Changes and continuities in police responses to domestic abuse in England and Wales during the Covid-19 ‘lockdown’

ABSTRACT Covid-19 and the associated public health response directing people to stay at home and/or shelter in place generated acute awareness of, and concerns about, the likely impact on violence(s) against women across the globe. Initial reports from support services suggested that such violence increased, and that its impact was more complex. Early evidence of increased demands in relation to domestic abuse on policing was however less clear. This paper, based on findings from a larger project, offers an analysis of the initial responses to domestic abuse by the police and the courts in England and Wales during the initial pandemic lockdown of 2020. These findings are situated within wider debates concerned with the nature and impact of the current organisational structure of policing and suggests that whilst police forces responded both quickly and innovatively in order to maintain a focus on domestic abuse, their capacity to continue in this vein will be limited in the absence of wider structural and organisational change.


Introduction
In his review of the state of policing in England and Wales for 2019 the Chief Inspector of Constabulary, Sir Tom Winsor, makes a compelling case for situating an appreciation of the policing demands emanating from the Covid-19 pandemic within the wider context of the need for policing reform in England and Wales. Whilst acknowledging that crime-driven demands for policing overall first declined and then stabilised in recent years, he goes on to note that domestic abuse is being increasingly viewed within a focus of policing vulnerabilities (of both victims and offenders) which has increased the complexity of demands on policing. Of course, the increasing focus on vulnerability needs to be considered in tandem with a wide range of other demand-led policing issues such as organised crime, terrorism, and managing the public response to natural disasters such as flooding. Embedded within these general observations there are specifics.
Domestic abuse has run somewhat contrary to the overall decline in crime in recent years. From August 31st 2013 to 30th June 2017 recorded incidents of domestic abuse increased by 88% (HMICFRS 2019) with the Office of National Statistics reporting another 24% increase by March 31st 2019 (ONS 2019). There may of course be a wide range of reasons underpinning such changes in recorded figures, and also a wide range of variations in practices between different police forces contributing to these overall figures (for a fuller discussion of this see HMICFRS 2019). However, for the purposes of the discussion here, figures such as these, set within the contextual observations of Sir Tom Winsor, provide the backcloth against which to begin to appreciate the impact of Covid-19 on police responses to domestic abuse in England and Wales.
In order to explore these issues further, the paper falls into four parts. The first offers a brief overview of the challenges still facing the police in response to domestic abuse despite over thirty years of policy interventions and recommendations on this issue. The second discusses the ways in which crises of different kinds, including public health crises, can impact on domestic abuse. The third section considers the specificities associated with the 'stay at home' directive in England and Wales and its impact on responses to domestic abuse. This draws on empirical data documenting policing responses to this directive from March 23rd 2020 to June 15th 2020 in England and Wales. The fourth and final part of the paper will return to some of the questions posed in the State of Policing Report (2019) by Sir Tom Winsor. In particular, it will consider the implications of the findings presented here for the questions of modernisation and parochialism referred to by Winsor, and will address these issues with policing responses to domestic abuse in mind.

