1Closed circuit television (CCTV) is defined as “a TV system in which signals are not publicly distributed but are monitored, primarily for surveillance and security purposes” [Pedersen, 2011]. Today it represents a huge financial market worldwide – nearly $15 billion in 2013, with an expected annual growth rate of 18 % [Sudip 2015] – and has become a widespread urban feature in European cities.
2Open-street CCTV refers to the cameras used to monitor public space by public authorities or the police. Over the last two decades, its introduction has been widely examined from various perspectives, in the fields of sociology, philosophy and surveillance studies, among others. However, geographical studies are not numerous and are often limited in scope or were carried out at a time when open-street CCTV was confined to city centres [e.g. Galdon Clavell, 2013]. Nonetheless, in our view, a spatial approach provides a better understanding of this technology. The geography of CCTV highlights its uneven distribution throughout the territory, and shows that it is aimed at protecting specific interests and at monitoring specific social groups.
- 1 See Debailleul and De Keersmaecker [2014: Appendixes] for a detailed account of the interviews and (...)
3In order to establish the CCTV profile of Brussels, we first adopted an empirical approach. We mapped the spatial distribution of public open-street cameras in the Brussels-Capital Region. Secondly, we reviewed reports by police and city officials in local newspapers and conducted semi-structured interviews. We met 16 interviewees ranging from police and security agents who monitor the cameras, to police chiefs and local and regional authorities. Our goal was to clarify the process and reasons behind the location of the cameras.1 Finally, we checked the reliability of our data through fieldwork (each camera location was verified on site). The research was carried out in 2014 and the data were updated in January 2016.
Figure 1. Open-street camera on a façade in Rue de la Sécurité in Koekelberg
Public cameras are usually 360° domes placed at a height of about 6 metres on façades, street lamps or poles.
Photo: Corentin Debailleul
4In this paper, we begin by delineating the historical uptake of public cameras in Brussels. We then present our empirical results and continue with their analysis in light of the Region’s social, legal, fiscal and political context. The concluding section wraps up our results and analysis.
5In 1968, the public transport company STIB/MIVB installed its first surveillance cameras to monitor metro platforms and sales booths. Cameras appeared only sporadically in the public space during the 1990s. The European Football Championships (UEFA) in 2000 provided an opportunity for cameras to appear in the surroundings of the Heysel stadium. Simultaneously, cameras were placed on the small ring road which surrounds the pentagon, for traffic regulation purposes. The technology truly pervaded the city in 2003 when the regional government earmarked a budget of € 1.5 million to help municipalities develop open-street schemes in their own territories. All municipalities, with the exception of Auderghem and Watermael-Boitsfort, replied favourably and applied for the grant [Lamquin and Voogt, 2003].
6The first CCTV schemes were thus initiated by municipalities, but the control of public cameras was shifted gradually to the police areas. At the end of the 2000s, the CCTV networks became unified and new control rooms were set up in the main police stations (except for Koekelberg, which retained its municipal system). The most recent development plans were more coordinated, such as the four-phase plan to install 230 cameras in the Bruxelles Capitale Ixelles police area (Polbru) between 2005 and 2015 (for a budget of over €10 million), the ongoing plan for 130 cameras in the Brussels-North police area (€6 million) and, finally, the 108 cameras planned by the Brussels-South police area, which was delayed due to a case of embezzlement, but should be completed in 2016 [Dupont, 2012].
Figure 2. Open-street cameras in the municipalities and police areas in the Brussels-Capital Region (in 2015)
Table 1. Camera count per police area
- 2 Interview with Christian Pauwels (Brussels-North Police Area), 2014
7These large-scale plans provide the opportunity for a qualitative jump in video surveillance technologies: indeed, the new cameras are said to be “intelligent”. This means that they will be associated with video content analysis (VCA) applications which are programmed to detect so-called suspicious behaviour automatically, such as people remaining still for more than 20 seconds in a car park (to track car theft), or people gathering (to prevent riots). VCA will also target “incivilities” such as waste dumping and graffiti. Though in practice, our interviews revealed that this technology is still in its infancy, as the applications tend to sound alarms continuously when used in public space. This requires the attention of an employee whom no one can or wants to pay for, i.e. the problem which VCA is meant to solve.2
8Besides these police and municipal initiatives, STIB/MIVB is now completing the refurbishment of its metro stations, which includes the setting up of an extensive video surveillance network, fully funded by the Brussels-Capital Region. Its aim is to have electronic eyes covering up to 95 % of the area it manages. The transport company is already the largest CCTV owner in the region, with 1,800 units in buildings (and 4,000 in vehicles), though this number is expected to double when the fully automated metro is implemented [La Capitale, 2014].
