Report WP4 - Fear of crime and insecurity in EuropeAnna Barker et Adam Crawford
Table of contents
The central theme for the meetings in Workpackage 4 of CRIMPREV has focused on coordinating research insights into perceptions of crime and insecurity in Europe. The six seminars and conferences organised within this workpackage were concerned with a variety of allied topics related to the development of contemporary insecurities and fear of crime from a comparative perspective. As well as exploring the research into fear of crime and the causes of crime-related insecurities, the workpackage considered and explored the extent to which perceptions of crime and insecurity inform diverse forms of governance, policy initiatives and social relations. In essence, the deliberations have revolved around three core concerns:
1. Questions of measurement and conceptualisation, including what we mean by public perceptions of insecurity and fear of crime, how we measure insecurity and interpret research findings, and how valuable these measurements are.
2. The causes of perceptions of insecurity involving the influence of the media, social and political transformations, cultural phenomena (including the role of specific cultural dramas and traumas).
3. The consequences of perceived insecurity forpolicy, politics and social relations, including the enactment of new laws, policing and criminal justice reforms, urban governance and allied developments as well as the implications of these responses for particularly marginal groups and social segregation.
This paper is primarily concerned with the first of these themes, in large part, because it is foundational to the subsequent discussions. Over the course of the deliberations of workpackage 4, we have sought to test and explore two loose hypotheses that routinely inform much public debate. These are:
1. The contention that we are living in an age of increased insecurity and uncertainty in which public perceptions of crime and insecurity are becoming more prominentand more salient.
2. The view and belief that perceptions of crime and insecurity are having a growing impact on social life and public policies, as a result of which societies are ‘governed through’ perceptions of crime and insecurity (Simon, 2007).
The deliberations of workpackage 4 have challenged but also confirmed these arguments, prompting more nuanced comparative insights and highlighting important correctives and countervailing tendencies that are apparent in relation to specific local contexts. The deliberations and reviews afforded by the workpackage have shown that there remain important divergences in experiences across Europe. They have demonstrated the importance of local political trajectories, social transformations and cultural forces. They have also identified that much attributed to the role played by perceptions of crime and insecurity actually is informed and influenced by wider cultural and social referents.
To inform the deliberations, this paper provides a summary and overview of the state of research and understanding about the nature and extent of fear of crime and perceptions of insecurity across Europe. In so doing, it seeks to assess the implications and limitations of traditional measurements of fear of crime and highlights the complexities associated with interpreting public perceptions of contemporary insecurities.
A useful starting point in assessing fear of crime across Europe is to consider the findings of Europe-wide crime and victimisation surveys that have now been conducted over more than a decade. Notwithstanding the obvious measurement and methodological limitations of ‘fear of crime’ survey findings, to which we return below, available data show an upward trend in most Member States in feelings of insecurity. Figure 1 plots Eurobarometer data showing the proportion of respondents saying that they felt unsafe (either a bit unsafe or very unsafe) over time. The data show that feelings of insecurity are not felt in equal measure across the whole of Europe. Respondents in Scandinavian member states, Denmark, Finland and Sweden, were most likely to report the greatest feeling of security. This was especially strong in Denmark, where the majority of respondents (60%) felt very safe. At the other end of the scale, respondents in Greece, the United Kingdom and Italy reported the greatest feelings of insecurity i.e. feeling either a bit unsafe or very unsafe (42% in the United Kingdom and in Italy, 43% in Greece). Nearly one in five respondents in the United Kingdom (19%) and over one in five of Greek respondents (22%) said that they felt very unsafe (European Commission, 2003, 3).
Percentage of people who feel a bit unsafe or very unsafe in response to the question: How safe do you feel walking alone in the area where you live (very safe, fairly safe, a bit unsafe, very unsafe).
Database: Eurobarometer 44.3 (1996), 54.1 (2000), 58.0 (2002)1.
- 1 EB 2002 was a survey base on face-to-face interv(...)
Across the European Union as a whole, there has been a small but consistent increase in the feeling of insecurity over the period, with 32% of respondents in 1996, 33% in 2000 and 35% in 2002 saying that they felt either a bit unsafe or very unsafe. The data show three broad clusters of countries: first, those countries where the feeling of insecurity has increased since 1996 (notably the United Kingdom, Italy, France, Luxembourg, the Netherlands and Sweden); second, those countries where insecurity increased between 1996 and 2000, but then decreased (notably Greece, Belgium, Denmark and Finland); and third, those countries where there has been an overall decrease in the feeling of insecurity (notably Germany, Spain, Portugal, Ireland, Austria and the US). The data show that across the European Union, women and the elderly are most likely to feel insecure. This has remained relatively stable over the period. Furthermore, the level of fear of crime in large towns is in almost every country higher than in rural villages.
However, country specific effects also explain different levels of measurements of crime fear. The only member state where there has been a continuous decline since 1996 in ‘fear of crime’ is Germany. This trend is particularly marked in the East German Länders that formerly comprised the German Democratic Republic where fear of crime has decreased from a relatively high point in the mid-1990s, having risen dramatically in the years that followed reunification (see Figure 2). Similarly, research in Poland shows a significant growth in measurements of fear of crime in the years after the fall of Soviet communism, followed by a modest decline.
Hence, the contemporary decline in recorded fear of crime in Eastern European countries, that bucks the wider global trend of increased insecurity - as evidence by the ICVS findings (see Figure 3) - might be understood as a product of regression to the mean.
Source: Van Dijk et al., 2006
Survey research highlights implications of high levels of fear of crime in that it correlates with on the one hand avoidance strategies and protective behaviours and, on the other hand has potential consequences for attitudes, including increased punitiveness, dissatisfaction with police and criminal justice and greater levels of intolerance and xenophobia.
For example, the results of an Ipsos Mori International Social Trends monitor of the US and the five largest countries in the European Union, namely Germany, Spain, France, Italy and Great Britain, provide some additional insights into public perceptions of insecurity (see Figure 4)2. The survey, conducted in September 2006, reveals that when asked to identify the three most worrying national issues from a list of eleven topics, over two-fifths of British people said crime and violence was one of the most worrying issues (43%), double the level in Germany (21%). Moreover, whilst British people appeared the most concerned about crime, they also have the lowest confidence in their government to tackle the crime problem (with only 29%). Conversely, whilst German respondents were least concerned about crime as a national issue - only identified by one-fifth (21%) - they were the most confident (57%) in their government’s capacity to manage the crime problem.
- 2 Approximately 1,000 telephone interviews were co(...)
Our knowledge about the fear of crime is largely a product of the way it has been measured and researched (Farrall et al., 1997). Since the 1960s, large scale victimisation surveys have attempted to measure the level and extent of fear of crime. As suggested above, these surveys have provided evidence to suggest that fear of crime is a widespread phenomenon, which is rising in a number of European countries. Through these surveys, it was observed that perceptions of crime outweigh the objective risk of victimisation and, as such, the emergence of the fear of crime has come to be regarded as a problem in its own right (Hale, 1996). To this end, tackling the perceived ‘reassurance gap’ has become central to contemporary policy in some European countries, notably the United Kingdom (Crawford, 2007). However, there are considerable methodological and conceptual concerns relating to the use of survey tools for measuring public perceptions about crime and insecurity. These can be loosely divided into (i) methodological concerns and (ii) conceptual concerns.
Methodological concerns with crime surveys include both issues of epistemology and the way the concept of fear of crime has been operationalised. Underpinned by a positivist epistemology, the intention of victimisation surveys are to objectively quantify the fear of crime. In doing so, they make the assumption that there is an absolute and unchanging social reality (Bryman, Burgess, 1999) in which fear of crime is clearly conceptualised and can be operationalised and measured through standardised, fixed choice questions. Surveys have been criticised at an epistemological level for transforming social processes into quantifiable events (Bowling, 1993), which ignore[s] the emergent, innovational and problematic character of everyday life by imposing a deterministic “grid” on it with its fixed-choice structure (Cicourel, 1964, 113). Interpretive scholars, who believe that social reality is constantly under social construction (Jupp, 1989, 29), argue that there cannot be an absolute measure of ‘things’ and are critical of the ‘facts’ obtained by crime survey findings for not reflecting this (Jupp et al., 2000). In addition, the positivistic preoccupation with measuring social processes has led to the accusation that surveys simply ignore or impose meaning upon those researched or assume that meaning is ‘self-evident’ and translates culturally (Cicourel, 1964, 113). Given these epistemological criticisms, there are concerns about the ability to generalise findings from the artificial situation of surveys to everyday life. This is supported by research which has shown that there tends to be discordance between what people say they believe or do when responding to surveys and what they actually believe or do in reality (see, for example, LaPiere’s (1934) study on ‘attitudes vs. actions’). Furthermore, research shows that people’s attitudes and beliefs are ‘extraordinarily unstable’ (Foddy, 1993, 4) and may vary considerably within a short space of time.