Policing domestic abuse
The 1970s witnessed a rise in wider concerns about violence against women (and children). Notably, the feminist movement had an undeniable presence in raising these concerns (see, for example, Wilson 1983) and, according to Goodmark (2018), prepared the ground for the demands of the criminal justice system to be more effective in tackling abuse. There emerged a strong view that more, and improved, implementation of the law would provide an answer to this problem. It was the Women's National Commission, established in the mid-1980s, which brought violence against women to the attention of the UK Government (Smith 1989). This commission pointedly recognised that domestic violence was most frequently not seen as 'real' policework and, even when all the conditions were met to make an arrest for domestic violence, this often did not happen (Dobash and Dobash 1980, Edwards 1986. Moreover, even in cases where the woman was keen to pursue a prosecution, the police often were unsupportive (Dobash and Dobash 1980). Instead, they chose to believe in the 'myth of the reluctant victim' (Stanko 1989). These concerns influenced Home Office Circulars 69/1986 and 60/1990, both of which marked turning points in the expectations associated with policing policy and practices on rape and domestic violence. Yet, despite these and many other policy developments and initiatives since the early 1990s, the view that domestic violence is not proper policework still lingers (for a review of the literature and interventions relevant to this period see, for example, Matczak et al. 2011, Sherman 2018. Indeed, the 2014 report of the HMIC Inquiry into the policing of domestic abuse concluded: Domestic abuse is a priority on paper but, in the majority of forces, not in practice. Almost all police and crime commissioners have identified domestic abuse as a priority in their Police and Crime Plans. All forces told us that it is a priority for them. This stated intent is not translating into operational reality in most forces. Tackling domestic abuse too often remains a poor relation to acquisitive crime and serious organising crime. (HMIC 2014, p. 6) This was a telling observation given that a pro-active response and support for such a response had been open to the police since Home Office Circular 60/1990. These observations were further endorsed in an updated report by the HMIC (2015) and the 2018 HMICFRS Report suggests that whilst control room responses to DA calls were improving there were still delays in first response; understandings of coercive control needed to improve (see also Barlow et al. 2020); initial investigations were improving but police officers' own statements reviewing the scene could improve; the use of body-worn cameras and the quality of risk assessments could improve (Robinson et al. 2016); there were still variations in police forces use of arrest, and a decline in the use of police bail was also noted. In addition, this report notes that there was an increase in the number of victims not supporting further investigations (from 15% to 58% depending on the force area) with no documented understanding of why. However, at the same time, the use of domestic violence protection orders (DVPOs) had increased by 16% from 2016 to 2017. It goes on to report that nearly all police forces now had MASH units (multi-agency safeguarding hubs) but, as with MARACs (multi-agency risk assessment conferences), there were variations in how these operated.
In addition, 33 police forces now had Operation Encompass in place (a process of informing schools when a child has witnessed domestic abuse). This report also states that 27 forces were already facing increasing demands in relation to domestic abuse and remarkably (given that flagging repeat incidents for domestic abuse on police systems has been a recommendation since the early 1990s), 28 forces still did not flag such victims.
This brief overview provides a flavour of both the progress made and the stagnation present in policing responses to domestic abuse in recent decades. Whilst the increase in inter-agency working is self-evident, some intransigencies remain, particularly in respect of the use of arrest (though the evident increasing use of DVPOs might be being thought of as a substitute for arrest), the quality of investigative processes and appropriate evidence gathering in support of prosecutions, as well as in the use and deployment of risk assessment tools (see Hoyle 2008, Barlow and. These might all be considered still to be works in progress. Thus, when Covid-19 became clearly identified as a public health problem requiring a robust response, the policing picture in relation to domestic abuse in England and Wales was already rather kaleidoscopic.

Covid-19 and domestic abuse
To be clear, it is well documented that domestic abuse takes its toll on victims' lives in ways other than marks of physical abuse. The number of lives shortened as a result of living with the constant insecurities of physical, mental and financial abuse are well-known (Walklate et al. 2020). That these experiences can be exacerbated by poor policing and criminal justice responses is also well-known (HMIC 2014). The interconnections between these experiences and their recognition as a public health issue has risen up the international agenda (WHO 2013). Recognition of these interconnections is also increasingly evidenced in relation to the consequences of a wide range of natural disasters. Crises like these all carry with them the potential to add significantly to the violence(s) experienced by women and children at the hands of men. Work in India (Rao 2020), the Philippines and Vietnam (Nguyen & Rydstrom 2018), Iran, (Sohrabizadeh 2016) and Japan (Yoshihama et al. 2019), all point to the increase in stresses placed on family life as a result of disasters. All of these consequences are gendered (True 2013). In addition, work by Lauve-Moon and Ferreira (2017) and Parkinson (2019) points to the ways in which, when disasters happen, the vulnerabilities of those living with violence in their lives can be compounded and their needs can become more complex. This finding is reiterated in the recent report by Pfitzner et al. (2020) for women living under lockdown in Australia. Parkinson (2019) also observed that when disasters occur, during which there are calls for everyone to pull together, domestic abuse can become invisible. Interestingly this has not yet proved to be the case during the Covid-19 pandemic. In many ways both in the UK and elsewhere the reverse has happened, as illustrated in some of the work cited below.
