The Abstract Police: A conceptual exploration of unintended changes of police organisations

Abstract

As a consequence of the 2013 police reforms in the Netherlands and Scotland, the police in both countries have made a shift towards a fundamentally different kind of organisation: the Abstract Police. The increasing abstract character resulted in changes in the internal and external relations of the police. The police became more formalised and dependent on rigid systems and system information. Citizens and communities became more at a distance. Gradual and long-term processes may have similar consequences. For that reason it may be expected that the increasingly abstract character of the police may also be found elsewhere.

This paper puts forward the thesis that over the past years the police in many Western European countries have made a shift towards a fundamentally different kind of organisation, a change with far-reaching consequences, but which have remained unnoticed until now. To understand this process, we introduce a new concept, Abstract Police. With this concept we mean that, both internally and externally, the police have become more at a distance, more impersonal and formal, less direct, and more decontextualised. The abstract police are also less dependent on personal knowledge of officer(s), as this is increasingly being replaced by ‘system knowledge’, framed within the ‘logic’ and categorisations of computer data systems (Ericson and Haggerty, 1997).

This break with traditional ways of organising the police can especially be found in countries such as the Netherlands and Scotland, where police reforms in 2013 have resulted in considerable organisational scale enlargement and highly centralised national forces (Terpstra and Fyfe, 2014, 2015). The main arguments for these reforms were to improve the effectiveness of the police (especially with regard to transregional problems), partially also their efficiency, and to find solutions for problems of organisational fragmentation (Fyfe and Scott, 2013; Terpstra, 2013). As we try to show in this paper, the rise of the abstract police can partially be seen as a significant unintended outcome of these reforms. However, more gradual and long-term social developments have also contributed to this new kind of police organisation. Seen from that perspective, the 2013 police reforms in the Netherlands and Scotland have mainly strengthened and accelerated this process, making it more prominent than elsewhere. This implies that it may be assumed that increasingly abstract police forces can also be found in other Western European countries, perhaps as yet in more modest and rudimentary forms.

We expect that in the future the abstract character of the police will gain in importance and have far-reaching consequences for the relations between members of the police services, for police work, and for the relations with citizens and local communities. It may also have an impact on the dominating views about what is ‘good’ policing and about police professionalism and leadership. For that reason we believe that the abstract police is also an important concept for reflecting on future developments of the police and about important challenges with which the police may be confronted over the coming years.

The central aim of this paper is to explore the concept of abstract police and to show its relevance for police and police research. We developed this concept in several of our empirical studies on diverging aspects of police reforms in Scotland and the Netherlands. Several findings of empirical studies in both Scotland and the Netherlands (such as Axiom, 2015; Fyfe, 2018; Hail, 2016; Salet and Terpstra, 2017, 2018; SIPR, WWS and ScotCen 2017a, 2017b; Terpstra, 2017, 2018; Terpstra et al., 2015, 2016) are used here to illustrate the increasingly abstract character of the police services.

The outline of this paper is as follows. First, we sketch two incidents to illustrate the relevance of the concept. Then we briefly deal with a definition, followed by an analysis of the most important internal and external aspects and consequences of the abstract police. Next, we go into the underlying processes resulting in the increasing abstract character of the police.

Two incidents

Before we will deal with our detailed exploration of the concept of the abstract police based on the findings of several empirical studies, first we present two incidents that happened in 2015, the first in the Netherlands, the second in Scotland. These incidents are used to illustrate the relevance of the concept and to show how far-reaching the consequences of the increasing abstractness of the police can be. Both incidents attracted a lot of media attention. Because of their dramatic consequences, in each case there were special investigations into why the police service had failed to respond adequately and why things went so terribly wrong. Although there are important differences between the two cases, there are also significant similarities. Our assumption is that both incidents reflect central features of the increasingly abstract police services.

Woman killed near hospital

On 10 August 2015, a woman aged 28 years, named Linda van der Giessen, was shot dead in the parking lot of Waalwijk hospital (Netherlands). At the time, Linda was walking to her car, at the end of her shift as a nurse in this hospital. It soon appeared that she had been killed by her ex-partner. In the days following her death, it was proved that in the previous weeks, Linda had reported several times to the police that she felt seriously threatened by her ex-partner, that he was stalking her and that she heard that he had bought a gun. Because this affair raised many questions about the way that the police had dealt with this case, an independent committee was asked to investigate it.

In May 2016, this committee published its report (Commissie Eenhoorn, 2016). It showed that in the three weeks before Linda was killed, she had reported several times to the police – by phone, by email, to patrol officers on the street and by a visit to a local police station (where she talked about her problems with an officer for more than three hours) – that she felt very threatened and that she was afraid that her ex-partner might use extreme violence, for instance by using his recently bought gun. The committee found that in these weeks at least seven different police officers had been involved in this case.

However, the work processes of the police (including their information systems) were so complex and fragmented that for the officers involved it remained unclear who was responsible for which tasks. In addition, the committee found that the police had had a one-sided focus on the criminal law aspects of the case. The compiling of a complete administrative file, with the assumption that this might contribute to sufficient evidence against the ex-partner to build a case for court, seemed to have been more important than protecting the threatened woman. This responsibility had been neglected.