Operational concerns largely relate to the internal validity of instrument tools and, as such, are directed at the design and phrasing of survey questions. The degree to which survey questions accurately reflect and measure the concept of fear of crime has been critically deconstructed and analysed by researchers over the last few decades. In particular, the question How safe do you feel walking alone in your neighbourhood after dark?, which has been used to indicate the level and extent of fear of crime internationally, has attracted a number of criticisms. These can be summarised as follows:
The question encourages negative responses and misleads respondents by starting with the word ‘How’ (Moser, Kalton, 1971, 325);
It refers only to safety in public space, excluding potential fear arising from the private sphere (Stanko, 1990) and the workplace;
It refers to an imprecise geographical area i.e. the ‘neighbourhood’, the meaning of which may vary by respondent (Garofalo, 1979, 82);
It does not explicitly cite the word ‘crime’ when asking respondents to judge how fearful they feel (Garofalo, 1979, 82) and it refers to ‘safety’ rather than specifically to ‘fear’;
It does not specify a time period about which respondents might answer the question, leading to wide variations (Farrall, Ditton, 1999, 59);
It asks respondents to comment on how safe they feel or would feel ‘walking alone after dark’ when they may not do this at all or may do it rarely therefore confusing the real with the hypothetical (Garafalo, 1979,82); and lastly
It confuses and conflates ‘fear’ with ‘risk’ (Ferraro, LaGrange, 1987, 77).
In short, while the widespread use of this question provides some consistency over time and across surveys (Baumer, 1985), Ferraro and LaGrange (1987, 76) suggest that this question ‘more accurately measures the risk to self of walking alone at night in one’s neighbourhood’ and hence find difficulty in justifying its use as a single and specific indicator of the fear of crime. It is here perhaps that research around multiple survey indicators of the fear of crime is important. Fear of crime is a complex concept and this complexity needs to be taken into account when attempting to measure it. Some researchers have cited the desirability of combining questions to produce a fear of crime index (Box et al., 1988).
In recent years fear of crime survey research has become more sophisticated, seeking to measure people’s worry about becoming a victim of specific offences. Questions which make reference to specific crimes, such as burglary or mugging, improve their design and validity by narrowing respondents’ frame of reference towards what LaGrange and Ferraro call ‘concrete’ fears (1989: 700). A good example of these types of question is How worried are you about being mugged? Measures that tap into ‘formless’ fears, LaGrange, Ferraro (1987) argue, leave it up to respondents to draw their own conclusions about what influences their response to a particular question which, in turn, threatens the reliability and validity of the measurement instrument. For example, a respondent answering the question How safe do you feel walking along in your neighbourhood after dark?, may base their response on any number of factors which might include being alone, being afraid of the dark, previous victimisation and fear of crime, all of which is unclear and ‘formless’. Despite these improvements to survey measures, questions which ask about ‘concrete’ fears still have the potential to encourage negative responses by beginning with the words ‘how worried’ and they do not address other contextual variables which may affect survey responses. That is, even though fear may be situated in relation to a specific type of crime, Farrall et al. (1997, 661) argue that these questions do not capture where, when and under what conditions the respondent is fearful and they fail to specify a time period in which people may express worry.
In addition to criticisms about specific operational measures, there is a more general concern over the use of survey questions which measure, on a scale, the extent of respondents’ perceptions of safety, fear or worry about crime. These questions typically require respondents to express how they feel on a five point scale from, for example, ‘very safe’ to ‘very unsafe’. These Likert-scale types of questions fail to capture important distinctions between respondents’ qualitatively different experiences and feelings of insecurity and their range and strength of emotions about particular crimes and disorders, both temporally and spatially (Farrall et al., 1997, 660). The ‘generalised levels’ of fear of crime reported by survey findings therefore lack important geographical, temporal and social context, which may be significant in understanding respondents’ actual level and extent of the fear of crime.
Together, these methodological concerns have led to the charge that surveys are only capable of capturing a ‘static and decontextualized snapshot’ of peoples’ insecurities about crime (Bowling, 1993, 232). More significantly, doubts about the internal validity of survey instruments, (alongside research which demonstrates that respondents’ levels of fear may vary depending on the way a question is asked i.e. open and closed questions), have led to the charge that existing survey tools misrepresent and exaggerate the incidence of fear of crime (Farrall et al., 1997; Farrall, Gadd, 2004). One explanation for this, provided by Farrall and Ditton (1999, 58), was that standard survey measures magnify people’s level of fear of crime because they tap into not only ‘fear’ but also a range of cognitive and affective states about crime including:
anger, outrage or annoyance
general thoughts or awareness
These are not ‘fears’ as such and concern wider feelings and attitudes about crime. This is supported by an emerging body of qualitative research, which has provided invaluable insights on this subject and has acted as a corrective to survey based research. Findings from qualitative studies suggest that ‘sensibilities’ towards crime are much more complex and more diverse than simply the fear of crime (Girling et al., 2000; Innes, Fielding, 2002; Farrall et al., 2006a; Enders et al., 2008). They find that localised concerns about crime and disorder are just one aspect of a number of issues which have the potential to cause anxiety and which might find a channel through ‘crime-talk’ (Sasson, 1995). That is, in talk about crime, people express concern about broader causes of insecurity. Theorists have commented that these might include ‘ontological’ security (Giddens, 1991), a ‘diffuse anxiety’ (Hough, 2004) or an ‘ambient experience of generalized insecurity’ (Bauman 2002: 55), all of which are only partially related to crime and disorder and gravitate around broader concerns about people’s position within society and (global) social change. As Farrall puts it: …people do not separate out the issue of crime from issues of cohesion, collective efficacy, social change and tension: rather than being about an irrational sense of crime, fear (whether it is everyday worry or anxiety) expresses and distils lay diagnoses about neighbourhood breakdown and stability and the violation of social rules (2007, 23).
If people cannot separate out these issues, it begs the question whether simplistic, single indicators of the fear of crime, often found in survey-based research, can do this. This emerging body of qualitative research is part of the shift in the theoretical ground from what has been a largely unproductive concentration on measuring the ‘fear of crime’ to that of understanding people’s ‘insecurities’. Together, these methodological concerns call into question the adequacy of existing operational survey questions to reflect and measure the concept of the fear of crime.
Conceptual concerns relate to the way in which ‘fear of crime’ has been understood and defined. These concerns are, therefore, closely linked to the way fear of crime has been operationalised by survey research. One of the major concerns with much survey-based research is that fear of crime has been ill-defined and this has led to it being inadequately operationalised and poorly measured. There are two main conceptual concerns. First, perceptions of ‘fear’ have been confused and conflated with perceptions of ‘risk’ of victimization. LaGrange and Ferraro (1987) suggest that ‘fear’ and ‘risk’ exist along a continuum, whereby the former represents emotional or affective components of perceptions and the latter represents a cognitive component to perceptions. Given that much survey research defines fear of crime in terms of perceptions of risk of victimization, surveys more accurately measure the concept of risk than they do fear, defined by Ferraro and LaGrange as an emotional reaction to crime or the symbols associated with crime (1987, 72). As they argue that the relationship between fear and risk is non-linear (fear both influences and is influenced by perceptions of risk), it cannot be a valid or an accurate measure of the fear of crime. Second, definitional problems are further confounded by the general reference to ‘crime’. Crime is a generic term which refers to many different acts. Perceptions of fear are likely to vary by crime type; fear of violence is likely to be different to fear of burglary, criminal damage, vehicle crime and robbery. Surveys rarely operate so that respondents may provide a considered and calculated answer in respect of their perception of fear in relation to crime generally and, as such, cannot indicate generalised levels of fear of crime, as the phrase implies.
In a review of the literature, Ferraro and LaGrange observed that the phrase “fear of crime” has acquired so many divergent meanings that its current utility is negligible (1987, 71). In essence, Girling and colleagues argue that the concept has been stretched’ so much so as to suggest that ‘“fear” ceases to be its primary organising idea (2000, 15).The diverse meanings given to the phrase, coupled with the preoccupation with measuring fear of crime, has hampered intellectual development and sophisticated theorisation of the concept and related concepts of risk, danger, security and safety (McLaughlin, 2006, 165-166). While it is not the intention here to render quantitative research on this subject invalid, it must be acknowledged that given these considerable epistemological, operational and conceptual concerns, what we actually ‘know’ about the fear of crime is questionable (Farrall, Ditton, 1999).