Both global and local voices have been keen to point out the potential problems inherent in 'stay at home' directives, especially for women and children. Phumzile Mlambo-Ngcuka, Executive Director of UN Women, stated that confinement would foster and increase tension and strain created by health and money worries for women isolated with violent partners. She described the situation as 'a perfect storm for controlling, violent behaviour behind closed doors,' naming the consequences of Covid-19 isolation restrictions on domestic abuse as 'The Shadow Pandemic' (UN Women 2020). To be clear, stay at home directives have the capacity to put everyone under stressnot least for those whose jobs were lost or might be threatened as a result. Moreover, these directives are not in themselves the cause of violence, though there is evidence to suggest that violence against women does increase when partners are put under financial stress (see inter alia Fraser 2020 and the work cited above). It is important to remember however, that the insecurities felt by everyone during the first half of 2020 and ongoing, about money, jobs, health, food supplies are the kinds of insecurities felt routinely by women and children living their lives with an abusive partner. Such insecurities can be multiplied when the spaces afforded by work, school, meeting with friends and so on are taken away. As Williamson et al. (2020) have pointed out, 'perpetrators can use the lockdown measures as a tool of control and coercion by, for example, either insisting on strict lockdown or failing to protect the health of family members.' Sometime this includes weaponizing children (Meyer and Fitz-Gibbon 2020).
To date, the extent to which these concerns have been realised has yet to be systematically established. Nonetheless, there has been much media coverage on the pressures being faced by support services, with UN Women (April 2020) reporting that domestic violence went up by 30% in France since the introduction of lockdown on March 17th, emergency calls for domestic violence went up in Argentina by 25% post lockdown on March 20th, and Cyprus and Singapore logged an increase in helpline calls of 30% and 33%. Similar increases in demands in reports and requests for shelter were reported in Canada, Spain, the UK, the US, and Germany (see also Usher et al. 2020). Parallel concerns have been aired in Australia (see inter alia Pfitzer et al. 2020, Women's Safety NSW 2020). In the UK, widely reported data from Refuge, a women's shelter organisation, showed that on average calls and contacts to the National Domestic Abuse Helpline seeking their help had increased by 49% for the week commencing 6th April compared to pre-lockdown (Refuge 2020). Similarly, in relation to male victims of domestic abuse, the UK Home Affairs Committee (2020, p. 8) reported, The Men's Advice Line for male victims of domestic abuse had an increase in calls of 16.6% in the week of 30 March, and a 42% increase in visits to its website and the Respect phone line, which offers help for domestic abuse perpetrators who want to change and stop being violent, had a 26.86% increase in calls in the week of 30 March, while its website received a 125% increase in visits in the same period compared to the week before.
Whilst reports to support services increased significantly, evidence on changes in reporting rates to the police globally is to date inconclusive, though data yet to be made publicly available may show otherwise. In a review of 17 studies, Peterman et al. (2020) point out that looking at criminal justice data on a month-by-month basis reveals little about wider trends over time nor anything about the accuracy of the data itself. Moreover, this kind of evidence can produce contradictory findings. For example, in two studies based in the US, one suggested a 10% increase in calls to the police for domestic abuse, largely driven by households with prior calls of such abuse (Leslie and Wilson 2020), while the second reported a decrease in such calls in the two cities studied (Mohler et al. 2020). In a study based in Dallas, Piquero et al. (2020) reported a short-term spike in reports followed by a decrease in reporting behaviour. Work by Campedelli et al. (2020) indicated no significant change in reported incidents, with Gerell et al. (2020) reporting a decrease in reports of indoor assaults in Sweden. Freeman (2020) also reported no evidence of an increase in recorded incidents of domestic assault on the introduction of social distancing in New South Wales, including the figures for more serious assaults for which it is suggested police involvement might still be expected. Indeed, Nix and Richards (2021, p. 1) in a study of six U.S. jurisdictions suggest caution in reading findings from data gathered early in this process pointing to data cleasing issues and the potential for what they term 'localized effects'.
