Police Discretion PDF

Cues What is police discretion? The **unwritten rule** that police officers have the right to be **selective** in how they do their jobs as long as they stay within widely prescribed departmental guidelines -deciding which rules to apply -deciding whether to apply them Why police use discretion?...

Cues What is police discretion? The **unwritten rule** that police officers have the right to be **selective** in how they do their jobs as long as they stay within widely prescribed departmental guidelines -deciding which rules to apply -deciding whether to apply them Why police use discretion? different types of crime/situations require different rules/judgements -there is no guidebook for every specific incident police will always need to apply discretion Police discretion = police judgement (all practices require discretion) Factors influencing whether police invoke criminal process **Criminal law**: vague definitions, conflicting explanations of criminal law Criminal law used for social and medical problems -the law can be hard/confusing to apply, is vague, short, changes all the time -decriminalisation / conflicting opinion on applying laws **Police environment:** limited supervision of officers, private encounters **Limited resources**: myth of full enforcement Recording crime, stops and searches, collecting evidence, making arrests, proactive investigations (e.g., investigation into increase in burglaries), case finalization (when do they close a case) **Every police decision is discretionary** -every decision requires judgement **Situational features of the incident:** -seriousness of crime, strength of evidence, -victim preference, relationship between victim and suspect, -suspect demeanour and condition, victim characteristics, reputation -number, volume and aggression Immediate work environment: -immediate environment = location, time -force characteristics = specific/formalised policy, professionalism/bureaucratization -officer characteristics = race, gender, education "The exercise of discretion lies at the heart of the policing function. It is undeniable that there is only one law for all; and it is right that this should be so. But it is equally well recognized that successful policing depends on the exercise of discretion on how the law is enforced. Discretion is the art of suiting action to particular circumstances." -Lord Scarman 1981 Police discretion is the defining characteristic of police work -is concentrated in the lower ranks of the police organisation ---many police-citizen encounters ('Street-level bureaucrats' - Lipsky 1980) some discretion is necessary/good - full enforcement is unattainable CJS lacks resources and time for full enforcement Examples of discretionary decision -people involved don't want to press charges -people call the police to diffuse a situation/for police presence instead of to invoke CJS :police may choose not to arrest Problems with Uncontrolled Discretion -denial of equal protection under the law -poor police-community relations -poor personnel management -poor planning and policy development Discretion needs to be guided and structured (not abolished) Cannot be abolished because features that cause it cannot be abolished/changed Who does policing-related research? -changes the motive of research Questions of policing research: Answers -university academics/students -national governments, local governments -the police service -national agencies (College of Policing) -independent commission for Complaints Against Police) -independent research institutes -commercial consultancies -Think Tanks -campaigning/third sector groups what do they do? how effective are they? what do they think? what is the relationship between police and public? how are they presented in media? how accountable are the police? EBP Lawrence Sherman -set up Cambridge Centre for EBP Three R → Triple T Targeting -what is the nature of crime -hot spots, repeat offenders, repeat victims Testing -does it affects crime Tracking -outcomes, outputs Policing Domestic Violence Examples: - Minneapolis Domestic Violence Experiment (field-experiment) Which police response is most effective for domestic assault? Experimental group -those incidents where officers made an arrest or separated the parties Control group -those incidents where officers offered advice or mediated the situation Dependent variable -re-offending by domestic violence perpetrators Reoffending was less when arresting the perpetrator Critique: Small sample; random assignment; short follow-up time; failure to replicate; differential impact based on offender's employment status - The Domestic Violence Arrest Decision (cross-sectional) "The Domestic Violence Arrest Decision: Examining Demographic, Attitudinal, and Situational Variables" Which type of variable will exert the most influence on an officer's decision to make an arrest at a domestic violence incident? -demographic -attitudinal -situational Situational variables provide the most explanation for the arrest decision. -Witnesses present (1.79) -Victim/suspect cohabitating (2.02) -Last hour of shift (-1.32) Why arrest? -There are other important police activities and decisions to study What about organisational factors? -None were included in the multivariate models (e.g. time, resource levels) Pseudo R² -Typically very low in social science research Cross-sectional versus longitudinal research -The 'chicken and egg' problem - Risk-led policing and the DASH risk tool (mixed methods) Domestic Abuse, Stalking and Harassment and Honour Based Violence (DASH, 2009) Risk Identification and Assessment and Management Model National Mapping Exercise -analysis of 43 forces to describe their approach -aimed to identify different approaches or models of risk-led policing -three forces representing three different models chosen for fieldwork Fieldwork in three forces -61 interviews with police and partners -120 hours of observation -1296 online survey responses -analysis of 2000 domestic abuse incidents Although the DASH 'concept' is valued, the operational delivery of it is being questioned. -Quality and completeness of DASH is a concern to both police and partners. Police understanding of domestic abuse and risk largely informed by physical violence (little understanding of coercive control). Critique: process rather than outcome evaluation; focus on IPV rather than DVA more broadly **Cop Culture** **Lecture** Cues What shapes the pattern of policing? What is police occupational culture? How has cop-culture been researched? Answers **Rational-legal** view of policing -hierarchical, disciplined police force that enforces the law impartially -police tasked to enforce the law and they did, nothing more to understand -dominated until 1970s Increasing sociological interest how '**law in action**' deviates from '**law in books'** Wide **discretion** of frontline policing -no full enforcement, finite resources, vague laws Social, economic and political processes -law is only a small factor that shapes policing - usually vague Early sociological police research moved away from 'rational-legal' perspective in explaining police conduct/misconduct -focus on role of **occupational cultures** on behaviour Different beliefs of what 'cop culture' is: The way police officers view the social world and their place in it -Reiner 2010 A distinctive list of **commonalities** of police **outlook**/similar patterns of **understanding** observed in police forces around the world A set of **beliefs, values, working practices** and **informal rules** shared between groups of working police officers 'Canteen culture' - **values** and **beliefs** in **off-duty socialising** 'Cop culture' - **norms and working rules** expressed and implied in course of **police work** -Waddington 1999 - **The Ethnographic Tradition** -Immersive fieldwork in policing -'Classic' studies (Skolnick 1966, Justice without Trial) -'Participant observation' by 'insiders' (e.g., Holdaway, Inside the British Police (1983)) -Non-participant observation by non-police researchers (e.g., PSI, Police and People in London (1983); Loftus, Police Culture in a Changing World (2009)) **Advantages** Immersion in back/front-stage worlds of policing Deeper/richer cultural understanding Direct observation of policing (\> post hoc police accounts) -first hand observer accounts Allows ongoing trust relationships with researchers **Limitations** Ethical challenges (informed consent) Practicalities - gaining access and time commitment The 'Hawthorne Effect' (knowledge of being observed) Going native (becoming like the police)/researcher bias Representativeness (depth against breadth) Replicability - **Qualitative Interviews** E.g., Roger Graef, Talking Blues (1990); Robert Reiner, Chief Constables (1991) **Advantages** Allows in-depth examination of account of key values/perspectives Richer accounts as researcher can probe/follow up First hand accounts of policing experiences/practices **Limitations** Researcher effects 'Sanitized' accounts for external audience Retrospective framing of police accounts - **Quantitative Surveys** E.g., Cordner (2017) Gutschmidt and Vera (2020) **Advantages** More objective, standardized measures Allows some statistical testing Generalization from representative samples to wider population **Limitations** Some aspects not amenable to quantitative measurement Reduction of complex ideas to a set of individual characteristics Validity of self-reported measures Recruiting a representative sample is difficult **Key Characteristics of 'Cop Culture'** - Sense of Mission - 'Not just a job but a way of life', action-orientation, 'noble, crime-fighter' - Cynicism & Pessimism bleak view of human nature - Suspicion suspicion of 'difference', particular groups or places - Isolation & Solidarity 'us versus them' mentality - Machismo assertiveness, aggression, physical strength, bravery, suppression of emotions - Prejudice sexism, racism, homophobia - Conservatism law and order, preserving status quo - **Where does cop culture come from?** Selection hypothesis Police culture reflects the **individual personality** characteristics of people who apply to and are accepted into the police Socialisation hypothesis Police culture is a **complex collective phenomenon**, generated by fundamental **nature of policework** Null hypothesis There is **no distinctive set** of 'police' personality characteristics, police officers are the **same as wider population** - **Individual Personality Traits** Most UK scholarship has tended to argue that police recruits' beliefs/attitudes largely **reflect** **wider population** (Bowling et al. 2019) (null hypothesis) But US research has suggested that police recruits demonstrate **distinctive psychological/personality** characteristics from wider populations (selection hypothesis) (TenEyk 2024) Recent official reports have also included some elements of the '**individual deviancy**' model (e.g. Casey Review 2023) (selection hypothesis) - **Structurally-shaped Phenomenon** - Nature of policework e.g., Skolknick's three features shaping 'working personality' of police officer -unique coercive authority -risk/unpredictable situations -pressure for results - Interactions with people -hostile encounters with particular groups -certain groups constructed as 'police property' - Socialisation -policing as a craft - learning on the job **How aspects of cop culture are shaped by police function and structural contexts of policework** - Mission -action-based/heroic narratives give an acceptable gloss to what is often a messy, difficult, depressing and mundane job - Cynicism/Pessimism -dealing with the raw end of society's problems - Isolation -coercive powers, unsocial hours, public hostility - Solidarity/In-group loyalty -risk/danger, need to rely on back up - Suspicion -emerges from crime control and order maintenance function - Machismo -traditional emphasis on stereotyped 'male' attributes of toughness, strength, violence - Prejudice -collective experience of hostile relations with particular segments of society - Conservatism -preserving established order, rule-based militaristic organisation **Problematizing 'Cop Culture'** Waddington 1999,2012 -'cop culture' no explanatory power on behaviour, not a useful concept Link between Cop Culture and police behaviour -'saying' versus 'doing' -(Waddington 1999, Policy Studies Institute 1983) -Social psychology research on attitudes/behaviour - 'good' people can do 'bad' things and vice versa (e.g., Milgram Prison Experiment) -disconnect between how people talk and what they do Cop culture as a distinctive police phenomenon -so called 'police' characteristics/values shared with wider population groups -Waddington 2012 Cop culture as universal, monolithic and homogenous -variations in cop cultures - by rank, by specialism, by force, by officer characteristics -Bowling et al. 2019 -major societal changes since 'classic' studies of 1970s/80s - **Changing Cop Culture & Reform** Standard Responses 1. Recruitment -targeted campaigns, recruitment targets, improved selection procedures, direct entry -problems - the reduction of complex cultural traits to 'problematic individuals' 2. Training -increased entry standards, formal training to address problematic attitudes, 'ethical' emphasis on training and development Wider Approaches 3. 'Micro level': individual accountability -rendering frontline policing 'more visible' via technology, more/better supervision -disciplinary codes and complaints mechanisms -'rule tightening' to restrict discretion (e.g., Police and Criminal Evidence Act 1984) 4. 'Meso level': Policing policy -community-based or 'problem-solving' approaches -organisational accountability (e.g., elected Police and Crime Commissioners) 5. 