Local Crime Prevention: ‘Breathing Life (Back) into Social Democratic and Penal Welfare Concerns’? 1

Fieldwork in the inner‐Sydney postcode area of Glebe (New South Wales, Australia) sought to understand how local community workers conceptualise crime causation and the approaches adopted to prevent crime. Observation of more than 30 inter‐agency meetings, 15 interviews and two focus groups with diverse local workers revealed that social‐welfare or ‘root’ causes of crime were central to explanations of local crime. Numerous crime prevention measures in the area respond directly to these understandings of crime (a youth diversion program on Friday and Saturday evenings, an alternative education program, a police‐youth exercise program, and so on). While other more surveillant forms of crime prevention were evident, the findings of this research suggest a significant social‐welfare orientation to crime prevention. These findings echo Brown’s (2012) observations of the resilience of penal‐ welfarism in Australia.


Introduction
observed that: … at least in the Australian context, and I suspect elsewhere, conversations with frontline criminal justice workers … often quickly turn to the material force of the social and economic determinants of crime, albeit not in the exact same terms ... Invariably the talk turns to drug and alcohol use and abuse, mental illness, poor education and job prospects, dysfunctional families, family violence, rehabilitation, resettlement difficulties, the inadequacy of post-release services. (Brown 2012: 88) This article explores, through the use of a local case study, explanations of crime and the responses to local crime by workers directly engaged in or indirectly contributing to the delivery of programs that seek to prevent crime. Despite the significant and growing body of JCJ&SD 42 Online version via www.crimejusticejournal.com © 2015 4(4) With respect to unemployment, 6.7 per cent of the Glebe residential population was unemployed in 2011 (compared with 5.9 per cent in NSW and 5.6 per cent in Australia). There is also diverse evidence of socially polarised populations. The two most common personal income categories are greater than A$2000 and between A$200 and A$299 per week and a significant proportion of the Glebe population (19 per cent in 2011, down slightly from 22 per cent in the 2006 and 2001 Censuses) resides in public or social housing. An almost equal number of Glebe residents live in properties that are owned outright, further highlighting the socio-economic differences across the suburb.
This area was selected because of the significant disparity in income and housing types (see Bottrell 2009 andVinson andRawsthorne 2013, for further discussion of these disparities), the presence of key crime generators and attractors (that is, shopping centre, large public housing estates, licensed premises), and the combination of significant residential transience (student, backpacker, and short-term accommodation) and stability. These characteristics provide important tensions and dynamics relevant to crime and its prevention in the area.

Research methods
A case study approach was adopted to explore how local actors understand crime causation and the local crime prevention practices and processes operating in the area. Simons (2009: 3) defines a case study as the 'study of the singular, the particular, the unique' and as an 'in-depth exploration from multiple perspectives of the complexity and uniqueness of a particular project, policy, institution, programme or system in a "real life" context'.
The findings presented here are from a larger research project (see Clancey 2014) which included:  Physical familiarisation -in the spirit of Connell's (2007: 206) arguments for 'linking theory to the ground on which the theorist's boots are planted', considerable time was spent walking the streets of the case study area. This allowed routine activities, pedestrian traffic, and the adoption of security and crime prevention practices to be observed.  Desktop reviews were conducted to gain an understanding of the history, social dynamics and service delivery systems operating in the case study area.  Crime data for an 18-year period were analysed.
 In excess of 30 inter-agency meetings, informal interviews, and community events in the case study area were observed and notes recorded.  Fifteen formal, semi-structured interviews with workers from various agencies in the area were conducted.  Two focus groups were conducted, with the express purpose of understanding falls in crime in Glebe in recent years and to discuss the impact of these falls on the Glebe Community Safety Plan 2009-2012 (developed specifically for this suburb by the City of Sydney Council, the local government body for the area).
For the purposes of this article, semi-structured interviews and focus groups with local actors generated the most relevant data.
Diverse inter-agency meetings were observed during the fieldwork period. These included Police Community Safety Precinct Committee (CSPC) meetings, the Mayor's local housing forums, Annual General Meetings of local services, meetings of a local inter-agency (the Forest Lodge and Glebe Group), and community events such as community safety barbeques. Rough notes were made as discreetly as possible during these meetings/events to ensure that meeting participants were not made aware of key observations and to ensure that they did not alter JCJ&SD 43 Online version via www.crimejusticejournal.com © 2015 4(4) their behaviour (Emerson et al. 1995;Maxfield and Babbie 2008). Handwritten notes were then retyped by the researcher some hours or days after the meeting. These notes taken during and directly after inter-agency meetings 'are subject to memory work and selectivity' (Cosgrove and Francis 2011: 213).