Early findings such as these need to be placed side by side with the increasing documented evidence concerning the impact of lockdown on the wider delivery of services for women and children. As has been mentioned above, Pfitzner et al. (2020) have pointed to the increasing complexity of needs for support women are presenting to services, a finding also endorsed by the work of Sibley et al. (2020) in New Zealand, in New South Wales (Women's Safety NSW 2020) and in England and Wales (Women's Aid 2020). Indeed, with an increase in calls to helplines being documented in a wide range of jurisdictions across the globe, some governments have been pressed into providing (at least) additional financial support. For example, on the 2nd May 2020, the UK Government announced a £76 million package for domestic abuse charities in recognition of the strains they were under. However, closer to the concerns of this paper and in the context of policing in England and Wales, the disparity between calls to helplines and reports to the police was noted in the minutes of National Policing Board held on May 6th 2020, with the suggestion that 'deep data dives' would be undertaken to try and make better sense of these differences. Making sense of the policing response to domestic abuse within these broader concerns is one of the foci of the project from which this paper emanates.

This study: methodological approach
This study analyses the impact of lock-down and the transition out of lock-down on criminal justice agency response to domestic abuse. The project commenced in June 2020 and ends in December 2021. 1 Our approach to this study adopts a mixed methods strategy and is informed by the 'rapid' research approach, predominantly used in health and clinical settings though equally applicable to exploring domestic violence and the criminal justice system. Rapid research can take many forms and usually involves short timeframes, team-based research and iterative data collection and analysis (Vindrola-Padros 2019). It can include collection of quantitative data (through surveys, and review of existing data sets), qualitative data (through formal and informal interviews with key informants) and mapping (to outline the population affected) (McNall and Foster-Fishman 2007).
For the purposes of this project the recommendations from reviews of rapid research methods followed include ensuring that data collection and analysis happen simultaneously, writing of interim reports early in the process, and aiming to ensure that data requests to respondents do not conflict with other requests for similar data (McNall and Foster-Fishman 2007). We are particularly committed to the early sharing of findings (Johnson and Vindrola-Padros 2017, Vindrola-Padros et al. 2020), and paying attention to 'reflexive interpretation of findings' (Vindrola-Padros and Johnson 2020) as this project unfolds over time. A survey of a wide range of academic, policy and grey literature is ongoing (some of which is reflected in the literature review above) and interviews with key stakeholders in the criminal justice process and the domestic abuse field is in progress.
As a first stage in the data gathering process an online questionnaire was distributed to all policing leads for domestic abuse in England and Wales (43 forces were approached). Of the 25 forces that responded, four were Metropolitan, fifteen were rural, and six were what we considered to be mixed in profile. The Metropolitan forces received the largest funding and policed the largest populations, if not physical areas. Three forces had a high level of ethnic diversity in the areas they police, six mid, and sixteen low (the rural forces). In terms of respondents to our questionnaire, we classified five as largely 'operational' (Detective Sergeants and Detective Inspectors) and seventeen as 'policy' focused (Detective Chief Inspectors, Superintendents, and Chief Superintendents), and one as operational and policy focused (two did not give this information). Thus the sample comprises a good mix of respondents from a variety of forces.
The purpose of the open-ended data elicitation used in this questionnaire was to focus on innovative practices. Respondents were asked to reflect on the initial lock-down period (March 23rd to June 15th 2020) and to address their responses to domestic abuse with three questions in mind.
1. Were they able to continue with normal practices? (if they could not, what were the reasons for that, and how did they adapt) 2. What, if any, were the changes introduced in relation to responding to domestic abuse under social isolation and what was the impact of these changes (respondents were asked to give examples) 3. In review, would they do anything differently, and, of the changes they made in response to the lockdown, which would they keep, and why?
This questionnaire received 26 responses from 25 different police forces (two officers responded from one force). In what follows, we offer an initial analysis of the findings generated by this moment of data collection (July-August 2020).
Perceived levels of service need during the lockdown period All respondents made some reference to the impact of lock-down on the reported incidence of domestic abuse in their force area. By far the majority noted that domestic abuse calls declined initially in March and then returned to normal by May (19/26 respondents) (to be noted here of course is the longer -term trend in increasing incidents of domestic abuse more generally cited above). Only one domestic abuse lead reported a year-on-year increase in domestic abuse calls by the end of the lockdown period (June 15th 2020). Two respondents offered data in support of their observations and this data indicates that, in their areas, reports of domestic abuse in May paralleled incidents recorded in December 2019 (prior to the Christmas period). Four respondents mentioned the knock-on effects of court closures and the absence of face-to-face men's behavioural programmes run by the Probation Service and other service providers, and the impact of the lack of refuge spaces for women wishing to leave their partners. Interestingly, none of the respondents referenced the wider changing legal context in which policing was being undertaken (namely the Coronavirus Act 2020). The four main themes identified in these questionnaires are discussed below under the headings of business as usual, context-led modernisation, furthering perpetrator focused practice, and what has been referred to elsewhere as 'justice by geography' (MaClenagan 2018).

