Insecurity and the feeling of insecurity in Brussels. What the Security monitor tells us

Whilst security issues have once again invaded the political and media agenda these past months, with Brussels being described as a dangerous city, we are forced to recognise that very little attention has been given to the tools which allow a rational evaluation of the situation experienced by the inhabitants of Brussels. The present article consists in a partial analysis of one of these tools, namely the latest Security monitor, published in 2008. Which offences were reported most by the in-habitants of Brussels who were interviewed? What percentage of the population is affected? Do the inhabitants of Brussels feel insecure? These are some of the questions which we shall try to answer.But our answers are just a beginning, on the one hand because questions regarding insecurity and the feeling of insecurity require the utmost caution and, on the other, because the Security monitor has limitations, such as not being able to provide a precise description of the situation in Brussels in terms of insecurity. One therefore wonders why there does not appear to be a pertinent scale of analysis for Brussels as regards security issues.


Introduction
At the end of January 2010, the question of security in Brussels returned to the heart of the debate. Among other events, the muggings around a higher education institution and the violent attack on an exchange office provided an occasion to examine the issue of insecurity in Brussels once again. As is often the case with events which elicit collective emotions, the debate flared up: alarmist speeches in the public debate, anecdotes on the topic of urban insecurity, a growing number of articles and debates in the press, the sudden involvement of a series of institutional stakeholders, etc.
Once again, there was a rediscovery of the issue of security, the development of a heated debate, the expression of a highly diverse range of well-founded and informed opinions and the declaration of political initiatives, followed by an inevitable loss of interest.
On 12 April, following two fatal robberies in jewellery shops, the question was revived, setting the same cycle in motion once again.
The present article does not intend to express an opinion on the measures which were touched on following the recent crises, i.e. merging of the police areas, zero tolerance, calling into question of police supervision standards, rigorous application of sentences delivered, increase in measures for the placing of minors, etc. Instead, we shall try to shed light on the situation in Brussels in terms of security.
In order to do so, we shall discuss one of the most interesting diagnostic instruments in the area of security: the Security monitor. By a stroke of luck, a new edition was published a few weeks before its feverish consultation in January. The odds are that it will teach us much more about the situation in Brussels than a street poll. This being the case, one should not expect miracles: as we shall see, in its present form, this instrument does not give us a precise enough picture of the situation in the neighbourhoods of Brussels.
We shall focus on three questions: How can the victimisation of the inhabitants of Brussels be characterised? To what extent do they experience a feeling of insecurity? What is their view of the problems which exist in their neighbourhood?

What is the Security monitor?
The Security monitor is a survey which has been conducted every two years since 1998 at the request of the Federal Public Service (FPS) of the Interior. It involves a telephone survey conducted with a large representative sample of people over the age of fifteen who live in Belgium, and is focused mainly on the offences committed against them during the previous twelve months, their feeling of insecurity and their relationship with the police services. The questions asked concern the experiences of the person being interviewed and certain offences committed against his or her household.
The Security monitor has a local and a federal version. We shall discuss the latter here. The last edition was based on data gathering conducted in 2008 and at the beginning of 2009. It is based on a sample of 12,000 people, to which is added the 21,352 surveys conducted in the framework of local monitors and the 3,596 surveys from local police areas. 1 Of a total of 36,948 surveys, 12,000 have been used to make up the figures which we shall analyse here. This reduction in the number of surveys allows a selection based on predetermined geographical criteria (prestratification) to be made so that the sample reflects the truth as accurately as possible. The distribution of the population within the different territorial subdivisions in Belgium was thus determined and a corresponding proportion of surveys were selected randomly among those conducted in each of these entities. 2 Poststratification was then carried out: the surveys used were weighted so that the overall sample would reflect the characteristics of the population in terms of age and gender. 3 Finally, let us mention that the federal Security monitor is freely accessible on the website of the federal police. 4 The Security monitor has three main strong points. The first is the size of its sample, which makes it a reliable tool. The size of the sample allows the confidence interval Brussels Studies the e-journal for academic research on Brussels 2 1 Possibility to participate offered to police areas which were not included initially in the Security monitor.
