More lesson learning, less risk aversion in England and Wales? Prospects for the police (conduct) regulations 2020

This article assesses police lesson learning in the context of complaints. First, it discusses their incidence and accounts of complainants’ and officers' perspectives. The genesis of ‘lesson learning’ and the provisions of the Police (Conduct) Regulations 2020 are summarised. Finally, it considers the extent to which the system is likely to replace ‘blame’ with ‘lesson learning’, taking account of psychological effects such as cognitive dissonance. The conclusions are drawn that whilst emphasis on lesson learning is welcome, its application requires an improvement-focussed approach requiring investment in training to reach its potential. The effect on police risk aversion will probably be modest.


Introduction
In the last 5 years, increasingly urgent questions about the core purpose of the police service in England and Wales have been raised, from calls to defund the police to increasing their resources in response to new demands.Although the police has been an omnibus emergency service since Victorian times, crime-fighting has traditionally been at the forefront of policing and public perceptions.The 21st century has seen the rise of a 'protection agenda' following the enactment of legislation such as the European Convention on Human Rights and a series of national scandals in the arena of child protection (e.g.Home Office, 2003;Bichard, 2004).In a 2015 study of demand upon the service, the College of Policing identified that whilst conventional crime had declined in the previous decade, there had been substantial increases in reports of vulnerability-related incidents such as sexual offences, mental health incidents and child protection cases (College of Policing, 2015).
Earlier, Sir Ronnie Flanagan's 2008 Review of Policing identified that policing involved a broader remit than previously, from policy focussing on 'anti-social behaviour' and 'community safety' to managing recently released prisoners (Home Office, 2008a, Home Office, 2008b: 1).The report also cited: …..the implementation of hefty processes which are modelled on worst-case scenario.Officers and staff then struggle to apply them to volume tasks on a daily basis (Home Office, 2008a, Home Office, 2008b: 52).
Flanagan identified a reluctance to reduce processes in cases of lesser perceived risk, in what he called 'risk aversion' (Home Office, 2008a, Home Office, 2008b: 52).'Risk aversion' refers to a combination of reputational risk to forces' (Heaton, 2010: 76) and disciplinary risk to officers.A further report by ex-Police Federation Chair Jan Berry was explicit about the reasons underpinning their stance.
Officers refer to a 'blame culture' and an intolerance of honest mistakes.They fear that at some point in the future, when new information comes to light, their actions will be heavily criticised.Some feel unsupported when 'measured risks' or reasoned judgements are made, and as a result, choose to follow the rules and procedures to the letter (Berry, 2009: 61).
From 2010, police numbers fell as austerity-driven budget cuts were made.The demands of public protection, together with crime-fighting, 'became increasingly difficult to sustain' (Higgins, 2020: 7).Policing in England and Wales suffered significant cutbacks with over 21,300 fewer officers and a reduction to the overall police workforce of 44,745 employees representing an 18.3% cut on 2010 levels of staffing (Sigurdsson and Dhani, 2010;Hargreaves et al., 2018).Chief Constable Dave Thompson commented (NPCC, 2018): Bluntly our ability to manage the big threats and protect the vulnerable yet still be the traditional police the public want and need is becoming ever harder.We are in danger of pursuing efficiency to the point of ineffectivenesswhere we can process the work, but we're not detecting crime as we should be and not meeting public expectations.
A report by Her Majesty's Inspectorate of Constabulary, Fire and Rescue Services, Diverging Under Pressure, drew attention to plummeting detection rates (HMICFRS, 2020).HM Inspector Matt Parr stated that 'the failure of the police to investigate highvolume crimes like car thefts, minor assaults and burglaries was having a 'corrosive effect on the public's trust in the police' (The Telegraph, 2020).The figures for 'charge or summons' as an outcome-based measure peaked in 2009 with 22% of outcomes resulting in charges or summons, declining to 12% in 2018, 8% in 2019and 7% in 2021(FitzGerald, 2018, Home Office, 2020a, 2021).The subject of the police role was revisited in 2018 by the Home Affairs Select Committee's 'Policing for the Future' Inquiry, identifying the complaints system as a critical driver of risk aversion.On 1 February 2020, new Police Conduct Regulations came into force, accompanied by Home Office Guidance.For the first time, a detailed procedure was specified to encourage the police to learn lessons from complaints in some cases, rather than pursue a disciplinary route by seeking to allocate blame.This article aims to assess the likelihood that the lesson learning procedure prescribed by the new Regulations and Guidance will achieve greater complainant satisfaction.