Business as Usual during the lock-down-continuing with normal practices
The open-ended nature of the questions asked offered considerable scope for a wide variety of answers both in the topics raised and the level of detail provided in the answers given. In these responses it is evident that some aspects of the challenges presented by policing during lockdown were made explicit and others not. For example, whilst a small number of respondents (4/ 26) highlighted the need to include in their risk assessment processes the health and safety of their officers when expected to engage in face-to-face contacts (availability of protective equipment, for example), the majority of responses seem to take these health and safety requirements as given. At the same time of course, these requirements were clearly directing the ways in which much 'normal' work was being conducted: As is regular practice, each call for service was graded but with the additional consideration of Covid-19 and the highly contagious nature of the virus. Emergency calls continued to be treated with the highest priority and despite the risk of infection to officers, if the situation dictated, they did enter properties and come into close contact with victims, their families, and offenders. In priority and standard responses, telephone statements and resolutions were the preferred options in an attempt to adhere to lockdown and reduce the risk for officers (Mid-sized Midlands, mixed rural/urban force).
Only three respondents made reference to the ways in which the overall reduction in crime documented during lock-down (National Police Chiefs Council press release 21/05/20), and the consequent increased availability of some policing resources, actually facilitated focused efforts to deal with already existing levels of domestic abuse. For example, one respondent reported an 8% increase in domestic abuse referrals to the Crown Prosecution Service and a 15.8% increase in perpetrators charged during lock-down as a result of their perpetrator focused activities during this time.
(Focused activities such as these are discussed in more detail below).

Context-led modernisation; changes to practice
One of the themes discussed in some detail in Winsor's Report (2020) is the question of the modernisation of policing. This aspect of his report will be returned to at the end of this paper. However, of interest at this juncture is his reference to police use of information and communications technology and the associated infrastructure as a means of providing the community with a consistent and generally accessible way of interacting with the police. He points to the value of the single 'online home' most police forces have now signed up to as an example of modernisation in this regard. Indeed, as an early scoping exercise, this project examined all police force websites and endeavoured to access information from them as a victim of domestic abuse. This was not always an easy process (See Richardson et al. 2020). Yet it is evident from our data that the context of Covid-19 resulted in a considerable impetus towards the use of information technology and other less 'advanced' technologies both to reach out to victims of domestic abuse and to do the criminal justice work associated with such abuse. This was manifest in all the responses received and is discussed in more detail in what follows.
In order to maintain a consistent and meaningful presence in relation to domestic abuse, police forces endeavoured to deploy the digital world (information and communication technology): A 'Safeguarding the Vulnerable' meeting was introduced to monitor all strands of vulnerability and chaired by Head of Public Protection. The purpose was to ensure police and support services were able to maintain business as usual and to identify proactive opportunities including enhanced communications strategy, commercial radio advertising, and introduction of Facebook Q and A (Small Southern rural force).
The biggest battle we have is to get the trust of the community when reporting domestic abuse. Although, as the police, we are law-enforcement, we ae also there to support victims, ensure they are supported, and give them a voice. This period has shown me how important this message is. DA Facebook chat was set up as an opportunity for victims and those concerned to get help. It involved partners from all four local authorities, Fire and Rescue, Corporate Comms, and Social Care. The first event reached 10,600 people, and we answered 13 live questions during the event. (Mid-sized Northern rural force).
In three force areas local radio stations were used to raise awareness of both domestic abuse and the availability of the police (noting that social media is not itself inclusive, and nor is access to the Internet in many rural areas). Other forces developed posters to be displayed in supermarkets (4/26 respondents), used outreach hotlines, and the expertise of other key workers in the community to recognise possible signs of domestic abuse (refuse workers 1/26, postal workers 1/26, supermarket employees 2/26; pharmacies, 4/26), and held domestic abuse surgeries in the local supermarket (1/26). New initiatives were launched, "safe at home" which was a supermarket surgery for anyone experiencing anything within the home that made them feel unsafe; excellent comms strategy, including pop-up ads on social media, which was a first for us (Large Southern rural force).