Semi-structured interviews were also conducted with 15 workers in the Glebe area throughout 2012 and 2013 to 'probe, typically with the use of follow-up questions' and to establish 'dialogue and exchange between the interviewer and interviewee' (Noaks and Wincup 2004: 79). These interviews were digitally recorded and transcribed by a professional transcription company. The semi-structured interviews ranged in duration from 31 to 99 minutes; the average interview lasted approximately 52 minutes. In total, 772 minutes of semi-structured interviews were recorded.
Broadly categorised, these interviewees were drawn from non-government organisations providing diverse social services (n=8), local government (n=3), law enforcement (n=3), and a voluntary organisation (n=1). Interviewees were recruited by direct contact, constituting a purposive sample (Maxfield and Babbie 2005: 238). Snowballing techniques were used to locate further organisations and personnel in the area (Maxfield and Babbie 2005: 241). A purposive sample was required because key participants were sought for their particular insights and knowledge of the Glebe area and crime prevention practices operating in the area. The codes (#01, #02) are used to refer to interviewees, and comments made by interviewees are used liberally, giving voice to their valuable insights and perspectives.
Three prospective interviewees from NSW government agencies with a presence in the area but with a considerably larger footprint than just Glebe were excluded because either they declined to be interviewed, failed to respond to correspondence inviting participation, or imposed excessive administrative requirements. The Sydney University Human Research Ethics Committee required organisational approval as well as individual consent for interviewees. One government organisation required completion of separate ethics procedures, which were excessive for a potential single interview and were not pursued. The non-participation of these three agencies is a limitation of the research. However, given the other data sources, information regarding the impact of these agencies on crime prevention in Glebe was still able to be generated.
While semi-structured interviews provide opportunities to explore issues in depth, there are limitations associated with this research method. Semmens (2011) suggests that difficulties can arise in building a trusting rapport with the interviewee and that there is a chance that the interviewee will not answer honestly, but instead give a response that is socially desirable or that will satisfy the interviewer in some way. Also, there is a danger of the interviewer misinterpreting the things people say.
Two focus groups were also conducted with diverse local stakeholders (some of whom were interviewees). The focus group discussions concentrated on recent crime trends and whether there was need to develop a new local crime prevention and community safety strategy (which was ultimately not recommended). The focus groups were also digitally recorded, professionally transcribed and analysed in a similar way to the interview transcripts.
Content analysis (also referred to as 'thematic qualitative analysis' by Cosgrove and Francis 2011) of the interview and focus groups transcripts and also the meeting notes was conducted manually. Given the relatively small number of interviews and focus groups, manual analysis was considered the most time-efficient method. All transcripts were closely read to 'support familiarisation with material' and 'a wide range of categories or themes were identified' (or coded) (Cosgrove and Francis 2011: 214). This process enabled the 'researcher to understand JCJ&SD 44 Online version via www.crimejusticejournal.com © 2015 4(4) the character of the data and to control for original assumptions' (Cosgrove and Francis 2011: 214).
Coding of the data 'entailed bringing a measure of organisation to the data and identifying conceptual categories' (Noaks and Wincup 2004: 130). This process was iterative. The transcripts were read on a number of occasions, allowing for reflection on the themes that emerged. The production of themes and sub-themes through this process is consistent with the whittling down of data, common in various forms of qualitative research (Cosgrove and Francis 2011). This process allowed for ordering and re-ordering of prominent emerging themes, which maintained closeness to the data.

Causes of (local) crime
The more than 30 inter-agency observed meetings and the 15 semi-structured interviews and two focus group discussions with local workers provided insights into their understanding of the causes of crime and subsequently the rationale for the approaches to preventing crime introduced later in this paper.
A small number of causes of crime -poverty, disadvantage, need, relative deprivation, limited life opportunities -consistently identified during the interviews and focus groups are explored here.