- 3 Interview with Vincent De Wolf (Etterbeek municipality) and Christian Mahieu (Brussels-East Police (...)
9The monitoring of public video surveillance in Brussels is currently very scattered, to the discontent of regional authorities who would prefer a centralised network.3 Steps have therefore been taken in recent years to integrate the existing public camera schemes in the capital city.
10In 2013, the Brussels Regional Informatics Centre (BRIC) was appointed to study the implementation of a “comprehensive, integrated solution for video surveillance to equip Brussels with a regional network.” It will connect the networks of the different police areas with those of the Port of Brussels, the fire department, the public transport companies (STIB/MIVB and SNCB/NMBS) and the Brussels Ministry for Mobility, also known as MOBIRIS [Carlot, 2013].
- 4 Interview with Brecht Debusschere (MOBIRIS), 2014
11The integrated network will be inspired by the one in Paris, which includes more than one thousand cameras [Doucet, 2013]. In March 2014, the regional government voted in favour of the project, and the platform will eventually integrate more than 3,000 surveillance cameras [Belga, 2014; Le Soir, 2014a]. There has already been some collaboration. For instance, the Automatic Number Plate Recognition (ANPR) cameras installed by MOBIRIS in the Leopold II and Rogier tunnels to ensure that drivers are not exposed to fumes for more than 20 minutes, are used each week by the police in their inquiries.4
12According to our interviews with the authorities in charge of CCTV (be it from the local police areas or from the public transport company), it emerged that there was limited enthusiasm regarding this politically driven plan. They feared that they would lose control of “their” cameras, stating that they wished to “keep their hand on the joystick”.
13By law, people are required to inform the Privacy Commission when setting up CCTV systems [Moniteur belge, 2007]. The Commission then adds a new entry in its online public registry. As we were aware that the registry was far from exhaustive, we decided to focus on public cameras, based on the assumption that public authorities would have fulfilled their duty by making the data easily accessible to us. The reality proved otherwise, however, as we eventually found twice as many cameras as there were in the registry.
14In the end, we had to use 22 different sources. Data were acquired mostly from the police, either through direct contact or from yearly reports (334 cameras); from the Privacy Commission registry (321); from newspaper articles (139); from the websites of municipalities or communications with civil servants (69); from previous studies (69); and finally from our own fieldwork (59) [Dessouroux et al., 2009; La DH, 2002; Jette, 2014; Nimy, 2014; Privacy Commission, n.d.; Romainville, 2005; Marlow Police Zone, 2013; Vlan, 2009 and personal communications].
15In total, we mapped 1,365 public open-street cameras in the Brussels-Capital Region. 378 cameras are monitored by MOBIRIS (symbolised by pink diamonds for the 348 in the tunnels and by green pentagons for the 30 outside). 984 cameras are monitored by the local police (or by municipal stewards in the case of Koekelberg). All of them were verified in the field or using Google Street View. 796 cameras were present (orange dots on the map), while 69 were not found for unknown reasons (black dots).
16The blue triangles represent the cameras already scheduled to be set up in the near future. The map also indicates a few cameras in Anderlecht which are permanently out of order (orange stars). The map used to indicate many more, but most of them will soon be replaced and are therefore represented by blue triangles.
17The municipalities of Saint-Gilles and Evere had considered installing cameras to fight against illegal waste dumping (white dots), however, the plans were abandoned. The 61 cameras owned by the Port of Brussels which are located along the canal were not represented, as we did not manage to obtain accurate data. We added the metro stations to our map as the STIB/MIVB has placed 1,800 cameras, but we do not know their exact distribution.