The first section of the article charts the development of regulatory and oversight provisions, from the inception of the Independent Police Complaints Commission (IPCC) in 2004 until the advent of the Police (Conduct) Regulations 2020, supervised by the Independent Office for Police Conduct (IOPC).The article describes the genesis of the lesson learning approach and summarises the provisions of the Police (Conduct) Regulations 2020.Finally, it considers the extent to which the system is likely to replace 'blame' with 'lesson learning', taking particular account of psychological effects such as cognitive dissonance.

Police complaints: development of legal and supervisory provisions since 2004
The Independent Police Complaints Commission (IPCC) replaced its predecessor, the Police Complaints Authority (PCA), on 1 April 2004, following the widespread perception that police activities required more intensive scrutiny (e.g.Strudwick, 2003;Smith, 2005;Porter and Prenzler, 2012).The IPCC was given powers under the Police Reform Act 2002 and the Police (Complaint and Misconduct) Regulations 2004.The Regulations required the most serious allegations to be referred to the Commission, which had the ability to carry out its own independent investigations.Alternatively, the IPCC could return the complaint to be investigated by the force concerned, with or without IPCC supervision.In relation to less serious cases, forces were expected to initiate their own internal investigations.Where, even if proved, the conduct complained of would not justify formal proceedings against the officer, a local resolution procedure could be initiated.Rights of appeal to the IPCC against police decisions in relation to complaints were created throughout the process, including the outcomes of local resolutions (Schedule 3, Police Reform Act 2002).The Police (Conduct) Regulations 2008 introduced the concepts of 'misconduct' and 'gross misconduct' (Regulation 5), the latter being conduct that would potentially attract a sanction of dismissal.
Earlier legislation, Section 87 of the Police Act 1996, had empowered the Home Office to issue detailed procedural guidance concerning how complaints should be dealt with, and it did so following the findings of the Taylor Review into police discipline (Taylor, 2005).The 2008 Guidance required that a 'severity assessment' determine the category into which the alleged conduct would fall (Home Office, 2008a, Home Office, 2008b: 31).The guidance also introduced the concept that a purpose of an investigation, was to identify any learning for the individual or the organisation (Home Office, 2008a, Home Office, 2008b: 36).Timeliness was also a concern.The guidance urged a 'proportionate and balanced investigation', adding that 'a frequent criticism of previous misconduct investigations was that they were lengthy, disproportionate and not always focussed on the relevant issues' (Home Office, 2008a, Home Office, 2008b: 36-37).Subsequently, these points were reiterated by revised guidance documents in 2012, 2014 and 2018, suggesting that they remained subjects of potential concern (Home Office, 2012Office, , 2014Office, , 2018a)).The 2014 document also stressed the need for the severity assessment to be 'demonstrably proportionate and fair' (Home Office, 2014: 25) and reiterated the point in its 2018 revision (Home Office, 2018a).
Section 33 of the Policing and Crime Act 2017 renamed the IPCC as the Independent Office for Police Conduct (IOPC), which came into being in January 2018.This move represented the 'culmination of a significant 'change programme' since 2012 for what was the IPCC' (Torrible, 2020(Torrible, : 1119)).The IOPC was given the power to initiate its own investigations into incidents without them being referred to it by a police force.The replacement of 12 Commissioners with a single executive head supported by a Board that included six non-executive directors was intended to speed decision-making and improve accountability (Section 33, Policing and Crime Act 2017).In order to increase the IOPC's independence, investigations would, when feasible, be undertaken by either a police force or by the IOPC as binary options.Intermediate 'managed' and 'supervised' investigations were replaced by 'directed' investigations which would be carried out using police resources, but in which the IOPC would make the key decisions.The presumption was that independent investigations would take place wherever possible (Home Office, 2016).