For a number of interviewees, poverty and disadvantage were considered to be significant contributing factors of crime in the area: [T]here's a correlation between poverty and crime. That sense of hopelessness and desperation. It starts to spiral I guess. (Focus Group Participant) [W]e knew that much of the offending behaviour was motivated by hunger, which again is probably an offending factor which isn't often considered, I think, by the mainstream. (Interviewee #2) A lot of them were quite desperate and poverty and their basic needs for food, clothing and shelter weren't necessarily being met. I'm not saying all crimes are committed to fulfil that need but certainly there was a degree of that. (Focus Group Participant) [Y]ou've got lower socioeconomic people, quite often they'll do what they need to do to make some money and whether it's feed the kids or feed their own habits or whatever it might be. (Interviewee #7) These explanations chime in accord with various criminological theories that highlight the importance of need, poverty and socioeconomic conditions to offending. Vold et al. (2002) note that attempts to connect crime to poverty 'go far back to antiquity' (2002: 84) and Watts et al. (2008) suggest that the history of criminology is replete with attempts to link poverty, unemployment and crime (2008: 113). While the complexities of establishing causal links between material conditions and crime raise numerous empirical and epistemological difficulties, it is nonetheless noteworthy that the causation of crime is couched in these terms rather than purely individualistic explanations. These views of crime causationdirectly animate approaches to prevention, as will be illustrated.
The experiences of poverty are also potentially exacerbated by the close proximity of people who are advantaged to those who are disadvantaged (a feature of the Glebe area). It is the relative depravation that matters with respect to crime causation: JCJ&SD 45 Online version via www.crimejusticejournal.com © 2015 4(4) I haven't approached this from an academic background but I believe anecdotally that when you sharper relief [sic] between wealth gaps or larger relief there's often a higher crime rate … I suppose the have nots get to see a lot of the haves right in front of them and there's that sense of, that marginalisation sense factor can be greater. The fancy phones, the nice cars. (Interviewee #4) Wilkinson and Pickett (2009) have made similar claims. The greater the inequality within a society, they argue, the greater the social problems (including crime). There is also more than a passing similarity to Merton's (1938) strain theory, where crime is considered to be an adaptation to frustrated attempts to acquire cultural goals such as consumer items, a good home, a new car.
Other factors like employment opportunities and access to drugs were also identified as possible causes of crime: … when you get people who are addicted to drugs, and drugs are really expensive, that's when they will start breaking into cars, stealing laptops and mobile phones and leather jackets and whatever they can get their hands on.
(Interviewee #7) The greatest thing that will keep the crime rate going down is getting the kids into employment … Already those first two boys that were working down at [building site] have got everyone's ears up and excited about -they were watching one of the boys went and bought a motorbike the other day. He didn't take a loan, he bought a motorbike. He got a licence, and he's legally riding a motorbike around, and the kids are just going where did you steal it from? And he says no, I bought it. (Focus Group Participant) These themes of socio-economic and structural factors contributing to crime also underpin the Glebe Community Safety Plan developed for this suburb by the City of Sydney Council in 2009. The Plan: … aims to take a genuinely preventative approach to Glebe's distinct crime and safety issues by tackling the underlying causes of crime. The aim extends to building on the existing strengths of the Glebe community and emphasises the importance of investing in those people identified as being vulnerable and in need of support. Compelling evidence suggests that those who feel excluded from participation in community life are more likely to offend against that community. The plan aims to address this sense of exclusion with a view to reducing the risk of offending by focusing on broad social outcomes. (City of Sydney 2009: 2) These observations are consistent with strands of various criminological theories that highlight poverty, inequality, strain, drug use and unemployment as contributing to crime. The social characteristics of the suburb and the experiences of those on the 'lowest rung' (Peel 2003) are elevated above individual determinants of behaviour and explanations of crime. While there are numerous difficulties in establishing causality of these structural conditions and crime, and concerns about the lack of attention given to offences committed by those from the 'good end' of Glebe (including white-collar crime), these observations nonetheless are consistent with Brown's (2012) observations regarding the resilience of penal-welfarism and the perspectives of frontline workers.