In addition, most police officers seemed to have underestimated the seriousness of the risks and dangers. Relevant information, such as the purchase of a gun by the ex-partner, had not been seen as reason to pay more attention to the case. According to the committee, it looked as if the police officers concerned had felt insufficient responsibility for the case and for the protection of the woman. Some of them had neglected their tasks, for instance a community officer who had been informed but had done nothing. He had not even warned the ex-partner to stop the stalking. The only exception had been a lower-ranked service officer, who not only talked extensively with the woman, but also tried to convince other officers of the seriousness of the case. However, this officer had not been able to find an adequate solution for the complexity, opaqueness and hierarchy of the system.

The committee concluded that the police should have paid more attention to the case and intervened with more urgency. According to the committee, the problems were caused by the highly ‘complex, bureaucratic and fragmented work processes’ of the police. Instead of looking for ‘meaningful interventions’, police officers seemed to be caught in ‘systemic communication’ and were mainly focused on following criminal law procedures. The committee noted that the perverse effects of the current organisation and processes seemed to be fully accepted by the police. Feelings of ‘unambiguous ownership’ had disappeared and ‘overall coordination’ was lacking. In general, the police showed too much confidence in the effectiveness of a ‘systemic approach’ and in IT communication (Commissie Eenhoorn, 2016: 20–21).

Car crash on M9

On Sunday 5 July 2015, a phone call was received by Police Scotland that a car had crashed off the M9 motorway and down an embankment not far from Bannockburn, near Stirling in central Scotland. Although the police logged the call, they failed to send a patrol to investigate. It took three days before police officers found the car near the road following another call from a member of the public who had seen the vehicle. One of the two persons in the car, 28-year-old John Yuill, had already died. The second person, Lamara Bell (25 years), was seriously injured but still alive and then died four days later in hospital.

This tragic incident attracted significant political and media attention in Scotland. Many wondered what circumstances had led the police to fail to follow up the initial report of the car crash and why it had taken another three days before the police responded. One of the independent investigations undertaken after this tragic incident, by HM Inspectorate of Constabulary in Scotland (HMICS, 2015), showed that there were significant weaknesses in the implementation of Police Scotland’s national call-handling system. The HMICS report concluded that the programme of rationalising control rooms following reform should stop until a thorough review of the new arrangements for a more centralised system involving fewer control rooms had taken place. According to HMICS, the problems were due to severe cuts in control room staff since 2013, with call handlers under pressure to resolve calls quickly. Meeting deadlines and increased productivity were said to have become more important than customer service. Members of the Scottish Parliament and representatives of the Scottish Police Federation said that the report showed that the creation of a centralised system of control handling, in combination with a target culture and budgetary cuts, had increased the risk of failing to respond adequately (HMICS, 2015).

The independent evaluation of the Scottish police reform also showed that it was not only lack of resources and the pace of reform that had caused these failures in policing. Both members of the public and local officers voiced concerns about the lack of local knowledge held by police control room staff. The public were concerned that staff from outside the local area and who did not know the locality now worked in control rooms, a concern shared by some officers ‘who felt that knowledge of the local areas was especially crucial to help direct them to remote locations, for example in relation to road accidents’ (SIPR, WWS and ScotCen, 2017a: 24, 2017b).

Conceptual notes

These two tragic incidents illustrate the shift towards a more abstract police service and some of the consequences that this may have. As we will explain in more detail below, a distinction can be made between two different dimensions of this concept. First, the concept of abstract police refers to changes within the police organisation. Both the relations between police officers, and those between officers and their chiefs, have become less personal, familiar and direct, and more formalised and governed by ‘systems’. Secondly, with the emergence of abstract police there have been comparable changes in the relations between the police and both citizens and (external) partner agencies: more at a distance, less personal and direct. It may be expected that the increasingly abstract nature of current policing may also have consequences for how the police operate and what they perceive as ‘good’ (‘professional’) policing.

It is important to emphasise that the shift towards the abstract police is not the same as the loss of local embeddedness. Even abstract police organisations can be oriented to local problems and partners, but increasingly this is being realised in more abstract or decontextualised ways, less dependent on local knowledge and personal and informal relations.

The abstract police should be understood as an ideal-typical concept (Weber, 1922). Ideal (or ‘pure’) types are abstractions that emphasise certain elements or perspectives on social reality. They should be helpful to describe and understand important elements, such as changes in contemporary police organisations.

The use of the ideal type of abstract police is not meant to suggest that the empirical reality of the current police completely corresponds with it. The empirical reality is more complex than that. Although the police may be seen as increasingly abstract, elements of former and more traditional organisational types and cultures can still be found, and may be in conflict with the abstract police. As a consequence, the issue is not if the concept of abstract police is (completely) ‘true’; what matters is if the concept is helpful to better understand increasingly dominant elements of contemporary police organisations.

In the following we will deal with both aspects of the abstract police. This concept should contribute to a better understanding of recent developments in police services and the interpretation of important findings of several empirical studies.