Figure 3. Open-Street Cameras in the Brussels-Capital Region
18The map shows that the neighbourhoods with the highest number of surveillance cameras are the city centre and the working-class and migrant neighbourhoods (e.g. Matongé and Molenbeek’s historic centre). Many public cameras can also be found in the area of the European institutions, in the traditional working-class neighbourhood of the centre, or in the northern part of the city (Schaerbeek). The camera density tends to decrease as we move away from the centre, with the exception of some local commercial hubs and town halls. A large share of Brussels is not subject to public video surveillance, notably the entire municipalities of Auderghem and Watermael-Boitsfort. With Woluwe and Uccle (which have rather low densities of cameras), they are part of the south-eastern quadrant of the city, characterised by its relative wealth.
19For 867 cameras (88 % of all listed open-street police cameras), we checked the elements surrounding them through quick fieldwork and using Google Maps, Open-Street Map or Google Street View. It is therefore quite a rough overview and is far from being an in-depth ethnographic study of places under surveillance. The chart below represents the features present in the surroundings of the cameras studied in over 4 % of cases.
Figure 4. Features likely to be found next to open-street cameras in the Brussels-Capital Region
20Many cameras were placed in secondary centres in the city, such as Town Hall squares, church forecourts, local retail nodes, or large intersections. Main roads and metro, tram and bus stops are sites of arrival in – and departure from – a particular space or municipality. Like train stations, they are places of high mobility and are very likely to be monitored. In the same way, places of very low mobility such as squares, bars, parks and parking lots are also often watched. Finally, there are places which require protection due to their intrinsic interest, such as monuments and tourist attractions (the Atomium, Manneken Pis and Grand’Place), national interest (Ministries, Royal Palace and national symbols), and municipal interest (Town Hall, libraries, schools and social housing). In the last case, local councils seem to behave like private owners protecting their belongings.
“Video surveillance simply modernises Newman’s defensible space principles” [Leclercq 2009: 144-145, our translation].
21We consider CCTV as part of defensive urbanism, a concept which allows us to understand its presence. The objective is to place video surveillance in a wider socio-economic context and to link it theoretically to other means of surveillance and social control, as well as to the urban environment in which it is embedded. This allows an understanding of the context, purpose and effects of video surveillance.
22The goal of defensive urbanism is to implement environmental design policies which make the city more secure both by deterring criminal behaviour and by enhancing the sense of ownership and responsibility of inhabitants with respect to their living environment. Its founding publication is the book Defensible Space. People and Design in the Violent City, by architect and city planner Oscar Newman . This theory is based on the questionable assertion that social problems can be solved by technical urban design and architectural measures.
- 5 Interview with Benjamin Goltzberg (Municipality of Schaerbeek), 2014
23In practice, the defensive approach has been limited in Brussels as compared to the American cities where the theory arose. Nonetheless, the principles of defensive urbanism are visible in the built environment, the most famous example being the benches designed to prevent homeless people from using them as shelters for a night. Defensive principles have also been used in the refurbishment processes of some public squares in Brussels, such the Carrefour de l’Europe in front of the Central Station, or Place Flagey in Ixelles. Typically, these new spaces were designed to avoid any potential hiding corners and allow rapid and efficient police interventions. To do so, the Ministry of the Interior provided grants to hire specialists in “techno-prevention”.5
Figure 5. A new level of defensive urbanism: at the cutting edge of CCTV in Rue de la Compétition, Anderlecht
Photo: Corentin Debailleul
24In this perspective, CCTV may be considered as a means of surveillance used to implement one of the goals of defensive urbanism, i.e. reducing crime through the shaping of public spaces. Indeed, CCTV clearly contributes to reducing the number of hidden spaces (one of the ways in which defensive urbanism operates), which in our view, illustrates an obsessive need for the authorities to minimise risks and maximise control.
25The increase in technical means of surveillance should not lead us to overlook traditional forms of control, among which human activity still has a central role. In addition to traditional police officers, private security agents are increasing in number and have recently received new attributions [Belga, 2012], municipal stewards are likely to have their power extended and join the traditional police force [Sadani, 2014], and civil volunteers may be hired more regularly to help with police action [Le Soir, 2014b]. The army started roaming the streets after the November 2015 Paris attacks. At the same time, municipalities have been employing experts to confront the “radicalisation process” and school teachers are asked to investigate and report early warning signs [Simonet, 2013]. The grey area which blurs the distinction between social and repressive actions is therefore expanding as a sign of the “integrated logic” of the social-securitarian state [Réa, 2007].