These national-level developments have played an important strategic-level role in setting out the 'rules' governing the complaints system.However, they have little to say about 'ground level' public and police perceptions of its operation, which determine views of its legitimacy.We will now attempt to create a fuller picture by considering complaints procedures in the context of their frequency and the reactions to them by the public and the police.

The incidence of complaints against police
In 1990, Egon Bittner described policing as responding to: 'something that ought not to be happening and about which someone had better do something now'.He further argued that the police are defined by their unique capacity, if persuasion fails, to use force (Bittner, 1990).Waddington argues similarly that policing involves 'the ubiquitous exercise of coercive authority' founded upon the monopoly of legitimate force (Waddington, 1999: 20).The level of confrontation in the events leading to a complaint against the police is likely to be far higher than a complaint about unsatisfactory retail consumer goods and services (Chapman, 2014, para 2.34, Waddington, 1999: 171).The scale of such confrontations is considerable.For example, England and Wales saw 563,837 stop searches (Home Office, 2020c) and 640,000 arrests for notifiable offences (Home Office, 2020d) in 2019-20.
In this context, complaints against police officers are nonetheless relatively uncommon events.In 2019-20, 28,223 complaints were recorded in England and Wales, a figure which has been reasonably stable for over a decade.The 122,404 police officers (Home Office, 2008b) could each expect to be the subject of a complaint, slightly more than once every 4 years.Although not all roles are public facing, the incidence is relatively low given the contested nature of policing.
However, the relatively low number of complaints cannot be simply equated with a low level of malpractice by police.A study by Smith (2009) pointed to persistent British Crime Survey results which showed that about 80% of people who were 'really annoyed' with the police, did not complain.The biggest single reason was that those concerned 'couldn't see any benefit' (Smith, 2009).

Public concerns
Academic descriptions of the complaints system have expressed persistent concerns for the imbalance in power between individuals and the police as state agents.The 'enduring problem' (Smith, 2005) has been identified as inadequate police accountability, particularly low levels of complaint substantiation and complainant satisfaction.The proportion of complaints that were upheld in 2019-20 was 11% (IOPC, 2020).The 2006-07 British Crime Survey found that of 100 complainants, 82 were dissatisfied with how their complaint had been dealt with (Grace and Bucke, 2009).Later, a smaller-scale study of 25 'local resolution' complainants found that almost two-thirds were dissatisfied with the outcome of their complaint (Hearnden and May, 2013).Attention has been drawn to potential bias in investigations conducted by the police, the low visibility of many policepublic interactions and inadequate supervision, both within the police and by scrutiny bodies.Several works review the evidence underpinning these problems more fully than is possible here (e.g.Waters and Brown, 2000;Sanders and Young 2008;Strudwick, 2003;Seneviratne, 2004;Smith, 2013;Torrible, 2020).
Although these concerns remain undiminished, some later studies have exposed some of the difficulties of reform.A nominally independent system established to enhance effectiveness and legitimacy would probably require police resources and support, for example, for scene preservation and the interpretation of forensic results (Savage, 2013).This dependence, together with the employment of ex-officers, could lead to attitude 'capture' whereby investigations would be insufficiently rigorous, for example, by treating officers as witnesses rather than suspects (Porter and Prenzler, 2012;Smith, 2013;Savage, 2013).Moreover, the terms 'effective' and 'legitimacy' are themselves open to variations of interpretation and measurement difficulties (Torrible, 2018), making the achievement of success potentially elusive.
There is little evidence that disquiet about complaints is reflected in general public confidence measures of the police.Such measures encompass a wider range of issues than the complaints system and are a result of complex factors such as 'expressive concerns regarding disorder and informal social control, as well as concerns about long-term social change in terms of belonging, trust and shared values' (Sindall et al., 2012: 747).It is not unreasonable to suppose that anxieties about impartiality are experienced more keenly by those who make a formal complaint, contributing towards high dissatisfaction rates.