Internal relations

The first important aspect of the increasing abstractness of the police can be found in the changing internal relations within the police services. With the establishment of the Netherlands’ National Police, and for reasons of flexibility and capacity management, the traditional small local police teams were replaced by much larger teams, often with between 150 and 200 members. Many local police stations were closed, especially in rural areas. A study on the local police teams in the Netherlands showed (Terpstra et al., 2016; Terpstra, 2018) that this resulted in much larger working areas for the police, and that this often had a negative impact on the traditional close relations between local officers. According to this study, as a result, since the introduction of the National Police many officers have felt the loss of the confident and familiar atmosphere of the former smaller teams and local stations. Now many of them do not know all the members of their team, which would have been unimaginable in the past. The plans for the National Police made it possible to introduce smaller subgroups in the large local teams. The main reason was that prior to the reform it was realised that police officers usually preferred to work in small groups with permanent colleagues, giving them a feeling of togetherness and protection. However, it was emphasised that this should not be at the expense of the flexibility of the large teams. As a consequence the small subgroups are largely only a paper reality. They do not provide recognisable and familiar work surroundings for officers (Terpstra, 2018).

This study also found that the relations between police officers have also been changing as a consequence of both the formalisation and increasing horizontal and vertical fragmentation of the police (Terpstra et al., 2016; Terpstra, 2018). In the past, in the Netherlands community police officers often had the discretion to work on problems on their own and in their own way. If they needed assistance from colleagues, they usually arranged this informally, for instance by asking them when they met in the police canteen. Today they have to follow detailed procedures, often determined nationally rather than locally, and have to involve other specialists and higher level units. Now a community officer must make what is termed a neighbourhood assignment in a certain computer programme and format. This assignment is assessed by a coordinating officer, who will decide if it will be given to patrol officers. Feedback is not communicated directly to the community officer, but by using the same computer programme (Terpstra, 2018).

An important goal of the Scottish police reform was to create specialist units at the national level and to promote more equal access to them. From the local perspective the general perception is that this aim of more equal access to specialised expertise and resources has largely been realised. Several studies (Fyfe, 2018; Hail, 2016), however, found that this new organisational arrangement also created more distance between local officers and the specialist units. It resulted in serious challenges for coordination and in boundary conflicts. In the past, local officers normally just phoned the specialist, often someone they knew personally, and asked if (s)he could assist. In the newly created national organisation, they often do not know the specialists personally. This specialist may be located elsewhere in the country, cannot be phoned directly any more by local officers and may only be contacted by sending an electronic form to some other department where some unknown, abstract ‘boundary guard’ will decide on the request.

The next example from the Netherlands shows that this combination of formalisation (or bureaucratisation) and fragmentation may cause that police officers do not operate any more in the direct and highly informal way to which they were accustomed.

A patrol officer is called to a reported housebreaking. The two elderly victims are highly upset, mainly because there were unknown persons intruding their house. The police officer is highly supportive and shows much understanding to the victims. He explains that he would like to ask the ‘forensics’ to come along to do some investigation. He apologises for not being able to phone them directly and to ask them to come along. Later on, when we are back in the car, he explains that nowadays he does not even have the phone number for the forensic department. Since the introduction of the National Police the only available way to contact them is by making his report on the housebreaking on the computer. After finishing the report it will automatically be sent to the forensic department, where it will be decided if forensics will take this case or not. The patrol officer is not pleased with this new procedure: he has lost the ability to arrange things for victims promptly, cannot inform them properly and has become highly dependent on a faceless computer system (Terpstra et al., 2016).

Both in Scotland and the Netherlands problems have arisen because of the large distance between local officers and the centralised contact centres. Studies in both countries show that local officers feel that the staff at the new contact centres have a lack of local knowledge and understanding of the problems and priorities of their communities (Terpstra et al., 2016; SIPR, WWS and ScotCen, 2017a). This was a main factor contributing to the failure of the police to respond adequately to the car crash on the M9. Cooperation between local officers and higher-level units may also be hampered because of differences in views, priorities and interests that often arise in fragmented organisations (see also Giacomantonio, 2015).

A community police officer in a rural team arranged for a woman living in his community to receive an emergency button in her house. Such a button is meant for women who feel threatened, for instance by their (ex-)partner. In case of emergency she can use the button to warn the police. For the system to work properly, local patrol officers should be informed about the button, so that they know about the situation if problems arise. A central information unit should arrange for this information to be delivered at the local team briefing. However, for many days the team briefings do not contain this information. The community officer thinks that members of this higher-level unit don’t give priority to this issue. In his view they do not understand the impact of these violent threats on his rural community (Terpstra et al., 2016).

Several studies showed that since the introduction of the national police services in both countries, local officers and their senior officers have become more at a distance. The scale enlargement of the local teams, especially in the Netherlands, makes it more difficult to maintain direct personal relations between senior and rank-and-file officers (Terpstra et al., 2016; Terpstra, 2018). The evaluation of the Scottish police reform found that many constables feel that now they have less access to and interaction with senior officers than before the reform. This is partly a consequence of the fact that the physical location of supervision and management teams are now at some distance from local stations (SIPR, WWS and ScotCen, 2017a: 26). In the Netherlands, many local officers feel that the chiefs of their teams are too remote, and that the new management structure is too complex and obscure. Many prefer the old situation with one chief who could be directly approached for all questions, instead of the new layer of different managerial positions, each with its own specialisation and domain (Terpstra, 2018).

This new situation differs radically from the traditional police culture of solidarity, strong feelings of togetherness with colleagues and social cohesion that used to dominate the local teams (Cockcroft, 2013; Loftus, 2009; for the Netherlands: Terpstra and Schaap, 2013). As Hail (2016) found in her study on two local police teams in the period immediately after the reform in Scotland, this may exacerbate the gap between street cops and management cops (Reuss-Ianni, 1983).