(CI) to be minimised, thus maximising the probability that the figures obtained are an accurate reflection of the truth. 5 The second strong point of the Security monitor is that it is based on surveys conducted directly with citizens. This means that it provides us with information about people's experiences and not about the activity of a state agency. Thus, as regards criminality, police figures reflect their activity essentially. If a few hundred police officers are hired, their presence in the field is increased and hence the number of offences reported by them. The police figures are therefore a reflection of the activity of this institution as much as -or even more than -the reality in the field. Offences which are not very visible or not often reported to the police (petty theft, fraud, etc.) will appear to be infrequent, unlike those for which good detection systems exist (automatic radars, obligations to report events to the police in order to be reimbursed by an insurance company, etc.).
The third strong point is the age of the instrument. It has been published since 1997, and now allows us to see the evolution of survey responses.
The Security monitor, however, is not a perfect instrument. Firstly, it is based on a telephone survey and therefore only gathers information about people with a landline. Furthermore, a telephone survey may only be conducted if it is short enough. It is therefore impossible to obtain a highly detailed vision of the respondents' experiences.
Secondly, there are limits regarding the population interviewed. It is not made up of lawyers, and therefore the perception of the criminal nature of certain types of behaviour may differ from that of the police, the public prosecutor's office or criminal jurisdictions. This particularity is specific to all surveys conducted among nonspecialists, and may even be considered as an asset in as much as it provides a view of the social perception of a series of phenomena which is relatively independent of pre-existing rationalisations. Furthermore, the lower age groups are underrepresented in the initial sample (which leads to the over-weighting described above). This situation has been worsening since 1998, as less and less Belgian residents under the age of 35 are represented. This is more than likely due to the decrease in landline connections among the population (KUUSELA, CALLEGARO, VEHOVAR, 2008). Finally, the survey is conducted in French or in Dutch, which may be an obstacle for residents with a poor grasp of these languages. We can see how this problem could be particularly crucial in a multicultural city such as Brussels.
Brussels Studies the e-journal for academic research on Brussels 3 Thirdly, the Security monitor does not provide free access to raw data, which are only available to the beneficiaries of the monitor 6 . This creates problems when trying to establish certain comparisons centred on the situation in Brussels. In the available tables, figures are provided for federal level and each of the regions, as well as for certain categories of city: big cities, Brussels-18, urban area municipalities, small cities, etc. The city-region of Brussels is dealt with in a special way, at three different levels: the Brussels-Capital Region (BCR) is considered on a par with the other regions, Brussels City (and not the region as a whole) is integrated into the category of 'big cities' with four other cities (Antwerp, Charleroi, Ghent and Liege), and the other municipalities of the BCR are included in the group 'Brussels-18' (whose pertinence may be questioned). We therefore have a view of the global situation in Brussels via the 'Brussels' regional category. Furthermore, we may attempt to make a comparison between the region and the 'big cities', but this amounts to comparing two partially secant entities which both include Brussels City. Finally, comparisons between the region and federal level boil down to comparing the situation of a (federal) entity and that of one of its sub-entities (BCR). From the point of view of Brussels, this structuring of figures is questionable and prevents detailed analyses to be made. It appears to us that it would have been preferable to consider the 19 municipalities of Brussels as a region and as a single city. More generally speaking, an analysis of the Security monitor data specifically centred on Brussels does not exist. When an offence is reported, the victim is asked to state its place of occurrence, however, these data are not made public. Tables indicate whether the offence took place in the municipality of residence, in another Belgian municipality or abroad. At this point, it is therefore impossible to know whether the offence took place in another municipality of Brussels, two streets away from the place of residence, or at the seaside.