However, public concerns that the complaints system may be weighted against them are not in the main, reflected by complacency on the part of officers.Instead, they share the public's perceptions that the complaints and discipline system may be unfair to them (May et al., 2008: 40).

Police concerns
Fear within the police service that complaints investigations may be biased in favour of the complainant, are particularly prevalent in respect of alleged 'neglect of duty' in the wake of 'risk' incidents involving potential danger to life (Black and Lumsden, 2019; Thomas and Forrester-Jones, 2019).'Neglect of duty' has become by far the greatest category of complaint over the last decade, comprising 41% of all complaints in 2019-20 (IOPC, 2020).It is now a significant driver of risk aversion (Heaton et al., 2019).
A study of call handling in police control rooms in 2016-17 found that risk aversion retained a significant role.The authors found that 'official discourses towards encouraging officers to be more confident in their decision-making………..have not filtered through to front line policing' (Black and Lumsden, 2019: 67).If officers were not sent to an incident, the fear of disciplinary investigation following a negative outcome affected decision-making.Resources were in the main, used in a precautionary rather than an objectively risk-assessed manner (Black and Lumsden, 2019).
In relation to public order, the College of Policing's guidance states that '…making risk decisions has become associated with blame, fear, internal and external inquiries and, therefore, something to be avoided (College of Policing, 2018).
Elsewhere, research interviews examining officers' experience of dealing with mental health incidents found similar attitudes.Using 2015 data and interviews of officers, Thomas and Forrester-Jones examined the reasons underlying persistently high levels of mental health detentions, in the face of efforts by forces and partner agencies to reduce their incidence.They found that The overwhelmingly strong view shared by 15 of the 17 interview participants (and recognised by the other 2) is that officers operate in fear of a 'death in police contact' which results in highly risk-averse behaviour and a high degree of compliance with force policy, even where the officers judge this is inappropriate' (Thomas and Forrester-Jones, 2019: 143).
Officers worry that conclusions may be drawn with the benefit of hindsight and be followed by unduly rigorous decisions by disciplinary authorities (Black and Lumsden, 2019; May et al., 2007;Waddington 1999).Moreover, there has been a specific perception that malicious complaints may be accommodated too readily (Chapman, 2014;Warburton et al., 2003).A survey of 30 police officers and staff who had been independently investigated found that 19 had 'a little less' or 'a lot less' faith in the complaints system after the process had been completed.The biggest single concern was that the IPCC was perceived to be '"seeking a result", not conducting an objective investigation' (May et al., 2008: 41).
In 2019, the BBC reported that in the previous 3 years, two-thirds of the gross misconduct cases held at the direction of the IPCC were found to be 'not proven'.Although this is a substantial proportion, it is problematic to contend that it is too high.The IOPC stated that its approach was 'appropriate', on the basis that the legal threshold for taking cases forward was the 'case to answer' test, which is relatively low (BBC, 2019).Moreover, the mission of the IOPC is to 'improve public confidence in policing by ensuring the police are accountable for their actions and lessons are learnt' (IOPC, 2018b: 4).From this perspective, the visibility of accountability might be viewed as of equal importance to its outcome.From a police perspective, it might appear to be a heavy-handed approach.Overall, the evidence points to shared and persistent police and public perceptions of insecurity about the objectivity of the complaints and discipline system.
In principle, 'lesson learning' could encourage officers to more readily admit to mistakes, promoting more positive dialogue, greater complainant satisfaction and confidence of all of those involved.We will now consider how the approach came to prominence and how well it might 'fit' into the complaints arena in practice, taking into account the process prescribed by the new Regulations and the nature of many policepublic interactions.

