Communicating security? Policing urban spaces and control signals

1. Visible criminal activities do not inevitably adversely affect perceptions of security

First, let us consider a dominant assumption of the SCP about the impact of visible criminal activities on urban spaces and insecurity. Urban populations construct perceptual maps of (dis)order that capture the times, places, and kinds of people and activities deemed to be threatening and which shape our perceptions of safety, behaviour and sense of ease in places (Merry, 1981). The SCP suggests that our perceptual maps are likely to be shaped by what we can see happening on the streets. A dominant assumption is that expressions of criminal incidents will mostly be read as producing adverse effects upon perceptions of safety because of their negative communicative properties. ‘Weak signals’ (incidents which do not have an impact on perceptions of safety by themselves), when persistent, have a cumulative effect amplifying in strength (Innes, 2004a: 346). Whilst ‘drugs’ were one of the two top signal crimes across the NRPP sites (the other being ‘young people hanging around’), my research suggests that routine familiarity with street dealers may serve to reduce their ‘signal value’ in particular neighbourhood contexts where they are observed as part of an area’s ‘normal appearances’ (Goffman, 1972). In this study, residents who had seen dealers, over the years, going about their business usually in quiet and efficient ways, points perversely to the important dimension of ‘normalisation’ in residents’ constructions of local social order and safety. Their visible existence did not necessarily equate with a sense of insecurity because residents’ trusted that dealers would not adversely affect their personal safety and had an expectation that derived from this:

Similarly, research by Bottoms (2006: 253) found that some residents ‘while agreeing that drug dealers did indeed frequent the area … they weren’t really a problem if you weren’t involved in the drugs trade’. In my research, it was largely the presence of drug users that were signals of unsafety. Drug users were considered to be ‘volatile’ because they were ‘always looking for money to pay for their next hit’, in contrast, dealers wanted to ‘keep to themselves and not attract attention’. A tension naturally remains given the relationship between dealers and users. Thus even though these criminal activities persisted on the ‘front-stage’, they were perceived as background ‘noise’. This finding connects with Hope and Foster’s (1992) work which illustrates how some kinds of criminal activity are part of the local social order. These insights inform us that crime can be orderly and challenge theories that assume criminal incidents will inevitably have the effect of reducing social activity in urban areas or that individuals will perceive themselves at greater risk. Where criminal activities are not automatically signals of insecurity, such as drug dealers, these ‘others’ can contribute to levels of natural surveillance, which was found to be a significant factor in perceptions of ‘safe’ spaces. Indeed, some types of crime serve positive surveillance functions (Bellair, 2000), some expressions of disorder are associated with ‘pride’ (e.g. artistic graffiti) and some criminal incidents (e.g. vigilantism) may reassure rather than rouse fear. My research identified a range of meanings, aside from risk and insecurity, attached by residents to visible expressions of crime and disorder. This complexity is not easily characterised by the concept of ‘noise’, which accounts for any incident that does not function as a signal crime. The concept of ‘noise’ and its implications for the reassurance function of policing are important, albeit less well-developed by the SCP.

Another example that emerged from the fieldwork was prostitution. Paradoxically, women street workers were seen by some residents as adding to the ‘buzz’ of informal activity that was important to their sense of safety:

I feel safe because there are a lot of people around and people come in cars, do their business, and take their lady away. I walk down there every single day and I feel safe, and maybe that is because of some of the activity going on. (Asian male, aged over 35, N1)

It would be a considerable leap to suggest that streets should be filled with drug dealers and prostitutes for the sake of additional eyes on the streets or more people around. However, this reminds us that there is nothing simple about order itself or the ‘bewildering number of components’ that go into and unite in their joint effect upon the safety of the sidewalk (paraphrase Jacobs, 1961: 111). The social life of the street includes the presence of strangers doing nothing or doing something, which may even amount to doing drugs and prostitution. The sense of security engendered by routine interactions with socially unlike ‘others’ suggests that perceptions of safety in some contexts may not necessarily depend on close bonds of togetherness or homogeneity, as implied by some social capital theorists (Putnam, 2000).

In explaining these findings, the SCP introduces the notion of ‘dissonance’ (Innes, 2004a: 353) as one aspect where neighbourhood context matters in regard to whether an incident functions as ‘noise’ or as a ‘signal’. This concept raises the fundamental issue of people’s levels of tolerance and social norms in different neighbourhood contexts. There remains an unresolved tension within the SCP between the implications of ‘dissonance’ and that of ‘amplification’. Whilst the former implies that the regularity of incidents in an area may normalise people to further acts (as was perhaps the case in the above drug-dealing example), the latter implies that the regularity or persistence of incidents may transform weak signals into strong ones. Further research might fruitfully explore these context effects in more depth and, in particular, at what point the regularity of specific criminal incidents mean they are interpreted with more or less significance for people’s sense of insecurity, which brings us to the second dominant assumption.

