Social responses to collective crime: Assessing the relationship between crime-related fears and collective sentiments

Abstract

The literature on the effects of fear of crime has investigated a number of the nuances of this phenomenon; however, how the fear of different types of crime influences communities has yet to be investigated. We hypothesize that the fear–decline model, which argues fear leads to a decrease in solidarity, applies to ‘routine’ street crimes; however, fearing crimes that attack the collective, such as school shootings, will increase community solidarity. Using two datasets collected in Finnish communities after they experienced tragic school shootings, our results indicate that the fear–decline model receives strong support but the fear–solidarity model does not.

Keywords

Solidarity fear of crime collective crime school shootings Finland

Crime can fracture a community’s social network, lead residents to withdraw from civic life, reduce levels of social capital and collective efficacy, and foster an environment that promotes more crime (Markowitz et al., 2001). Yet, crime can also galvanize a community, bringing its members together in an outpouring of social solidarity (Garland, 1990; Hawdon et al., 2010). It appears that different types of crimes produce different social reactions. Our research addresses a related question; however, we focus on the fear of crime instead of the occurrence of crime. Specifically, we ask, ‘Do different types of fear of crime produce different consequences for a community?’

Using data from two Finnish communities, we analyse if fear of collective crimes such as terrorist acts or school shootings produces different social responses than fear of ‘routine’ street crimes such as homicide, assault or rape. We use data collected after tragic school shootings in Jokela (2007) and Kauhajoki (2008) Finland. Both offenders targeted their own communities and claimed to have societal motives for their actions. In total, 20 people died in the shootings, which are the worst cases of mass shootings in Finnish history. The Jokela data were collected in the spring of 2008, six months after the tragedy but prior to the Kauhajoki shooting. The Kauhajoki data were collected in the spring of 2009, six months after the shooting occurred there. These data are well suited for our research because they enable the exploration of fear of crime and social response in communities that experienced mass tragedies.

Our research contributes to the study of crime in several ways. First, as noted by Gray and her associates (2011), there is a need to analyse the emotional responses to collective crimes, and Muschert (2007) argues for research analysing school shootings in different cultural contexts. Our research addresses both of these issues. Second, most fear of crime research focuses on fear as a dependent variable, and the studies using fear of crime as an independent variable analyse how it affects individuals’ wellbeing or behaviours (e.g. Cobbina et al., 2008; Jackson and Gray, 2010; Jackson and Stafford, 2009; Stafford et al., 2007). Our research focuses on the consequences of fear on community wellbeing. Finally, our research on two Finnish communities provides an important cross-national case study.

Literature review

As the research on fear of crime has expanded, scholars have explored the nuances of the topic. Researchers have noted those who fear crime are not those who have experienced it (e.g. Skogan and Maxfield, 1981), and the relationships between victimization and fear depends on both the type of victimization and the timing of victimization. Similarly, while a gender difference is found in the fear of violent victimization, men and women appear to fear economic victimization equally (Roundtree, 1998; also see Callanan and Teasdale, 2009). Scholars also distinguish between diffuse anxieties about crime and everyday worries about crime (Farrall et al., 2009; Gray et al., 2011). While fear can lead to dysfunctional worries that reduce quality of life, it can also lead individuals to adopt activities that protect them from crime and buffer them from the generally negative effects of crime fears (Gray et al., 2011; Jackson and Gray, 2010). Moreover, fear can also stimulate community action (Hawdon and Ryan, 2009). Thus, a broad range of existing literature identifies different causes and consequences of fearing crime; however, the consequences of fearing different types of crime have not been researched.

Two basic models can relate fear of crime to community-level responses (see Liska and Warner, 1991). The first perspective, the fear–decline model, argues that fear of crime leads to community decline. The second model, the fear–solidarity model, argues that fear generates a collective condemnation of crime that latently increases social solidarity. Thus, these perspectives predict opposite consequences resulting from fear. After reviewing the logic of and supporting evidence for each model, we offer a means of possibly reconciling these contradictory positions.

The fear–decline model

According to the fear–decline model, fear leads individuals to withdraw from community life, which eventually results in the atomization of the community (Hale, 1996; Skogan, 1986, 1990). The mechanisms by which fear leads to community decline are complex and likely include reciprocal relationships. In fact, the model typically begins with physical and social disorder leading to fear, which in turn leads to community withdrawal, weakened bonds and lower levels of collective efficacy, which in turn leads to more fear and increased crime (Markowitz et al., 2001; Robinson et al., 2003; Taylor, 2001; Wyant, 2008). Disorder probably leads to fear more than fear leads to disorder (Brunton-Smith, 2011); however, fear of crime and disorder are clearly related. While we are aware of the complexities of these relationships, we focus our discussion on the factors relating fear of crime to social solidarity.

First, as crime and fear mount, residents become dissatisfied with their neighbourhood, begin to distrust neighbours and strangers, and form intentions to move from the community (see Carro et al., 2010; De Donder et al., 2012; Ferguson and Mindel, 2007; Hartnagel 1979; Liska and Bellair, 1995; Skogan, 1986, 1990). Fear of crime also affects residents’ daily activities. Not only does fear reduce physical activity (see Stafford et al., 2007), it also influences what people do and when they do it. Fear can lead residents to adopt routines aimed at preventing victimization, such as limiting exposure to public areas or modifying where their voluntary activities occur (Adams and Serpe, 2000; Ferraro, 1995; Jackson and Gray, 2010; Liska et al., 1988; Mesch, 2000; Rader et al., 2007; Rengifo and Bolton, 2012; Woldoff, 2006). Fear of crime also reduces interacting with friends and participation in social activities (Britto et al., 2011; Cobbina et al., 2008; De Donder et al., 2012; Ferguson and Mindel, 2007; Ferraro, 1995; Markowitz et al., 2001; Oh and Kim, 2009; Stafford et al., 2007).