The rise of the 'lesson learning' agenda
The notion of 'lesson learning' is not unique to the police, having gained traction elsewhere in the public sector and offering a positive way forward for all parties.Whilst the term was not defined, it was made explicit by Home Office Guidance in the context of police complaints, that it encompassed both organisational learning and that which might be applied to individuals (Home Office 2020d: 25).In 2016, it was the subject of a speech by the Health Secretary to the Global Patient Safety Summit in London (Department of Health and Social Care, 2016).Arrangements in the airline industry were held as an exemplar of the right way forward.The Health Secretary noted with approval the ideas expressed in Matthew Syed's 2015 book Black Box Thinking (Syed, 2015).Although Syed did not define the term 'black box thinking', his thesis was that mistakes are a rich source of learning opportunities.He saw the aviation industry as a rare example of the application of this concept.Syed contended that the investigation approach to accidents and near misses could identify underlying problems.For example, in a particular aeroplane, the side-by-side location of two levers with entirely different functions had contributed towards pilot error.Further intervention and redesign reduced the opportunity to make mistakes, and aircraft safety had improved.
In contrast, the reaction in most organisations was the simplistic blaming of the individual concerned.Awareness of this culture led to euphemisms such as 'unexpected outcome' and the deliberate covering-up of mistakes.According to Syed, a 'growth mindset' was required to be unafraid of mistakes and initiate incremental improvement (Syed, 2015).Syed recommended using aviation's approach in healthcare but cautioned that 'it is also more complex……for example there are 300 models of the surgical pump but just two models of long-distance aircraft' (Syed, 2015: 63).He could perhaps have added that flying an aircraft is predominantly a series of interactions between a pilot, mechanical systems and fixed procedural protocols.The fact of an accident is obvious, and a 'near midair collision' is closely defined and identified by radar.At this point, it might be surmised that policing, on the contrary, consists in the main of interpersonal, frequently adversarial interactions, raising the question of whether 'black box thinking' is readily transferable from aviation.Examples of the challenges of adversarial events including Bloody Sunday Inquiry and Hillsborough Investigations, where responses from the fallout of the tragic events failed to satisfy those affected.We will consider the transferability of the approach of aviation later.However, the concept of lesson learning presented a useful and constructive approach that offered 'intrinsic appeal' (Torrible, 2020(Torrible, : 1120)).
The IOPC Director General, Michael Lockwood, told the Police Federation's annual conference in May 2018 that he supported using genuine mistakes as opportunities to learn, with disciplinary sanctions being reserved for the most serious cases (Lockwood, 2018).In June of that year, a Police Foundation/KPMG report How Do We Move From a Blame Culture to a Learning Culture in Policing?addressed the topic, and again, arrangements in the airline industry were viewed favourably.Contributed to by representatives of the police, academia, police scrutiny bodies and others, the report identified several contributors to 'blame culture'.These included historic police culture, a strong regulatory framework including the disciplinary system, and perceived public expectations.The report made several recommendations aimed at replacing the dominant 'blame culture' with one of learning from mistakes, in common with the aviation and nuclear industries (Police Foundation, 2018).Such a move was unlikely to be easy.A perceived element of police culture is resistance to change, and the move to a learning culture has been shown elsewhere to be a taxing process.The resistance to the police professionalisation agenda is an example of the challenges of the police cultural change of shift to a reflective learning culture (Chan, 1997(Chan, , 2003;;Wood, 2020aWood, , 2020b)).
The Home Affairs Select Committee Policing addressed the same subject in its Policing for the Future report, published in October 2018.The various contributions of the Police Federation, Police Superintendents' Association, National Police Chiefs' Council (NPCC) and the Association of Police and Crime Commissioners (APCC) were united in advocating a less personal blame-seeking approach in order to promote the learning of lessons and service improvement.The Committee's report identified that: 'Strikingly, policing witnesses were almost united in lamenting a risk-averse "culture of blame"' (Home Affairs Committee, 2018, para 211).
The police conduct regulations 2020a paradigm change?
The rise to prominence of the lesson learning agenda was reflected in the Police (Conduct) Regulations 2020 which introduced a formalised process for the first time.In addition, the accompanying Home Office Guidance attempted to set a new tone, being explicit that henceforth, 'misconduct and performance procedures should be dealt with at the lowest appropriate managerial level' (Home Office, 2020b;para. 1.5).