A significant new development in our understanding of the fear of crime has come from the findings of a large-scale research project ‘Experience and Expression in the Fear of Crime’ (Farrall, 2007). Building on the earlier research, which suggested that the concept of fear of crime was ill-defined and that the incidence of fear of crime had been misrepresented and exaggerated by past survey research, Farrall and colleagues sought to re-conceptualise the fear of crime and develop new and more robust survey measurements with which to test it.
The first part of the research involved re-conceptualising the fear of crime. Through qualitative in-depth interviews exploring everyday emotions about crime with a sample of 88 people (a large proportion of which reported high fear and high risk of victimisation), the researchers make a key conceptual advance in our understanding of the concept by finding that: what has previously been identified as “fear of crime” can usefully be carved up into two phenomenon: everyday worry about crime (which is relatively rare and typically affects those who live in high crime areas and who have direct and indirect experience of crime) and anxiety about crime (which is more widespread and typically affects those who lead more protected lives) (Farrall, 2007, 23).
That is, they found that the lived experience of the fear of crime in the lives of ordinary citizens exists on a continuum. At one end, emotional reactions to specific situations represent the experiential dimension of the fear of crime. At the other end, more general attitudes or anxieties about crime represent the expressive dimension of the concept. They theorise that surveys generally tap into the latter and hence produce greater estimates of the fear of crime. This returns us to the distinction made earlier by LaGrange and Ferraro (1987) between the concepts of ‘fear’ and ‘risk’. The implications of this distinction is that if you want to measure actual ‘fear’, understood as an emotional reaction, you need to disentangle and separate it from related concepts of risk and general attitudes about crime.
The second part of the research involved testing new measures designed to tap into the experiential dimension of the fear of crime and comparing these with existing survey measures. The research builds on best practice in survey research and implements the recommendations of critiques of existing crime surveys. In particular, the new measurement strategy improved the validity of past measures by:
tapping into the emotional component of the fear of crime;
making an explicit reference to particular crime types;
providing a specific time period;
avoiding hypothetical questions; and
employing a ‘filter’ question which avoids the word ‘how’
In terms of the latter, a ‘filter’ question was used to more accurately measure if respondents have actually worried about crime in the past year. Arguably, the new measurement strategy advances on old survey measurements which have routinely led respondents to answer in a negative way by leaping straight into asking how worried they are about crime. The filter question avoids this by asking In the past year, have you actually worried about…being mugged/burglary/car crime? Only if respondents confirm that they have actually worried about crime were they asked a frequency question How frequently have you felt like this in the last year?’ and an intensity question ‘On the last occasion how fearful did you feel? For these reasons, the researchers assert that the new measurement strategy provided a more precise estimation of the level of fear of crime (Gray et al., 2006).
In a rigorous and systematic examination, the researchers fielded their new measurement strategy, alongside the old measures of fear of crime, to the same respondents at different intervals to a sample of 4,448 in the 2003/2004 British Crime Survey. Their results found that the old measurement strategy produced a higher estimate of the level of worry about crime mugging, burglary and car crime than the new measurement strategy (see Figure 5). According to the researchers, the results of the new measures reflect more closely neighbourhood crime rates and personal victimisation (Gray et al., 2006, 2). This has important implications for our interpretation of the so-called ‘reassurance gap’ identified through a deviation in survey measures with recorded crime data.
Note: The old measure comprises those who responded ‘fairly’ or ‘very’ worried to the question How worried are you about…being mugged, burglary, car crime? The new measure comprises those who responded ‘yes’ to the filter question above.
A further aim of the research project was to explore how often people worried and the intensity of people’s worry about crime. Farrall and colleagues report that traditional victimisation surveys have largely concerned themselves with measuring the scale of the problem of the fear of crime and have not tended to examine the actual frequency and intensity of fearful events. This gap means that we consequently lack information on how often people worry about crime (i.e. daily, weekly, monthly, yearly etc.) and the intensity of recent fearful episodes (i.e. not very worried - very worried). The findings of their research show that worry about crime is a relatively infrequent occurrence rather than a part of everyday life for most people. That is, only a small proportion of all respondents reported worrying about crime once a week or more for all three crime types (see Figure 6). The researchers claim that this evidence suggests that old measures exaggerate or magnify the everyday experience of fear.
Of those respondents who had actually worried about crime in the past year, the majority of recent episodes of worry were ‘not very’ or ‘a little bit’ intense. However, a significant minority were ‘quite’ or ‘very’ intense (see Figure 7). They found that people’s intensity of worry about mugging was higher than for burglary or car crime.
By combining the filter and intensity questions, the researchers found that the level and intensity of fear further decreases (see Figure 8). The study concludes that people’s everyday experience of the fear of crime may have been overstated by previous research (Gray et al., 2006, 9).
By combining the old and new measures, the researchers created a typology of the sample in terms of their experiences of the fear of crime (Farrall et al., 2006b). According to this typology, 15 - 20% of the sample fall into the ‘unworried’ category who reported not being worried to both old and new survey questions. A further 15-30% of the sample fall into the ‘worried’ category who had reported some level of concern about crime to old and new survey questions, indicating both expressive and experiential elements to their concerns. The majority of the same (50 - 65%), however, fall into the ‘worried well’ category, reporting no direct experience of crime but some expressive concerns about crime in their responses to the old survey measures. This supports their re-conceptualisation of the fear of crime as a ‘relatively rare’ experience, which ‘typically affects those who live in high crime areas and who have direct and indirect experience of crime’ (Farrall, 2007, 23). Linking this research back to the first of the two loose hypotheses workpackage 4 sought to explore, the contention outlined at the beginning of this report that we are living in an age of increased insecurity is not supported by this study.
Despite attempts to refine the concept and measurement of ‘fear of crime’, to render it more meaningful, we are reminded both that fear of crime is in essence a political construct and that perceptions of insecurity serve political uses and have social implications. In many senses, the current debate about insecurity is trapped in the conceptual and methodological frame of the social survey. The invention of ‘fear of crime’ as a methodological artefact and subject of enquiry, after all, was prompted by the introduction of victimisation surveys first in the US in the late 1960s (Ennis, 1967), then in the UK in the form of the British Crime Survey (BCS, first initiated in the early 1980s) and then progressively across the globe via the International Crime and Victimisation Surveys (ICVS). These instruments have become established features of the criminological landscape informing and constraining debate. This applied criminological focus has spawned a whole research and prevention industry (Lee, 2007). The very notion of the ‘fearful subject’ has become a politically useful construct informing and being used in support of diverse policy initiatives (Lee, 2001; Sessar, 2008).
As the work of Farrall and colleagues testifies, modifying and re-conceptualising the fear of crime measurements in the BCS are of importance precisely because the BCS findings matter politically. However, this methodological fine-tuning work fails to attend to the wider question of the purpose of such research and its political uses. In the UK, public perceptions of insecurity and fear of crime as measured by the BCS not only structure public debate but also have become key performance indicators for public services such as the police. Despite their evident methodological and conceptual limitations, they have become means of holding organisations to account for their performance. This is clearly evident in the new single national confidence target for the police and local authorities in England and Wales, whereby government has committed to raise public perceptions from a baseline (measured in 2007/2008) of 45% of the adult population to 60% by 2012 (with local variations). The question asked is: How much would you agree or disagree that the police and local council are dealing with the anti-social behaviour and crime issues that matter in this area? Those who agree are deemed to be ‘confident’. Despite criticisms that such measures are crude, premised on subjective interpretations of salience and locality, conflate the work of different organizations as well as ‘anti-social behaviour’ and ‘crime’ and the factors that inform such perceptions lie far beyond the control of the agencies whose performance is being measured, nevertheless, considerable effort will be spent over the next three years or so in the quest to influence these findings. Such has been the politicisation of the BCS findings that the National Statistics Commission (2006) in a report into crime statistics not only noted a significant lack of trust as a major issue, but also recommended a clearer separation of the production of crime survey statistics from policy and that consideration be given to moving the BCS out of the Home Office to the more independent Office of National Statistics (ONS)3.
- 3 A subsequent review, published later the same ye(...)