An indication of the changing relations is that many local officers feel that since the introduction of Police Scotland there has been heavy reliance on email and internet in the communication between them and their senior officers. A workforce survey found that there was a strong appetite for more personal forms of communication (Axiom, 2015). Since the reform many Scottish police officers have wondered why senior officers ‘can’t (-) come in and engage with us, and actually explain to us the reasoning behind why they want these things done’ (SIPR, WWS and ScotCen, 2017a: 26). Hail (2016: 217–233) found that local officers experienced being ‘bombarded’ with e-mails and felt overwhelmed by information. For them, this created a very different working environment to what they were used to before reform in terms of more personal and direct relationships with their chiefs.

Since the introduction of Police Scotland, there has been more emphasis on performance targets. Instead of direct, personal and context-dependent forms of supervision, increasingly abstract quantitative targets are used. These suggest that the police can work on general goals, regardless of specific context and local situation. The evaluation of the Scottish reform suggests that the stronger emphasis on performance management must also be seen as a mechanism for enhancing centralised control over a large organisation (SIPR, WWS and ScotCen, 2017a).

Senior officers who operate at a distance are often not well informed about local situations and may not know the individual officers personally. This may make them hold strictly to formal rules and procedures instead of adapting their strategies to local circumstances and needs. This may have negative side-effects on local police work, as one of the studies in the Netherlands showed.

One of the Netherlands’ control centres gets a call that in a small and remote village a man is walking down the street with a gun. As the new procedure says, the responsibility is automatically taken over by a senior police chief, who is not from this region, but working in a large city 60 kilometres from here. Three local patrol officers are sent to the village with the strict order to wait until a heavily armed unit arrives (which will take at least half an hour, maybe more). The three officers, waiting just outside the village, do not see anything special, except that they notice that their presence, including the two police cars, causes a lot of curiosity and unrest among the village people. Some of them are getting nearer to see what the officers are doing. At a certain moment the most experienced officer decides that this is not the proper way to handle this case. He walks on his own to the farmhouse where the man with the gun was said to be. He knocks at the back door, opens it and asks: ‘Anyone there?’ Then the farmer comes and says: ‘Hi, copper, you need a cup of coffee?’ He proved to be the man with the gun, which he used because he had the usual problems with crows hanging around his farm. According to this officer, in this region every farmer has such a gun to scare the crows. The armed unit is informed that they can return. Later on, the senior police chief who had ordered the armed unit also seems to realise that his approach might have been a bit exaggerated given the local circumstances in this rural area. Still, he tries to justify his strategy: ‘Yes, well, this is the general procedure and this is how we do it in the city’ (Terpstra, 2017).

External relations

The increasing abstract character of the police cannot only be found in the internal relations of the police, but also in their external relations. Both in Scotland and the Netherlands, the transition to national police systems has resulted in greater distance between the police and their audiences. In both countries the closure of police stations and the reduction of opening hours, especially in rural areas, have contributed to this. However, the increased distance is not only a matter of available offices and opening hours. For instance, Hail (2016: 240–241) showed that since the start of Police Scotland there has been a loss of local orientation and local knowledge of the police, especially felt by lower-rank officers. The loss of small and personal beats has contributed to a perceived declining (moral) ownership and responsibility of local officers for their areas and for helping residents with their problems. Part of this change has been a shift in the style of policing, with less attention paid to local engagement activities and more emphasis on response and enforcement. Hail suggested that the loss of local orientation may also result from the dominating performance management systems and the shift of tasks to central units, such as the newly created central control centres. These findings are confirmed by several other studies (SIPR, WWS and ScotCen, 2017a, 2017b; Terpstra, 2018). In Denmark Holmberg and Balvig (2013) found that after the Danish police reform the police lost local knowledge and affiliation.

In the Netherlands an evaluation study of local policing (Terpstra et al., 2016; Terpstra, 2018) found that the introduction of a new model of service provision to citizens has also contributed to the growing distance between the police and citizens. According to this so-called Multichannel Model, citizens can use different channels to contact the police, report a crime or ask for information: by internet, by phone, by a visit to the police station or by meeting a police officer at home. The model suggests that citizens as ‘customers’ of the police have the room for ‘multichannel shopping’ (Rangaswang and Van den Bruggen, 2005). However, for the police this model is primarily ‘a convenient way to manage their workload’ (Welch et al., 2004). Normally citizens are supposed to use the internet or telephone; only in exceptional cases should they come to the police station (as they used to in the past). Citizens who still come spontaneously to the police station, should be refused or are told that first they should make an appointment by phone or internet. As can be expected, since the introduction of this model the number of internet contacts with the police has increased. However, the number of citizens who come to the desk of the police station has reduced sharply. As many local police officers have noticed, this model has created considerable distance between police and citizens, especially those who do not have the relevant computer skills. What is even more important, the system does not take into account that many citizens have the emotional need to tell their story in person and not by internet or teleservice system (Terpstra et al., 2016). The model is an example of what Welch et al. (2004: 388) call ‘keeping citizens at arm’s length’. It may improve the predictability and efficiency of the organisation, but as a study of local policing in the Netherlands showed, it also means that the police receive less information from citizens and at a later time (Terpstra et al., 2016). For the local police it is now more difficult to have an up-to-date overview of all incidents in their area. In addition, much information now coming by internet is incomplete and hard to use by the police (Terpstra, 2018). In Scotland, the introduction of centralised contact centres for the police and of a single non-emergency number for the whole country, in combination with reduced opportunities for direct telephone communication between the public and the local police, has had similar consequences. The evaluation of the Scottish police reform shows that local officers now receive less information from the public about what is happening in their area (SIPR, WWS and ScotCen, 2017a: 23).