In order to encourage the use of the new processes, the guidance emphasised that 'the procedures are intended to encourage a culture of learning and development for individuals and the organisation' (para 2.9) and that 'the police service will support any manager who has exercised their judgement reasonably and adhered to the guidance provided' (para 2.10).Disciplinary proceedings should only be applied to 'serious allegations that justify a formal disciplinary outcome, at a minimum a written warning' (Home Office, 2020b;para. 4.28).The initial 'severity assessment' would continue to determine whether the alleged conduct amounted to misconduct or gross misconduct.If it was neither, the Home Office Guidance stated that the complaint could be handled 'in any reasonable and proportionate manner', including the use of informal action by line management or the taking of no further action (para 4.12).'Practice Requiring Improvement' (PRI) was identified as the conduct of a minor disciplinary nature that previously had outcomes of management advice or no action.It was defined as 'underperformance or conduct not amounting to misconduct or gross misconduct, which falls short of the expectations of the public and the police service as set out in the Code of Ethics' (para 4.44).Furthermore, the Guidance stated that 'the principal focus of following this process is to learn and to develop by improving from mistakes, poor judgement and low-level wrongdoing through early intervention' (para 13.10).Lesson learning remained inextricably linked with substandard officer behaviour.PRI would trigger a 'Reflective Practice Review Process' which would be led by officers' line managers ('reviewers') (para 4.45).The process was intended to be constructive and where possible, restorative (para 4.52).
Reflective practice would consist of fact-finding and discussion stages, followed by a 'reflective review development report'.The discussion stage would identify 'key lessons to be learnt by the participating officer, line management or police force concerned' (para 13.58).The report would identify actions to be taken, including timescales which would be followed by a review and would form a part of the officer's annual performance and development review.In the event of an officer failing to engage with the process, the reviewing officer could refer to the failure for re-assessment in relation to the severity test.Complainants were allowed the right to appeal against the process, in most circumstances to the Police and Crime Commissioner for the force (Home Office 2020b; para 13.87).
We will now consider in more detail how well the procedure might work in the policing environment.
Learning not blame: realistic or optimistic?
In the face of challenge, the notion that the police should uncritically and routinely accept their actions as mistakes to be remedied is less than obvious.In other contentious environments such as political voting or a decision to be made in an appeal court, those taking part may reach entirely different conclusions based upon the same information and after extensive debate.There is no expectation that those on the 'losing' side will express contrition or plan some form of self-improvement.Instead, differences in perspective are accepted as the primary cause of disagreement by all parties who move on to other business.Where lesson learning is sought as a means to individual and organisational improvement, a primary requirement is for the identification and acceptance of clear error.Whether Syed's 'black box' approach could be transferred to police complaints depends upon whether the psychological dynamics of an interaction between two people is similar to interactions in the aviation world.Syed contends that 'we progress fastest when we face up to failure and learn from it' (Syed, 2015: 282).Identifying misconduct or a mistake is a prerequisite for 'Practice Requiring Improvement' under the Regulations.On the face of it, the outcome of this requirement is that the lesson learning process depends upon whether such behaviour can be readily identified and whether an agreed view of events can be reached.We will consider these in turn.
Identification of misconduct and mistakes.Police actions are products of many considerations, including the law, force and national policy together with officer discretion, which may include factors such as the nature of an interaction with individual and external pressures.This complexity means that 'police wrongdoing' can be a highly contestable concept.Ambiguity can persist to high levels within the disciplinary and criminal court systems.For example, a Kent custody officer was alleged to have struck a violent and disruptive prisoner on the arm, to make him withdraw it through the cell flap.No injury was caused.The officer was charged with assault and convicted at court (IOPC, 2018a).The verdict was later overturned by the Appeal Court which stated that it could not be sure the use of force was unreasonable.The officer was then referred to a disciplinary panel that concluded that no misconduct had occurred (Kent Online, 2018).