As implied earlier (see Figure 4), public perceptions of insecurity and fear of crime may be loosely related (inversely) to confidence in the government and particular organisations to manage crime rather than the level of crime itself. Thus public policies aimed at addressing perceptions of insecurity may be more concerned with quests for state legitimacy. Fear is pivotal in politics in that it explicitly demands or implicitly prompts a response. The ‘politics of fear’ has become a pervasive feature of contemporary culture and public debate as well as a major driver of governmental action (Furedi, 2002; Sunstein, 2005, 2007). On the one hand, fear can be disabling, inciting social withdrawal and flight. Drawing inspiration from Michel de Montaigne (some three and a half centuries before), this was recognised by Franklin Roosevelt who, in his Presidential inauguration speech in 1933 during the Great Depression, asserted: let me assert my firm belief that the only thing we have to fear is fear itself - nameless, unreasoning, unjustified terror which paralyzes needed efforts to convert retreat into advance (cited in Rosenman, 1938, 11). On the other hand, however, fear can provoke action. It implies the need to make decisions about the future in the face of danger and risk. In so doing, it prompts the allied allocation of responsibility for decisions taken. It insinuates the decidability and management of danger and hence raises expectations about its governance. As such, fear can mobilise activity and constitute a focal point for social change and policy innovation.
Public perceptions of insecurity and fear of crime have also become more important and salient in criminal policy debates due to the theoretical and conceptual linkages that have been made between perceptions and their consequences for crime, disorder and neighbourhood decline. Here perceptions matter, regardless of their rational basis, precisely because they are believed to shape behaviour. In a reworking of W.I. Thomas’s famous dictum, perceptions are real in that their consequences are real. Three key theoretical models or hypotheses have become central to this field: ‘broken windows’ thesis, ‘collective efficacy’ theory and the ‘signal crimes perspective’. While each view disorder, fear and crime as inextricably connected, the conceptual linkages and emphasis placed on each of these three elements in the nexus differ. While the ‘broken windows’ thesis developed from an attempt to explain effective crime reduction in US neighbourhoods through the policing of minor disorders, the signal crimes perspective gained its impetus in the concern for effective police reassurance and bridging the so-called ‘reassurance gap’ by systematically targeting local perceptions of insecurity, central to current policy debates and performance targets in the UK.
Collective efficacy theory, however, has its roots in the rejection of the ‘broken windows’ hypothesis and the testing of an alternative understanding of the disorder-crime link. On account of the different theories of change they each advance, their implications for criminological policy debates and interventions directed at reducing or managing perceptions of insecurity have also differed.
Wilson and Kelling’s (1982) ‘broken windows’ thesis has been one of the most influential hypotheses linking the disorder-fear-crime nexus and, arguably, has become an ‘unquestioned orthodoxy’ in governmental thinking in the UK (Jones, Innes, 2005, 4). While early empirical evidence for the ‘disorder thesis’ found that disorder, fear and crime co-exist temporally and spatially, the ‘broken windows’ thesis assumes that the three components are causally related in a particular developmental sequence. The sequence begins with the idea that minor disorders, such as a broken window, if left unchecked or in disrepair, will generate the perception that ‘no one cares’. In turn, this perception will activate a series of linked responses resulting in the attraction of serious predatory crime (Wilson, Kelling, 1982; Kelling, Coles, 1996). The first of these responses, heightened levels of fear, is theorised to be a result of daily exposure to untreated disorder (disorder-fear link). The consequences of heightened fear are to set in motion a number of de-stabilising area processes including flight from the neighbourhood by those with the economic capital to move, avoidance of public spaces and lesser involvement in everyday neighbourhood life. Consequently, levels of natural surveillance and informal community controls are weakened, allowing further disorder and some crime to occur unregulated (disorder-crime link). Eventually, it is argued, this process will reach a tipping point, whereby levels of disorder and fear are high and informal guardianship of the area is low, resulting in more serious crimes being committed (disorder-fear-crime link). Thus, the ‘broken windows’ thesis assumes that disorder is ‘criminogenic’ in that it causes higher levels of crime to occur in an area. Perceptions of insecurity are fundamental within this link precisely because the consequences of fear lower community defence to the extent that, the proponents of the thesis argue, within a short period of time even a ‘nice’ neighbourhood can become an inhospitable and frightening jungle (Wilson, Kelling, 1982, 32).
Two main sources of evidence lend support to the thesis. First, Skogan’s (1990, 1986) study of disorder and decline combined quantitative data from 40 American neighbourhoods collected as part of six studies in five cities and identified ‘spirals of decay’ in five of those cities of a similar trajectory as that hypothesised by the ‘broken windows’ thesis. A key finding from his research is that disorder plays an important role in sparking urban decline (Skogan, 1990, 2). That is, the influences of structural characteristics including inequality, residential instability and ethnic heterogeneity on crime were found to be largely mediated by neighbourhood disorder. Skogan’s data appeared to offer evidence of a significant causal and independent link between disorder and crime. Recent evidence, however, has now largely refuted his findings. Harcourt’s (2001) re-analysis of Skogan’s data found only limited support for the thesis. He suggests that the trajectory of change described by Skogan only held true for robbery, rather than disorder per se. In addition, Taylor (2000) argues that Skogan’s cross-sectional research design limits our ability to infer that the causal chain runs from disorder to fear to crime. Put simply, Skogan’s test of the ‘broken windows’ hypothesis does not account for prior crime levels [which] might contribute to unrepaired signs of incivility in the first place (Taylor, 2000, 123). Taylor’s (2000) study sort to establish, using longitudinal data, the temporal ordering in the disorder-fear-crime nexus. He did not find that increases in levels of minor incivilities for sixty-six areas of the Baltimore city in 1981 predicted higher rates of crime for thirty of the same areas measured in 1994. While he does not rule out the indirect role that disorder could have on overall neighbourhood well-being and fear, he does suggest that the thesis is over-simplistic and its causal sequence remains largely unproven.
The second piece of evidence in support of the thesis came from the close associations that have been made between ‘broken windows’-inspired policing of minor disorders and the drop in serious crime in New York during the 1990s. William Bratton, the then Police Commissioner of the New York City Police Department, and George Kelling, one of the authors of the thesis, argue that the central cause for the decline in serious crime was zero-tolerance policing (Bratton, 1997; Kelling, Bratton, 1998). However, many other analyses do not support this contention. A more nuanced (and arguably more reasoned) understanding of the fall in crime is offered by Blumstein and Rosenfeld (1998). Their analysis concentrated on ‘multiple’ and ‘countervailing influences’ including the contraction of the crack cocaine market (this is explored in more depth by Bowling, 1999), economic expansion, incarceration effects which began in the 1980s and reductions of intimate partner homicides, as well as a number of other reactive forces such as the decline in gun availability.
Setting aside the link to crime, several researchers have questioned the disorder-fear link proposed by the ‘broken windows’ thesis. Implicit within the thesis is the understanding that perceptions of insecurity map neatly onto objective levels of disorder in a neighbourhood. However, a number of studies have found that not all high-crime areas have correspondingly high levels of fearful subjects (Taub et al., 1984). To explain this, several studies have sought to examine the extent to which fear is a product of ecological differences between neighbourhoods, as hypothesised by the thesis, or rather a result of psycho-social differences between individuals (Garofalo, Laub, 1978; Lewis, Salem, 1986; Taylor, 1999). A significant finding from Taylor’s research is that individuals who perceive more local problems than their neighbors are more fearful than their neighbors even within the same neighbourhood where the same ecological conditions are being observed (1999, 72). Taylor (2000, 368) argues that fear of crime has been ‘overecologized’ by Wilson and Kelling and suggests that fear is as much about residents responding in different ways to the same observable conditions, supporting Garafalo and Laub’s (1987) model. Hence, a ‘fixing broken windows’ strategy (Kelling, Coles, 1996) would fail to reduce fear of crime for all. More recent research has found that cultural stereotypes in American society are important in influencing perceptions of places as disorderly. That is, Sampson and Raudenbush found the ‘racial, ethnic and socioeconomic structure of neighbourhoods shapes perceptions of disorder over and beyond what people see in the streets’ (2005, 7). These researchers, concerned with structural factors and collective social processes on crime and perceptions, discussed next, have sought to explore a different understanding of the conceptual linkages within the disorder-fear-crime nexus.