It is not only the relations of the police with the general audience that have been changing, but also relations with partner agencies. Several studies showed that in the early phases of reform in Scotland, local partnership working deteriorated and that the communication and exchange of information with partners declined (Hail, 2016: 225, 297–301). It was suggested that financial cutbacks in both the police and partner services had made it more difficult to continue local cooperation (SIPR, WWS and ScotCen, 2017a: 24-25; Hail, 2016: 298).

Studies in the Netherlands showed that in the past the effectiveness of local security networks often strongly depended on the close and informal relations and trust between the individual representatives of the participating agencies (Terpstra, 2008). Now the relations between cooperating agencies are often more remote, formal and dependent on communication by email and computer systems (Terpstra et al., 2016).

An example of this development is ASAP (As Soon As Possible). In the Netherlands in 2011 this programme was introduced to make the processing of cases of high-volume crime much faster and more efficient (Salet and Terpstra, 2018). It resulted in a restructuring and far-reaching standardisation of the work processes of the police and some other agencies, including the public prosecution agency and the probation service. With ASAP, the cooperation between partner agencies shifted from the local or regional level to a much higher level (of the units of both the national police and the public prosecution service), areas with between 1.1 and 3.1 million inhabitants.

An empirical study of ASAP (Salet and Terpstra, 2017, 2018) found that from the perspective of the local police, the partner agencies are now remote, both physically and symbolically. Before the introduction of ASAP, problems of petty offenders were discussed in local security networks, with local police officers as participants and in close and personal cooperation with local partners. In many cases there was even a representative of the public prosecution agency working at the local police station for two or three days a week, allowing direct and personal relationships with the local police. Now with ASAP, the collaboration is at a higher level, local police officers are not involved any more in this decision-making and often do not know the officers working at the ASAP unit. They usually only communicate with the ASAP unit by e-mail and computer systems.

Although there is also a police officer working at the ASAP unit, usually this officer lacks detailed information about the individual case and the personal circumstances of suspects and victims. The only information usually available at the ASAP unit is provided by the computer systems. In other words, the cooperation in ASAP has been taken out of the local context, away from the local police, has become highly depersonalised and dependent on computer systems.

As a consequence ASAP decisions are usually made without direct information available about the suspect, victim and context of the case. Some local police officers try to compensate for the resulting rigidity of the ASAP procedure, for instance by trying to avoid the IT communication and computer systems. For instance, they try to provide relevant information about the suspect to the ASAP unit directly by phone or even informally try to convince the ASAP unit of their own view, even if this is not considered to be their task (Salet and Terpstra, 2017, 2018).

Organisational changes and changing views

Several factors seem to have contributed to the increasing abstract nature of the police. We assume that especially two of them are relevant to understand this process: organisational changes and shifts, and changing views on police and policing.

Organisational processes

First, the considerable scale enlargement of the newly created police organisations (especially in the Netherlands resulting in large local teams, extensive working areas and the closure of local police stations) has changed fundamentally both the internal and external relations of the police. The police often used to work in small teams, with close and intimate relationships between officers sharing more or less similar cultural and social backgrounds, with family-like relationships and the values of mutual solidarity and isolation from the outside world dominating their interactions (Cockcroft, 2013; Loftus, 2009). Now, as the study of local policing in the Netherlands showed (Terpstra et al., 2016; Terpstra, 2018), relations between police colleagues and with their chiefs have become more at a distance and more impersonal. The scale enlargement has contributed to more distance, both factually and symbolically, between the police and local communities, a process of local disembedding (Terpstra et al., 2015, 2016).

Second, the establishment of national police forces has resulted in standardisation of work processes and more uniform organisation, although local differences have remained more important than was originally envisioned (Terpstra and Fyfe, 2015). To realise this standardisation there has been an increasing formalisation of organisation and work processes (Terpstra, 2018). This is especially relevant in combination with two other developments mentioned below: fragmentation and increasing reliance on IT.

Third, a main aspect of the two national police forces is their fragmentation, both from an organisational perspective and with regard to many work processes. The merger of the former, smaller regional police forces created organisational integration at a higher level, but also resulted in new internal boundaries (Giacomantonio, 2015; Fyfe, 2018), which are more or less comparable to interorganisational relations (Benson, 1975). Many organisational and work processes are horizontally fragmented, with tasks and responsibilities divided among several actors and units within the organisation. The dramatic incident near the Waalwijk hospital is an example of the perverse effects that such fragmentation may have on police work (Commissie Eenhoorn, 2016).

The new national police forces also have considerable vertical fragmentation. Some tasks have been transferred to higher organisational levels, often with specialised expertise and powers. This may create new challenges for coordination and the transfer of information. It may make it hard for police officers to have an overview of relevant activities, and may create room for diverging views and interests and for new boundary conflicts (Giacomantonio, 2015; Fyfe, 2018). The M9 incident mentioned before dramatically illustrates what this vertical fragmentation may lead to. Both forms of organisational fragmentation may have a negative impact on feelings of responsibility and ownership by police officers for certain problems (Commisie Eenhoorn, 2016).