However, arguably the biggest growth area for all respondents was in the use of social media to engage with, and respond to, victims. All respondents referred either to Facebook pages, online surgeries, the development of webchat facilities, and/or means of maintaining contact with victims by telephone and/or Skype calls. Respondents were careful to mention the importance of assessing the risks of this approach, for example, checking whether the victim is alone. All reported feeling that the use of online facilities in this way would be maintained in the coming months. Some aspects of this use of digital and other technologies however reached beyond maintaining a wider community presence and their availability to victims. All respondents commented on the significance and value of being able to maintain inter-agency working relationships virtually using online platforms (MS Teams for the police, Skype for Business by the courts). Indeed, these platforms and access to them have been essential in maintaining what work the criminal justice system as a whole was able to sustain. Moreover, with one exception, all respondents thought that virtual meetings/ video conferencing would remain the way in which this aspect of their domestic abuse work would be conducted in the future. The one exception felt that there more to be learned in 'real' face-to-face meetings as opposed to virtual ones. The reasons lying behind the enthusiasm for maintaining virtual meetings ranged from the practical (it was much easier to get everyone in the same room at the same time with no travelling issues to negotiate) to observations pointing to the improvement in meeting quality, efficiency and frequency. All of which, when taken together, are clearly suggestive of a potential for more timely and speedier responses to the victim at risk. Whilst less was made of the use of virtual platforms in relation to victims and offenders (though as becomes clearer below, they were used in relation to these contacts too), there is an interesting synergy here between these changes in practice and the observations made by Winsor (2020) in relation to the question of modernisation discussed below.
Non face-to-face interaction with victims was another interesting development reported on by 10/26 respondents. This covered a range of different types of communication strategies from taking statements over Skype, to delivering non-disclosure decisions in relation to the Domestic Violence Disclosure Scheme (Clare's Law) 2 over the telephone, to accepting electronic copies of ID documents when making a request under Clare's Law, to providing a telephone response as the first response to the report of an incident. HMICFRS (2019) had advised extreme caution in a telephone based first response. They suggested, then, that this was only to be used as part of the College of Policing pilot study and then only when there was no significant history of domestic abuse, the incident being reported was only for minor offences, there was no evidence of coercive control, there were no immediate safeguarding concerns, and when the partners were not cohabiting. This is despite the evidenced efficacy for a response of this kind established in one police force area (Robinson 2017). Again, whilst these changes in practice are relatively small, it will be interesting to reflect on what their cumulative impact might be over time.
Clearly these thematic findings smooth out differences between force areas, the different emphases within force areas given to each of them, and the kind of work prioritised as a result. Of interest and perhaps striking in these themes, is both the capacity for innovative responses emanating primarily from the demands of this global health emergency and the speed with which many of these responses were put in place. Indeed, it is not possible to capture in a paper of this kind the huge variation within and between forces in developing these responses and the speed with which they were undertaken (though this is discussed in a little more detail below). However, within this overall picture it is also clear that a small number of forces (three in this data set) used the lockdown opportunity in 'perpetrator focused' ways. This is the second theme to be discussed here.

Furthering perpetrator focused practice; consolidating practice
During 2020 three domestic abuse leads, in response to our questionnaire, foregrounded their work in furthering perpetrator focused practice. One began by stating that good practice in this respect was a force-wide policy. As cited above this respondent reported an increase of 8% in domestic abuse referrals to the Crown Prosecution Service, an increase of 15.8% in perpetrators charged and a 20% improvement in case turnaround during lock-down. Another respondent reported on the value of proactively targeting perpetrators, particularly for addresses where repeat incidents were known to have occurred, and where no contact had been made with those addresses during the first four weeks of lock-down. They reported that this approach had 'worked well' (making 383 contacts out of a potential 539) and felt that this kind of targeted approach would be maintained. A further respondent similarly reported focussing on 220 couples who were repeat callers prior to lock-down and who had not called from 23rd March until the beginning of May. Taking an inter-agency approach where no agency had heard from this group, specialist domestic abuse officers were sent to complete a face-to-face 'safe and well' visit. Thirty visits (Covid-19 compliant) were completed. A further project focusing on 200 couples where a child was present during domestic abuse was (at the time of the response) also being undertaken. Given the longstanding evidence that all a perpetrator had to do before the police arrived was to disappear (Edwards 1986), the evidence on repeat victimisation in cases of domestic abuse (Farrell and Pease 1993, Hope 2007, Pease et al. 2018, and the importance of flagging repeat incidents on call systems (HMICFRS 2019), practices such as these are welcome.