Whilst mistakes and misconduct are readily identifiable in some instances, there are others where the position is beset by ambiguity and different but firmly held perceptions in reaction to the 'necessarily contested nature of policing' (Torrible, 2020(Torrible, : 1129)).We will now consider why people are reluctant to change their initial perceptions and how this may affect a lesson learning process.
Reaching an agreed view of events.When a complaint is made, police officers and complainants are likely to find themselves in a position that is unfamiliar and uncomfortable and emotionally charged.The complainant may feel that their rights have been violated, whilst the officer may believe that no misconduct has occurred and that his or her livelihood is at stake.In these circumstances, people tend to become defensive and are loathe to change their views of events.Risk consultant David Ropeik illustrated the point.
You argue the facts, as thoughtfully and non-confrontationally as you can, but the facts don't seem to get you anywhere.The wall of the other person's opinion doesn't move.They don't seem to want it to move (Ropeik, 2010, online article).
Psychological terms for this effect include 'cognitive dissonance' (e.g.Festinger, 1957), 'confirmation bias' (e.g.Nickerson, 1998), 'motivated reasoning' (e.g.Westen et al., 2006) and 'motivated memory' (Conway, 2005).In brief summary, Festinger's theory of cognitive dissonance proposed that people who experience internal psychological inconsistency feel stressed when their belief system is challenged by new information.They reduce this dissonance either by justifying the behaviour that stemmed from their belief or denying contrary information and believing what they want to believe, amounting to 'confirmation bias'.This is psychologically easier than altering underlying beliefs (Festinger, 1957).Further research suggests that people experience 'aversive arousal', which motivates the affirmation of control beliefs following violations of personal control (Proulx et al., 2012), with 'exaggerated zeal' (McGregor, 2010: 301).
Nickerson considered confirmation bias: … to be sufficiently strong and pervasive that one is led to wonder whether the bias, by itself, might account for a significant fraction of the disputes, altercations and misunderstandings that occur among individuals, groups and nations (Nickerson, 1998: 175).
In a review of the subject, he found that the bias could be 'motivated' or 'unmotivated' depending on whether or not a personal interest was in dispute.Motivated bias increased in strength, with the degree to which the evidence related directly to a dispute in which a person had a personal stake (Nickerson, 1998: 175-177).Moreover, a series of experiments by Gal and Rucker found that people whose confidence in closely held beliefs was undermined engaged in additional advocacy of their beliefs.In other words, people tend to dig in their heels when challenged (Gal and Rucker, 2010).
These theoretical studies, which raise serious doubts about the likelihood of reaching agreed views of complaints in the context of policing, are mirrored by practice.Waddington (1999: 169) cites an earlier study that found that often, in the absence of independent evidence, 'each party flatly contradicts the other' (Skolnick and Fyfe, 1993).That finding was supported by a later fieldwork study of 64 complainants and 76 police officers who had taken part in local resolution (LR) processes (May et al., 2007).Slightly over half of both complainants and officers were 'dissatisfied' or 'very dissatisfied' with the outcome.Officers tended to believe that the system favoured complainants and failed to exonerate their actions.Complainants were dissatisfied with end results, for example, the lack of an apology (May et al., 2007).
Similar research several years later was carried out on behalf of the IPCC, in relation only to complainants but not limited to LR cases (Hearnden and May, 2013).The proportion of dissatisfaction was found to be about two-thirds of the sample size of 25 respondents.The authors noted that: For many complainants, satisfaction with the outcome tends to hinge upon whether the expectations and aims they have at the outset are realised……….Whilst the professionalism of the IO (investigating officer), the speed with which the process reaches its conclusion and the clarity of the information provided are all important components, these tend to pale into insignificance if the outcome is not what the complainant expected or wanted to achieve.(Hearnden and May, 2013: 10-11).
In their conclusions, they added that: 'Complaints suitable for LR tend to be about perception…………..The view of the officer is almost always diametrically opposed to that of the complainant' (Hearnden and May, 2013: 18).