'Macro-level': The societal context of policing -tackling broader patterns of inequality: social and economic policies outside field of policing **Disorder, Crime and Policing** - **Zero-tolerance Policing** - **'Broken Windows' (Wilson and Kelling, 1982)** - **The effectiveness of police patrol** \*\*-\*\*1970s/80s research suggested little impact on crime rates (randomly patrolling police rarely come across a crime) =Police resources put into emergency response/serious crime compared to random patrolling - **Links between disorder and crime** growth of low level disorder/crime argued due to lack of visibility of police officers -Disorder/incivilities raise the fear of crime (despite personal risk of victimisation) =Law abiding people avoid public spaces/leave the area =Erosion of informal community controls, less natural surveillance fear of public - don't want to interfere =Graduation from incivilities to serious crime encourages crime from people who might not normally commit a crime - **Spirals of decline** =Communities reach 'tipping points' after which disorder and crime spiral out of control and feed off each other - **Community controls and order-maintenance policing** how to fix disorder/crime spiral -'Broken Windows' policing, 'order maintenance' policing, 'quality of life' policing -Police should act to 'reclaim' the streets -Clamp down on incivilities/disorder (things that may not be criminal) -Trigger '**virtuous circle**' of reduced disorder, reduced fear of crime, more informal social control (via citizen confidence, greater use of public space) and eventually reducing serious crime rates - **Concerns** - Empirical evidence for disorder-crime relationship -**No clear causal connection** demonstrated (Harcourt 1999) -"Disorder" and "crime" both related to more fundamental conditions in deprived neighbourhoods -Crime reduction effects may be due to **traditional deterrence** (greater police presence) or incapacitation (offenders caught and imprisoned), NOT the 'developmental sequence' suggested by BW -**Varied findings** depending on style of policing intervention (e.g. intensive law enforcement v community policing) -Ongoing debates about BW mechanism and the effects of 'disorder policing' on crime and insecurity (see Weisburd *et al* 2015 for a summary) - Conceptual concerns -Simplistic dichotomies -- "Respectable/problematic" populations; "orderly/disorderly" behaviour -Disorder/incivility as a political/social construction what is disorder/incivility is down to subjective opinion unclear conception of what counts as disorder - difficult to police - **Emergence of Zero Tolerance: The New York City 'Miracle'** Rudy Giuliani and Bill Bratton Policing NYC-style: -Charismatic leadership -Media management -'Quality of Life' Policing -Devolved accountability and COMPSTAT -Expansion of police numbers (36K-47K 1990-1995) - **Impact** Changing crime rates -homicide 2,262 (1990) to 767 (1997) (-66%) -murder rate 26.5/100,000 population (1993), 4/100,000 (2014) -car theft (-70%) -burglary & robbery (-61%) -rape (-35%) Explanations -General crime drop across US cities from early 1990s -economic trends -demographic shifts -imprisonment -drugs markets (decline in crack) -policing - **Zero Tolerance in Britain** Limited applications in UK policing policy (localised experiments in a small number of forces during 1990s focusing on minor crimes and incivilities) 'Broken Windows' ideas have been influential in the broader political arena, particular in relation to political **rhetoric** within law and order debates **Symbolic** use of ZT terminology -- particularly associated with Tony Blair and 'New Labour' but used extensively by Conservative also politicians in recent times Focus on '**disorder**s' or '**incivilities**' became a core part of 'reassurance policing' and the Neighbourhood Policing Programme (see Part 2) Broader influence of 'Broken Windows' thinking visible in the growing legislative focus on  'anti social behaviour' and 'disorder' -The Crime and ***Disorder*** Act 1998 -***Anti-social behaviour*** orders (ASBOs) -The ***Anti-Social Behaviour*** Act 2003 - **Debate about Zero Tolerance** United Kingdom August 2011 Riots Cameron -- "We haven't talked the language of zero tolerance enough" Bill Bratton mooted as next Commissioner of Metropolitan Police Appointed as PM's 'advisor' on gangs USA Academic debate continues Bratton re-appointed as New York City Police Commissioner in 2013 Killings by police in US cities from 2012 on re-ignite debate (eg. Mike Brown, Eric Garner \#blacklives matter movement) Summer 2020 -- police murder of George Floyd sparks mass protests across USA and across the globe - **Benefits and Limitations** Benefits -symbolic/expressive value -policing and crime reduction linked Limitations -limited impact -treating symptoms not causes -**police brutality and corruption** -community relations -labelling - **The Signal Crime Perspective and Reassurance Policing** - The emergence of 'community policing' The 'softer' side of policing? Public contact and relationships, peacekeeping and order in local neighbourhoods Foot Patrol -- visibility historically considered the 'backbone' of public policing 'Omnipresence' -- local 'Bobbies' living and working within and for communities Accessibility and familiarity seen as key to public reassurance - The decline of the 'local bobby' 1960s/70s -- Home Office driven changes in policing practice to widen area coverage Unit Beat Policing: Introduction of patrol cars and personal radios "created an unprecedented gulf between police and public and... limited, or even removed, opportunity for day to day interaction" (Weatheritt, 1986) - Growing criticism and divisions over policing Urban riots in many UK cities from 1980s Insensitive policing and failing relationships between police and communities (Scarman, 1981) Big impact on public perceptions of security, the legitimacy of policing and community reassurance 1980s/90s various initiatives to reduce the gulf that had emerged between police and public - The Reassurance Gap Sharp declines in recorded crime from the mid-1990s, but crime surveys (e.g. Crime Survey for England and Wales, CSEW) showed high proportions thought it was rising Ongoing falling public trust and confidence in the police Analysis of CSEW data found that fear of crime was significantly correlated with visible 'signs of disorder' in local neighbourhoods (independently of actual crime rates) Particular incidents of 'disorder' seem to play a crucial role in transmitting signals that shape people's risk perceptions (youths hanging around, litter/graffiti/vandalism, public drinking) - British Crime Survey (1991-2003): Trends in Crime, Fear of Crime and Incivilities - Crime, disorder and fear of crime **Public perceptions** of crime/disorder are crucial in the wider symbolic construction of social space Particular incidents of 'disorder' seem to play a crucial role in transmitting signals that shape people's risk perceptions (youths hanging around, litter/graffiti/vandalism, public drinking) These send a powerful message to residents that the local area is 'out of control' (in a way that burglary figures, for example, do not) Physical and social signs of disorder appear to be concentrated in poorer areas This formed the basis of the Signal Crime Perspective - The Signal Crime Perspective (Innes 2004) The **communicative properties** of particular incidents and social control responses Particular incidents of crime/disorder and social control have a **disproportionate impact** upon how individuals/communities experience and construct their beliefs about local area Signals include expression, content and effect Types of effect -emotional -cognitive -behavioural Control signals -visible authority figures or physical security -measures convey notions of security and guardianship Formal control signals (police, private security) Informal control signals ('organic' features of local areas, natural surveillance, mixed use of public space) - The Signal Crime Perspective: Implications for policing Citizen focus -engage with communities to assess what the public's priorities are Diagnosis of how and why people fear certain incidents and issues Identifies problems to target problem-solving, enforcement and preventative efforts upon to achieve maximum, often disproportional, perception management impact Opportunity to engage with the public via community intelligence gathering methodology - Reassurance Policing and Neighbourhood Policing National Reassurance Policing Programme 2003-2006 pilots Expanded numbers of police officers Strategies to increase 'visibility' Introduction of Police Community Support Officers (PCSOs) Roll-out of Neighbourhood Policing Programme from 2008 Locally based embedded policing teams in each neighbourhood Local citizen panels advising on policing problems Problem-solving and co-production approach Significant impact of austerity and cuts in police numbers following 2010 **Reading** **Question 1: Policing, race and accountability** **Policing Ethnic Minorities** - **Over-Policing** **Three Dimensions of policing-BAME relations:** - Under-representation: of ethnic minorities as police officers - Under-protection: of ethnic minorities as crime victims - **Over-policing: of ethnic minorities as suspects** overrepresentation of: -stop and search -arrests and use of force -prosecution decisions -deaths in police custody **Police stop and search per 1000 population** (England and Wales) Ethnic Group White Black Asian Mixed Stop & Search per 1000 population 6 27 9 9 - Black \[Caribbean and Black African\] rate is 4-5 x higher than that of white people - Asian people and Mixed heritage people stopped about 1.5 x more than white people - Arrest rate (arising from stop) higher for black and mixed groups **Arrests per 1000 population and use of force** Ethnic Group White Black Asian Mixed Arrests/1000 population 9 27 10 16 - Black people arrested 3 x more than whites - Mixed heritage people arrested about 2x more than whites - Black people 3 x experience 'use of force' compared with white people **Charging/Prosecuting Decisions** - The role of the Crown Prosecution Service (CPS) -- Two tests ('evidential' and 'public interest' inform decision to charge and proceed with prosecution - Previous research documented **higher 'discontinuity rates'** for ethnic minority suspects (ie. possibly insufficient evidence to prosecute or not in public interest) - This may be suggestive of **police bias** against ethnic minority suspects at **earlier stages** - Lammy Review (2017) found that for every 100 white male suspects charged by CPS, there were 98 Black and 92 Asian suspects charged **Deaths in Police Custody** 2008/9-2018/19 Characteristic White Black Asian Mixed Percentage of deaths in custody 85% 8% 3% 2% Percentage of population 86% 3% 8% 2% - **Explaining over-policing** Three distinct explanations of **patterns of differential treatment** - Police racism and discrimination (a1 - direct discrimination, a2 - indirect and/or institutional discrimination) - Differential offending rates, police statistics reflect some degree of 'reality' - Wider structural factors (demography, deprivation, areas of residence, etc) - **Forms of racial discrimination** **Direct discrimination** (Equality Act 2010) - police officer treats members of group disadvantageously purely because they belong to that group. (bad apple theory - problems caused by racist individuals in the police) **Indirect discrimination** (Equality Act 2010) - where policing follows rules that apply to everyone equally, but which in practice always place ethnic minority people at a disadvantage (and no adequate justification can be provided) e.g., height restrictions, uniform requirements **Institutional discrimination** (MacPherson Report 1999) the collective failure of an organisation to provide an appropriate and professional service to people because of their colour, culture or ethnic origin. It can be seen...in processes, attitudes and behaviour which amount to discrimination through unwitting prejudice, ignorance, thoughtlessness and racist stereotyping which disadvantages minority ethnic groups - Individual prejudice and direct racial discrimination (a1) - Long history of reports of **racial prejudice** in occupational police culture - Major ethnic disparities in **stop & search/arrests** etc - Higher proportions of **'discontinue' decisions** for ethnic minority people when assessed by Crown Prosecution Service (suggests that police more likely to process ethnic minority suspects **without sufficient legal basis**) - Racial disparity highest in **offence categories** with greatest discretion for police (eg. public order offences, drugs possession) - "The evidence points to **racial discrimination** being a significant reason why black and Asian people are more likely to be **stopped and searched** than white people" (Equality and Human Rights Commission 2010; see also recent reports by the Independent Office for Police Conduct (IPOC) - (Bowling and Phillips (2007)) However... - These factors are strongly suggestive of **direct discrimination** as a **factor** shaping the overall picture of **disproportion** - Cop culture is something **more than individual** (deviant) dispositions and also the observed disconnect between stated **prejudice and actual behaviour** (Saying v. Doing) (PAJ Waddington 1999) - Stretches credibility to suggest that racial prejudice in hostile views/attitudes have no impact on practice, but the **relationship is not straightforward** - On balance, it does appear that direct discrimination is at work here, but **other factors** are also important in shaping the overall pattern of racial disparity - 'Institutional' and 'indirect' forms of discrimination (a2) - Problems with **conceptual clarity** of notion of **Institutional Discrimination** (IR) (Lea 2000, Dennis et al 2000 for different critiques) - Lack of **precision**: IR does not identify precise policies and processes leading to unequal outcomes - Conflates individual & collective racism - 'Indirect' discrimination is **defined in law** -- clearly important in relation to police policies and practices (e.g. deployment decision, recruitment and promotion procedures within the police, former police rules on height and uniform) - Internal **police policies