26Sensational events are always used as an argument in favour of the implementation or extension of CCTV networks [see e.g. Coleman and Sim, 2000; Hier, Walby and Greenberg, 2006; Carr, 2016]. In the case of Brussels, the murder of Joe Van Holsbeek at the Central Station in April 2006 is emblematic of the use of sensational events to justify open-street CCTV [TV Brussel, 2008; brusselsnieuws, 2009; De Wolf and Mahieu, 2014]. The use of such events causes the debate to become irrational. When those who oppose video surveillance are accused of being on the side of the murderers, critical thought is made impossible. Moreover, it is fundamental to investigate the reasons behind the success of video surveillance, as the usual legitimisation for its implementation is groundless for the most part. Indeed, the argument of the struggle against crime is unwarranted, as studies show that it has a limited effect on nuisance and that there is a spatial shift of offences rather than a reduction in crime [IAU, 2008; Mortelé and Deprins, 2012].
27In their analysis of the worldwide distribution of CCTV in public spaces, Norris, McCahill and Wood  identified the socio-economic, legal, fiscal and political contexts as the four decisive factors in determining the stage of development of a local CCTV scheme. We decided to adopt this multivariate approach as we deemed it particularly appropriate in order to grasp the complexity of the uptake of video surveillance in Brussels.
28Brussels can be described as concomitantly very rich and very poor – in short, unequal. As the third region in the European Union in terms of GDP [Eurostat, 2012], the youth unemployment rate in some of its neighbourhoods is nevertheless over 50 % [IBSA, 2015]. Norris, McCahill and Wood [2004: 121] point to the use of CCTV as a “package to revitalise the city centres by mimicking the security measures found in the malls and could be used to monitor and regulate an increasingly visible underclass”. The most deprived districts are concentrated within walking distance of the commercial and tourist centre, making social unrest latent at the city gates.
Figure 6. Income distribution in the Brussels-Capital Region
29When cameras were installed in the Bruxelles Capitale Ixelles police area, Matongé was among the first and most monitored locations. This historically Congolese neighbourhood is home to a relatively poor population and faces issues involving drug trafficking, but it is also located next to Avenue Louise and its luxury shops, the shops in Chaussée d’Ixelles, and the European Institutions. Similarly, the poorest neighbourhoods (Cureghem, Molenbeek, West Station) are located just next to the city centre and its commercial and tourist activity. The reason for fearing such situations is based on the fact that when inequalities are strong, visible and stand alongside each other, social problems including crime have proven to be higher [Brennan 2012: 7-8]. According to this logic, poverty itself is considered less of a problem than wealth disparities, especially if spatially close.
30The importance of socio-economic factors was confirmed through quantitative analysis: significant correlations were highlighted at the neighbourhood scale (as defined by the Monitoring des Quartiers) between the density of open-street police cameras and the
share of Africans (r=0.534, p=4.3x10-10)
unemployment rate (r=0.522, p=4.3x10-10).
Inversely, negative correlations were found with the
share of elderly (65+) (r=-0.525, p=8.9x10-10)
average income per inhabitant (r=-0.494, p=1.2x10-8).
[Data source: IBSA 2015]
31In some countries, laws or constitutions have acted as a barrier against the proliferation of CCTV, at least temporarily. This is not the case in Belgium, where the legal framework is rather permissive and has simply evolved with technological innovations and police practices.
32In Belgium, the Privacy Act adopted in 1992 oversees the protection of privacy in relation to the processing of personal data. The Privacy Commission was established together with this act to ensure the fair and protected use of personal data. The 2007 Camera Act regulates the installation and use of surveillance cameras. It distinguishes between open spaces, publicly accessible closed spaces, and closed spaces which are not accessible by the public. A camera can only be placed in an open space with the agreement of the municipal council, which first has to consult with the local police chief [Moniteur Belge, 2007: Art. 5]. There are restrictions on the purposes legally fulfilled by cameras in publicly accessible closed spaces, but their installation does not otherwise require any particular authorisation [Art. 6]. While cameras in closed spaces can be installed freely [Art. 7], all CCTV cameras, regardless of the type of space, require a declaration made to the local police chief and the Privacy Commission [Art. 5, 6 and 7].