The provisions for a lesson learning process seem to be broadly similar to trials of restorative conferences concerning informal resolution processes in the Thames Valley force area in 2002-3.These were compared with the results of traditional processes in Hampshire.The study found that overall, complainant satisfaction was 'moderately better' in relation to restorative justice, in particular where a trained, experienced facilitator was used and officer misconduct was admitted.However, this amounted to a minority of cases.For example, officers admitted to some kind of wrongdoing in 19% of cases that involved a restorative meeting, compared with 9% of officers in cases dealt with by the traditional informal resolution process (Young et al., 2005).
The conclusions from research studies are consistent and clear, having significant implications for a lesson learning approach to resolving complaints.Unless the circumstances are straightforward and uncontested, the identification of 'mistakes' may be an uphill task.According to Hoyle and Young concerning the Thames Valley initiative, 'Police are being asked to contemplate taking on the role of a quasi-offender rather than, as under the traditional complaints system, the role of an accused' (Hoyle and Young, 2003: 697).There is no such requirement on the part of complainants.This may produce a process that may be perceived as somewhat one-sided by officers and undermine their authority even if it is not formally about the dispensation of justice.

Discussion and conclusions
A key aim of the complaints system is the maintenance of police discipline (Waters and Brown, 2000), and as a deterrent, it may be more successful than it is sometimes allowed credit for.Although far from representing all dissatisfaction, complaints against police are relatively rare.
Persistent concerns of complainants are mirrored by anxieties within the police about the disciplinary system.This may be expressed as risk aversion in decision-making, particularly in potential 'neglect of duty' cases.Despite repeated Home Office advocacy of lesson learning, the tendency has been to invoke formal investigation processes, even though the substantiation rate remains low.This is not surprising within the context of the inquiries, reports and regulations which have been given the subject a high public profile over the last 20 years and in the main, tightened accountability.
The complaints process could arguably be seen as dealing with a spectrum of interactions between the police and the public.At one end of the spectrum, police wrongdoing can be clearly identified, and a view may be reached that the complaint is upheld.If allowed by the severity assessment, there is potential for the lesson learning process to be initiated and for the police to become a more improvement-focussed service.In this respect, the emphasis of the new regulations upon timely resolutions at the lowest appropriate level seems to be a move in the right direction.At the opposite end of the spectrum are complaints that are unwarranted.If there is clear evidence, these may be completed as 'not upheld'.In the middle ground are cases where the evidence is indecisive.The prevalence of these cases, often dominated by different perceptions and cognitive dissonance in the form of both officers and complainants wanting to be vindicated, may be substantial.
The Home Office Guidance appears to rely upon the process of reflection and discussion to overcome cognitive dissonance, and it is not clear how successful efforts will be in the face of a commonplace and profound psychological effect.Much may depend on the ability of the reviewing line manager to decouple lesson learning from wrongdoing and encourage the taking of responsibility in place of blame.The Chapman Report stated: 'what should we do about it' is often a better response than 'did he (or she) do it' (Chapman, 2014, para 5.1).This amounts to the adoption of a forward-looking rather than a backwards-looking approach.In that respect, the Guidance's explicit association of the lesson learning process with 'mistakes, poor judgement and low-level wrongdoing' (para 13.10) may be unhelpful.At best, a substantial training effort may be required to achieve officer buy-in to the process and greater complainant satisfaction.At worst, the process may become or appear to be a tokenistic one that is satisfactory to neither police nor the public.
Concerning the wider problem of risk aversion in policing, it is unlikely that the 2020 Regulations will give the service the confidence to manage workload by according a lower priority to some types of work.The provisions for the most weighty complaints cases, such as deaths following police contact, are substantially unchanged.The outcomes of disciplinary and court cases can turn on the perception of small details.In order to reduce police risk aversion, substantial legal protections would be required.These would require profound changes of political and public awareness and are not on the foreseeable public agenda.
The function, role and practice of police complaints is a central component to police engagement with the public, but the context in which it operates is inevitably linked to resource, regulation, role and culture.The importance of a balanced complaints system, a learning culture and a police service structured and resourced appropriately is crucial to sustaining police legitimacy at a time of significant change and reform.

Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/ or publication of this article.

Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.