Justice by geography; differences in practice
The concept of justice by geography has a history arising in the United States and has had some resonance over the last decade for experiences of the criminal justice process in England and Wales. It focuses attention primarily on the differential experiences of, and access to, the criminal justice system and the work of criminal justice professionals depending upon which part of the country you live in, or even which postcode address an individual might have. (For example, one of the respondents to our questionnaire reported different response levels and practices in their force area depending upon policing division). In the context of responses to domestic abuse ONS statistics regularly point to the variations between police forces in arrest rates alongside the differential use of both different offence categories and different legislative procedures also found between them. One example of this is found in the differential responses to and use of the domestic violence disclosure scheme (Clare's Law) in which the term as 'justice by geography' (MaClenagan 2018) has been utilised.
The nature and extent of the variable responses to domestic abuse both within and between forces has been a concern for HMIC Reports over the last decade. This variation is evident in the responses to our questionnaire and is implied in the presentation of the themes discussed above. Not all police forces respond to domestic abuse in the same way either as a priority in terms of their mission statements or as a demand-led priority in relation to service delivery. This is not new and serves as a reminder that criticisms of service delivery in relation to the policing of domestic abuse can be context specific. In this vein it is of value to reflect upon the comments offered by Shapland (1988) who proffered some detail on the difficulties facing general victim-centred initiatives and the capacity for them to be embraced within the criminal justice system. She used the analogy of the relationship between 'fiefs and peasants' as one way of understanding the challenges faced in centring victim interests. Getting each 'fiefdom' to talk to each other let alone with the same priorities was a challenge then and remains a challenge contemporarily (HMICFRS 2019). One of the fiefdoms to be negotiated was, and is, the police. Of course, during the 1980s the role of different Chief Constables in setting local agendas according to their own predelictions is well documented (see for example, Reiner 1991). Yet the extent to which there are local variations for policing priorities, taking account of local demands but within the purview of Chief Constables and their senior officers, continues to have some resonance. Winsor's (2020) discussion of parochialism and the lack of fit between a 1960s policing organisational structure and a twenty-first century policing world is illustrative of this ongoing challenge. One example of the kind of the variation in prioritisation and its mediation by parochialism being alluded to here can be discerned in the priority statements set by local Police and Crime Commissioners (PCCs) at the beginning of the lock-down period when compared with the responses to our questionnaire provided by police force domestic abuse leads.
The Association of Police and Crime Commissioners Report published in April 2020 contains the statements of 34 PCCs' priorities for protecting the vulnerable during lock-down. Twenty of these statements tally with 20 police areas covered by our data. Of these twenty, only 11 PCCs mention domestic abuse support in the broadest sense in their list of priorities in relation to such protection. This observation is interesting given the crucial role afforded to such commissioners in commissioning local victim support services including the provision of refuge spaces. It is also interesting insofar as it serves to remind us of the complex and kaleidoscopic world in which local service responses to domestic abuse are constructed, prioritised and not always in the gift of the police alone. An appreciation of this complexity adds a further dimension to what is already neither a simple nor a straightforward canvas for policing responses to domestic abuse in which our respondents operate.