As fear leads residents to withdraw from the neighbourhood, it eventually reduces social cohesion (Bellair, 2000; Oh and Kim, 2009). Several researchers have found that the more cohesive a community, the lower levels of fear in that community (e.g. Gibson et al., 2002; Jackson, 2004; Moore and Shepherd, 2007; Wyant, 2008), yet, longitudinal research indicates that these variables are reciprocally interrelated (Markowitz et al., 2001). Results from a non-recursive model where cohesion affects disorder, which affects fear of crime, which in turn affects cohesion indicate ‘that neighbourhood cohesion, crime/disorder, and fear affect each other as part of a reciprocating cycle’ (Markowitz et al., 2001: 311).

The fear–solidarity model

While the fear–decline model predicts fear will lead residents to withdraw from their community, a second model, largely based on Durkheim’s insights regarding the social response to crime, predicts that fear leads to heightened solidarity. Here it is important to note that when we say ‘solidarity’ we mean Durkheim’s concept of ‘mechanical solidarity’. Mechanical solidarity refers to the social integration of community members based on similar and shared values and beliefs, which Durkheim refers to as a ‘collective conscience’ (see Durkheim, [1893] 1964). The collective conscience, while internally held by the group’s individual members, is intense and highly determinate, thereby leading residents to cooperate. This solidarity corresponds to penal law with repressive sanctions that reinforces the commonly held sentiments upon which it is based¹ (Durkheim, [1893] 1964).

Indeed, according to Durkheim, an act is a crime because it offends collective sentiments. Since crime disturbs collective sentiments, it produces a collective response, and this response has the unanticipated consequence of strengthening normative consensus and solidarity. Durkheim ([1893] 1964: 102) states, ‘Crime brings together upright consciences and concentrates them.’ Thus, when crime strikes a community, it represents a threat to that community’s collectively held norms. This threat results in people ‘com(ing) together to talk of the event and to wax indignant in common’ ([1893] 1964: 102). Durkheim ([1893] 1964: 102) continues by stating:

From all the similar impressions which are exchanged, from all the temper that gets itself expressed, there emerges a unique temper…. which is everybody’s without being anybody’s in particular. That is the public temper.

Thus, as the collective’s shared norms are violated, the collective, in turn, responds to the threat by acting collectively. As Durkheim argues ([1893] 1964: 103), ‘since it is the common conscience which is attacked, it must be that which resists, and accordingly the resistance must be collective.’ Through this collective action and expression of mutual outrage, solidarity is promoted and the group’s unity is enhanced.

There is support for Durkheim’s well-known contentions with respect to how the occurrence of crime stimulates solidarity (Collins, 2004; Hawdon et al., 2010, 2012); however, would the fear of crime also lead to heightened solidarity? While Durkheim did not address this issue directly, it can be argued that a fear–solidarity model can be logically derived from his basic insights. According to this model, crime produces a collective response. This collective response, the expression of the ‘public temper’, intensifies social interaction. As the community comes together to confront their common enemy, the group becomes increasingly united and solidified (Cohen, 1966). Thus, fear of crime can contribute to group solidarity by generating intense interactions among the community members (see Liska and Warner, 1991; also see Erickson, 1966).

There is evidence that fear of crime promotes solidarity, at least among some residents. For example, Lauderdale (1976) found that an external threat was associated with the rejection of deviant individuals and heightened solidarity. More recently, Oh and Kim (2009) found that, relative to younger adults, fear of crime among elderly residents increases interactions with neighbours and their perceived level of social cohesion. Still, the overwhelming majority of studies that investigate the relationship between fear of crime and solidarity find fear reduces solidarity (see Liska and Warner, 1991).

Nevertheless, there is reason to believe the fear–solidarity model may still be valid. In addition to the need to differentiate between diffused anxieties and everyday worries about crime, it is possible that the fear of different types of crime lead to different outcomes.

Fear of individual and collective-targeted crimes

Garland (2001: 147) argues that crime has become a ‘normal social fact’ in most modernized societies. Indeed, homicide, rape, assault, robbery, burglary, and larceny are ‘routine’ in many societies. In Finland, for example, the 2011 murder rate was 2.1 per 100,000 population, and the assault rate per 100,000 population was approximately 743. These are relatively high levels of crime; among the highest in the industrialized world (see UNODC 2011; see also The National Research Institute of Legal Policy 2011). Not only are these crimes daily occurrences, the media frequently report these crime, which can distort the reality of crime and lead to overestimating the ‘crime problem’ (see Callanan, 2012; Farrall et al., 2009; Kupchik and Bracy, 2009). While these crimes are ‘normal social facts’ in many nations, crimes that target collectives are not, at least in the industrialized world. Collective-targeted crime occurs when an individual or group uses violence against another group or set of individuals, typically to achieve political, economic or social objectives. Acts of terrorism, rampage school shootings, and mass murders are examples of collective-targeted crimes. Although these acts typically attract considerable media attention, they are rare (see Muschert, 2007; Newman et al., 2004).

There is evidence that collective-targeted crimes increase the solidarity of the communities in which they occur. An outpouring of solidarity was witnessed after the Columbine school shootings (see Cullen, 2009), the terrorist attacks of September 2011 (Collins, 2004), the Virginia Tech school shooting (Hawdon and Ryan, 2011; Hawdon et al., 2010), and the mass murders in Jokela and Kauhajoki Finland (Nurmi, 2012; Nurmi et al., 2012; Oksanen et al., 2010). In fact, an outpouring of solidarity after collective-targeted crimes is relatively common. In contrast, homicides and other routine crimes typically target individuals, and individual-targeted crimes rarely produce mass displays of solidarity. While people will express sympathy for the victims, these acts rarely result in public vigils, collective memorials, or other symbolic displays of collective grief and solidarity (Ryan and Hawdon, 2008). It is therefore possible that fear of routine street crime and fear of collective-targeted crime have different consequences for community solidarity.