2. Perceptions of safety can coexist with high levels of crime in particular urban spaces

A study of two neighbourhoods in Chicago conducted nearly 30 years ago is an important reminder that high-crime areas can have high levels of ‘satisfaction with safety’ (Taub et al., 1984). As outlined above, a dominant assumption of the SCP is that exposure to a series of spatially or temporally proximate criminal incidents may be interpreted as a strong signal through their ‘amplification effect’ (Innes, 2004a: 346). An interesting finding from the multilayered mapping work is that perceptions of safety can coexist with high levels of crime in particular types of urban spaces. ¹ Main roads and high streets that experienced high levels of crime were perceived as some of the safest types of spaces, even after dark where a general ‘unsafe’ pattern was found across the study areas. One explanation is that not all incidents of crime in these spaces would appear to be functioning as genuine ‘signals’ – a key reason (other than the presence of effective control signals) for why high crime may not equate with high fear. Yet by emphasising the coexistence of higher crime with perceived safety, the research also informs us that some aspects of the social and physical environment act as resilience factors. Focus group discussions revealed that these were not necessarily additional police control signals (as suggested by Taub et al., 1984) but greater levels of natural surveillance, through increased pedestrian and traffic flow. This reinforces the crucial role played by ordinary citizens going about their everyday routines in structuring the contours of security. These findings are consistent with research in two Australian cities where busy thoroughfares were perceived as distinctly safe despite being hotspots for personal crime (Doran and Burgess, 2012). It may be that the reassurance value of police patrols reduces where there are already high levels of ‘organic’ control signals acting as capable guardians (Crawford et al., 2005: 58). These findings are interesting given that much debate has focused on places that are ‘overly fearful’ and experience low crime and prompt new avenues of enquiry into the ways formal and informal cues – to both safety and danger – interact in particular spaces and times of day. There may be much to be learned for the reassurance function of policing by the study of urban spaces that experience high crime yet engender low fear.

3. Control signals can be ambivalently interpreted

A dominant assumption of the SCP is that uniform police foot patrols targeted toward local concerns perform a reassurance function (Innes, 2007). While alternative readings and contrary effects of formal control signals are acknowledged and were evident in the NRPP, my research suggests that contested and paradoxical interpretations that beset ‘control signals’ have been potentially underestimated. In practice, it is very difficult to convey the message that citizens should feel reassured by formal control signals simply by virtue of their visible presence. This is because the processes of signification, through which control signals communicate security, are fraught with ambivalence. The messages sent by police patrols were not read as self-evident. While some residents derived reassurance from police presence, they also had the capacity to act as reminders of potential threats. A discussion about policing in a local park illustrates this ambivalence:

Communicative policing strategies may be interpreted by onlookers as double-edged; reminding people simultaneously of their protection by the presence of formal figures of authority and the need for their existence. They impart messages about the crime risks and the fragile nature of social order that demand visible police personnel. Unlike the examples noted above of associating a tap marked red with hot water or smoke with fire, the sight of a police officer was frequently indicative of both crime and of protection; conveying ‘worry’ and ‘relief’. Similar unintended effects were noted by the NRPP where half of respondents to a small survey associated high-visibility uniforms with signs of ‘danger’ and ‘emergency situations’ (Innes, 2007: 136).

Hence, the absence of visible patrols was, consequentially, interpreted by some residents as a positive signal about the wellbeing and safety of the neighbourhood as the presence of police often signalled that ‘something bad is going on’. These unintended effects are noted by studies in other European cities (Winkel, 1986). Therefore, a reduction in visible patrols, rather than being interpreted as a ‘control deficit’, could send the opposite signal:

You very rarely see police patrolling the streets. If you don’t see them, you know there isn’t much crime going on. They used to always be about but now crime has gone right down. (White male, aged over 35, N1)

Again, we are reminded that a healthy social order does not require the routine presence of police personnel.