Conclusion: modernisation, parochialism and responding to domestic abuse
The tensions between modernisation, parochialism and what might be thought of as the public interest situates responding to domestic abuse, at whatever point in time, within the larger context of policing more generally. The themes of modernisation and parochialism facilitate an ability to make sense of the responses reported in this paper at this point in time, As cited earlier, Winsor (2020) makes a sound case for the development of information and communication infra-structure in relation to modernisation. In many ways the findings reported above cast considerable light on the extent to which policing responses to domestic abuse under lockdown have made good use of technological developments to keep service responses in view. However, Winsor (2020) goes on to elucidate the ways in which the push towards modernisation and the pull of parochialism remain intertwined, sometimes to the detriment of service delivery. He states: With some exceptionssuch as discipline arrangements and police officer ranksforces are largely free to operate in the way they consider most suitable for their local demands. This arrangement gives police forces considerable freedom, which is generally welcome. But it can come at a cost. For example, differing procedures and practices between forces can cause operating standards and performance to become unacceptably inconsistent. Or procedures can vary so much that forces find it hard to work together when they need to, such as during major incidents and civil emergencies (Winsor 2020, p.40).
As is well-documented in ONS data, HMICFRS data and in a wide range of academic literature, the variations between police forces in their responses to domestic abuse manifestly speak to the problems of inconsistency in operating standards and performance as expressed by Winsor above and illustrated by the findings of this paper. Whilst much of the academic literature on policing domestic abuse has focused attention on the difficulties inherent in its 'incident-led' approach, the difficulties of 'seeing' coercive control, the reluctance to listen to women and so on, little of this work has framed these problems as an inherent feature of the tensions between the freedom given to police forces to respond to local demands and the costs of that freedom. Some features of those tensions have been illustrated in this paper in comparing and contrasting the responses of domestic abuse leads under lockdown and the priorities given to policing in their force areas by their local Police Crime Commissioners and carry with them the persistent risk of justice by geography in terms of service delivery for domestic abuse.
Winsor (2020) has his own recommendations for a route out of this cul de sac. These include (amongst other things) according a more central role for the College of Policing, taking another look at the role of the Home Office, reviewing the procedures for the appointment of Chief Constables and how they work with each other and other key actors within the criminal justice process. These observations interestingly return us to Shapland's (1988) observations on criminal justice fiefdoms. Moreover, as with any fiefdoms, the relations between them and within them are neither simple nor straightforward. The findings here demonstrate that in some police force areas domestic abuse leads have clearly made both good use and best use of the resources available to them during 2020 to maintain service level responses to domestic abuse. The demands on policing under the conditions of the public health crisis have been central to their ability to do this. However, it is also evident that this work has been done with different emphases in mind in different force areas. This is perhaps indicative of the extent to which the parochialism inherent in local force areas creates the space for domestic abuse leads to be entrepreneurs in their field of expertise. As entrepreneurs they play a crucial role not only in service delivery, but also in their capacity to develop and share good practice with others within their own force area and between force areas. Indeed, some of this sharing has gone on under the auspices of the College of Policing and with support from the office of the national policing lead for domestic abuse. The question remains, and this is an ongoing question for the project on which this paper is based, as to the extent to which this learning has a recursive effect on practice in the future.
To summarise, policing response to domestic abuse in England and Wales have been subjected to considerable critique over the last forty years. From the pro-arrest stance influenced by the work of Sherman and Berk (1984) to the legislation introducing coercive control as a criminal offence in 2015, much, not unreasonably, has been expected of the policing response to domestic abuse. These expectations have been sustained within an organisational policing context which has largely remained the same. The tension this poses has been expressed in this paper as the problematic tension between modernisation and parochialism. Thus police officers committed to, and concerned about, domestic abuse (in this paper domestic abuse leads) can forge a space (or not) to become entrepreneurs for their field of expertise. Unfortunately, the organisational structures of policing in England and Wales can be a barrier for such expertise to become widespread. However, it is not only policing organisational structures that can stand in the wayother fiefdoms can do this too. The global pandemic has constituted a moment in which violence(s) against women and children have risen in terms of public discourse., The response to this issue is, however, not a task for policing alone.
Notes 1. Ethical approval gained on 9th June 2020 (University of Liverpool ref. 7858). 2. Domestic Violence Disclosure Schemes (commonly referred to as Clare's Law) were introduced in 2014 and commonly referred to as Clare's Law. Clare Wood was murdered by a partner who had a record of domestic violence against women. The scheme enables women and other to ask for information about a partner's prior criminal history. For further information see Fitz-Gibbon and Walklate (2017).

Disclosure statement
No potential conflict of interest was reported by the author(s).