According to Durkheim, crimes arouse collective sentiments against the infringement of strongly held norms. Collective-targeted crimes such as terrorist attacks and school shootings are rare and excessively shocking, and these are clear attacks against the community and its shared norms. The terrorist attacks of September 2001, for example, were clearly an attack on the United States, not just the Twin Towers and Pentagon. They were an attack against America’s strategic interests and protests against American foreign policies. Moreover, the targets were highly symbolic. Similarly, the school shootings in Jokela and Kauhajoki were attacks on the entire community, and they disrupted residents’ sense of what their towns were: safe and quiet (Nurmi, 2012). The response to these incidents and other collective-targeted crimes is exactly as Durkheim predicted. It is possible that in modern society, it takes acts that target collectives for people to ‘wax indignant in common’. If this is indeed true, fears over collective-targeted crimes may produce a solidarity-generating response in a similar manner that experiencing such crimes does. Fear of crimes that threaten the collective may cause the group to respond collectively in an attempt to protect their group.

In contrast, it is possible that ‘routine street crimes’ no longer shock the common conscience sufficiently to evoke a collective response that results in solidarity. Street crimes such as most homicides, rapes, and assaults typically injure one person and the victim’s primary networks, not the entire community. While a homicide demonstrates a lack of value consensus in the community (i.e. some community member does not believe ‘Thou shall not kill’), this value has been violated frequently enough that such transgressions no longer stimulate collective action. Since ‘routine crimes’ target individuals, the response to these attacks is likely individualized. The repeated violation of the law signifies that numerous residents do not hold those laws sacred, and this perception, regardless of its validity, will probably result in residents fearing each other. Individuals may still react functionally, but not how Durkheim predicted. Instead, they become cautious; they curtail their interactions with potentially dangerous community members; they limit their activities that occur in potentially dangerous places; they withdraw into a defensible space (see Jackson and Gray, 2010). These precautions are functional for the individual because they make him or her feel safer and reduce the likelihood of victimization; however, they do not meet the functional needs of the community. On the contrary, these actions lead to an atomized community and reduced levels of solidarity. If the above argument is accurate, we should find support for the following two hypotheses: (1) fear of collective-targeted crimes increases solidarity, and (2) fear of individual-targeted crimes decreases solidarity.

Data and methods

We used data from two mail surveys collected in the Jokela and Kauhajoki communities after they suffered school shootings in 2007 and 2008, respectively. Jokela, located in the Tuusula municipality approximately 50 km from Helsinki, is a town of approximately 5600 residents. Kauhajoki is approximately 350 km from Helsinki in the Ostrobothnia region of Western Finland. With a population of approximately 14,000 inhabitants, Kauhajoki is a larger community than Jokela both geographically and in population.

The surveys were fielded in May–June 2008 in Jokela (n=330) and March–April 2009 in Kauhajoki (n=319), approximately six months after each shooting incident took place. Using addresses from the population register database, we selected a simple random sampling of 700 18-to-74-year-old Finnish-speaking residents. The response rates were 47 percent (Jokela) and 46 percent (Kauhajoki). Comparing the gender and age structure of the data with population figures of the Tuusula and Kauhajoki municipalities (all population data from Statistics Finland, 2009) indicates that the samples are biased with respect to gender. According to 2008 and 2009 estimates, 50.4 percent of the Tuusula population aged 18–74 and 51.8 percent of the population in Kauhajoki municipality aged 18–74 are male. However, 55.3 percent of the Jokela sample and 43.7 percent of the Kauhajoki sample is male. Younger respondents are also underrepresented in both samples. In 2008, 45.6 percent of the Tuusula population between the ages of 18 and 74 were 31–50 years old, but only 40.6 percent of the Jokela sample is in this age range. In the Kauhajoki municipality, 43.5 percent of those 18–74 are between the ages of 31 and 50; however, only 29.1 percent of the sample is in this age range. Although these figures are not within the expected margin errors for the samples, they are relatively close, especially for Jokela. We also emphasize that the population figures are for the larger Tuusula and Kauhajoki municipalities and not directly applicable to the towns from which the samples were drawn. Population data for these small towns are unavailable. Given the lack of directly comparable population data, we do not attempt to weight the data.

Measures

Our dependent variable, social solidarity, is measured by combining six Likert questions that range from 1 ‘strongly disagree’ to 5 ‘strongly agree.’ The items are: (1) I am proud to be a member of my community; (2) I feel I am part of the community; (3) People in my neighbourhood share the same values; (4) My neighbourhood is a good place to live; (5) I trust my neighbours; and, (6) People work together to get things done for this community. The Cronbach’s alphas for the indices were .85 and .89 for the Jokela and Kauhajoki samples, respectively. This measure of solidarity, derived from Bachrach and Zautra’s (1985) sense of community scale, has produced similarly high reliability coefficients in studies using different datasets from different countries (e.g. Hawdon et al., 2010, 2012).

Independent variables

Our predictors of central theoretic interest include three measures of fear of crime. First, we use two measures of fear of collective-targeted violence. We measure fear of school shootings with an item asking respondents how worried they were about a school shooting reoccurring. The item ranged from 1 ‘not worried at all’ to 5 ‘extremely worried.’ Our second measure of collective-targeted fear is fear of terrorism, which is measured by an item asking if terrorism poses a threat to Finnish society (1=‘strongly disagree’; 5=‘strongly agree’). We considered combining these items into a combined measure; however, the correlation between these items (0.38 in Jokela and 0.43 in Kauhajoki) indicates they are interrelated, but different, types of collective fears. Because of this and the exploratory nature of our research, we analyse these as separate indicators of fear of collective-targeted crimes. To measure fear of individual-targeted crime, or what we refer to as ‘street crime’, respondents were asked how worried they were about becoming a victim of street violence, burglary, violence at work, domestic violence, and sexual harassment. The responses to these items ranged from 1 ‘not worried at all’ to 5 ‘extremely worried.’ The items were combined into an index with Cronbach’s alphas of .73 and .82 for Jokela and Kauhajoki, respectively.