Jacobs (1961) astutely noted that order is held by ‘a web of public respect and trust’ where formal authority is best restricted to where order has been seriously breached or irretrievably broken down. Hence, changes to the presence of formal control require an explanation precisely because it sends signals about the levels of informal control in an area. Therefore, the overarching trend to formalise the delivery of social control, including the provision of supplementary policing personnel on the streets, can be problematic for reassurance endeavours. For these reasons, understanding the contradictory effects and limitations of police control signals are of critical importance for policy-makers and service providers.

Similar ambivalent interpretations emerged in relation to environmental control signals installed by both public authorities and private individuals. A common critique relates to their ‘troublesome trade-offs’ (Ekblom, 2005: 215), notably between their instrumental value and their capacity to remind people of the risk of crime. In this research, the aesthetics of environmental controls influenced how they were interpreted and whether these conveyed the image of a place as ‘safe’ or, adversely, were read as ‘anxiety markers’ (Zedner, 2003: 165). Residents felt that highly obtrusive ‘prison-like’ security made streets appear dangerous. A resident commented on an alley-gating structure described by many as intimidating:

It makes the area look unsafe. When people see them, they think there is more crime. (Black female, aged over 35, N2) ²

These findings have relevance for research exploring the semiotic properties of architecture and their implications for reassurance (Millie, 2012).

The installation of security grilles on the doors and windows of houses by private individuals were widely seen to have a detrimental effect on communal feelings of safety, contributing to the communication of a fragile and insecure order. This supports Loader’s (1997) contention that private security may be an oxymoron, in that security is a ‘public’ matter. Likewise, Davis’s research in the USA points to an untenable situation in which ‘the social perception of threat becomes a function of the security mobilisation itself, not crime rates’ (1998: 224). An implication of these observations is that the ‘reassurance gap’ may be widened by an overt focus on security. Where control signals undermine a collective sense of wellbeing and social order, they have the potential to weaken local social relations.

4. The reassurance function of policing is highly contested; visible patrols and police-initiated contacts were seen by some groups as unwarranted

Problematically, the SCP implies that police actions calibrated towards local concerns serve a public reassurance function. This assumption reflects the idea that the pursuit of security is a universal good, in need of little or no qualification, which requires limited attention to the ethical issues and dynamics of security strategies (Zedner, 2003). One such issue relates to the manner in which police strategies may privilege the security concerns of some groups over others and treat people differently. A widespread concern, consistently reflected in the CSEW findings and one of the top disorder signals identified in the NRPP, is of ‘young people hanging around’. A key question arising from the NRPP is whose concerns are accorded prominence in normative discussions about how the police should define social order. The citizen-focused discourse of reassurance policing evades the issue that some people – often defined as ‘police property’ – are the objects of police control signals and are more likely to find themselves in an adversarial relationship with the police, notably young people. It is often a small and unrepresentative proportion of the public who are involved in consultation processes aimed at influencing local policing activity.

Young people who took part in the focus groups identified wide-ranging concerns about visible patrols and the procedural fairness of police-initiated contacts. Many young people expressed that they did not find the presence of police reassuring. Rather, their visibility on the streets was described as having ‘no effect’, that it ‘does not make a difference to the neighbourhood’ and, regularly, that they ‘don’t like the police’ or that there were ‘too many police’. It was not uncommon for young people to refer to the police as ‘stuck up’, ‘arrogant’ and ‘superior’, judgements which appeared to be shaped by perceptions of their encounters with the police, as the following extracts illustrate:

They [police] talk to you like they are better than you just because they’ve got a uniform on their back. (White female, aged under 19, N1)

They [police] say stuff to us to make us scared of them. (Asian male, aged under 19, N1)

Generally, the above comments were informed by their concerns about fair treatment and respect by which the police exercised their authority to stop, search or move them on in the course of their patrols. A routine concern was the manner in which young people felt unfairly stopped and questioned while in public spaces, as the following extracts illustrate:

We hang around ginnels ³ and they [police] always get out of the car and say, ‘What are you lot up to?’ (Black male, aged under 19, N2)

Police are called all the time. Say you are walking down the street with some of your mates, you’ll see a man looking out the window on the phone and the next minute you’ll see the police down the street. They’ll come that quickly, even if we’re not doing anything. (Asian male, aged under 19, N1)

According to Tyler, personal encounters with the police are ‘teachable moments’ (2013: 11) in which the public learn about the legitimacy of the law and authorities. While some young people interpreted the above contacts as the police ‘just doing their job’ and seemed to accept this strategy as a means of community protection, many others expressed that the police ‘stop you for no reason’ or ‘search you to make it look like they’ve done better work’. These young people evidently did not feel that the police treated them fairly or could be trustworthy. The perceived absence of procedural fairness in some young people’s encounters with the police shaped the reassurance function of policing as a highly contested concept in practice. Moreover, these interpretations refute a dominant assumption of the SCP that there is some form of consensus in the support for a particular policing style and definition of policing priorities. An important implication of research on procedural justice for the SCP is that police can build or undermine their relationship with the public by the manner in which they exercise their authority when interacting with people on the streets (Tyler, 2013).