Fourthly, a victimisation survey such as the Security monitor essentially provides information regarding the offences experienced by the population. Information about certain offences without victims is only obtained indirectly, such as drug use and trafficking, driving offences, or offences which do not affect people directly, such as economic and financial criminality. In the same way, offences which concern mainly state agencies are under-represented (tax evasion, environmental crime, etc.). Finally, this instrument provides little information on serious yet infrequent offences, such as murder and armed attacks, for example.
Let us also mention a fifth limit, inevitably linked to the use of a closed questionnaire: a limited number of types of behaviour are considered. The population is questioned regarding theft and burglary, but not regarding swindling or fraud committed against them.
Having made these remarks, it is possible to continue with a better understanding of what we may expect from an analysis of the Security monitor.

4
Are the inhabitants of Brussels victims?
Brussels has been described as a dangerous city and the inhabitants of Brussels as being the regular victims of crooks of all sorts. It is therefore important to try to give a review of the offences committed against the population of Brussels.
It should be mentioned that the Security monitor provides us with information about the offences committed against the inhabitants of Brussels and not about the offences committed in Brussels. This is why -as indicated above -we are incapable of saying whether the offences were committed in BCR or elsewhere (the municipal level does not seem pertinent to us here). In return, we do not have access to figures concerning the non-inhabitants of Brussels who have been victims of offences in Brussels. Certain offences are certainly or in all probability committed close to the place of residence, and others are not. A burglary cannot take place just anywhere; furthermore, 86.31% of offences against households reported by inhabitants of the police area Brussels-Capital-Ixelles took place in the municipality of residence, compared with only 64.88% of offences against people. The remaining offences were committed in another municipality, but we do not know whether they took place in Brussels: the figures exist but are not public. It would therefore be very rash to allow ourselves to speculate on where the offences discussed below were committed. We shall limit ourselves to an overview of the victimisation of the inhabitants of Brussels and not of the offences committed in Brussels.
The Security monitor distinguishes between two types of victim: people and households. The theft of a family car affects an entire household and not only the person who is entitled to drive it. On the other hand, blows and wounds are inflicted upon a specific individual. According to the nature of the act, a household or a person will therefore be the victim. The figures from these two categories are therefore not comparable. There are very significant variations from one qualification to the other. Among the offences against people, threats of physical violence clearly prevail. If we consider that in 2008, 34.10% of declared offences involved threats of this type and 11.45% involved physical violence, 7 we cannot fail to notice the importance of the problem of violence for the inhabitants of Brussels. For the rest, the distribution is relatively uniform.
As regards crimes against households, damage caused to cars clearly predominates, with a proportion of 39.59% of all offences declared by households. It should be noted that 17.29% involved the theft of objects in a car.
Burglaries also account for a significant share: 11.53% involved burglary attempts and 6.27% involved burglaries. The phenomenon therefore carries considerable Brussels Studies the e-journal for academic research on Brussels 5 weight within the phenomenon of criminality against the inhabitants of Brussels, as reported by them. Furthermore, by comparing the significance of burglaries within criminality against households and that of offences against property within criminality against people (see the three categories in fig. 1), we are able to assess the importance of criminality involving theft.
We have just discussed the clear prevalence of four types of offence among all offences reported by the inhabitants of Brussels. Two of them fall within the category of offences against people: theft, on the one hand, and violence and threats of violence, on the other. Two belong to the category of offences against households: burglaries and burglary attempts and damage caused to cars. For these types of offence, we have access to figures which allow us to make comparisons between Belgium as a whole and the category of 'big cities'.
The figures which we shall present here no longer represent a proportion of the reported offences, but a prevalence, i.e. a percentage within the basis considered.