33It is forbidden for private persons to specifically film areas which they are not responsible for, except with the agreement of those in charge [Art. 5, §3]. In theory, this denies people the right to monitor public space, but the vagueness of the term “specifically” [Atas and Deroover, 2009: 90] as well as experience in the field show that the opposite is true. This situation implies that an exhaustive study of video surveillance in public space should not limit itself to the analysis of public cameras, and highlights the main restriction of this article.
34The aim of the Camera Act was to simplify the use of CCTV and, as such, it was a success. Its most problematic aspect was modified the very year of its creation: the obligation to produce a “security and efficiency study” before any CCTV implementation in the public space was simply removed [Dumortier, 2009: 19]. However, the obligation to show some restraint and comply with the principles of proportionality, appropriateness and minimum intervention, which are supposed to preside in the choice of installing a camera, is simply neglected. Similarly, it is mandatory to declare and specify the aim(s) of a surveillance system to the authorities. However, stating that a camera will be used for “security matters” or for “surveillance purposes” is enough. These types of justification are so broad and vague that they actually allow an almost unlimited use.
35While laws are expected to determine the framework of police action, the opposite is often true: first the police act, and then laws are adjusted to legitimise practices [Neocleous, 2000]. For example, recent on-board camera and “bodycam” systems in use in Brussels-North simply do not comply with signage laws. Instead of respecting the law and abstaining from using these tools – as their lack of legal compliance should imply – the police use them anyway, as they assume that the law will soon be adapted to suit their practices.
- 6 Interview with Chris Cortoos (STIB/MIVB), 2014
36As mentioned above, the Brussels-Capital Region allocated a budget of € 1.5 million in 2003 to help the municipalities develop their open-street CCTV schemes. 17 out of the 19 Brussels municipalities showed interest in this opportunity and benefitted from this grant. For example, Berchem and Jette both received € 49,000, while Etterbeek, which is more central, received € 148,000 [Lamquin and Voogt, 2003; Degreef, 2009]. CCTV systems are extremely costly, which means that municipalities rarely take a unilateral decision (and the budgetary responsibility) to implement one. Instead, they try to establish collaborations within their police areas and take advantage of funding. Similarly, STIB/MIVB cameras in metro stations are fully funded by the Region.6
37In addition to the aforementioned regional grants, the European Union also subsidises the municipalities of Brussels where the European Institutions are located and European summits are held. The Eurotop Fund amounts to € 20 million per year, divided between the different municipalities and police areas [AVCB, 2012: 7], although of course not exclusively for CCTV implementation.
- 7 Interview with Sylvie Murengerantwari (Ministry of the Interior), 2014
38At federal level, the Ministry of the Interior subsidises Belgian municipalities in order to improve “security and prevention”. To benefit from this aid, municipalities must have a population of over 30,000 inhabitants, conduct a “security diagnosis”, and have a prosperity index below the national average [Moniteur Belge 2013: Art. 2]. In total, the municipalities of Brussels receive € 6.8 million (20 % of the national budget), and up to 10 % of these grants can be spent on “investments”, including CCTV schemes.7
- 8 Interview with Benjamin Goltzberg (Municipality of Schaerbeek), 2014
39It is important to note that open-street CCTV is a costly technological tool and is mainly used for police activities, which means that these funds raise police budgets indirectly8 and constitute money transfers from the state to private corporations [Carr, 2016].
40When municipalities plan to implement a CCTV scheme, the Ministry of the Interior first advises them to define a goal for their system [Mortelé and Deprins, 2012: 37]. This exemplifies the typical absence – or at least vagueness – of reasons prompting politicians to have cameras installed. We agree with the authors who claim that cameras are first and foremost a political tool used by decision makers in an attempt to do something about insecurity issues, even if it is purely cosmetic [Norris, 2009; Carr, 2016]. Influenced by the media and the pressing demands of local shopkeepers, there is an overpowering need for policy makers to take action against insecurity [Bétin, Martinais and Renard, 2003]. At the same time, public officials tend to be stuck in the short-sightedness inherent to the regular deadlines of the election process. They have to look for simple, vote-catching solutions to include in their platform, which may be fulfilled effectively if they are elected. Our study shows that the political will is the driving force behind the implementation and expansion of open-street CCTV, while the police are often less convinced about it (as they are more aware of the technological limits). Nonetheless, the police welcome the projects when they do not have to fund them.