Finally, over the past decades the police have become more dependent on computers and computer systems, resulting in important changes in relations and work processes. For several decades a shift has been going on from what Lipsky (1980) called street-level bureaucracy, first to screen-level bureaucracy then to system-level bureaucracy (Bovens and Zouridis, 2002; Buffat, 2015; Reddick, 2005). Traditionally street-level police officers had considerable discretion and direct relations with citizens. Then the relationship between the police and citizens became increasingly mediated by the computer screen. This still left officers with considerable factual discretion. Next, with system-level bureaucracy citizens only have to deal with the computer (system). The discretion has been largely replaced from street- or screen-level officers to computer specialists and designers of the systems. It is only in exceptional cases that there is still room for direct and personal contact with citizens. The Multichannel Model of the Netherlands’ National Police illustrates this practice. As a consequence the (local) police and citizens are now more at a distance (Terpstra et al., 2016; Terpstra, 2018), with reduced direct communication between them (Tollenaar, 2014) and limited possibilities for individualising services to citizens (Danziger and Andersen, 2002).

This development has also made the police more dependent on ‘system information’ at the cost of direct and ‘personal’ information. The information of the police has become increasingly dependent on the frames and categories of computer systems (Ericson and Haggerty, 1997). This creates the risk that decisions by the police will often not be able to escape the (ir)rationality of (police) computer systems: decisions are only made if the computer system asks for it, and the question of whether the right decision was made is assessed by looking at what the computer (system) says. As a result, the unique boundary-spanning role of street-level officers (Prottas, 1979) that makes it possible to combine system information with direct information about what is happening outside the police may be in danger.

The increasing use of and dependence on IT systems may also have consequences for the relations of the police with partner agencies. In the past, local security networks were often dependent on informal relations and trust between the representatives of the partner agencies. As the study of ASAP showed (Salet and Terpstra, 2017, 2018), now the communication between agencies has become more dependent on computer systems, resulting in strong formalisation, depersonalisation and de-localisation of partner networks.

It is not only the external but also the internal relations of the police that have become more dependent on system information and IT communication. The study of local police teams in the Netherlands (Terpstra et al., 2016) found that communication by internet and email has increasingly replaced direct and personal interaction. The strong emphasis on performance management (especially in Scotland) (Hail, 2016) is usually realised by means of computer systems and formats. It has promoted highly abstract forms of management with the suggestion of decontextualised universal targets that can be measured, evaluated and communicated with a minimal personal relationship between officers and their chief.

Changing views

It may be assumed that the shift towards a more abstract police organisation is also the outcome of changing views on the police and police organisations. Since the 1980s, these views have gained much influence. Although there is a certain, often implicit relation between each of these views and the notion of abstractness, this outcome is largely unintended. Not any of the then relevant actors had the aim to realise abstract policing. Still, in practice these views have contributed to this outcome, gradually, and step by step. Three of these changing views on the police and police organisation seem to be most relevant for understanding the creation of abstract police: the New Public Management (NPM), information-led policing and changing views on the core tasks of the police.

First, in many Western countries since the 1980s, the police (as so many other public sectors) have been confronted with the pressure to realise so-mentioned NPM values: economy, effectiveness and efficiency (Terpstra and Trommel, 2009). In Scotland, the 2013 police reform was largely motivated because of huge budget cuts. The political promise that this should not be realised by reducing the number of operational officers (Fyfe and Scott, 2013) meant that other measures had to be taken, such as the closure of police stations, reduction in opening hours, scale enlargement, centralisation of specialist units and a stronger reliance on IT (in the Netherlands resulting in the replacement of service desks by internet communications for citizens who wanted to contact the police). In some cases the use of NPM instruments was mainly motivated by the wish to imitate the private sector: in the Netherlands one of the main arguments for closing rural police stations was that if the banks closed their rural offices, why should the police not follow (Terpstra et al., 2016), a form of isomorphism as DiMaggio and Powell (1983) called it. In the end, these NPM-inspired measures contributed to an increasingly abstract police organisation.

Second, the growing abstractness of the police has also been promoted by a belief in what can be summarised as intelligence-led policing (Ratcliffe, 2016; Fyfe et al., 2018), a more sophisticated offspring of Goldstein’s (1979) problem-oriented approach. Both are based on the assumption that police work should be led by systematic information and analysis of problems. This view on police work has resulted in the collection and analysis of huge amounts of data, the proliferation of new specialists in the police focused on intelligence and crime analysis, and in more risk-focused perspectives in policing. It also contributed to a gradual underestimation of the value and importance of direct, personal and often informal knowledge of police officers. These processes have increasingly made intelligence more or less reified, isolated from actual police work. It has led to new forms of distance in police work: intelligence was developed by officers who were not directly acquainted with the objects of their work, and officers who were directly involved in policing were confronted with the fact that their personal knowledge was superseded by abstract information from some specialist unit far away (Terpstra et al., 2016). One of the factors contributing to this process is that this kind of information is also used to account for police performance.