As regards offences against people, the percentages represent an estimation of the number of people in the population concerned who have been victims. As indicated  figure 3A, the inhabitants of Brussels are victims of theft more often than Belgians and the inhabitants of big cities. 8 The same observation seems possible as regards (threats of) violence ( fig. 3B), but we must point out the fact that the difference between Brussels (11.62 -15.58%) and the big cities (10.36 -13.76%) is not statistically significant. It is, however, between Brussels and Belgium (8.52 -9.54%).  3D) have been victims of a burglary or a burglary attempt, compared with 6.08% (5.65 -6.51%) of Belgians and 10.23% (8.03 -12.43%) of the inhabitants of a big city. The difference between Brussels and the big cities is insignificant here.
The situation of the inhabitants of Brussels therefore seems less favourable each time than that of the inhabitants of Belgium as a whole. The comparison with the big cities should be more detailed: only the categories of damage to cars and theft present a statistically significant difference. Let us mention in passing that the situation of the inhabitants of the big cities is also less favourable each time than that of Belgians, and that the differences are significant each time.
It is not up to us to try to find an explanation for the differences observed. We do feel, however, that it is important to underline the fact that the figures need some interpretation: there are many factors which may explain the variations observed. Therefore, as road traffic is notoriously difficult in Brussels, part of the damage to vehicles is perhaps due to hit-and-run offences with parked cars. In the same way, it seems logical that building characteristics in Brussels may have an influence on burglaries and burglary attempts. As regards crimes committed against people, the realities specific to big cities should be considered.
Another noteworthy question is that of the evolution of the prevalence of these four categories of offence in Brussels. 9 In this respect, the situation is particularly simple: even if slight changes emerge, none of them are statistically significant for a confidence interval of 95%, from one year to the next or from 2002 to 2008.

Do the inhabitants of Brussels feel insecure?
Let us underline the importance of making a clear distinction between insecurity and the feeling of insecurity. Insecurity is an objective situation of a risk of victimisation, whereas the feeling of insecurity is a subjective perception of this risk. One might think that the feeling is linked to an objective situation of exposure to insecurity. However, this is not necessarily the case. The factors at the origin of the feeling of insecurity are very diverse and are not automatically linked to the commission of offences: poverty, degraded environment, social exclusion, etc. The many reasons for feeling unhappy may involve situations which are easy to point the finger at, such as being the victim of an offence. We must therefore be extremely careful when comparing insecurity and the feeling of insecurity (Lagrange, 1984;Pottier, Robert, 1997;Roché, 1998, Vanneste, 2000. 10 The fact that insecurity and the feeling of insecurity are separate notions does not mean that the latter is not a social reality. It is indeed experienced by part of the population and has very real consequences in terms of behaviour and the quality of life. The feeling of insecurity is therefore a true social problem which must be taken into consideration. However, we would without a doubt fail to meet our objective if Brussels Studies the e-journal for academic research on Brussels 11 10 As regards the constructed character of the fundamental discursive categories of insecurity, see (Milburn, 2000 we attempted to reduce this feeling by limiting our action to repressive strategies. If this problem is to be taken seriously, we must try to understand its complexity and implement various actions.
But what is this feeling of insecurity in Brussels? Do the inhabitants of Brussels really live in terror, as some of them have claimed? And if they are afraid, what do they fear essentially?
As regards the levels of the feeling of insecurity in Brussels, the situation does not appear to be catastrophic. Firstly, in 2008, 51.91% of the inhabitants of Brussels who were interviewed stated that they never or sometimes had a feeling of insecurity ( fig. 5), 35.69% rarely, 10.54% often, and 2.67% always. Only 13% of the population experience a strong feeling of insecurity. This figure is not insignificant, but it puts the problem into perspective.
Furthermore, if we examine the evolution of the figures, we notice that the situation is stable. Between 2002 and 2008, there was an increase in the number of people who stated that they rarely or never had a feeling of insecurity, but the variation remains just within the confidence intervals and is therefore not statistically significant.