5. The value of PCSOs as key control signals may be exaggerated

The PCSO was designed to act as a pivotal control signal central to the delivery of reassurance policing. A dominant assumption is that ‘PCSOs … will be able to deal with minor crime and disorder problems as a visible sign of authority by virtue of their presence’ (ACPO, 2008: 4–5). Going back to the example of the passenger on the bus in the NRPP, police presence was assumed to reassure by ‘subtly persuading the children to behave in a more orderly fashion’ in part because of their ‘potential for the enactment of legitimate force’ (Innes et al., 2004: 213). Implicit here is some recognition of what Bittner identified as the ‘special competence’ of the police, namely their capacity for ‘decisive action’ (1974: 35). This raises an important debate in relation to the introduction of PCSOs and the extent to which they may only provide the illusion of security without having the powers to be able to restore order when it is threatened or breached. Research on procedural justice also informs us that compliance is not constructed through coercion or force (i.e. police powers) but by legitimate authorities (Tyler, 2013). In the past, this may have been a bus conductor who is without coercive powers!

The role of the PCSO recognises that order is largely maintained through ‘soft’ policing (Innes, 2005) without recourse to the use of force wherever possible (Bittner, 1974). The potential of PCSOs for creative powers of negotiation and persuasion of others, rather than ‘hard’ powers of coercion and enforcement, have been thus identified as a critical institutional strength for policing local communities. It is these ‘soft’ powers that are thought to make PCSOs valuable members of the reassurance policing family (Innes, 2005). Moreover, despite their limited powers on the streets, they have significant institutional resources to call upon.

However, it was evident in this study that the manner in which PCSOs were perceived to deal with minor disorder that arose in the course of their patrols had implications for their capacity to signal reassurance to onlookers:

A gang hangs around outside that shop, they are underage, they are smoking and drinking, and the community police don’t say anything to them. (White female, aged under 19, N1)

This supports wider lessons from research on policing in the USA (Skogan and Hartnett, 1997) that highlights the importance of what officers are seen to be doing when they are undertaking patrols. Moreover, the perception that PCSOs need to wait for back up to make an arrest rather than take decisive action fostered a limited confidence in PCSOs’ ability to increase orderliness in public spaces:

Police make people feel more secure because they have the rights to arrest somebody and take action there and then. PCSOs have to get somebody to come on the scene. (Asian male, aged over 35, N1)

These experiences reinforced the view of some residents that PCSOs lacked the capacity to secure compliance from those who were visibly engaging in low level crime and anti-social behaviour. It also raises concerns over whether PCSOs produce legitimacy deficits, as residents’ noted:

Police make you feel safer than PCSOs because police sort of do do stuff about things. (White female, aged under 19, N1)

These constraints affect their reassurance value with many residents calling for a return to sworn officer patrols.

6. Control signals can be a risk factor in neighbourhood decline

Urban studies have long been concerned with the consequences of crime and disorder on the trajectories of neighbourhoods. An important finding from a systematic body of work in the USA is that subjective perceptions of disorder have objective consequences: ‘social perceptions reinforce later disorder and potentially poverty absent an exogenous intervention’ (Sampson, 2009: 18). This argument is that social perceptions of a place as dangerous or disorderly affect the social structure through influencing the behaviour of corporate investors, businesses and residents. Bottoms (2009) recently pointed to the possibility that effective perceptual interventions, performed by appropriate control signals, might be one of the ‘exogenous interventions’. Whilst it is acknowledged that control signals can convey negative messages, the dominant assumption is that where appropriately tailored they act as a ‘recovery factor’ working against the forces of neighbourhood decline (Innes and Jones, 2006). As such, SCP proponents affirm that those who dismiss reassurance policing fail to consider the importance of managing perceptions in shaping the fortunes of neighbourhoods.