Third, especially in the Netherlands, over the past few decades there has been a debate on whether the police should concentrate on certain ‘core tasks’. Although the Netherlands’ government decided that the formal tasks of the police should not be changed, in practice over the past two decades there has been pressure on the police to concentrate on their ‘core tasks’ (Terpstra et al., 2010). This has had two consequences. On the one hand, there has been a degradation in the status of community policing and police work in rural areas. Rural policing is often seen as ‘old-fashioned’ and not focused on the ‘real problems’ (Terpstra, 2017). On the other hand, the core of police work is now often redefined as the gate to criminal law procedures. It looks increasingly as if problems are only perceived as worth the attention of the police if they are defined as criminal law issues. The Waalwijk incident is only one example of the disastrous effects that this view can have on police work. This concentration on core tasks contributes to the increasing abstractness of the police: seen from this perspective the visibility, local presence, and service to citizens, let alone the symbolic aspects of policing, are gradually eroding.

Consequences

What are the consequences of the enhanced abstractness of the police? Again we look at the findings of several (evaluation) studies about the Scottish and Netherlands’ police reforms. However, these studies only provide a first and limited view on these consequences, implying that the following should be seen as partially hypothetical.

The abstract character of the police may have diverse and contradictory consequences. On the one hand, organisational changes such as scale enlargement, standardisation, formalisation and the stronger reliance on IT systems and communication were introduced because they were assumed to contribute to a less fragmented, more effective and more efficient police organisation that would be better equipped for problems such as organised crime, cybercrime and terrorism. However, these organisational measures and innovations may also have a range of unintended and negative effects, both on the police and on their external relations. In the preceding sections many of these effects were mentioned. The increasing abstractness may make the police less informed, less flexible, and more dependent on rigid systems. It may lead to vulnerability of the police because they have less means to notice if things go wrong or if available information in the police systems is incorrect or incomplete. Abstract police may be less able to realise adequate solutions for the problems of citizens and communities, because internal processes and organisational procedures are fragmented, creating a lack of overview and having a negative impact on feelings of ownership and responsibility. Internal fragmentation may contribute to new problems of coordination and communication. The combination of scale enlargement, distance and fragmentation may result in new challenges for management and leadership. Local policing may be strongly dependent on central units that have a lack of knowledge and understanding of local situations and relations. Because of the distance of the police from citizens and partner agencies, the abstract police may have problems realising partnership working and effective collaboration with groups of citizens and communities.

The increasing abstract character of the police may also disrupt the close, direct and personal relationships between police officers and between officers and their chiefs, resulting in feelings of isolation, fatalism and powerlessness (Terpstra et al., 2016), or in feelings of (work or policy) alienation (Tummers et al., 2009). It may undermine traditional police culture, with its emphasis on solidarity and togetherness, and the related traditional collective responses and solutions of police officers to the risks and problems they are faced with in their daily work.

In the long run, with increasing abstractness not only the organisational aspects of the police may change, but also the kind of policing that is delivered. The distance between the police and citizens, communities and partner agencies may be hard to bring into line with notions of community policing that dominate the policy ambitions of many police services in Western Europe. Relations with citizens and communities may become less personal and direct and more dependent on abstract police information systems.

The creation of abstract police may also have an impact on the dominant notions and understanding of police professionalism. With the increasing dependence on information systems, IT communication and the shift towards system-level bureaucracies, other skills are valued more than those of the traditional police craft, that put a high value on the police as street-corner politicians (Muir, 1977) and the creative use of discretion in the handling of all kinds of practical situations and problems at street level (Bittner, 1970).

One may wonder what consequences the increasing abstractness of the police may have from the perspective of citizens. Earlier studies showed that citizens expected the police to be visible, approachable and available (Van der Vijver, 2004). To what extent do citizens, in an age of increasing abstractness, change their expectations, not only of banks (the institution that the Dutch police try to imitate in their retreat from the rural areas), but also of the police? There are several reasons why personal and direct relations between citizens and the police may be still very important. Citizens with low trust in the police, often have a much better relationship with the individual community officer, if (s)he is personally known (Van der Vijver, 2004). Victims of serious crime often prefer to report their problems to an officer in person instead of using abstract teleservice systems (Boekhoorn and Tolsma, 2015). Finally, the police have a strong interest in maintaining close relations with citizens and communities because they are highly dependent on their information, also to prevent escalation of problems.

Abstractness, McDonaldisation and the iron cage

The increased abstractness of the police can also be seen as the largely unintended outcome of wider and long-term processes of rationalisation and (late) modernisation. Following Max Weber’s analysis of modernisation as a process of (formal) rationalisation, Ritzer (1993) introduced the notion of McDonaldisation. In his view this form of hyper-rationalisation is aimed at improving efficiency, calculability, predictability and control, but the ‘downside’ of this process is that it can become irrational in its consequences, increasing the inefficiency, unpredictability, incalculability and a loss of control. In the view of Ritzer, the strong emphasis on control leads to the replacement of human judgement by the ‘dictates of rules, regulations and structures’, with the result that work processes become less dependent on human skills, abilities and knowledge. In other words, according to Ritzer, McDonaldisation contributes to ‘dehumanising’, deskilling and ‘unreasonable systems’ (Ritzer, 1993: 21, 118–122).

In many respects the creation of abstract police can be seen as a process of McDonaldisation of the police. ¹ This is especially visible in the case of the large-scale police reforms in Scotland and the Netherlands, aimed at improving the efficiency, predictability and control of police organisations, but also resulting in irrationality, loss of efficiency and predictability, and in problems of control. The greater distance (in many respects), the fragmentation of the organisation, of work processes and responsibilities, the stronger dependence on IT communication and on system information and system-level bureaucracy, mean that the police are confronted with what Ritzer called dehumanising and deskilling trends.