In addition to the fear variables, we control for known correlates of solidarity including neighbourhood interactions, prior victimization, and depression. Interacting with neighbours is consistently found to be related to residents’ attachment to their communities (e.g. Hipp and Perrin, 2006; Sampson, 1988; Wilkinson, 2007). Similarly, solidarity is related to criminal victimization (e.g. Lee, 2000; Sampson and Groves, 1989; Sampson et al., 1997) and emotional depression (e.g. Hawdon and Ryan, 2012). Given the high correlations among these factors, fear of crime, and social solidarity in other studies, we include these in our models to isolate the independent effects of fear of crime on solidarity.

Interactions with neighbours is measured using a four-point scale concerning how often respondents associate with their neighbours, with the responses ranging from 1 (hardly ever) to 4 (daily). A dichotomized item (0=no, 1=yes) asking if respondents had been attacked or threatened by a stranger in the last five years measures prior victimization. We use the Finnish modification of the Beck Depression Inventory (BDI) to measure respondents’ emotional depression. The modified version of Beck’s inventory, the R-BDI (Kaltiala-Heino et al., 1999; Raitasalo, 2007), has been widely used to measure depression in the general population. The 13 questions of this depression screener ask about respondents’ general wellbeing, life satisfaction and outlook for the future. It also includes questions about sleeping, appetite, anxiety, and suicidal ideation. The alpha reliability for the depression inventory was .88 in both the Jokela and Kauhajoki samples. The overall indicator was converted to a dichotomous variable using inventory guidelines to separate those who did not report depressive symptoms from those who have at least mild depression (0=no depression; 1=at least mild depression) (Raitasalo, 2007: 61).

Finally, we include a number of socio-demographic variables, including gender (0=female, 1=male), marital status (0=married/cohabit, 1=others), household composition (0=has school-aged children living in household, 1=no school-aged children living in household), and age (measured as a continuous variable). The items, variable coding, and descriptive statistics are reported in the Appendix.

Analysis and results

Since the solidarity indices were approximately normally distributed (M=21.44 in Jokela [SD 4.46] and M=22.09 [SD 5.26] in Kauhajoki), ordinary least squares (OLS) regression is an appropriate analytic method. We treat the samples separately because the communities’ demographic structure and geographical locations are different and the samples were collected at different times. We begin our analysis with an initial model that includes only the effects of fear of collective-targeted violence.² We then add fear of individual-targeted violence and frequency of meeting neighbours to the base model. The final model includes all predictors and control variables. The three regression models predicting the level of solidarity in Jokela are presented in Table 1.

Table 1.

Level of solidarity in Jokela by crime-related fears and other independent variables.

	Model IUnstandardized coefficients B (SD)	Standardized coefficients β	p	Model IIUnstandardized coefficients B (SD)	Standardized coefficients β	p	Model IIIUnstandardized coefficients B (SD)	Standardized coefficients β	p
Constant	15.24 (.71)		.000	11.26 (.99)		.000	9.82 (1.53)		.000
Fear of school shootings	–.62 (.22)	–.17	.005	–.39 (.23)	–.11	.093	–.52 (.24)	–.15	.028
Fear of terrorism	.45 (.23)	.12	.053	.53 (.23)	.15	.020	.64 (.22)	.18	.004
Meeting neighbours				1.28 (.27)	.27	.000	1.10 (.26)	.23	.000
Fear of street crime				–1.25 (0.29)	–.27	.000	–1.17 (.29)	–.25	.000
Sex (male)							–.74 (.50)	–.09	.144
Age (in years)							.01 (.02)	.02	.696
Marital status (married)							1.47 (.56)	.15	.009
Household type (school-aged children)							–0.94 (.55)	–.10	.088
Experienced an assault							1.50 (.76)	.11	.050
R-BDI (no symptoms)							2.86 (.68)	.24	.000
R² adj.	.02			.13			.21

Note: Dependent variable: solidarity index.

As Model I in the table indicates, fear of collective-targeted crimes significantly predicts solidarity in Jokela (F=4.34, df=2, p<.01). Fear of school shootings produces a stronger effect than does fear of terrorism; however, what is noteworthy is that the effects are opposite of each other. While fear of terrorism increases solidarity (B=.12; p=.053), fear of school shootings clearly weakens solidarity (B=-.17; p=.005). This finding contradicts our hypothesis.

Model II is statistically significant (F=11.49, df=4, p<.001), and all of the variables in it except fear of school shootings are statistically significant. Fear of terrorism remains significant and the effect is positive (B=.15; p=.020). As predicted, fear of street crime is negatively associated with solidarity (B=-.27; p<.001). In addition, solidarity is considerably higher for those who meet their neighbours often (B=.27; p<.001) compared with those who meet neighbours less often. This finding is in line with several previous studies.

Model III includes the effects of all independent variables. This model is statistically significant (F=8.29, df=10, p<.001) and accounts for approximately 21 percent of the variance in solidarity. All of the fear variables are statistically significant, with the effect of fearing terrorism remaining positive (B=.18; p=.004) and fearing street crimes remaining negative (B=-.25; p<.001). Fear of school shootings again achieves significance in the full model (B=-.15; p=.028), but the effect is in the opposite direction than predicted. Depression and having experienced an assault are inversely related to solidarity. Marital status is the only demographic factor significantly related to solidarity, with married respondents reporting higher levels of solidarity than others do.