Whilst this research takes seriously the role of social perceptions and their centrality to the (re)production of a sense of place, the SCP leaves uncertainty about the range of conditions under which control signals aid processes of recovery or, conversely, convey negative messages that could increase the risk of an area declining. The findings of this research indicate two main ways that control signals have implications for fostering rather than tempering perceptions of a place as disorderly. First, environmental control signals could be perceived as physical manifestations of disorder transmitting negative symbolic messages to residents that could engender neighbourhood stigma:

Everybody thinks robbers come from here; they must think this is a bad area, that’s why bars and gates are needed. (Asian male, aged over 35, N1)

It was also not uncommon for residents to comment that ‘an abundance of policing’ may communicate and reinforce the message that ‘it is a bad place’. Other research supports this notion in relation to the provision of supplementary policing:

The purchase of additional policing cover may suggest a place to be disorderly both to its inhabitants and outsiders. But it can serve to remind people that there is a perceived problem to which policing is the preferred solution. (Crawford et al., 2003: 47)

Second, the tone and agenda in which controls were introduced by authorities and the ‘problems’ they sought to address, had implications for their communicative properties and how the study areas could be labelled. One of the dangers, evident in this study in relation to alley-gating, for example, is that places may become stigmatised as dangerous if interventions are resourced under a crime reduction agenda (as opposed to a community development agenda), since the official message that could be communicated to residents is that these places are vulnerable to high levels of crime. Alley-gated streets were interpreted as ‘bad streets’, despite the range of improvements they were perceived to foster, clearly outlined in a residents’ discussion:

These discussions suggest that control signals have the potential to contribute to processes of negative ‘neighbourhood coding’ (Hope, 1995: 56). According to Hope (2001), ‘stigmatic images’ have implications for long-term neighbourhood decline by reducing an area’s ‘exchange value’ defined as the relative attractiveness to newcomers and businesses. Exchange values are important because they affect the position a neighbourhood has within the wider urban framework. This casts control signals as a ‘risk factor’ in neighbourhood decline.

Some residents suggested that authorities should avoid generating stigma by embedding security seamlessly into the urban landscape. For example, a resident cited the use of unobtrusive ‘extra protective’ front doors that, unlike security grilles that do not fade into the background, may continue to influence the possessor of them to feel safe without compromising wider community perceptions. Such views align with Flusty’s (2001) ‘quaintification’ arguments, which propose that the growth in security hardware that has become a mainstay of modern streetscapes may be rendered publicly acceptable by normalising and embedding security, such that it blends into the background, does not offend the aesthetics of place or does not suggest the place is insecure. The ‘quaint policing’ of urban spaces stands in stark contrast to the policy of high-visibility policing in the name of reassurance.

7. The SCP fails to engage with the real factors that inform insecurity

The above discussions leave us questioning to what extent public authorities can and do engage with the real factors that inform insecurity and safety through communicative policing strategies. The dominant assumption that a visible police presence will perform a perceptual intervention and increase feelings of safety was evidently not upheld in any direct way. This seriously questions any sense that visibility is a baseline by which reassurance can be meaningfully conveyed. Moreover, the limited capacity of control signals to improve perceived security are perhaps also due to the finding of this research that factors aside from crime and disorder were prominent sources of insecurity, while the concept of ‘control signals’ refers specifically to the targeting incidents of crime and disorder. Public consultations undertaken by community safety professionals emphasised that insecurity in these areas related to structural social and economic changes whereby efforts to police urban spaces may be only interpreted as a ‘buffer’:

We got some feedback from residents and the comments that were made were things like, ‘Well how do you expect us to feel safe when you close our post office’ and ‘We don’t have a bank, we don’t have a cash point’. They still feel very much on the edges and marginalised … The interventions that we are doing were good, all good things, but they did not really have such a huge impact on people’s perceptions of what it was like living in that area … So actually, it’s not crime and disorder, it is the bank and the post office in some respects … But insecurity is broader than fear of crime isn’t it? (Community safety professional, N2)

The decline of neighbourhood shops and services was felt by residents to have negative implications for casual public contact, which was an important informal source of reassurance in these socially diverse areas that translated to perceptions of relations with others on the streets:

We’ve lost seeing local people buying their goods and they become familiar. And then you might see them on the street and it just makes that difference, seeing people. You see them in the shop so therefore you know they’re not a threat, they’ve been buying their groceries at the same place you have. (White male, aged over 35, N1)

These examples highlight an important point that public reassurance, rather like community cohesion, needs to be viewed in structural and institutional terms linked to the political economy, not just in symbolic terms ‘as a set of attitudes we can implant or mobilise’ (Currie, 1988: 282–283). In this research, professionals indicated that too much focus had been on crime without addressing the wider social and economic issues that might more deeply shape a sense of place.