On the other hand, this process of rationalisation of the police resulting in irrationality can also be seen as the outcome of more gradual long-term social processes. The two police reforms have accelerated this process and made the outcomes more visible and explicit. Important changes within Western societies becoming more visible over the past few decades, such as time-space distantiation, the disembedding of institutions, globalisation, individualisation and the increasing dependence on abstract systems (Giddens, 1990, 1991) have created new challenges for the police, but also new options for further modernisation. According to Giddens this process of ‘late modernisation’ is not only resulting in new forms of unification and integration, but also in new tensions. With the abstract police new opportunities have been created for dealing with a more complex, globalised world, but similar to other abstract systems this has also resulted in deskilling, alienation, fragmentation and the loss of local control (Giddens, 1991: 136–138). With the scale enlargement, merger of smaller organisations and the integration at higher levels, the police organisations have created greater ‘security’, but as Giddens noted, ‘there is also a serious price to pay for these advances’. Abstract systems depend on trust, but this ‘trust in systems takes the form of faceless commitments’ (Giddens, 1990: 88). The trust in abstract systems does not provide ‘the moral rewards which can be obtained from personalised trust, or were often available in traditional settings’. In addition, the ‘whole-sale penetration of abstract systems into daily life’ is creating new risks for the individual (Giddens, 1991: 136), in this case for the police, both for the organisation, and for individual officers.

However, the greatest risk of this process of increasingly abstract police may be in the unintended creation of an iron cage, to use another term originally introduced by Max Weber (1905). It is quite usual to interpret this iron cage as a sort of Kafka-like experience of the individual being caught up in an ‘iron bureaucracy’, where s(he) is dependent on red tape, anonymous powers and increasingly on IT systems (see for example Ritzer, 1993). Originally Weber had another interpretation of the iron cage in mind, which is even more relevant for understanding the risks of abstract policing: not an external iron cage, but a flexible ‘shell as hard as steel’, which is not external but a part of the living being (see Baehr, 2001). Following this interpretation, an iron cage is not an external prison for the individual police officer, but a sort of complete acceptance of the abstractness. It is resulting in a belief of police officers in their work as an abstract task, with abstract responsibilities, working for abstract communities, abstract citizens and evaluated by means of abstract performance targets and procedures. The negative outcomes of this process can be illustrated in many ways, but once again the case of the young woman killed near the Waalwijk hospital is most relevant here: many police officers just living in their own abstract reality, with their complete belief in system information and IT communication, without realising any more that in the ‘real’ world the life and protection of this young woman was the core of what they should be working on.

Concluding remarks

The preceding analysis has focused on the shift that over the past years the police have made to a very different kind of organisation. It is remarkable that this shift happened after more than two decades of a strong belief in the importance of small, flexible, informal, customer-oriented organisations with much emphasis on bottom-up processes of professionalism, change and governance and a general antipathy to highly centralised and bureaucratic organisations (Reed, 1993). Not only the two police reforms mentioned, but also the creation of abstract police, are in a sharp contrast to these notions.

Given the analysis presented here, there is no need for nostalgia. Many of the innovations mentioned here, such as the two police reforms or the introduction of new organisational structures and the use of modern technologies, were motivated by the aim of improving the effectiveness and efficiency of the police in a globalising and increasingly abstract social context. The point is, however, that this process of rationalisation and modernisation created large numbers of unintended side-effects, in sum the abstract police.

One of the main concerns here is that this shift, despite its far-reaching consequences, has been barely noticed until now. A public debate on how to deal with this increasing abstractness of the police has not taken place. Central issues for such a debate should be a reflection on the consequences of the analysis presented here for the future of the police, for the kind of police work delivered, for the position of community policing and the value of policing principles, for the relations with citizens and communities, for the legitimacy of the police, and for the dominating views on police professionalism and police management. In the end, the main question should be: is this the kind of police that we want, at a distance from the public and partners, fragmented, and largely dependent on system knowledge and IT systems?

If this debate will not be held, it may be expected that the increasing abstractness will cause new conflicts within police organisations and with communities. Many police officers are deeply concerned with the consequences of many of these changes and they may try to resist the abstract systems in more hidden ways. This may have a negative impact both on the effectiveness and transparency of the police. The study on local policing in the Netherlands found several examples of these resisting practices (Terpstra, 2018). Communities will feel that the police are less accessible and will have less confidence that the police have a good understanding of local problems. This will impact on information sharing and trust, which are vital to police effectiveness and legitimacy.

A public debate on abstract policing is also needed because police forces may have a rather naïve belief in new technologies. The 10-year future strategy of Police Scotland (‘Policing 2026’) provides some examples of it. It puts great trust in new IT technologies for improving the police and police organisation. However, main questions that arise from this perspective are not raised, probably because of a lack of awareness, such as what the relations should be between, on the one hand, the increasing emphasis on technology and the belief in ‘systems’, and on the other hand, the fact that the police have to work for communities and are dependent on ‘the human factor’. Any future strategy of the police should pay attention to these issues. If not, the growing abstractness of the police will continue, with ever more negative consequences for police organisations, the public and partner agencies.