But what are the inhabitants of Brussels afraid of? They were asked to assess the risk of being a victim of four types of offence: burglary, physical violence or threat of physical violence, theft with neither threat of violence nor violence and traffic offences ( fig. 6). There was a remarkably stable percentage of people who felt that the risk was low. Likewise, the differences in the numbers of people who felt that there was a very high risk are relatively unimportant. The highest variations are seen among people who are moderately pessimistic and those who are very optimistic. Physical threats and burglaries are feared least, unlike theft and traffic offences.
Although the objective of the present article is not to examine the relationship between insecurity and the feeling of insecurity, figure 6 presents the victimisation figures for the three categories of offences for which they are available. 11 It is interesting to note that there is a reverse relationship between the fear of victimisation and victimisation. In this figure, the categories of offence are listed from left to right according to a growing perception of risk, and the actual risk of victimisation follows the exact opposite progression. 12 This is therefore an additional indication of the distinction between the feeling of insecurity and actual insecurity, which has been underlined time and again.
Apart from being asked to estimate the frequency of offences committed against themselves and their families, the respondents were asked to state which types of behaviour they considered to be problematic in their neighbourhood, from a list of Brussels Studies the e-journal for academic research on Brussels 12 11 In order to obtain the victimisation figures used, we added those related to threats of physical violence, physical violence, burglaries, burglary attempts, petty theft and theft outside of the house, victim absent. We have thus established categories which correspond relatively well to those used in the questions regarding the perceived risks of victimisation. 12 However, we must note that for burglaries, victimisation involves a household, i.e. a group of several people. This being the case, the significance of offences in the percentages of offences reported ( fig. 1 & 2) is presented in a similar way: the least likely offences are the ones which are most expected to occur.
seventeen. 13 These data reflect a concern with respect to a situation, be it qualitatively significant or not.
Eight problems received the 'absolutely' response more than 20% of the time in Brussels ( fig.7). Among them, three involved road traffic, two involved theft (burglaries and theft of objects in cars) and three involved the degradation of the neighbourhood (objects left lying in the street, dirty walls and damage caused to street Brussels Studies the e-journal for academic research on Brussels 13 13 Bicycle theft, theft of objects in cars, aggressive driving, noise pollution caused by traffic, other types of noise pollution, threats, dirty walls and/or buildings, problems caused by groups of young people, inappropriate speed of traffic, people bothered in the street, objects left lying in the street, road accidents, damage caused to street furniture, burglaries, violence, problems related to drug use and car theft.  (2008). Question: Over the next twelve months, what do you feel is the risk that your household will be a victim of… Source: Security monitor 2008, tables C.G. Ins C1 to 4 Very high High Low Very Low Victimisation furniture). 14 Of the eight problems, six therefore concern questions which are more directly related to the quality of life than to criminality, according to the common definition of the word. These figures show that a significant proportion of the inhabitants of Brussels indicate a series of problems which they experience daily, and that they are of little relevance with respect to questions of criminality strictly speaking.
Brussels Studies the e-journal for academic research on Brussels 14 14 As regards the link between 'disorderliness' and the feeling of insecurity, see (Milburn, 2000;Roché, 2000

Conclusion
Moments of mass emotion are not conducive to a reasonable evaluation of concrete situations. The recent events have thus provoked strong reactions and analyses which are rather unrelated to the reality in Brussels. Between crying out in fear and denying problems, it is certainly possible to develop a comprehensive vision and a position which does not deny the difficulties experienced by the inhabitants of Brussels in fact and in feeling, without stigmatising the city-region as a dangerous place. It is probably in some people's interest to decry this blot on the Belgian political/ institutional landscape, but if we wish to develop rational and potentially effective solutions to the benefit of citizens, we must be wary of caricatures.
A (small) part of the construction of an accurate image of reality may consist in the analysis of figures which are -let us recall -available to all: politicians, researchers and citizens. These figures paint the picture of a city which appears to be confronted with more security problems than in the country as a whole and slightly more than in all of the big cities. It would be advisable to examine the situation in Brussels in order to understand the deciding factors and to determine possible actions. Nonetheless, there is nothing to support the image of a Belgian Chicago.
