen’s (2019:485) findorganis-ings when examinorganis-ing the Danish

multisectoral approach to counter violent extremism (CVE). Two parallel changes distinguish the new model from previous multisectoral work in Denmark; the

en-largement of the information network and directions of agency interaction. The first refers to the extended reach of state agencies tasked to report concerns of

radicalisation. The new model has generated an expansion of security infrastruc-ture as the responsibility to monitor and report deviation has been partly

decen-tralised (Johansen 2019:484). The second refers to directions of information ex-change. Groups previously tasked with the welfare of children – a social matter – now report concerns to the police on potential violent extremism – a state security

matter (Johansen 2019:485). Including multiple agencies, the ghetto package, and CVE measures alike, extend and enlarge rhizomatically, by including more state

agencies and other actors. Thereto, the ghetto package and CVE measures have made suspicion the mode of operation by making security their overall organising

(34)

principle, rather than social equality, welfare or perhaps justice. As securitising

techniques, rhizomatic surveillance structures attend to everyday activities. They build on the social texture of unease to render bits and pieces of everyday life

po-tentially suspicious. A child’s absence from day-care may be a conscious parental choice, but it may also be an indication of risk for isolation. As suspicion is enacted the principle of relating in the ghetto package, when living in ghetto areas, it is by

default considered the latter. As such, parents need to be subjected to surveillance and reporting by day-care personnel (Børne-og Undervisningsministeriet 2019:2). It

is important to emphasise that not all Danish children will be subjected to this form of attendance and performance reporting. Day-care from a certain age is mandatory

only for children living in ghetto areas (Regeringen 2018b:25). The following section will hence engage further with practices of purposely reassembling surveillance data

in order to enact population segments and pursue targeted interventions.

5.1.3 Reassembling Surveillance Data

A financial report on parallel societies from February 2018 voices the paradox in detecting the occurrence of parallel societies. Parallel societies are physically or

mentally isolated, not seeking to actively engage with the Danish society (Regerin-gen 2018c:1). The report stresses that “parallel societies are about the individual’s

identity and values” (Regeringen 2018c:1, translated by author). It is hence difficult to detect by statistical data alone how many persons of non-Western origin live in

parallel societies (Regeringen 2018c:8). The report however states that indicators of isolation – a central feature of parallel societies – can be detected by register data. “On the other hand, on the basis of register data, it is possible to set up various

indicators for persons of non-Western decent who live relatively isolated from the surrounding community” (Regeringen 2018c:1). With these selected indicators, the

Danish government estimates how many immigrants and descendants live isolated from the Danish community (Regeringen 2018c:8). The indicators focus on living,

working and going to school in a non-Western majority group, individual’s criminal records and long-term unemployment. Out of 180.000 families of non-Western

de-cent, 11.000 families are affected by at least three out of eight indicators (Regeringen 2018c:9).

(35)

From statistical data, it seems that the Danish society indeed suffers from isolation

and parallel societies. Persons potentially isolated are as many 74.000 (Regeringen 2018c:11). Residential areas that qualify for the extended list of vulnerable areas

are 55, with 22 of them being ghettos (Regeringen 2018c:11). A conclusion one may draw from this is that some policy is indeed needed to resolve the issues of isolation and ghettos. From a critical view however, the data making up these

cri-teria could be read as a process of purposely reassembling surveillance (Ruppert 2011:222). Surveillance data is no accurate reflection of individual citizens. Rather,

data are abstract pieces of information retrieved from persons in standardised form. Hence, they are stripped of personality, history, consciousness or context (Huysmans

2014:97). Data can then be intentionally reassembled to identify new patterns or correlations (Ruppert 2011:222). The ghetto package may be read as the framework

deciding what ways data shall be reassembled. In the package, data is clustered and given meaning. One of the five criteria for ghetto listing is that the number of

non-Western residents living in the area exceeds 50 percent (Regeringen 2018c:8). Another is that more than 60 percent of the residents aged 30-59 at most have a pri-mary education (Regeringen 2018b:11). Making a distinction between immigrants

from West and non-West mirrors a perception of difference between the two. It is not immigration per se that makes up the criteria for ghetto areas. The selection

displays a conscious division between some countries and other. In the words of David Lyon (2003:22), complex information infrastructures “has not ensured that

the identities and data doubles are classified free from stereotypes or other prej-udicial typing”. Thereto, data measurements are not static. Helle Lykke Nielsen

(2019:7) accounts for the changes made in the ghetto criteria. When deciding on the education criteria in the ghetto package, the government excluded education

obtained outside of Denmark, contrary to what was done just a few years earlier. With the new measurement, only education obtained in Denmark or officially ap-proved in Denmark would count as valid (Nielsen 2019:7). Nielsen (2019:7) further

points out that the education criteria changed in between the previous preparatory reports and the final ghetto package policy document. With this change, the

(36)

from more than 50 percent (Regeringen 2018c:8) to more than 60 percent

(Regerin-gen 2018b:11). Alterations like the ones identified by Nielsen disclose the selective dimension to reassembling data. They cannot be read as “raw data” or objective

accounts of reality (Ruppert and Savage 2011:88). Instead, they are made into cat-egories by politically selected criteria. The ghetto package enacts these catcat-egories by deciding ways to reassemble surveillance data. In doing so, the ghetto package

enacts segmentation.

Segmentation

As a securitising technique, segmentation distinguishes areas, population groups or

transactions by reassembling surveillance data. From data, state agencies classify population groups as potential security risks by using various criteria of income,

citizenship, ethnicity, criminal record, family relations, etc. The segments are then subjected to targeted interventions. Policies are put in place to mitigate identi-fied segments, either by altering their composition or by monitoring them closely

(Huysmans 2014:125). The securitising practice of segmentation is hence dual. It refers to the assembling process of clustering data that produce areas or groups

of particular interest – the segment. It equally entails governing interventions tar-geting these segments (Haggerty and Ericson 2000:606), often differently from the

way society is governed at large. With the Danish ghetto package, the former en-tails the five criteria for identifying ghetto areas (Regeringen 2018c:6). The latter

entails policies that make up the ghetto package, as they differ from the way Dan-ish society is governed at large. The ghetto package has made it mandatory for

children living in ghetto areas from the age of one to attend day-care (Regeringen 2018b:24). If absent from day-care, parents will be deprived of state child support (Børne- og Undervisningsministeriet 2019:2). The package has also made it

possi-ble for the Danish police to decide on increased penalty zones (skærpet strafzone), where penalties for committing criminal offenses are raised severely or even doubled

(Regeringen 2018b:23). The government suggests a “scheme that allows the police to decide on special areas plagued with crime and unsafety where the penalties for

certain forms of crime are significantly increased for a period (increased penalty zone)” (Regeringen 2018b:23). Criminal acts in ghettos are hence governed

Security and Limits to Democracy : A Critical Study of the Danish Ghetto Package