and structures** perpetuate/fail to challenge biased attitudes within the police organisation (e.g. Casey Review) - **Deployment decisions** about which areas and which crimes to prioritise could indirectly discriminate against ethnic minority populations (e.g. Vomfell and Stewart 2021) - Disproportionate involvement in crime and disorder (b) - Association between some **ethnic groups and particular crime types** (e.g. organised crime in particular communities) - 'Strike rate' of arrests arising from stops -- higher for black people (see Ministry of Justice statistics) though may also reflect **biased** use of **police discretion** (e.g. police work harder to find a reason to arrest a black suspect) - Known structural relationship between **socio-economic deprivation** and **higher risks of offending** in street crimes (e.g. poverty disproportionately affects particular ethnic groups) However... - Ultimately we cannot know the **true level of offending** in any ethnic group - **Self-report studies** do not indicate higher rates of drug use/other offending amongst the black populations who experience the most unequal policing - Official crime statistics **reflect police and CJS bias** (Bowling and Phillips 2007) - Structural and demographic factors (c) - **Age** structure: Ethnic minorities have younger population profile, and on average young people more likely to come under police attention - **Socio-economic** factors (unemployment, poverty, school exclusions etc not evenly distributed between ethnic groups) - **Geographical** factors: Inner city concentration of ethnic minority populations into poorer 'high crime' and more intensively policed areas - '**Availability**' for police attention -- variations between ethnic groups in levels of presence in 'public space' (Waddington et al. 2004) - Ethnic minorities' higher 'availability' in some public spaces reflects wider processes of discrimination and exclusion (Bowling and Phillips 2007) - **Ethnic minority people are statistically more likely to fall into those social groups & geographical areas that are the targets of adversarial policing** - **Tackling racial disparities in policing** **Transforming police culture** - **Recruitment** -targeted **recruitment drives** (with national targets) -more vigorous **vetting schemes** (Casey 2023) -professionalisation (e.g., policing degrees) and direct graduate entry schemes - **Training** -'cultural awareness' / anti-racism training **Problems** -**Implementation** failure - practical limitations and slow progress -Theory failure - **'individualises'** the problematic aspects of cop culture to the idea of 'bad apples' (rather than a 'rotten barrel') or fundamental structural inequalities (Waddington 2012) **Restricting discretion via internal police policy** Greater **visibility** of routine policing -(e.g., body worn cameras) **Record** keeping and review -(e.g., officers required to record reason for stops) Tighter line management/supervision **Stricter** disciplinary code -(increase risk for officers tempted to act on prejudiced views) **Changing the pattern of policing** **Stop and search** -reduce overall use: not an effective method of crime control but has huge impacts on poor community relations -ensure when it is used it is properly intelligence-led Improving **police policies** on racial harassment and attacks **Community policing** -shifting from an 'enforcement' to a 'problem solving' ethos - **'Defunding' the police** - **Rationale** Key function of state police forces since their foundation has been to **control marginalised groups** Repeated **failure of liberal reform** efforts -discriminatory, oppressiveness is built into function of policing Mission creep and expansion of policing -police used in mental health cases, involved in schools, etc Need **alternative** and more **effective** and just **approaches** to tackling victimisation and community conflicts - **In practice** Campaigning **slogan** versus 'real world' change Diverts police funding into c**ommunity crime prevention**, mediation, social and welfare services **Civilian responders** for community conflicts Reduce overall scope and power of police (e.g., reverse 'mission creep') Enhance **local accountability** and oversight mechanisms US-driven movement, but arguments can be applied to UK policing - **Problems** Centralised police organisation and funding in England and Wales Austerity and racialized policing Crime, disorder and inequality Limitations of community alternatives to policing If not police, then who? **Facial Recognition Technologies** - Policing Context 2008 financial crisis =Budget cuts and police disinvestment -(Greig-Midlane 2019) 'The austerity era' -'a political and socio-economic backdrop to...policing reforms and change' Less resources = more pressure on resources remaining Retirement of experienced officers -(Bullock et al., 2020) Shift away from Neighbourhood Policing Need to respond to changing 'demand' on policing with shrinking resources Impacts on local knowledge in policing At the same time... Technological developments Growing availability of FR systems Supporters argue it held the promise of supporting 'austerity-afflicted law enforcement agencies' (Fussey et al 2021) Began being used by: -Metropolitan Police Service -2016 -South Wales Police - 2017 - **What is Facial Recognition** Biometric processing of captured images for the purposes of matching to a database and identifying individuals (Fussey et al 2021) 1. Analyses key facial features 2. Generates a mathematical representation of features 3. Compares them against the mathematical representation of known faces in database 4. Generates possible matches - Police uses of FR **Live Facial Recognition (LFR)** Compares a live camera feed of faces against a predetermines watchlist to find a possible match that generates an alert **Retrospective Facial Recognition (RFR)** Post-event use of facial recognition technology, which compares still images (e.g. CCTV) of faces of unknown subjects against a reference image database in order to identify them **Operator Initiated Facial Recognition (OIFR)** A mobile phone app use of FR technology, which compares a photograph of a person's face taken on a mobile phone to the predetermined watchlist to assist an officer to identify a subject - **Deployment** Factors impacting FR: 1. Organisational Policing routines, policies, strategic choices and standard operating procedures that directly impact deployment (especially LFR; Fussey et al., 2021) - Watchlist decisions - When and where to deploy - Whether to tell the public (marked/unmarked vans, signage) - Positioning of the cameras and street teams (e.g. Queen Street and Principality Stadium -- high footfall) A flag seller arrives on scene to set up stall. There is a large Italy flag which at times obscures the camera view as it waves in the wind. Operator discussion about whether to ask nearby PCSOs to intervene and ask to move on. Over time, it appears the stall might be a bonus because people tend to look up as they approach it. (fieldnote, SWP, 31 May 2017). 