Table 2 shows the three models predicting levels of solidarity in Kauhajoki. Model I, which only includes the effects of fearing collective-targeted violence, is barely significant (F=3.84, df=2, p<.05) and only explains two percent of the variance. Fear of terrorism increases solidarity (B=.16; p=.01), but fear of school shootings decreases solidarity (B=-.12; p=.054). The initial models are therefore similar across the two communities. Model II, which adds fear of street crimes and meeting neighbours to the base model, is statistically significant (F=13.69, df=4, p<.001) and accounts for 15 percent of the variance in solidarity. However, neither fear of school shootings nor fear of terrorism approach significance. Fear of being a victim of a street crime decreases solidarity (B=-.18; p<.001).

Table 2.

Level of solidarity in Kauhajoki by crime-related fears and other independent variables.

	Model IUnstandardized coefficients B (SD)	Standardized coefficients β	p	Model IIUnstandardized coefficients B (SD)	Standardized coefficients β	p	Model IIIUnstandardized coefficients B (SD)	Standardized coefficients β	p
Constant	13.54 (.76)		.000	9.08 (1.10)		.000	10.43 (1.81)		.000
Fear of school shootings	-.54 (.28)	–.12	.054	–.01 (.27)	–.02	.720	–.07 (.27)	–.02	.794
Fear of terrorism	.76 (.29)	.16	.010	.31 (.28)	.07	.264	.21 (.28)	.05	.449
Meeting neighbours				1.93 (.30)	.36	.002	1.64 (.31)	.30	.026
Fear of street crime				–.99 (0.32)	–.18	.000	–.79 (.35)	–.14	.000
Sex (male)							.35 (.59)	–.09	.552
Age (in years)							–.04 (.02)	–.13	.041
Marital status (married)							1.03 (.69)	.09	.136
Household type (school-aged children)							–0.94 (.68)	–.08	.167
Experienced an assault							–.30 (.92) ns	–.02	.741
R-BDI (no symptoms)							2.58 (.79)**	.19	.001
R² adj.	.02			.15			.22

Note: Dependent variable: solidarity index.

Finally, all variables were added in the Model III, which is statistically significant (F=8.37, df=10, p<.001) and accounts for 22 percent of the variance in solidarity. Neither of the fear of collective-targeted violence variables is significant, but fear of street crime is (B=-.14; p<.001). Depression and interacting with neighbours produce similar effects as in Jokela; however, the effects of the socio-demographic variables differ across the samples. In Kauhajoki, age is the only significant predictor and indicates that younger respondents report higher levels of solidarity than older residents do.

Comparing the results from the final models, we see notable differences. First, the two variables measuring fear of collective-targeted crime are significant predictors of solidarity in Jokela but not in Kauhajoki. Second, the effects of basic socio-demographic characteristics differ between the communities. In Jokela, we find that marital status is significant, whereas age and gender are not. In Kauhajoki, there is a weak, negative effect of age, but other demographic factors are unrelated to solidarity. Depression and interacting with neighbours are good predictors of solidarity, and these operate similarly across the samples.

Discussion

This article explored the consequences of fearing crime in local communities that suffered dramatic collective crimes. Specifically, we analyse if the fear of collective-targeted crimes produces different social responses than the fear of more prevalent, individual-targeted street crimes (e.g. homicides, rapes and assaults). We use two competing models based on earlier research (e.g. Liska and Warner, 1991): the fear–decline model, which proposes that fear reduces social solidarity, and the solidarity model, which argues that crime solidifies communities in opposition to it. We extend the crime–solidarity model and propose that the fear of collective-targeted crime can increase solidarity in a similar manner that experiencing collective crimes does.

Our argument that fears of collective-targeted crime will increase solidarity receives very limited support in our analysis. Although the fear of terrorism increases solidarity in Jokela, fear of terrorism was unrelated to solidarity in Kauhajoki. Moreover, the fear of school shootings clearly weakens solidarity in Jokela and is unrelated to solidarity in Kauhajoki. Conversely, the fear–decline model receives support in both analyses. As predicted, fear of individual-targeted crimes such as assault, burglary, and rape reduces solidarity in both communities. Furthermore, background factors such as interacting with neighbours and depression function similarly in both communities, and these findings replicate results from other studies. Therefore, we conclude that the fear–decline model is better for understanding the consequences of fear of crime. Although there is evidence that Durkheim’s ideas accurately predict the immediate reactions to severe collective crimes when they occur (e.g. Collins, 2004; Hawdon et al., 2010), this model cannot be extended to the fear of crime, even the fear of collective-targeted crimes.

One interesting difference between the analyses is that fear of school shootings reduced solidarity in Jokela but was unrelated to solidarity in Kauhajoki. While there are several possible reasons for this conflicting finding, we believe one is particularly likely. Our analysis focused on two small Finnish communities that experienced violent mass tragedies. The school shootings that occurred in these towns were both unexpected and shocking crimes; however, before the Jokela tragedy, few thought a targeted armed assault directed at one’s own community was possible in Finland. Because of this, Jokela became synonymous with school shootings in Finland (Nurmi et al., 2012). Since the Jokela shooter was a community resident most of his life (Nurmi, 2012; Oksanen et al., 2010), his act of violence, as extreme as it was, had similar effects that individual-targeted crimes do. The fact that the shooter was ‘one of their own’ symbolized a decline in the community’s ability to control its members. Fear of this occurring again, therefore, meant fearing a community member, and such fears decrease solidarity.