Community and social approaches to crime prevention
Based on these understandings of crime, it is unsurprising that many of the local programs and services that have direct or indirect crime prevention objectives are akin to community (Hope 1995) and social crime prevention (Tonry and Farrington 1995), as reflected below: What are the causes of crime or what leads to crime -there are so many factors within that. The approach to preventing crime or crime prevention would be to look at each of those causes and that's supporting families and support children, young people having diversionary programs. If your view is that the causes are multi-pronged then the solution to it needs to address each of those factors. (Interviewee #8) Some relevant local programs contributing to crime prevention are briefly described here (see Clancey 2014 for a fuller description of these programs). These programs are community-based and socially-oriented and directly respond to the perspectives of the causes of crime mentioned previously although many operate with little direct or no expressed intention of preventing crime per se. In fact some workers would reject the labelling of their programs as 'crime prevention'. However, their potential for 'serendipitous' crime prevention (Challinger 1992), justifies their inclusion here.
First mention is of a Breakfast Club run by a local out-of-school-hours child-care program (Centipede) and the Australian Red Cross at the local primary school. This is to help the children (generally aged between five and 12 years and predominantly from disadvantaged backgrounds) prepare for the school day. By ensuring that they have sufficient food, there is greater likelihood that the children will be able to concentrate in school and actively engage with their school work. The provision of food is a notable component of numerous local programs.
Also school-based is the Schools as Community Centre (SaCC) (now known as Glebe Tree House) funded by the NSW Government and which has operated out of the local primary school for over a decade. This initiative 'is a universal prevention and early intervention initiative supporting families with children aged birth to eight years in communities facing marked challenges' (NSW Department of Education and Communities 2012: 2) and is delivered across numerous sites in NSW. SaCCs support families raising children in partnership with local human service agencies, the local community and the school. SaCCs provide a range of projects, including: playgroups for children and parents, parenting workshops, supporting transition to school, bringing other services into the school and child health screening (NSW Department of Education and Communities 2012). This initiative and the specific programs delivered by the Glebe Tree House reflect the tenets of developmental crime prevention (Farrington and Welsh 2007;Homel 2005).
The Glebe Pathways Project started operating from the Glebe Youth Service in October 2009 (replacing a similar previous program that operated between 2004 and 2008). The program operates five days per week from 10:00am to 1:00pm for up to 14 young people aged 13-16 years (as many as 90 per cent of whom are Indigenous young people, according to Hayes (2011)). The main target population is young people residing in or having strong connections to the Glebe area in years 8 and 9 of secondary school and who are having difficulties at school. This alternative education program aims to help young people at risk of being de-schooled and operates on a student-centred learning model. In so doing, this program addresses an enduring characteristic of juvenile crime -negative school experiences (see Agnew (2009)  Another program aimed at the young is the After Dark Program run by the Glebe Youth Service and funded by the City of Sydney Council. The program involves the provision of recreational activities from 6:30pm to 10:30pm on Friday nights and between 7:30pm and 11:30pm on Saturday nights. After Dark 'is a space for young people to enjoy a healthy meal and participate in a variety of sports and activities. This program is both a diversion from youth anti-social behaviour, but also a refuge from hardships'. The activities are 'designed to divert young people from drinking, crime and other unhealthy or antisocial behaviours' (Glebe Youth Service 2013: 14). Again the provision of a healthy meal is a central part of this program and directly responds to concerns raised by local workers about hunger and need being a factor in some local crime. This program evolved in response to specific local problems, but shares programmatic features with numerous other youth diversionary programs (including Midnight Basketball).
The Friend in Hand Program was developed by the local police in 2011 to positively engage with local young people. Police and local young people considered to be at risk of involvement in crime exercise together three mornings per week. Sporting facilities close to the area are utilised, including the Glebe-Leichhardt Police Citizens Youth Club (PCYC). After training, a healthy breakfast is served at the PCYC and the young participants are escorted to school or vocational education facilities. Again, this initiative evolved in response to specific local dynamics, but is reflective of police-youth engagement programs operating in many jurisdictions.
Other available programs whose impacts have not been considered here are local Alcoholics Anonymous, Narcotics Anonymous and Al-Anon meetings, a methadone clinic adjacent to the case study area, a residential rehabilitation program for alcohol and other drug dependence (Glebe House), a domestic violence refuge (Elsies) 4 and a residential program for ex-offenders (Rainbow Lodge) operating in the case study area (which, as has been stated, is just over two square kilometres). Also excluded are the programs and activities of the local PCYC, which strongly focus on preventing crime and working with young people at risk of offending; and the court support work provided by the chaplain attached to the local secondary school. This is not an exhaustive list of programs or approaches to crime prevention operating in the area. It does, however, illustrate a significant commitment to social-welfare approaches. Crime prevention is (one of) the key objectives of some of these programs and an indirect objective or possible outcome of others. A recap of these programs serves to illustrate the diverse array of actors and organisations potentially contributing to crime prevention activities, many of which provide multiple programs and services that have minimal or no specific agenda aimed at preventing crime.