2. System Technical aspects, software and hardware - Algorithm -- black box (Pasquale, 2015), commercial, algorithm updates impact performance - 'Threshold score' -- how certain the system is that it's a positive match before it sends an alert (to the operator) - Faces -- how many faces the system can process per 'frame' (load) - Image quality (watchlist, custody images) - Cameras steaming up - Camera zoom ('zone of recognition') -- bicycle example (Fussey et al., 2021:337-8) 3. Operator human--machine interactions that shape how the outcomes of FR are produced and used (Fussey et al., 2021) Human operator essential --'assisted facial recognition' - Discretion of individual officers - Officer judgements about the credibility of alerts - Trust in the system - Types of human deference to computational outputs - Availability of intervention teams - Length of time working in LFR vans Sometimes, one operator would be looking for the person in the crowd while the other was describing them aloud from the image captured on the screen: operators reported using key facial features, such as eyes, nose, mouth, jawline and hairline to inform their decisions. While not relevant to a subject's appearance, some officers also recruited background information (e.g. offence type) for their deliberations. - **Outcomes** 2017 Champions League Incorrect - 2554 alerts, → 5 investigations Implementing a new technology 1. Implementation in a **live police environment** 2. Estimated 310,000 people in Cardiff on the 3rd of June (very high load) 3. Watchlists containing images from international partners -- poor quality 4. Top 15 individuals accounted for 25% of all incorrect matches 5. Led to a force-wide programme to improve the quality of custody images 6. Understanding how many faces FR system can process at a time 7. Score -- learning where to set 2023 and 2024 Algorithm updates Increased the threshold score Publishing more detailed data that provides more insight -- i.e., number of faces seen (scanned), record of threshold score **Trial** Key recommendation from the 2018 evaluation work was a **large-scale controlled trial of FR** to comprehensively test system accuracy across different demographic group Alleged technological biases deriving from how technical accuracy recedes for subjects who are older, female and for some people of colour (Buolamwini and Gebru 2018; National Institute of Standards and Technology 2019). National Physical Laboratory equitability study Looked at the True / False Positive Identification Rate across different groups Key findings include: 1. Equitability is dependent on the face-match threshold settings and on the size and demographic composition of the LFR watchlist or RFR reference database. 2. TPIR of the system at face-match threshold 0.6 is **equitable across gender and ethnicity groups.** 3. Variation in TPIR for **different age groups** (statistically significant) -- higher TPIR for older groups; lower TPIR for younger groups - **In Court** SWP taken to court for AFR -infringes on privacy, etc LFR in the courts 1. SWP case: the court highlighted the presence of **excessive discretion** 2. Two "impermissibly wide areas" of discretion: - **Who** becomes targeted for surveillance (included on a watchlist) - **Where** the technology is deployed Issues: - Suspicion and Discretion - Suspicion - Normative formations of suspicion -- surveillance routinely **steered towards segments** of the population dubbed 'police property' (Reiner, 1992) - Certain neighbourhood policing officers acquire '**cop cultural capital**' through their ability to identify 'on sight' (e.g. recognise) **members of street gangs** and have **knowledge** of their patterns of behaviour and association (Innes et al., 2020) - Early ethnographies of street policing showed police 'gaze' not distributed uniformly or equally... - 'Bureaucratic' and 'incidental' suspicion (Matza, 1969): - Incidental = crimefighter myth of police work, **skilled investigator** determines 'whodunit' by linking aspects of the **incident to a suspect** with the means, motive and opportunity to commit the offence. Favoured **fictional idea** of police work -- but not that common. - Bureaucratic = rounding up the 'usual suspects'; police search strategies are based upon **knowledge of individuals** previously displaying forms of criminal behaviour similar to the incident under investigation; **suspect identification** becomes based upon specific **socio-demographic characteristics.** - FR Reinvented '**bureaucratic suspicion**' (Fussey et al., 2021) - e.g. watchlists are (typically) comprised of **police-held custody images** - FR specifically targets police attention towards **individuals already known** to the authorities (the 'usual suspects') **Recognition** -- PACE 1984 says recognition may constitute grounds for stop and search - But when introducing FR, this takes the initial recognition **'work' away from officers** Scores -- sometimes discussed when officers were deliberating system matches -e.g. high score = more 'likely' FR technology performs a framing and priming role in **how suspicion is generated** - Discretion Can be summarised as **how police interact** with citizens when deciding **whether to operationalize** legal powers (Fussey et al., 2021) - Previous research shown officers on the street possess **high decision-making power** in terms of against whom, when, how and why criminal law is enforced (Skolnick, 1966) - Policing delivery found to be grounded in finely grained officer **judgements influenced by various factors,** such as the values and precepts of 'cop culture' (Fussey et al., 2021) - 'Law in books' vs. 'law in action' - Police as active decision makers - Led to analysis of injustices and inconsistencies arising from police discretion - FR **Police discretion influences the technology:** - **Deploying** LFR/OIFR -- discretionary decision - Formulating **watchlists** -- discretionary decision - Where to set the **score** -- discretionary decision **The technology influences discretionary decisions:** - Adjudication (deciding if there is a correct match) - 'Reformulated discretion' - Linking back to suspicion, initial recognition does not originate from the officer exercising discretion -- initial recognition work is done by the system, then the officer exercises **discretion to decide if it's a 'match'** - Similar process for RFR -- operator sees a list of possible matches - Existing bias and disproportion issues - Existing Bias **LFR** receives more attention in **academic, policy and media** debate -- live deployment is typically more **contentious** than post-event (post-crime) deployment of the tech, as in RFR. But the same underlying issues exist... Facial Recognition -- brings bureaucratic suspicion ('the usual suspects') to the fore ↓ Drawbacks (Matza, 1969): bureaucratic suspicion has a tendency to frame **persons of interest as different** to others, regardless of any available evidence ↓ Individuals listed on watchlists and databases are cast as **warranting suspicion** and the FR surveillant 'gaze' is specifically oriented towards them LFR Decisions about **where to deploy** Watchlists made up of 'mugshots' of people **previously arrested** Who is in the 'space' (**available population**) RFR (Retrospective) not searching through images of whole population Only custody database - people **known to police** - Discretion and Discrimination Discrimination is most likely under certain circumstances, including: - Where there are **no clear guidelines** on criteria for decision-making - Where decisions depend on **subjective judgements** rather than/not in addition to objective criteria - Where there is considerable **scope for individual discretion** to be exercised -(Bowling et al., 2008) - Explanations of discriminatory policing (policing ethnic minorities lecture) Indirect discrimination present in deployment decisions -- which areas and which crimes to prioritise indirectly discriminate against ethnic minority populations Structural and demographic - Age structure: Ethnic minorities have younger population profile, and on average young people more likely to come under police attention - Socio-economic factors (unemployment, poverty, school exclusions etc. not evenly distributed between ethnic groups) - Geographical factors: Inner city concentration of ethnic minority populations into poorer 'high crime' and more intensively policed areas - 'Availability' for police attention -- variations between ethnic groups in levels of presence in 'public space' (Waddington et al., 2004; Bowling et al., 2008) -Ethnic minorities' higher 'availability' in some public spaces reflects wider processes of discrimination and exclusion (Bowling and Phillips, 2007) -Ethnic minority people are statistically more likely to fall into those social groups & geographical areas that are the targets of adversarial policing - FR and underlying biases Existing social biases of police activity - Disproportionate focus on young people and members of African Caribbean and other minority ethnic groups (inter alia The Lammy Review 2017) - Surveillance routinely steered to certain segments of the population 'police property' -police 'gaze' not distributed uniformly or equally Alleged technological biases - Specific concern around FR is that biases are exacerbated by technology - Alleged lower accuracy for subjects who are older, female and for some people of colour (Buolamwini and Gebru 2018; National Institute of Standards and Technology 2019) - We know ethnic minorities (especially black people) are already disproportionately: stopped and searched, subject to force (e.g. restraint, taser) and arrested (Jones and Wyn Jones, 2022) - Technology has now been shown to be equitable for ethnicity and gender (not proven to be equitable for age; National Physical Laboratory, 2023) - But other underlying disparities remain i.e. people already known to police on custody databases/watchlists, the 'available population' in the spaces where FR is deployed by police - Regulation College of Policing Authorised Professional Practice -- setting out guidance for **LFR** deployments. Includes stating that: The nature of the deployment should be informed by the force's **policing requirements** and their use case for LFR, with all deployments being: - **targeted** - **intelligence-led** - **time-bound** and **geographically limited** when set within the context of the relevant use case But this has still been criticised for being **too broad** -- forces **still have discretion** for deployments and watchlist construction. There is **no set offence severity threshold**, for example. -Other jurisdictions have more stringent rules -- such as needing a judge's approval for LFR. We know 'vagueness' has been a problem in other contexts: e.g. legal regulation of stop & search does not prevent the abuse of discretion -'**reasonable suspicion**' is vague and interpreted differently (Bowling et al 2008) Regulatory bodies such as the Information Commissioner's Office (ICO) Continued scrutiny from civil liberties groups **South Wales Police** are now deploying again. Efforts to be transparent include: - Information pages and FAQs - Issuing notices before deploying - Publishing results from deployments - Social media updates **Question 2: Policing domestic violence** **Domestic Violence** - The Changing Police Response: Historically, policing domestic violence was non-interventionist - Police should help keep **'family unit'** together - Domestic violence considered a '**private matter'** - Police culture view of domestic violence as **'garbage work'** - Training encouraged officers to mediate the situation or separate the parties, not to formally intervene i.e. making arrests was a last resort. The approach to policing domestic violence has changed over time - Police forces in US and UK now have policies to promote '**positive action**' in an attempt to structure police use of discretion at these calls. - Changing Police Policy **Home Office Circular on Domestic Violence (60/1990)** - Police should adopt a more interventionist approach by arresting assailants where an offence has been committed. **National Policing Improvement Agency Guidance on Investigating Domestic Abuse (2008)** - The Human Rights Act 1998 places positive obligations on police officers to take reasonable action, which is within their powers, to safeguard the following rights of victims and children: - Right to life (Article 2, ECHR); - Right not to be subjected to torture or to inhuman or degrading treatment (Article 3, ECHR); - Right to and respect for private and family life (Article 8, ECHR). - Failure to make an arrest when there are grounds to do so may leave a victim at risk from further offences. It may also mean that the police force is vulnerable to legal challenge under both the Human Rights Act 1998 and the law relating to negligence. **College of Policing Authorised Professional Practice (2015)** - Updated to reflect changes in policy and law. - Updated to address recommendations from HMIC (2014) national inspection. - Reinforces requirement for positive action in all cases of domestic abuse. - Expanding Policy and Law - "Any incident or pattern of incidents of controlling, coercive, threatening behaviour, violence or abuse between those aged 16 or over who are, or have been, intimate partners or family members regardless of gender or sexuality" Home Office (2013) - Section 76 of the Serious Crime Act 2015 - Controlling or Coercive Behaviour in an Intimate or Family Relationship. Home Office (2015) - Domestic Abuse Act 2021 created a statutory definition of domestic abuse. Home Office (2021) - Explaining the Change **Women's liberation movement** - Media attention on domestic violence - The Burning Bed - First refuges for battered women set up **Court cases** - Victims not provided with equal protection under the law - Thurman v. City of Torrington - Police sued for failure to protect **Criminological