This chapter will describe the moment at which the fear of crime research discourse was imported into Sweden. This is a story that began in the late 1970s, with the addition of a fear of crime indicator to the Survey of Swedish Living Conditions,21 developed rapidly during the 1990s, and was institutionalized in the 2000s. Fear of crime was, by then, already a sprawling beast of a research discourse, laden with methodological conventions and theoretical assumptions.
This chapter will chart the institutional story of this moment of importation.
Which instruments, in the form of surveys, were imported and from where? What theoretical assumptions did they bring with them? Which Swedish institutional actors carried out this research? The methodological approach in this chapter is reminiscent of the institutional ethnography of DeVault and McCoy (2001)in the sense that it joins together interview data and documents in the form of government reports and academic research. The chapter will finish with a discussion on the conjunction of the moment of import. What kind of penal politics developed in Sweden during the 1980s and 1980s, and how does fear of crime fit into this construction of a new common sense of crime?
21 More on this in chapter 7
The first fear of crime survey – The Stockholm Project
While questions on fear of crime have been part of the Swedish Living Conditions (ULF) survey since 1978, the first study dedicated entirely to crime, victimization, and fear of crime in Sweden was the Stockholm Project. It collected data in 1989 and published its results in 1990 (Wikström, 1990). Two American studies inspired the methodology of the 1990 study; Skogan and Maxfield (1981) and Taub, Taylor, and Dunham (1984), according to Wikström, Torstensson, and Dolmèn (1997).
The Stockholm Project was a research project initiated in the late 1980s through a collaboration between the Crime Prevention Council and the Department of Criminology at Stockholm University. The project aimed to study crime and victimization in the city of Stockholm using a variety of methods. One of these was a structured quantitative phone interview with adult inhabitants of eight Stockholm neighborhoods (Wikström, 1990). This study included a variation of the Ennis operationalization,22 which was likely the first time this indicator was used in Sweden.23 Participants were also asked how much of a problem they considered the following in their neighborhoods: littering, drunken people, fighting and violence outside, young people fighting or making trouble, women or children being harassed, homes for addicts, disruptive children, and disruptive neighbors (Wikström, 1990, pp. 210-232). The project was highly focused on the ecological perspective, and on studying the relationship between crime and the city. At times it proposed an (unreferenced) general correlation between urbanization and crime, and a correlation between the size of a city and the level of crime (Wikström, 1990). The term “traditional crime” was used repeatedly to denote theft, vandalization, and violence. Two theoretical models were presented, shown here in Figure 29 and Figure 30.
22 “Om du går ut ensam sent en kväll i *ditt område* känner du dig trygg, eller otrygg, eller går du aldrig ut sent på kvällarna?” (P.-O. Wikström, 1990, p. 215)
23 The Survey of Swedish Living Conditions (ULF) also asks participants if they have avoided going out because of fear of crime. This question was first included in 1978.
Figure 29 Theoretical model presented in Wikström (1990, p. 7) Source: (Wikström, 1990, p. 7)
Figure 30 Theoretical model presented in Wikström (1990, p. 18) Source: Wikström (1990, p. 18)
The first model, depicted in Figure 29, is clearly inspired by Clarke and Felson (1993) theory of routine activity, a type of control theory which explains crime as an effect of a lack of social control and shares with other types of control theory the assumption that people will commit crime if not sufficiently controlled. In the first model, urbanization is thought to lead to weaker social control and increased opportunities for crime. These two factors lead to an increased number of motivated offenders and increased crime.
The second model, depicted in Figure 30 aims to explain the level of crime in a neighborhood primarily as a function of the type of housing in the area, which is thought to explain the socio-economic status and the sizes of the families living there. The type of housing, socioeconomic status and type of families gives the level of “problem homes” in the neighborhood and thus its crime level. Problem homes are homes receiving social welfare (Wikström, 1990). Thus, the crime
URBANIZATION
WEAKER SOCIAL CONTROL
INCREASED OPPORTUNITY
FOR CRIME
MORE MOTIVATED OFFENDERS
INCREASED CRIME
TYPE OF HOUSING
TYPE OF FAMILIES SOCIO-ECONOMIC
STATUS
PROBLEM HOMES
LEVEL OF CRIME
levels of a neighborhood are thought to be caused by the type of housing, where large, poor families are considered criminogenic. Multi-family homes with large, cheap apartments cause high crime levels in a neighborhood, the model suggests.
The Stockholm Project was redeveloped into Stockholm’s local fear of crime project. The three largest Swedish cities, Stockholm, Göteborg and Malmö, all have their own local fear of crime surveys. They are not examined in detail, but some data from the contemporary Stockholm fear of crime survey will be dealt with in chapter 7.