The importance of local, small-scale interventions responding to particular local conditions has been highlighted by this case study. However, critics might question the efficacy of programs of this kind, suggesting that the evidence for social crime prevention is not especially positive (see Homel 2007); and they might also caution that these programs represent 'the criminalisation of social policy' (discussed later in some detail). Nevertheless, three issues which arose during consideration of these programs and approaches to crime prevention are dealt with briefly here. Cherney and Sutton (2003: 345) noted that 'most crime prevention 'success stories' have arisen out of unique circumstances': these frequently relate to local conditions, needs and resources. The programs operating in the Glebe area reflect the importance of local conditions in shaping local crime prevention responses. Despite the developments in evidence-based crime prevention (Sherman et al. 2002(Sherman et al. , 2006, considerable local innovation in addressing local issues continues. As has been shown, some programs reflect or embody tenets from the emerging evidence-based crime prevention literature but also reflect local innovation (as recommended by Cherney and Sutton 2007 This focus on local conditions does not diminish the influences of wider socio-political factors on crime causation and prevention. Local conditions and institutions will have limited capacity to correct devastating losses of local industry or global trends in the movement of labour. However, these regional and global forces equally do not erase local characteristics that should be understood in responding to local crime issues, which might well relate to small geographical areas (such as an access route through a public housing estate) or a small number of known offenders (which is consistent with the tenets of environmental criminology -see Groff et al. (2010) for a discussion of the importance of micro level analysis).
Listing and describing these local programs suffers from the absence of analysis of their effectiveness, which is not well understood in the case study area. Few programs have been rigorously evaluated, which is a persistent problem for crime prevention in Australia (English at el. 2002;Morgan and Homel 2013). The complexity of evaluating social prevention initiatives has, according to Hope, contributed to an 'anti-social bias' (Hope 2005) in the growing crime prevention evidence-base. Programs that have been evaluated have tended to be analysed in isolation of the other local initiatives. Narrow evaluations of this kind fail to recognise the connections between local workers that result in referrals between agencies, the co-delivery of programs, the promotion of local programs across the area and the efforts to coordinate local programs and services. These intricate networks, referral pathways, pooling of resources and collaborative efforts have great potential to contribute to the 'collective impact' (Kania and Kramer 2011) of these programs and activities. As more funding regimes provide for narrowly defined service-delivery packages and outcome measures, these wider structures and networks will be further undermined.
Despite these challenges, there is some evidence that programs of this kind can positively contribute to the prevention and reduction of crime. As Currie has noted: … it is highly probable that the growth of community-based prevention has contributed to lower crime rates. For although prevention has taken a back seat to incarceration in recent years, it is also true that, mostly in quiet and unheralded ways, a variety of prevention programs have taken root around the country. (Currie 1998: 189) Therefore, there is considerable merit in better understanding the individual and collective impact of localised forms of crime prevention in Australia and any contributions to falls in crime in recent years. One the one hand, this will only be achieved through the development of evaluation approaches that recognise the interaction and intersection of service delivery systems. On the other hand, some commentators have raised concerns about monitoring programs with this objective labelling such an approach the 'criminalisation of social policy' (Blagg et al. 1988;Crawford 1998;Evans 2011;Gilling and Barton 1997;Knepper 2007;Rodger 2008;Wincup 2013). According to Knepper, the 'criminalisation of social policy' refers to: … [the] situation in which social welfare issues become redefined as crime problems. When goals of providing affordable homes, improving health, and providing incomes through employment become secondary to crime reduction in social policy, criminalisation of social policy has occurred. (Knepper 2007: 139) I argue that these concerns have limited traction in the context of the case study site. The programs described are adjuncts to the main programs and activities of the organisations delivering these programs. For example, the Glebe Youth Service provides a host of programs in addition to the After Dark and Pathways programs. Afternoon drop-in, homework assistance, individual casework support, cooking programs, sporting activities, overnight camps during the school holidays, and an array of other programs operate to support and assist local young people (and their families). Moreover, programs such as After Dark and the Glebe Pathways Project have multiple objectives, the prevention of crime being only one of them.