After the Kauhajoki shootings, however, the national rhetoric about such crimes changed, and school shootings were considered a national problem (Oksanen et al., 2010). Although the Kauhajoki shooter was a student, he was not a long-time resident and was considered an outsider in the Kauhajoki community (Nurmi, 2012; Oksanen et al., 2010). Therefore, while their community was attacked and they had been victimized, the attacker was an outsider, not a community member. The attack may represent a decline of morality in Finland; however, it did not necessarily represent a decline of the Kauhajoki community. School shootings were interpreted nationally as a risk scenario, but the shootings were not considered to be acts of domestic terrorism, despite the politically orientated motives of the offenders (Kiilakoski and Oksanen, 2011; Malkki, 2011). Even in Kauhajoki, the offender, although an outsider, was not seen as a terrorist. Therefore, fearing another school shooting meant fearing community outsiders, but not outsiders from distant lands. This fear apparently does not reflect poorly on the community and hinder its sense of cohesion; but it is also not sufficient to draw the community together.

Our study has several limitations. First, the samples are from small communities and are not nationally representative; thus, the findings may not be generalizable beyond these communities. However, the samples are from different type of communities that experienced similar tragedies, so we can likely generalize to at least small communities that experience mass tragedies. In addition, our main finding that the fear–solidarity model does not apply is likely to extend to other communities. The fear–solidarity model is based on Durkheim’s notion of ‘mechanical solidarity’ or the solidarity that emerges from likeness and shared values, and this type of solidarity is more likely in small communities. As Liska and Warner (1991) conclude in their test of the fear–solidarity model, Durkheim was referring to rural and highly cohesive societies where mechanical solidarity would prevail, but in urban societies where mechanical solidarity is largely replaced by organic solidarity based on interdependence, fear likely decreases the mechanical solidarity to which we refer. Although Jokela and Kauhajoki are far from undifferentiated rural communities, they are likely more close-knit and cohesive than larger and more ethnically diverse cities such as Paris, London, Berlin, or Amsterdam. If the fear–solidarity model is not supported in our smaller communities, it is highly unlikely it would be supported in larger areas.

Next, we lack pre-tragedy data and therefore we cannot analyse how fear of crime changes in communities that experience mass tragedies. While we recognize our data are not ideal for testing general criminological theories, we nevertheless are confident in our findings. The data provide a good setting for testing the accuracy of the fear–solidarity model. Since these communities suffered collective-targeted crimes shortly before the data were collected, it is likely that residents there were more cognizant of these types of crimes than resident of most other communities. Therefore, the model would likely be supported in these communities if it were accurate. The fact that we find little support likely means that Durkheim’s argument about how crime can stimulate solidarity cannot be extended to the fear of crime.

Finally, we recognize that it is possible that the difference between fear of collective-targeted crime and fear of street crime may be a result of differences in perceived risks. While fear of street crime is likely fear directly resulting from one’s assessment of their own personal risk of being victimized by such a crime, fear of collective-targeted violence is likely more about perceived general risks. Respondents are likely concerned about the possibility of these more general risks more than they are the likelihood of personally being a victim of a collective attack. This difference in perceived risks may very well influence the relationship between fear and solidarity, at least between the theorized relationship between collective-targeted violence and solidarity. The generalized risk and sense of worry about an attack on the collective indeed reflects a more other-centred perspective than does worry about personal victimization. It is therefore likely that worry about others would result in heightened solidarity far more than worrying solely about oneself. The fact that our findings fail to support the fear–solidarity model is therefore even more striking. Even among those who are clearly worried about the risk that collective violence poses to their community, this worry and fear does not stimulate solidarity.

Consequently, despite several limitations, we are confident in our conclusion that the fear–solidarity model is not an accurate depiction of social life. The limitations of our analysis would make it easier for this model to receive support. The fact that the model is still unsupported therefore increases our confidence in rejecting it. The fear–decline model, by contrast, receives clear support. Since this model is widely supported in the literature, we gain confidence in our findings.

Conclusion

We analysed data from two small communities collected in the aftermath of mass tragedies. In our study, different operationalizations of fear of collective crime were moderately related to social solidarity, and the results contradicted our hypotheses. Yet, this study examined a phenomenon that has largely been unexplored. The relevance of this type of data is important, since in recent years we have seen collective-targeted attacks in Norway (2011), Belgium (2011), France (2012), Finland (2012), and the United States. Collective-targeted attacks are statistically rare, but they have serious social consequences. After all, the fear of crime affects significant numbers of people, and if we understand fear of crime as an indicator of community decline and solidarity as an indicator of community cohesion, we can understand why collective-targeted crimes often produce both solidarity and conflict. Given these considerations, future studies should focus on how crime and fear operate not only in everyday life but also after mass tragedies.

Footnotes

Appendix

Appendix.

Descriptive statistics of independent and dependent variables used in the analysis.

Variable	Jokela (n=330)		Kauhajoki (n=319)
	Measurement	%	Mean (SD)	%	Mean (SD)
Gender	Nominal
Women		48.3		55.3
Men		51.7		44.7
Age	Scale (18–74)		50.21 (13.59)		48.71 (15.08)
Marital status	Nominal
Married or cohabit		70.6		72.4
Not married or cohabit		29.4		27.6
Household composition	Nominal
Has school-aged children within household		26.4		21.0
No school-aged children within household		73.6		79.0
Victimization experience	Nominal
Has been attacked or threatened by stranger		11.0		10.3
No experienced victimization		89.0		89.7
Depressive mood	Nominal
No symptoms		84.5		83.7
At least mild depression		15.5		16.3
Fear of school shootings	Ordinal	1=‘not worried at all’ 5=‘extremely worried’	3.12	1.26	3.74
Fear of terrorism	Ordinal	1=‘strongly disagree’ 5=‘strongly agree’	3.53	1.20	3.86
Fear of street crime	Ordinal	1=‘not worried at all’ 5=‘extremely worried’
Street violence			2.54	1.12	2.34
Burglary			2.54	1.10	2.51
Violence at work			1.23	0.71	1.37
Domestic violence			1.68	1.03	1.83
Sexual harassment			1.55	0.90	1.62
Meeting neighbours	Ordinal	1=‘hardly ever’ 4=‘several times a week’	2.52	0.94	2.17

Notes

References

Adams

Serpe

(2000) Social integration, fear of crime, and life satisfaction. Sociological Perspectives 43: 605–629.