The project (and its methodology) was later absorbed into the Crime Prevention Council (BRÅ) and inspired the methodology of what was to become the Swedish Crime Survey (NTU), launched in 2006. This is a typical example of institutional practices that shape what the research discourse becomes. American studies are copied in terms of methodology and theory, are empirically tried out in Sweden on a smaller scale, with only small differences in methods, and become the blueprint for fear of crime studies with impressive scope and longevity. The Swedish Crime Survey has collected data for over 15 years, with the survey sent to 200,000 participants annually (more on Swedish Crime Survey in chapter 7).
Between the governmental and academic
Another actor with an active role in the establishment of the fear of crime research discourse in Sweden was the research unit of the Swedish police. The origin of fear of crime research in Sweden, as in the American case, was entangled with the political and governmental. The Research Unit at the Swedish National Police College was first established in 1988 (Tham & von Hofer, 2014), and then had an interesting history of being abolished and re-established several times, depending on how the police leadership felt about research as a matter for police engagement. An informant describes working at the police research unit in the mid-1990s as a “very strange period”. They were supposed to do research and had very generous resources (“we never applied for any grants”), but research was looked down upon by senior police and was met with considerable resistance in the police organization.
An informant says the initiative to start measuring fear of crime was motivated by change in how the mission and purpose of the police was formulated. The government sends the police a letter of regulation, a document stating what the
police are meant to do, annually.24 Since 1992 this has stated that the police are supposed to “minska brottsligheten och öka tryggheten” – to decrease crime and increase trygghet. The goal of increasing trygghet was introduced in 1992. The exact formulation of this mission is shown in Table 2.
Table 2: Formulation of the purpose and mission of the Swedish Police in its Letter of Regulation25
Year Formulation (Swedish) English translation
1991 Polisen ska särskilt rikta in sin verksamhet på åtgärder som går ut på att värna om människors trygghet till liv, hälsa, integritet och personlig egendom.
In particular, the police shall focus their activities on measures aimed at safeguarding people’s security of life, health, integrity and personal property.
1992/1993 Målen för polisverksamheten ska vara att minska brottsligheten och öka människors trygghet i samhället
The objectives of police activities shall be to reduce crime and increase people’s feelings of safety.
1997 Det övergripande målet för polisen är att minska brottsligheten och öka människors trygghet
The primary goal of the police is to reduce crime and increase people’s feelings of safety.
1998 Målet för kriminalpolitiken är att minska
brottsligheten och öka tryggheten. The aim of penal policy is to reduce crime and increase feelings of safety 1999
2000 Målet för kriminalpolitiken är att minska brottsligheten och öka människors trygghet.
The aim of penal policy is to reduce crime and increase people's feelings of safety.
2001 2002
The survey the research unit imported and implemented was the Skogan and Maxfield (1981) survey discussed in chapter 5. The reason why this study provided the blueprint for Swedish fear of crime surveys was both circumstantial and structural. Skogan was a representative of the Chicago School, which has greatly influenced Swedish criminology and sociology. An informant remarks that they (the researchers) were trained in urban sociology and socio-ecological theory that placed great emphasis on the physical environment and thus had a theoretical affinity with Skogan’s work. Interestingly, the informant also remarks that the focus on the individual, which was to become a defining feature of fear of crime research, wasn’t necessarily part of this theoretical socialization, but rather something they “fell into". Signs of Disorder is often described in the interviews as the "only available theory at the time".
According to Wikström, Torstensson, and Dolmèn (1997, pp. 4-6), local fear of crime measurements had the following three purposes: to describe the local crime problem and use this description as a basis for crime prevention and
24 Before the reorganization of the Swedish police into a single agency in 2015, Letters of Regulation were sent to the National Police Board.
25 The Letters of Regulation were collected from the Department of Justice in spring 2021.
allocation of resources; to use the description of local crime to educate the police and other local crime prevention actors; and, through repeating measurements, obtain information on how well central goals for the police were met locally, in terms of decreasing crime and increasing trygghet. The main problem with fear of crime surveys, according to the authors, is that they describe the crime problem as something experienced by individuals, and do not consider how businesses experience crime. The authors also write that they are aware that it is mainly the established and conventional citizenry who are represented in these surveys, and that marginalized groups are under-represented. The respondent rate is a respectable 80 percent (Wikström, Torstensson, & Dolmèn, 1997, pp. 4-6).
Some aspects of these early surveys align well with how Tham and von Hofer (2014); Wikström (1996) describe the “new” criminology of the 1990s. They all have a strong focus on the local, the neighborhood and factors related to it. Their theoretical model is depicted in Figure 31.
Figure 31 Theoretical model presented in Per-Olof H. Wikström et al. (1997, p. 11) Per-Olof H. Wikström al. (1997, p. 11).
Figure 31 depicts a version of control theory with socio-ecological aspects, where crime is thought to be caused by urbanization and segregation, which causes (lack of) social integration, and leads to a lack of informal control, which in turn causes crime. We can note that crime is thought to be a result of both segregation and informal social control in turn. Wikström, Torstensson, and Dolmèn (1997) argue that a “lack of reaction” from society leads to a downward spiral. This orients the