To date, Sampson and Raudenbush’s (1999) research into collective efficacy arguably offers the most significant advance in our understanding of structural constraints and collective social processes on crime. Unconvinced by the ‘broken windows’ hypothesis and concerned with the lack of independent assessments of disorder they argue are essential for interpreting the linkages, Sampson and colleagues sought to test whether the disorder-crime link was spurious and whether differences in neighbourhood crime rates were the result of collective efficacy. Collective efficacy is defined as the linkage of cohesion and mutual trust with shared expectations for intervening in support of neighbourhood social control (Sampson, Raudenbush, 1999, 612-613). They hypothesised that if Wilson Kelling’s thesis held true, disorder should mediate the effects of any wider community processes on crime. In a rigorous and systematic social observation of 196 Chicago neighbourhoods, they were only able to find support for the causal connection between disorder and robbery. Their key finding is that both disorder and crime were found to be symptoms of the same cause - the concentration of inequality and lowered collective efficacy. Therefore, they propose that policy interventions should be directed at ‘disempowering the forces that produce both’ (Sampson, Raudenbush, 1999, 614). Given that concentrated disadvantage, immigration concentration, and residential stability explained 70% of the neighbourhood variation in collective efficacy, the proponents argue that the structural origins of crime and disorder cannot be ignored (Sampson et al., 1997, 923). In terms of the impact of fear, Sampson and Raudenbush maintain that perceptions are important because informants might reasonably infer low collective efficacy from their perceptions of disorder in the neighborhood (1999, 625) and, consequently, act accordingly. Variants on collective efficacy theory have, more recently, found that it enhances perceptions of security (Gibson et al., 2002).
Moving from the neighbourhood-level to insights from the macro-level, trans-cultural analyses of low and high crime societies have drawn parallels with collective efficacy theory. For Komiya (1999), one of the key characteristics distinguishing high and low crime societies, taking America and Japan as an example, are differences in collective social processes and, specifically, the level of ethnic homogeneity and cohesiveness of its citizens. In Komiya’s analysis, our attention is drawn to Japan’s idealisation of the group in comparison to America’s cult of the individual and theorises that there is little crime in traditional Japanese society because of the value placed on close-knit homogeneous groups which promotes security over selfish acts such as crime and conflict. In turn, it is assumed that citizens’ do not rebel against this high level of informal social control but strive for self-control to contain wants and desires and to live by the traditional culture. When crime does occur, the underlying practices of informal and formal social control place importance on shaming over stigmatisation, sympathy over blame and apology over self-justification (Hayley, 1998; Braithwaite, 1989). Here, the police are part of the moral discipline of citizens (Bayley, 1991). At a societal level, the historical ideals of freedom, democracy and autonomy that are ingrained in the lives of American citizens do not promote processes of collective efficacy as seen in Japan. Bayley proposes that social and self control are not presumed like in Japan but require conscience; the lack of regimentation that Americans value in personal life may affect the amount of criminal disorder in public life (1991, 177). The implication being that valuable insights might be drawn from low crime societies, like Japan, and to look at how they could be applied in a Western, capitalist society. In thinking about how collective social processes could be established in this context, one could observe shifts from individualist tendencies to collective processes in the example of President Barack Obama’s recent web-based election funding campaign, which shifted the emphasis away from corporate donations to donations made by the people. The election yielded a record voting turnout, particularly among those who had not previously participated. As Putnam (2000) draws parallels between participation in formal civic affairs and voting as key ingredients in social capital, it would be interesting to explore the impact, of what has widely been described as the largest grassroots campaign in American history, on collective social processes and the implications this has for the disorder-fear-crime nexus.
The signal crimes perspective continues in the same vein as Sampson and Raudenbush in thinking about disorder as part and parcel of crime itself (1999, 608). However, unlike ‘broken windows’ and collective efficacy hypotheses, which seek to explain the disorder-crime link (and the role of fear within this), the signal crimes perspective is concerned with how crime and disorder together explain perceptions of insecurity and fear (crime-disorder-fear link).
Conceptually and empirically, Bottoms (2006, 258-259) suggests that the signal crimes perspective enhances our existing understanding about the relationship between the linkages in the nexus in two main ways. First, Innes and Fielding (2002, 5.2) make a key conceptual shift from the ‘broken windows’ thesis by differentiating between the differential impacts of disorder (and crime) on perceptions of insecurity. Rather than disorder having a generic impact on fear and assuming that different disorders such as graffiti and young people hanging around are equally important in shaping perceptions of insecurity, their central proposition is that some crimes and disorder incidents matter more than others to people in terms of shaping their risk perceptions (Innes, 2004a, 336). They call these ‘signal crimes’ and ‘signal disorders’ due to their communiticative or visual properties. It is argued that these particular crimes and disorders act as ‘warning signals’ to users of the area and shape individuals’ beliefs about their degree of safety and security either retrospectively, indicating signs of danger occurring in the past, or prospectively, indicating signs of a future risk. Crimes and disorders that function as ‘signals’ are distinguished from those which function as ‘noise’, defined as the plethora of incidents and occurrences that… assume no real significance to people in the routines of daily life, and consequently are not consciously attended to (Innes, 2004a, 342). Drawing on Eco’s (1976) idea that ‘a sign does something’, a crime or disorder must have an ‘effect’ for it to be defined as a ‘signal’ (Innes, 2004b, 15). Innes proposes that such effects are important because they change the way people think, feel or behave. In making these links, Bottoms (2006) notes, the signal crimes perspective presents an advance on the more simplistic and ambiguous role allocated to disorder by the ‘broken windows’ thesis.
Second, the signal crimes perspective advances our understanding about the way perceptions of insecurity are formed and shaped through both exposure to multiple signals and the differential functioning of signals for different people and in different neighbourhood contexts. In terms of the former, drawing on Wilkins’ (1964) notion of an ‘amplification effect’ and Vaughan’s (2002) distinction between ‘weak’ and ‘strong’ signals, Innes suggests that exposure to multiple weak signals may function as a strong signal serving to amplify ‘the perceived significance of other previous and subsequent incidents’ (2004a, 346). It is here that similarities can be drawn between the signal crimes perspective and the ‘broken windows’ thesis, in that both claim that incivilities, which to the police may seem trivial or low-level, can represent a serious threat to local safety (Millie, 2003). The idea of an amplification effect is also very similar in tone to that of Frank Field’s notion of ‘persistence’ whereby he argues that it is in its regularity that anti-social behaviour wields its destructive force (2003, 77).
In terms of the latter, Innes and Fielding (2002, para. 5.5) do not assume that everyone will interpret a signal in the same way. They suggest that socio-demographic factors such as social class, age, gender, ethnicity and other variables, such as previous victimisation and lifestyle, shape how a signal is interpreted. Innes and colleagues’ notion of ‘attentiveness’, defined as integration in local social networks and routine activitie’, is also important because residents will be more or less tuned in to crime and disorder happening locally and hence more or less likely to fear it (2004, vxiii). This theory of insecurity is very different to that proposed by variants of collective efficacy theory, which argue that integration in social networks decreases fear because they benefit from the norms of reciprocity and trust that arise from them (Gibson et al., 2002). In addition, Innes and Fielding (2002) suggest that signals will be contextually situated. While it has become accepted within the research literature that perceptions of crime and disorder are mediated by neighbourhood context, the signal crimes perspective offers a specific interpretation of these ‘context effects’ with reference to the concept of ‘dissonance’. For the signal crimes perspective, it is the ‘dissonance value’ which determines the impact of a crime or disorder on perceptions of insecurity within a given neighbourhood. They theorise that in affluent areas residents may become more fearful of disorder if it is out of place with ‘normal appearances’ (Goffman, 1972) while in deprived neighbourhoods, disorder may be regularly visible. In this way, Goffman’s (1972) notion of ‘normal appearances’ contributes to our understanding of why ‘signs of disorder may signal to users of a location either an absent, weakened or fragile local social order’ and, in turn, why people modify their attitudes or behaviour in ‘tune’ with these ‘signals’ (Innes, 2004a, 341). This is also important in understanding why social and physical disorders, rather than crime per se, may be significant in generating fear within a neighbourhood. Despite these qualifications, the empirical work found that there is a collective dimension to risk perception, termed ‘signal coherence’, defined as agreement between residents about whom or what is driving insecurity (Innes et al., 2004, 347).
However, other perspectives argue that there is a different relationship between neighbourhood context, crime and fear. Counter to the above, Taylor et al. (1985) argues that disorder has the most impact on the perceptions of people who live in ‘moderate-income’ neighbourhoods rather than affluent areas. While in affluent areas disorder will be read as ‘nonthreatening’ precisely because of its ‘dissonance value’ (i.e. the fact that disorder is atypical) and in deprived areas it will be sidelined because of more serious concerns, in moderate-income neighbourhoods the appearance of disorder can heighten insecurity because of their fragile status, especially given the effects of disorder on local housing markets (Little 1976). In addition, drawing on the ideas around ‘desensitisation’,4 there is theoretical support to suggest that in neighbourhoods where people already experience high levels of crime and disorder, the ‘amplification effect’ described by Innes could actually work in reverse. That is, cumulative exposure to ‘weak signals’, rather than combining to function as a ‘strong signal’, could in theory weaken or desensitise people to previous or further acts of crime and disorder in their neighbourhood. While this does not render their insecurities any less valid, if this were true, it may go some way to explain why some researchers have found that levels of insecurity are not always high in run-down, disorderly neighbourhoods (Taub et al., 1984). This is something that has not been fully explored in the literature (except, see reference to the argument by Rogerson, Christmann, (2007)). While Innes recognises that a crime or disorder may only become a ‘signal’ depending on its ‘dissonance value’ (sensitisation), incorporating these ideas to some extent, the perspective does not fully account for the possibility of ‘desensitisation’ to disorder and crime. Much is left unsaid about the concept of ‘noise’, noted above, which could equally provide valuable insights about why and under what circumstances particular crimes and disorders are not attended to by the public and hence why they are not considered a key driver of their insecurities.