research** - Deterrent effect of arrest - Minneapolis Domestic Violence Experiment (Sherman & Berk, 1984) - National Picture 2023 2,124,000 adults (aged 16-59) experienced domestic abuse. ↓ The police recorded 889,918 domestic abuse-related crimes. ↓ The police made 69,634 arrests for domestic abuse-related crimes. ↓ The CPS prosecuted 47,361 suspects for domestic abuse-related crimes. ↓ Resulting in 39,198 offenders convicted for domestic abuse-related crimes. ↓ Of which, 566 were convictions for coercive and controlling behaviour. "Any incident or pattern of incidents of controlling, coercive, threatening behaviour, violence or abuse between those aged 16 or over who are, or have been, intimate partners or family members regardless of gender or sexuality." - The Arrest Decision: - Most police forces in US and UK now have 'positive action' policies in an attempt to structure their use of discretion at these calls. - "The decision whether or not to arrest a suspect rests with the police officer....." APP (2015) - Myhill & Johnson (2016) police discretion and domestic violence - Police interpretation of the official policy definition - Not considered a domestic - Police decision to "cuff" incidents - Taking no further action - Police decision to "cuff" crimes - Not recording (all) crimes - Influencing Factors **Decreases** likelihood of arrest - Suspect left the scene - Victim uncooperative - Victim unlikely to prosecute **Increases** likelihood of arrest - Seriousness of the incident - Weapon involved - Witnesses present - Research of effect of arrest **The Minneapolis DV Experiment** - Headline finding: arrest reduced the recidivism rate by half against the same victim within the following 6 months (i.e. arrest produces a 'deterrence effect') **The U.S. Dept. of Justice** sponsored several "replication" studies, with variable results. - Omaha, NB: no significant difference between arrest, separation or mediation - Milwaukee, WI: arrest deters employed but not unemployed offenders - Charlotte, NC: did not find that arrest was more effective in deterring offenders - Colorado Springs, CO: mixed results from official vs victim interview data - Miami, FL: arrest plus follow-up services reduces revictimization - Atlanta, GA: findings not published **Pooled analyses** showed: - Arrest produced less revictimization on average; - one found as much as a 25% reduction in re-victimization. - Benefits of arrest - giving the victim some time to feel safer and improve the likelihood of them engaging with the police and other services - disrupting an established pattern of abusive behaviour - allowing time to pursue other lines of enquiry - providing an opportunity to put in place services to support the family or relationship - generating a PNC ID and obtaining DNA and fingerprints if the offence is recordable -- this may assist with resolving outstanding or future offending - being able to impose bail conditions to protect the victim - sending a message to the perpetrator that their behaviour is not acceptable and will not be tolerated, and that the victim is not alone - giving an indication to the victim that they are being taken seriously - Victims' Perspectives: - An **ecological model** depicts four levels - **Individual** - **Interpersonal** - **Community** - **Societal** - Each of these levels poses a particular set of challenges for effectively addressing domestic violence - Taken together, they help us to understand the multi-faceted and intersecting obstacles facing victims - And the inherent limitations of any 'one- size-fits-all' approach - What Victims Expect of Police - Robinson & Stroshine (2005) - Victim satisfaction is determined by the extent to which policing measures up to their expectations - Positive / zero / negative disconfirmation - Exceed / meet / worse than expectations - **Police behaviour:** collecting evidence, writing a report, making an arrest, etc. - **Police demeanour:** demeanour: courteous, respectful, taking time to listen, etc. - When victims expected and received certain police behaviours and/or demeanours, they were more likely to be satisfied - Expectations regarding police demeanour more important determinants of satisfaction than expectations regarding behaviour - Arrest - a victim's perspective 'I was strangled by my ex-husband a few years ago. He released his hands when he thought I was dead. He did that in front of my son. My son, who was terribly scared and was threatened not to call 999 by his Dad, managed to call 999... The police officer was informed that I was strangled, but he asked me if I wanted my husband to be arrested. I was very scared not only for me but also my son, I said I do not know. I needed to go to hospital first, because I could not breath properly and I was not able to feel most of my body.' 'I finally divorced my ex-husband in 2003. It was an acrimonious and difficult divorce in which he really showed his true colours by assaulting our daughter and myself. At one stage I was in fear of my life..... He stole my cash card but when I went to the Police they just said to \"order another\". I was referred to our local DVU and although they were good to talk to, they didn\'t really do much or were very helpful (e.g. they advised me to go to a refuge with my daughter). How could I do that when I had sole responsibility to pay all the bills on the home. Why wasn\'t he arrested?' - What matters to victims? - Procedural justice concepts - Listening to people - voice - Decisions based on objective criteria - neutrality - Treating people with respect - ethicality - Competent decision-making - accuracy - Ability to fix mistakes - correctability Pure procedural justice describes situations in which there is no criterion for what constitutes a just outcome other than the procedure itself. -John Rawls, A Theory of Justice (1971) - Lack of Voice / Enabling Voice And this policewoman rang me back.... and she made me freeze... with her I just felt like a rabbit in the headlights and I couldn't tell her exactly what he was doing to me. Because the one or two things that I said to her about how he was you know making me feel and what his family had done -- to her that was nothing. I just thought okay so if that's nothing then whatever else I tell you next is going to be nothing, and I just froze. And she said 'Well look, you're not giving me anything, I can't do anything'... But I think if it hadn't have been... I won't say if it had been another police person I perhaps wouldn't have done, but they were brilliant. She just sat and listened to me, there was no rush. She wasn't 'Oh why won't you press charges?' -- she really understood.... (crying) You know she didn't force me, but she was there going 'Come on, it \[abuse\] isn't right is it? Right, let's have a cup of tea, let's have a talk about this'. You know there was none of this 'Right well if you don't want to press charges, you're wasting my time'..... There was none of that, there was like 'Hang on a minute, she needs to get her head round this'. (crying)