The figures also paint the picture of a relatively low feeling of insecurity among the population. Brussels is liveable; after all, why should this be such a surprise? Certain inhabitants of Brussels are very worried and an examination of the reasons for their feeling of insecurity would be necessary, but on the whole, the situation does not appear to be catastrophic. Furthermore, the inhabitants of Brussels are mainly concerned about issues which have little to do with widespread insecurity. Public cleanliness and issues related to road traffic are ranked first among the problems they identified in their neighbourhoods. These concerns are more directly related to the quality of life and urban management than to problems regarding the maintenance of order and the crackdown on crime. In this respect, let us mention in passing the fact that more than one out of five inhabitants in Brussels finds his or her neighbourhood to be 'not very well-kept' or 'not well-kept at all'. 15 Admittedly, we have not considered the issue from all angles -not even the Security monitor itself. But we feel that certain conclusions may already be drawn. The first one -which is perfectly simple -is that it would be useful to examine the available sources of information about Brussels and therefore point public policies in a better direction. This truism was apparently not considered as such in the eyes of the majority of participants in the recent debate regarding security in Brussels.
The second conclusion is that if insecurity is a social problem, it should not be dealt with roughly, but rather moderately, with thought given to the reasonable objectives which may be pursued: what are the crime rates which we can hope to lower, via which means and at what cost (financial as well as human)?
The third conclusion is that the feeling of insecurity is indeed a truly complex reality. It is a reality because it causes suffering and leads to specific behaviour. It must, as such, hold our attention and be the object of voluntarist public policies. It is complex because it appears to be resolutely multifactorial. The quality of life thus seems to play an essential role, as does -most likely -overall everyday security (employment, health, education, etc.). In order to define exactly what troubles the inhabitants of Brussels, it is important to examine the aetiology of this feeling of insecurity.
Among others, the questions of security and the feeling of insecurity could be considered based on a vision of the city and life in the city. The fact stands out that, once again, questions regarding knowledge and projects are being asked. Is it after all so surprising that a safe city which is also felt to be safe is above all a nice city to live in? Be it the evidently dangerous character of Brussels -according to some -or the obvious need for a collective project -which is probably more reasonable -it is now up to each and every one of us -politicians and citizens -to draw our own outlines of the city.
Finally, let us go back to a problem which was mentioned at the beginning of this article, namely the incapacity of the Security monitor to describe the reality in Brussels. We have figures regarding the experiences of the inhabitants of Brussels without knowing where the reported offences took place (apart from the category of 'in/ outside of my municipality', which is of little relevance with respect to BCR). We do not have figures regarding the offences committed in Brussels against noninhabitants of Brussels. There are, however, several hundred thousand people who come to work in the city every day. We therefore do not know what happens in the region's territory.
The figures do exist, yet they are not made public or used by the Federal Public Service (FPS) of the Interior.Precise data regarding the location of offences are collected, but are part of the raw data which are not communicated automatically. And as a regional security monitor does not exist, we know nothing in this respect, for want of the financing required to carry out a more detailed analysis of the figures concerning Brussels. Everything takes place as though the social and political reality of BCR were of no relevance to the analysis of security questions or for the elaboration of public policies.
On the one hand, it is hard to justify the withholding of figures collected using public funds which, in so doing, prevents researchers from analysing them. On the other hand it is difficult to understand why no one has considered producing a relevant Security monitor for Brussels, a region and unified city which is divided at administrative level into police areas and distinct municipalities.
Whilst taking full responsibility for this text, the author wishes to express his sincere thanks to Sylviane Bogaerts (Federal Public Service (FPS) of the Interior -ZP Vesdre) and Els Enhus (VUB) for their careful rereading and constructive comments.
Brussels Studies the e-journal for academic research on Brussels