- 4 Here, the term is used in a similar way to that(...)
Turning the debate from the causes of fear to its consequences, the signal crimes perspective argues that ‘public perceptions matter at least as much as objective conditions’ in their consequences for crime, disorder and neighbourhood decline (Innes, Jones, 2006, 27). To mitigate the consequences of fear and counter public anxieties and insecurities, Innes et al. (2004) develop the notion of ‘control signals’. A ‘control signal’ is defined as an act of formal or informal social control that functions to communicate a message about the presence or absence of effective security mechanisms (Ibid., 14). In order to enhance perceptions of security and foster reassurance, the signal crimes perspective implies a broader role for the police. Ditton and Innes (2005) argue that the police have to engage with what they call the ‘logic of perceptual intervention’. In terms of policing this means that attention should be paid to the processes of symbolic communication, impression management and the ways in which communities interpret crime and policing on a routine basis (Innes, Fielding, 2002, 8.4). In doing so, the signal crimes perspective gives due significance to the symbolic and communicative aspect of crime and disorder (as well as policing), which has been ‘often overlooked’ in criminological theory and practice (Crawford, 2007, 155). However, this contention has led to complex and contested arguments about how and where to concentrate police and other public resources. We know that crime hot spots and fear hot spots do not always align, with some areas comprising high levels of fear and low levels of crime. The danger is that the signal crimes logic may create a situation where the distribution of policing resources lies in an inverse relationship to the distribution of crime risks (Hope et al., 2001), especially as some communities or individuals may not have sufficient social capital to be heard by the police (Loader, 2006). Taking into consideration arguments on desensitisation and the ethics of fear reduction, Rogerson and Christmann question whether it is fair to direct interventions towards a community because it is seemed [to be] over sensitised to crime or to ignore crime in a community because it has “grown used to it” and no longer voices its concern (2007, 83). In terms of criminal justice policy debates then, as Crawford aptly puts it …the key normative question therefore is to what extent, in the name of reassurance, should public policing as a residual public resource, be drawn into low-level incivilities in areas where fear of crime may be high but the incidence of crime low, and hence, away from areas with more serious crime problems? (2007, 161). In a similar vein, FitzGerald et al. (2002) argue that signal crimes methodology is at risk of real scrutiny if it is used only as way of making people feel safer. They argue that the most important goal of initiatives must be to affect objective material conditions - levels of crime and disorder - rather than manage people’s subjective impressions. However, for the proponents of the perspective, the concern with targeting perceptions of insecurity is central precisely because insecurities are thought to have consequences for community defence. Here, Bottoms draws similarities with Sampson’s work in stating that positive control signals should encourage, as an effect, residents’ willingness to engage in the interventions that are a key element in collective efficacy (2006, 269). To this end, proponents of the signal crimes perspective repeatedly reaffirm that those who dismiss criminal justice policy interventions designed to engage with ‘impression management’, or what has come to be regarded as the ‘just perceptions’ debate, fail to consider the importance of perceptions in shaping how neighbourhoods change (Innes et al., 2009, 31).
Returning to the two hypotheses outlined in the introduction to this report, we see that public perceptions of insecurity have become central to criminal policy in the UK and other European countries precisely because we are believed to be living in an age of increased insecurity and because this pervasive sense of insecurity is thought to have a growing impact on, and consequences for, social and public life. The consequences of fear on policy debates are especially apparent over the last decade or so. In the UK there has been a shift in governmental thinking towards ‘fear reduction’ as an object of policy in its own right (Hale, 1996). Similar developments are to be found in France (Roché, 2000, 2003), Italy (Bricocoli, Savoldi, 2008) and the Netherlands (van Swaaningen, 2005). In the UK, this has taken a particular form given the emphasis on measuring perceptions and the use of such measurement in holding organisations accountable for their performance. This has seen a realignment of traditional organisational priorities towards public demands (Crawford, 2007), a new national ‘confidence target’, an increasing focus on the ‘communicative’ or visual properties of policing and the policing of anti-social behaviour and environmental ‘signs of crime’. Collectively these trends affirm that public reassurance and ‘impression management’ are at the heart of contemporary criminal policy. As discussed above in reference to the politics of fear and insecurity, these agendas have not however been wholly unguided. Given that there is now arguably ‘more fusion than faction’ (Downes, Morgan, 2002, 317) on law and order issues between the major political parties, persistent jostling for political advantage (Noaks, Wincup, 2004, 22) continues over levels of investment in the criminal justice system, police numbers and effective reassurance and security provision rather than ideologically different policy options. Fear of crime and public perceptions arguably play a lead role. In this way, we see that fear of crime, cast as a social problem that needs to be managed, has become profitable for both political purposes and the security industry.
Fear can be a good thing in the sense that it can lead to protective behaviours which may reduce risk. In a sense, the misperception of crime - under estimating risk - may mean we fail to take adequate precautions to protect our safety. However, fear has been cast overwhelmingly as a destructive force. As early as Skogan (1986), connections have been made between the consequences of fear and the structuring of American society through processes of ‘white flight’, leaving typically inner city neighbourhoods devoid of the kinds of conditions which protect against crime and disorder. Since this time, the driving force behind the rise in gated communities in various parts of the world is theorised to supply the fearful with protection from the fearsome and have subsequently been cast as ‘refuges’ from urban crime and fear (Atkinson, Blandy, 2005, 180). For Simon (2007), as well as the rise in gated communities, the implications of ’governing through crime’ in America can be observed in the use of mandatory drug testing by employers as well as the routine screening of school children for weapons. He argues that fear has systematically become institutionalised within our everyday life. In the UK and elsewhere in Europe, the ‘criminalisation of social policy’ is reconceptualising dominant views of welfarism.
By way of concluding this discussion, we observe that the disorder-fear-crime nexus goes some way towards explaining urban decline and change and this recognises the importance of public perceptions in these processes. As we have seen, the disorder-fear-crime nexus is primarily concerned with the ‘local social order’ (Bottoms, 2006) which, in essence, suggests that order within the public sphere, especially relationships between strangers, is central to ‘human flourishing’ (Wrong 1994) and, by implication, certain crimes and disorders are privileged in this nexus. However, several scholars have noted that multiple urban conditions, besides crime and disorder, may inspire fear (Taylor, 2000, 66) as the work of Giddens (1991) regarding ‘ontological security’ and Bauman’s (2002) orientation towards an ‘ambient sense of generalised insecurity’ testify. Neither is fear perhaps the most important factor explaining how or why particular communities decline. Urban change processes are complex, and in this respect, several scholars bring our attention back to the fundamental question about the extent to which structural factors, and particularly concentrated disadvantage, influence community variations in crime and neighbourhood decline (see Morenoff et al., 2001).
The foregoing overview of the literature and research highlights a number of issues that challenge many dominant policy prescriptions. First, the complex manner in which generalised anxieties, uncertainties and sentiments of well-being and identity infuse and confound expressions of fear of crime, not only raises methodological and conceptual challenges of measurement and meaning but also questions universal or easily transferable policy solutions. Not only do perceptions of crime and insecurity differ significantly between nation-states but so too, more fundamentally, they take on a decidedly local character within and between cities, regions, countries and social groups in the population. Despite pressures of globalisation and the instantaneous flows of mass information that link apparently disparate parts of the world, our insecurities and sense of social order remain refracted through local structures and cultural sensibilities.