41The specificity of Brussels is that many different authorities are involved when it comes to implementing, developing and funding CCTV. The multitude of local interests and power struggles at stake shape the complex geopolitical dynamics whereby municipalities, police areas and other local stakeholders compete with each other. The most striking example is that of Koekelberg, where there is the highest CCTV concentration in Brussels and where many cameras have been placed at the “border” with the disadvantaged and ill-famed municipality of Molenbeek. Philippe Pivin, the mayor, declared [quoted by Moyart, 2012, our translation]:
“As for the theory that the network would only shift the problem to the surrounding municipalities, I am the mayor of Koekelberg and in charge of security for this territory. The others should take action or bear the consequences of their lack of action.”
42However, as we have seen, there is a current trend towards centralisation, and these local skirmishes are likely to lose significance in the coming years and eventually vanish. Let us not be mistaken: this does not mean the end of competition, but a change in scale. We live in a context where globalised competition between cities is a reality, and where public authorities tend to import techniques and policies from abroad, sometimes even without any pressure to do so. “A degree of international policy convergence has occurred” [Galdon Clavell 2013: 201].
43These “policy transfer” dynamics are particularly visible in Brussels, where the proliferation of CCTV and the future integrated network are inspired explicitly by the case of Paris. Right-wing MP and mayor of Etterbeek Vincent De Wolf – one of the biggest supporters of the project – took part in a parliamentary delegation in Paris with other officials and police chiefs. The object was to visit the city’s installations and use them as an inspiration. When we interviewed Mr. De Wolf, he expressed this logic of competition very clearly by stating that Brussels used to be ahead of Paris in terms of CCTV but that the trend has been reversed: the opposite situation prevails now, hence the “necessity” to take action.
- 9 Unfortunately, this account of the political context disregards the opposition to CCTV, as it goes (...)
44The same intercity competition can probably explain the prevalence of cameras in the city centre where the tourism and retail sectors as well as European and other governmental institutions have all succeeded in dictating the necessity to convert Brussels into a “safe” city, which is notably safe for investments. High concentrations of cameras are also found in the district of Laeken in the north-west, which hosted the Euro 2,000 football competition and which is currently the scene of major urban developments. This brings us to the western area along the canal where the rationale of the control of “racialised” poor populations prevails, together with a slow gentrification process. Such processes of “beautification under surveillance” [Dessouroux, Van Criekingen and Decroly, 2009] were recently pinpointed in other cases of formerly disadvantaged neighbourhoods adjoining city centres, such as Nørrebro in Copenhagen and Lavapies in Madrid [Ruiz Chasco, 2014].9
Figure 7. CCTV pole welded in order to prevent sabotage (Rue de la Compétition, Anderlecht)
Photo: Corentin Debailleul
45An analysis of the public video surveillance scheme in Brussels was carried out with the use of self-produced maps and semi-structured interviews. First, it appears that a series of contextual elements have created a favourable environment for CCTV development in Brussels. Second, the cameras are mainly concentrated in the consumption hubs as well as in the most socio-economically deprived areas, especially if these areas have a large “non-western” population and are located within walking distance from the city centre.
46Moreover, there seems to be an increased surveillance of these two types of space when they coexist. This can be explained by the fact that, in a context of a capitalist globalised economy and increased competition between cities, part of a city’s strategy to attract international capital is to make its environment “safer” [Galdon Clavell 2013].
47In conclusion, CCTV may be conceptualised at the crossroads of private economic interests, local political stakes, and faith in technological fixes. As Stephen Graham puts it, policies are shaped by “technophilic dreams and fetishistic urges for mastery and control. [...] Superior technology is thus the key to all locks, the silver bullet” [Graham, 2010: 162]. Nonetheless, the actual uptake of CCTV exists in different forms and intensities according to the city, and must be analysed in light of the local political economy [Galdon Clavell 2013].