Bachrach

Zautra

(1985) Coping with a community stressor: The threat of a hazardous waste facility. Journal of Health and Social Behavior 26: 127–141.

Bellair

(2000) Informal surveillance and street crime: A complex relationship. Criminology 38: 137–167.

Britto

Van Slyke

Francis

(2011) The role of fear of crime in donating and volunteering: A gendered analysis. Criminal Justice Review 36: 414–434.

Brunton-Smith

(2011) Untangling the relationship between fear of crime and perceptions of disorder. British Journal of Criminology 51: 885–899.

Callanan

(2012) Media consumption, perceptions of crime risk and fear of crime: Examining race/ethnic differences. Sociological Perspectives 55: 93–155.

Callanan

Teasdale

(2009) An exploration of gender differences in measurement of fear of crime. Feminist Criminology 4: 369–376.

Carro

Valera

Vidal

(2010) Perceived insecurity in the public space: Personal, social and environmental variables. Quality and Quantity 44: 303–314.

Cobbina

Miller

Brunson

(2008) Gender, neighborhood danger, and risk-avoidance strategies among urban African-American youths. Criminology 46: 673–707.

10.

Cohen

(1966) Deviance and Control. Englewood Cliffs: Prentice-Hall.

11.

Collins

(2004) Rituals of solidarity and security in the wake of terrorist attack. Sociological Theory 22: 53–87.

12.

Cullen

(2009) Columbine. New York: Twelve.

13.

De Donder

De Witte

Buffel

. (2012) Feelings of unsafety in later life: A study on the influence of social networks, place attachment, and civic participation on perceived safety in Belgium. Research on Aging 34: 425–448.

14.

Durkheim

(1964). The Division of Labor in Society. New York: The Free Press (published originally in French in 1893, De la Division du Travail Social).

15.

Erickson

(1966) Wayward Puritans. New York: Wiley.

16.

Farrall

Jackson

Gray

(2009) Social Order and the Fear of Crime in Contemporary Times. Oxford: Oxford University Press, Clarendon Studies in Criminology.

17.

Ferguson

Mindel

(2007) Modeling fear of crime in Dallas neighborhoods: A test of social capital theory. Crime & Delinquency 53: 322–349.

18.

Ferraro

(1995) Fear of Crime. Interpreting Victimization Risk. New York: SUNY Press.

19.

Garland

(1990) Frameworks of inquiry in the sociology of punishment. British Journal of Sociology 41: 1–15.

20.

Garland

(2001) The Culture of Control: Crime and Social Order in Contemporary Society. Oxford: Oxford University Press.

21.

Gibson

Zhao

Lovrich

(2002) Social integration, individual perceptions of collective efficacy, and fear of crime in three cities. Justice Quarterly 19: 537–564.

22.

Gray

Jackson

Farrall

(2011) Feelings and functions in the fear of crime. British Journal of Criminology 51: 75–94.

23.

Hale

(1996) Fear of crime: A review of the literature. International Review of Victimology 4: 79–150.

24.

Hartnagel

(1979) The perception and fear of crime: Implications for neighborhood cohesion, social activity, and community affect. Social Forces 58: 176–193.

25.

Hawdon

Ryan

(2009) Social capital, social control and crime. Crime and Delinquency 55: 526–549.

26.

Hawdon

Ryan

(2011) Neighborhood organizations and resident assistance to police. Sociological Forum 26: 897–920.

27.

Hawdon

Ryan

(2012) Wellbeing after the Virginia Tech mass murder: The relative effectiveness of face-to-face and virtual interactions in providing support for survivors. Traumatology. Online First http://intl-tmt.sagepub.com/content/early/2012/04/22/1534765612441096.full.pdf+html

28.

Hawdon

Oksanen

Räsänen

Ryan

(2012) School Shootings and Local Communities: An International Comparison between the United States and Finland. University of Turku Press: Turku, Finland.

29.

Hawdon

Ryan

Agnich

(2010) Crime as a source of solidarity: A research note testing Durkheim’s assertion. Deviant Behaviour 31: 679–703.

30.

Hipp

Perrin

(2006) Nested loyalties: Local networks’ effects on neighbourhood and community cohesion. Urban Studies 43: 2503–2523.

31.

Jackson

(2004) Experience and expression: Social and cultural significance in the fear of crime. British Journal of Criminology 44: 946–966.

32.

Jackson

Gray

(2010) Functional fear and public insecurities about crime. British Journal of Criminology 50: 1–22.

33.

Jackson

Stafford

(2009) Public health and fear of crime: A perspective cohort study. British Journal of Criminology 49: 832–847.

34.

Kaltiala-Heino

Rimpelä

. (1999) Finnish modification of the 13-item Beck Depression Inventory in screening an adolescent population for depressiveness and positive mood. Nordic Journal of Psychiatry 53: 451–457.

35.

Kiilakoski

Oksanen

(2011) Soundtrack of the school shootings: Cultural script, music and male rage. Young 19: 247–269.

36.

Kupchik

Bracy

(2009) The news media on school crime and violence. Youth Violence and Juvenile Justice 7: 136–155.

37.

Lauderdale

(1976) Deviance and moral boundaries. American Sociological Review 41: 660–676.

38.

Lee

(2000) Community cohesion and violent predatory victimization: A theoretical extension and cross-national test of opportunity theory. Social Forces 79: 683–706.