Arguments in favour of the 'criminalisation of social policy' also flounder in the context of current key local narratives. Crime was rarely the focus of local inter-agency meetings. Rather, the state and quality of public housing properties dominated many local forums and discussions (amongst other pressing issues). Given the significant proportion of local residents living in public or social housing (nearly one in five), the growth in waiting lists and times (exceeding 10 years for some property types in the local area), the unsuitability of old housing stock in the face of demographic changes (including the rise in the proportions of single occupants and elderly tenants), the significant maintenance requirements of public housing properties (NSW Auditor-General 2013), and Sydney's 'housing affordability crisis' (Begley 2014), it is perhaps unsurprising that the shortcomings of the public housing issue dominated many local meetings and events. In this context, crime was a peripheral concern of many local meetings and discussions, which suggests that any claims about the 'criminalisation of social policy' have limited utility.
While the social and community-based approaches were most prominent, other more punitive approaches were also evident. These tended to directly focus on surveillance as the means of controlling crime. Some examples are considered here.

Techniques of control and surveillance
While it is argued that there is a strong social-welfare ethos permeating the study area and many approaches to the prevention and control of crime, there are also many examples that overtly seek to enhance surveillance. In part, this reflects views expressed by local workers engaged directly or indirectly in localised crime prevention initiatives about the potential impact of a small number of individuals on local crime statistics: In a small area like this the stats can get skewed by one individual. (Interviewee  #4) [W]e know now that there were actually two or three people committing probably 75 to 80 per cent of the crime. Well, it was seriously … the amount of people … if you count those smashed windows, one person seriously was doing 110 a night. (Focus Group Participant) Some of the strategies designed to monitor and surveil include the policing of bail conditions, the establishment of alcohol-free zones, and the actions of private security guards employed at the local shopping centre.
The policing of bail has increased dramatically in NSW in recent years (Booth and Townsley 2009;Brown 2013;NSW Law Reform Commission 2012 Bail compliance checks involve police visiting a home of a person on bail and checking that they are complying with the various bail conditions. If there is non-compliance, then bail is breached. Figure 2 shows There was considerable commentary about the policing of bail throughout the fieldwork including some of the following: … we had the police enforcing very stringent bail conditions on the young people in question. They were knocking on doors at all hours of the evening. Some people saw that as harassment; some people said that's our modus operandi for this particular type of offender … So enforcement of that bail condition obviously had the intention of restricting the movements of that young person to their domestic setting and also to where they were able to go at other times. (Interviewee #2) [W]hile at times I think the policing of those curfews was bordering on abusive in that some houses were getting visited seven, eight times a night by patrol cars saying, 'I need to see  1995 1996 1997 1998 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 Breach bail conditions JCJ&SD 51 Online version via www.crimejusticejournal.com © 2015 4(4) It is not just state agencies that initiate contact with local residents. Security providers and systems also operate in the area. The local shopping centre is an example of 'mass private property' (Shearing and Stenning 1983) that has an active private security force, closed circuit television (CCTV) cameras in operation, and numerous forms of security designed into the products and shop design. Banning notices are issued to offending individuals by the shopping centre security guards to prevent their return to the shopping centre following various misdemeanours and joint police-security operations combat retail theft.
So while there is generally a strong commitment to social-welfare approaches to crime prevention, other more intrusive regimes co-exist in the area. The policing of bail, the use of private security to patrol the local shopping centre, the use of alcohol-free zones, and the operation of CCTV, are just some of the practices, hardware and policies which routinely operate in the area. Various other situational and environmental design crime prevention techniques have also been adopted (see Clancey (2015) for a summary).

Discussion
Local workers engaged directly or indirectly in localised crime prevention initiatives in the case study area very much understand crime causation in light of structural disadvantage and reduced life opportunities. These views, shaped by their disciplinary backgrounds and their often lengthy involvement with the area, inform their beliefs about how crime can be prevented. Breakfast clubs, parenting programs, structured recreational activities, alternative education classes, mentoring and employment pathways are provided in Glebe. There is strong lobbying, often by residents from the 'wealthy end' of Glebe, of NSW government agencies to improve housing conditions for public, social and Aboriginal housing tenants. This advocacy and commitment to community development, socially just and socially progressive services reflects the long-standing activism found in the area (Solling 2007), as well as the commitment of local workers to these ideals.