Second, the above discussion reinforces the pivotal role played by ordinary citizens and civil society institutions in structuring order and a sense of security. As Jane Jacob’s astutely noted nearly half a century ago: The first thing to understand is that the public peace… of cities is not kept primarily by the police, necessary as police are. It is kept primarily by an intricate almost unconscious, network of voluntary controls and standards among the people themselves, and enforced by the people themselves (1961, 31-32). This should reinforce the central role that the public play in understanding and shaping crime control and local social order, and place public engagement at the centre of a politics of security. And yet, conflating security with criminal justice and policing risks what Simon (2007) has aptly described as ‘governing through crime’, where crime becomes an organising concept central to the exercise of contemporary authority. Positing a policing solution to the problems of security, well-being and order is not only misleading but, given the capacity of crime and policing to excite fears and anxieties, may not be the most appropriate vehicles around which to organise open, tolerant and inclusive communities. The quest for policing and security solutions to local problems of order may nevertheless fail to tackle the more fundamental and structural social issues that often lie behind and inform these problems.
Finally this leaves us with questions about the nature of appropriate responses to public fears and insecurities. If public perceptions matters, both in themselves and in their consequences, how should governments respond to them? Given the disjuncture between subjective anxieties and more objective categorisations of risk, and the identification of the ‘worried well’ and the ‘stoically fearful’, however, we would be wrong to treat public perceptions as unproblematic. Simply responding to public insecurity by acceding to demands for greater security and policing through the provision of more visible patrols and security hardware may fail to engage with and negotiate the nature of the demands voiced by different publics. As a consequence, it may miss the opportunity to subject public demands and expectations to some kind of reasoned debate and local dialogue. In such a debate, governments have a pivotal role in informing the public about different risks through measured evidence and rational harm assessment. Rather than exploiting a politics of fear, governments need to take careful steps to ensure that policies, initiatives and laws reduce, rather than inflate or replicate, the errors to which fearful people are prone (Sunstein, 2005, 226).
ACPO, Reassurance – Civility First: A Proposal for Police Reform, London, ACPO, 2001.
Bauman Z., Violence in the age of uncertainty, in Crawford A. (ed.) Crime and Insecurity: The Governance of Safety in Europe, Cullompton, Willan, 2002.
Baumer T. L., Testing a General Model of Fear of Crime: Data from a National Sample, Journal of Research in Crime and Delinquency, 1985, 22, 3, 239-255.
Bayley D., Forces of Order: Policing Modern Japan, University of California Press, England, 1991.
Blumstein A., Rosenfeld R., (Recent Trends in U.S. Homicide Rates, Journal of Criminal Law and Criminology, 1988, 88, 4, 1175-1216.
Bottoms A., Incivilities, Offence, and Social Order in Residential Communities, in Simester A.P., A. Von Hirsh A. (eds), Incivilities: Regulating Offensive Behaviour, Oxford, Hart, 2006.
Bowling B., Racial Harassment and the Process of Victimization, British Journal of Criminology, 1993,33, 2, 231-250.
Bowling B., The Rise and Fall of New York Murder: Zero Tolerance or Crack’s Decline?, British Journal of Criminology, 1999, 39, 4, 531-554.
Box S., Hale C., Andrews G. Explaining Fear of Crime, British Journal of Criminology, 1988, 28, 3, 340-356.
Braithwaite J., Crime, Shame and Reintegration, Cambridge University Press, New York, 1989.
Bratton W., Crime is down in New York City: Blame the Police, in Dennis N. (ed.), Zero Tolerance: Policing a Free Society, London, Institute of Economic Affairs, 1997.
Bricocoli M., Savoldi P., Villes en Observation: Politiques locales de sécurité urbaine en Italie, Paris, PUCA, 2008.
Bryman A., Burgess R.G., Qualitative Research, London, SAGE, 1999.
Cicourel A., (1964) Method and Measurement in Sociology, New York: Free Press.
Crawford A., Reassurance Policing: Feeling is Believing, in Henry A., Smith D.J. (eds), Transformations Of Policing, Aldershot, Ashgate, 2007, 143-168.
Ditton J., Innes M., The Role of the Perceptual Intervention in the Management of Crime Fear, in Tilley N. (ed.), The Handbook of Crime Prevention and Community Safety, Cullompton, Willan, 2005.
Downes D., Morgan R., The Skeletons in the Cupboard: The Politics of Law and Order at the Turn of the Millenium, in Maguire M., Morgan R., Reiner R. (eds), The Oxford Handbook of Criminology, Oxford, Oxford University Press, 2002.
Enders M., Jennett C., Tulloch M., Revisiting Fear of Crime in Bondi and Marrickville: Sense of Community and Perceptions of Safety, in Farrall S., Lee M. (eds), Fear of Crime: Critical Voices in an Age of Anxiety, Abingdon, Routledge, 2008.
Ennis P.H., Criminal Victimization in the United States. A Report of a National Survey, National Opinion Research Center, Washington DC, US Government Printing Office, 1967.
European Commission, Analysis of Public Attitudes to Insecurity, Fear of Crime and Crime Prevention, Results of Eurobarometer 58.0, Brussels, European Commission, 2003.
Farrall S., Experience and Expression in the Fear of Crime: Full Research Report, Swindon, ESRC, 2007.
Farrall S., Bannister J., Ditton J., Gilchrist E., Questioning the Measurement of the Fear of Crime: Findings from A Major Methodological Study, British Journal of Criminology, 1997, 37, 657-678.
Farrall S., Ditton J., (1999) Improving the Measurement of Attitudinal Responses: An Example from a Fear of Crime Survey, International Journal of Social Research Methodology,1999,2, 1, 55-68.
Farrall S., Gadd D., Evaluating Crime Fears: A Research Note on a Pilot Study to Improve the Measurement of the “Fear of Crime” as a Performance Indicator, Evaluation, 2004, 10, 4, 493-502.
Farrall S., Jackson J., Gray E., Everyday Emotion and the Fear of Crime: Preliminary Findings from Experience and Expression, Experience & Expression in the Fear of Crime Working Paper No. 1, Swindon, ESRC, 2006a.
Farrall S., Jackson J., Gray E Combining the New and Old Measures of the Fear of Crime: Exploring the ‘Worried-Well’, Experience and Expression in the Fear of Crime Working Paper No. 4, ESRC Grant RES 000 23 1108, Swindon, ESRC, 2006b.
Ferraro, K.F. and LaGrange, R. (1987) ‘The Measurement of Fear of Crime’, Sociological Inquiry, 57(1): 70–101.
Field F., Neighbours from Hell: The Politics of Behaviour, London, Politico’s, 2003.
Fitzgerald M., Hough M., Joseph I., Qureshi T., Policing for London, Cullompton, Willan, 2002.
Foddy W., Constructing questions for interviews and questionnaires: theory and practice in social research, Cambridge, Cambridge University Press, 1993.
Furedi F., Culture of Fear, London, Continuum, 2002.
Garofalo J., Victimization and the fear of crime, Journal of Research on Crime and Delinquency, 1979, 16, 1-97.
Garofalo J., Laub J.H., The Fear of Crime: Broadening Our Perspective, Victimology, 1978,3, 242-253.
Gibson C. L., Zhao J., Lovrich N.P., Gaffney M.J., Social Integration, Individual Perceptions of Collective Efficacy, and Fear of Crime in Three Cities, Justice Quarterly, 2002, 19, 3, 537-564.
Giddens A., Modernity and Self-Identity, Cambridge, Polity Press, 1991.
Girling E., Loader I., Sparks R., Crime and Social Change in Middle England, London, Routledge, 2000.
Goffman E., Relations in Public: Microstudies of the Public Order, New York,Harper Colophon, 1972.
Gray E., Jackson J., Farrall S., Reassessing the Fear of Crime: Frequencies and Correlates of Old and New Measures, Working Paper No. 3, Swindon, ESRC, 2006.
Hale C., Fear of Crime: A Review of the Literature, International Review of Victimology, 1996, 4, 79-105.
Haley J.O., Apology and pardon: learning from Japan, American Behavioral Scientist, 1998, 41, 6, 842-868.
Harcourt B.E., Illusion of Order: The False Promise of Broken Windows Policing, London,Harvard University Press, 2001.
Hope T., Karstedt S., Farrall S., The Relationship between Calls and Crimes, London, Home Office, 2001.
Hough M., Worry about Crime: Mental Events or Mental States?, International Journal of Social Research Methodology, 2004, 7, 2, 173-176.
Innes M., (2004a) Signal Crimes and Signal Disorders: Notes on Deviance as Communicative Action, British Journal of Sociology, 2004a, 55, 3, 335-355.
Innes M., Crime as a Signal, Crime as a Memory, Journal for Crime, Conflict and the Media, 2004b, 1, 2, 15-22.
Innes M., Abbott L., Lowe T., Roberts C., Weston N., Signal Events, Neighbourhood Security Order and Reassurance in Cardiff, Universities Police Science Institute, University of Cardiff, 2009.