39.

Liska

Sanchirico

Reed

(1988) Fear of crime and constrained behavior specifying and estimating a reciprocal effects model. Social Forces 66: 827–837.

40.

Liska

Bellair

(1995) Violent crime rates and racial composition. American Journal of Sociology 101: 578–611.

41.

Liska

Warner

(1991) Functions of crime: A paradoxical process. American Journal of Sociology 96: 1441–1463.

42.

Malkki

(2011) Suomessa ei ole terrorismia, koska “Suomessa ei ole terrorismia”? [There is no Terrorism in Finland, because “There is no Terrorism in Finland”]. In: Kullberg

(ed.) Suomi, terrorismi, Supo. Helsinki: WSOY, pp.204–224.

43.

Markowitz

Bellair

Liska

. (2001) Extending social disorganization theory: Modeling the relationships between cohesion, disorder, and fear. Criminology 39: 293–320.

44.

Mesch

(2000) Perceptions of risk, lifestyle activities, and fear of crime. Deviant Behavior 21: 47–62.

45.

Moore

Shepherd

(2007) The elements and prevalence of fear. British Journal of Criminology 47: 154–162.

46.

Muschert

(2007) Research in school shootings. Sociological Compass 1: 60–80.

47.

National Research Institute of Legal Policy (2011) Crime and Criminal Justice in Finland. Research Report 257. English Summary. Helsinki: Ministry of Law, The National Research Institute of Legal Policy. Available at: http://www.optula.om.fi/Etusivu/Julkaisut/Tutkimuksiasarja/1302674231639

48.

Newman

Fox

Harding

. (2004). Rampage: The Social Roots of School Shootings. New York: Basic Books.

49.

Nurmi

(2012) Making sense of school shootings: Comparing local narratives of solidarity and conflict in Finland. Traumatology doi: 10.1177/1534765611426787.

50.

Nurmi

Räsänen

Oksanen

(2012) The norm of solidarity: Experiencing negative aspects of community life after a school shooting tragedy. Journal of Social Work 12: 300–319.

51.

Kim

(2009) Aging, neighborhood attachment, and fear of crime: Testing reciprocal effects. Journal of Community Psychology 37: 21–40.

52.

Oksanen

Räsänen

Nurmi

. (2010) This can’t happen here!’ Community reactions to school shootings in Finland. Research on Finnish Society 3: 19–27.

53.

Rader

May

Goodrum

(2007) An empirical assessment of ‘the threat of victimization’: Considering fear of crime, perceived risk, avoidance, and defensive behaviors. Sociological Spectrum 27: 475–505.

54.

Raitasalo

(2007) Mielialakysely: Suomen oloihin Beckin lyhyen depressiokyselyn pohjalta kehitetty masennusoireilun ja itsetunnon kysely. Helsinki: Kela, Sosiaali- ja terveysturvan tutkimuksia 86. [The Finnish standardization of Beck’s depression and self-esteem inventory].

55.

Rengifo

Bolton

(2012) Routine activities and fear of crime: Specifying individual-level mechanisms. European Journal of Criminology 9: 99–119.

56.

Robinson

Lawton

Taylor

. (2003) Multilevel longitudinal impacts of incivilities: Fear of crime, expected safety, and block satisfaction. Journal of Quantitative Criminology 19: 237–274.

57.

Roundtree

(1998) A reexamination of the crime-fear linkage. Journal of Research in Crime and Delinquency 35: 341–372.

58.

Ryan

Hawdon

(2008) From individual to community: The “Framing” of 4–16 and the display of social solidarity. Traumatology 14: 43–52.

59.

Sampson

(1988) Local friendship ties and community attachment in mass society: A multilevel systemic model. American Sociological Review 53: 766–779.

60.

Sampson

Groves

(1989) Community structure and crime: Testing social-disorganization theory. American Journal of Sociology 94: 774–802.

61.

Sampson

Raudenbush

Earls

(1997) Neighborhoods and crime: A multilevel study of collective efficacy. Science 277: 918–924.

62.

Skogan

(1986) Fear of crime and neighborhood change. In: Reiss

Jr. Tonry

(eds) Communities and Crimes. Chicago: University of Chicago Press, pp.203–209.

63.

Skogan

(1990) Disorder and Decline: Crime and the Spiral of Decay in American Neighborhoods. New York: Free Press.

64.

Skogan

Maxfield

(1981) Coping with Crime: Individual and Neighborhood Reactions. Beverly Hills: Sage.

65.

Stafford

Chandola

Marmot

(2007) Association between fear of crime and mental health and physical functioning. American Journal of Public Health 97: 2076–2081.

66.

Statistics Finland (2009) Statistics Finland PX-Web Database. Available at: <http://pxweb2.stat.fi/database/StatFin/vrm/vaerak/vaerak_en.asp>

67.

Taylor

(2001) Breaking Away from Broken Windows: Baltimore Neighbourhoods and the Nationwide Fight against Crime, Grime, Fear, and Decline. Boulder, CO: Westview Press.

68.

United Nations Office on Drugs and Crime (UNODC). 2011. 2011 Global Study on Homicide: Trends, Contexts, Data. Vienna: United Nations Office on Drugs and Crime.

69.

Wilkinson

(2007) The multidimensional nature of social cohesion: Psychological sense of community, attraction, and neighboring. American Journal of Community Psychology 40: 214–229.

70.

Woldoff

(2006) Emphasizing fear of crime in models of neighbourhood social disorganisation. Crime Prevention and Community Safety 8: 228–247.

71.

Wyant

(2008) Multilevel impacts of perceived incivilities and perceptions of crime risk on fear of crime: Isolating endogenous impacts. Journal of Research in Crime and Delinquency 45: 39–64.