These views of local workers regarding crime causation are similar to those of criminal justice personnel in Australia and also in England and Wales. In talking with frontline workers in Australian criminal justice agencies, Brown observed (2012) that 'there are signs within criminology that life is being breathed back into social democratic and penal welfare concerns, habitus, and practices' (2012: 78). Coleman and McCahill (2011: 87) made a similar observation in regards to England and Wales, when they stated that 'empirical research has shown that "front-line" practitioners in criminal justice agencies and security networks continue to be guided by the "old" concerns of "welfare" and "reform"'. Similarly, Hallsworth (2002: 212-213) suggests that '[p]eople like community safety officers, workers in organisations such as Youth Offending Teams and Drug Action Teams as well as outreach workers, youth workers and other support workers do not see themselves as agents of an oppressive state'.
It is therefore unsurprising that, in this context, much crime prevention activity is delivered through social-welfare programs rather than through punishment or patrolling. This community development orientation and socially progressive programs and services, some directly seeking to impact on crime, are also consistent with the original forces that influenced the emergence of local crime prevention in NSW (and across Australia more broadly). During the late 1980s and early 1990s, numerous inquiries, studies, reports and conferences were conducted that supported and called for the development of localised crime prevention models (Hogg 1991;Juvenile Justice Advisory Council 1993;Youth Justice Coalition 1990). When local crime prevention arrangements were enshrined in NSW legislation in 1997, community participation, consultation and social and community crime prevention measures were very much at the forefront (Shaw 1997 crime prevention have partially waned through growing centralisation and focus on opportunity-reduction, some limited research suggests that there has been resistance to these trends. The observations of Sutton and Wilson (2004) that local government crime prevention staff are committed to community-based crime prevention continues to resonate, as does the observation by Morgan et al. that 'the emphasis on a community-based approach has influenced the range of crime prevention strategies implemented in Australia over the past two decades ' (2011: 20).
This is not to suggest that there are no signs of more punitive, controlling and surveillant forms of prevention. As has been shown, there is clear evidence of heavy policing of bail, operation of alcohol free zones in the areas with the highest concentration of public and social housing, and bans imposed by private security personnel attached to the local shopping centre. However, it would be difficult to argue that all crime prevention is necessarily animated by intentions to control, surveil and monitor, as has been suggested by some. For example, Borch has argued that: … in the name of prevention, ever-new social and material technologies are invented to regulate the life of ordinary citizens ... it focuses on virtually all dimensions of life: our health, the way we live, our identities, how we play, the way we move, our relations to neighbours, etc. (Borch 2005: 91) Establishing an alternative education program for young people experiencing difficulties with mainstream schooling, hosting activities for young people on Friday and Saturday nights and providing healthy meals for participants, and conducting an exercise program are not equivalent to the 'principally all-encompassing and boundless power' that Borch (2005: 102) suggests crime prevention represents. Consequently, the findings from this case study serve to challenge dystopian depictions of crime prevention that have come to dominate some commentaries.

Conclusion
As demonstrated by this case study, there is and continues to be a strong social-welfare orientation to local crime prevention in Australia. This orientation is borne out of the beliefs of local workers engaged directly or indirectly in localised crime prevention initiatives that crime is caused or exacerbated by poverty, need, disadvantage, drug use, limited employment opportunities and relative deprivation. These findings echo Brown's (2012) observations about the resilience of penal-welfarism in Australia due to the views held by frontline criminal justice workers regarding the causes of crime, and also echo the observations of  regarding the primacy of community-development approaches to crime prevention in Australia.
However, this is not to suggest that these predominantly small-scale local programs necessarily achieve the prevention of crime or avoid the unintended negative consequences that McCord (2003) importantly demonstrated arose in other contexts. The absence of evaluation makes it difficult to ascertain either the effectiveness or the unintended consequences of these programs. Greater focus on evaluating programs and the adoption of evaluation methods sensitive to both the positive and negative unintended consequences of local crime prevention programs is required. Given the potential contribution of social crime prevention initiatives to the 'crime decline' (Currie 1998;Skogan 2006) and the volume of such programs operating across Australia, there is considerable merit in better evaluating these local crime prevention programs. In an era of 'evidence-based policy', programs that are not evaluated, nor captured in meta-analyses, stand diminishing chances of survival.
More sophisticated understanding of the contribution of local social-welfare approaches to the prevention of crime is also vital given the financial pressures on and growing requirements of