Innes M., Fielding N., From Community to Communicative Policing: “Signal Crimes” and the Problem of Public Reassurance, Sociological Research Online, 2002, 7, 2.
Innes M., Hayden S., Lowe T., MacKenzie H., Roberts C., Twyman L., Signal Crimes and Reassurance Policing Volume 1: Concepts and Analysis, Guildford, University of Surrey, 2004.
Jacobs J., The Death and Life of Great American Cities, New York, Random House, 1961.
Jansson K., British Crime Survey – Measuring Crime for 25 years, London, Home Office, 2007.
Jones V., Innes M., Crime-Disorder-Fear Nexus and Urban Development Trajectories. Paper presented at the Securing the Urban Renaissance: Policing, Community and Disorder Conference, 16-17 June 2005.
Jupp V., Methods of Criminological Research, London, Unwin Hyman, 1989.
Jupp V., Davies P., Francis P., Doing Criminological Research, London, Sage, 2000.
Kelling G., Bratton W., Declining Crime Rates: Insiders views of the New York Crime Story, Journal of Criminal Law and Criminology, 1998, 88, 4, 1217-1231.
Kelling G., Coles C., Fixing Broken Windows, New York, Free Press, 1996.
Komiya N., A Cultural Study of the Low Crime Rate in Japan, British Journal of Criminology, 1999, 39, 3, 369-390.
LaGrange R.L., Ferraro K.F., Assessing Age and Gender Differences in Perceived Risk and Fear of Crime, Criminology, 1989, 27, 4, 697-720.
LaPierre R., Attitudes vs. Actions, Social Forces, 1934, 13,2, 230-237.
Lee M., The genesis of “fear of crime”, Theoretical Criminology, 2001, 5, 467-485.
Lee M., Inventing Fear of Crime: Criminology and the Politics of Anxiety, Cullomptom, Willan, 2007.
Lewis D.A., Salem G., Fear of Crime,New Brunswick, NJ, Transaction Books, 1986.
Little J.T., Residential Preferences, Neighborhood Filtering and Neighborhood Change, Journal of Urban Economics, 1976, 3, 68-81.
Loader I.,) Policing, Recognition and Belonging, Annals of the American Academy of Political and Social Science, 2006, 605, 1, 201-221.
McDonald M.P., The Return of the Voter: Voter Turnout in the 2008 Presidential Election, The Forum, 2008, 6, 4, 4.
McLaughlin E., Fear of Crime, in McLaughlin E., Muncie J. (eds), The Sage dictionary of criminology, (2nd Edition), London, Sage, 2006.
Miethe T., Lee G.R., Fear of crime among older people: A reassessment of the predictive power of crime-related factors, Sociological Quarterly, 1984, 397-415.
Millie A., Monitor’s report, in Police Foundation (Ed.), The Reassurance Policing Pilot Project: An Interim Report on Independent Monitoring and Mentoring Activity, London, Police Foundation, 2003.
Morenoff J., Sampson R., Raudenbush S.W., Neighborhood Inequality, Collective Efficacy, and the Spatial Dynamics of Urban Violence, Criminology, 2001, 39, 3, 517-559.
Moser C., Kalton G., Survey Methods in Social Investigation, (2nd edition), Aldershot, Gower, 1971.
Noaks L., Wincup E., Criminological Research, London, Sage, 2004.
Putnam R., Bowling Alone: The Collapse and Revival of American Community, New York, Simon and Schuster, 2000.
Roché S., La théorie de la “vitre cassée” en France. Incivilités et désordre en public, Revue française de science politique, 2000, 50, 3, 387-412.
Roché S., En quête de sécurité. Causes de la délinquance et nouvelles réponses, Paris, Armand Colin, 2003.
Rogerson M., Christmann K. Burglars and Wardrobe Monsters. Practical and Ethical Problems in the Reduction of Crime Fear, British Journal of Community Justice, 2007, 5, 1, 79-94.
Rosenman S. (ed.), The Public Papers of Franklin D. Roosevelt, Volume Two: The Year of Crisis, 1933, New York, Random House, 1938.
Sampson R., Raudenbush S., Neighborhood Stigma and the Perception of Disorder, Focus,2005,24, 1,7-11.
Sampson R., Raudenbush S., Systematic Social Observation of Public Spaces: A New Look at Disorder in Urban Neighbourhoods, American Journal of Sociology, 1999, 105, 3, 603-651.
Sampson R., Raudenbush S., Earls F., Neighbourhoods and Violent Crime: A Multilevel Study of Collective Efficacy Science, 1997, 277, 918-924.
Sasson T.,) Crime Talk: How Citizens Construct a Social Problem, New York, Aldine de Gruyter, 1995.
Sessar K., Fear of Crime or Fear of Risk?Some Considerations Resulting from Fear-of-Crime Studies and their Political Implications, in Kury H. (ed.), Fear of Crime – Punitivity. New Developments in Theory and Research, Bochum, Brockmeyer, 2008, 25-32.
Simon J., Governing Through Crime, Oxford, Oxford University Press, 2007.
Skogan W., Disorder and Decline, New York, Free Press, 1990.
Skogan W., Fear of Crime and Neighborhood Change, Crime and Justice: A Review of Research, 1986, 8, 203-229.
Smith A., Crime Statistics: An Independent Review, London, Home Office, 2006.
Smith S., Donnerstein E., Harmful effects of Exposure to Media Violence: Learning of Aggression, Emotional Desensitization, and Fear, in Geen R., Donnerstein E. (eds), Human Aggression: Theories, Research, and Implications for Policy, New York, Academic Press, 1998.
Stanko E.A., Everyday Violence: How Women and Men Experience Sexual and Physical Violence, London, Pandora Press, 1990.
Statistics Commission, Crime Statistics: User Perspectives, London, Statistics Commission, 2006.
Sunstein C., Laws of Fear: Beyond the Precautionary Principle, Cambridge, Cambridge University Press, 2005.
Sunstein C., Worst-Case Scenarios, Cambridge, MA, Harvard University Press, 2007.
Taub R.P., Garth Taylor D.G., Dunham J.D., Paths of Neighborhood Change: Race and Crime in Urban America, Chicago, University of Chicago Press, 1984.
Taylor R., The Incivilities Thesis: Theory, Measurement, and Policy, in Measuring What Matters, Washington, D.C., National Institute of Justice, 1999.
Taylor R., Breaking Away from Broken Windows, Boulder, CO, Westview Press, 2000.
Taylor R. B., Schumaker S., Gottfredson S.D., Neigborhood-Level Linkages between Physical Features and Local Sentiments: Deterioration, Fear of Crime, and Confidence, Journal of Architectural Planning and Research, 1985, 21, 261-275.
Van Dijk J., Manchin R., van Kesteren J., Nevala S., Hideg G., The Burden of Crime in the EU, Research Report: A Comparative Analysis of the European Crime and Safety Survey (EU ICS) 2005, Turin, UNICRI, 2006.
Van Swaaningen R., Public safety and the management of fear, Theoretical Criminology, 2005, 9, 3, 289-305.
Vaughan D., Signals and Interpretive Work: The Role of Culture in a Theory of Practical Action, in Cerulo K. (ed.), Culture in Mind: Toward a Sociology of Culture and Cognition, New York, Routledge, 2002.
Wilkins L., Social Deviance: Social Policy, Action and Research, London, Tavistock, 1964.
Wilson J.Q., Kelling G.L., Broken Windows, Atlantic Monthly, 1982, 29-38.
Wrong D.H., The Problem of Order: What Unites and Divides Society, New York, Free Press, 1994.
1 EB 2002 was a survey base on face-to-face interviews with 16,067 respondents aged 15 years and over across 15 member states, with a sample of 1,000 respondents for each member state.
2 Approximately 1,000 telephone interviews were conducted with representative samples of adults in each country [http://www.ipsos-mori.com/content/britons-most-worried-by-crime-and-government-is-le.ashx] (last accessed 31.12.2008).
3 A subsequent review, published later the same year, also recommended the separation of statistical and political commentary on BCS figures (Smith, 2006). It concluded that it is no longer possible to have a rational debate on the basis of agreed facts.
4 Here, the term is used in a similar way to that of the learned emotional desensitisation to media violence, whereby empirical studies have found that consumers of violence become accustomed or comfortable with it following repeated exposure (Smith, Donnerstein, 1998). While some have attempted to link this concept to crime, it has been less discussed in relation to fear.
Date of publishing :
Paper ISBN :
To cite this document
Adam CRAWFORD – Director - Centre for Criminal Justice Studies - University of Leeds - UK - LEEDS LS2 9JT. E-mail : Law6ac@Leeds.ac.uk