Policing Revision-PDF

[Police & Policing Revision] **[Police Discretion & Decision Making]** - **Definition** -- unwritten rule that police officers have the right to be selective in how they do their jobs if they stay within departmental guidelines (deciding WHICH rules and WHETHER to apply them) - **...

[Police & Policing Revision] **[Police Discretion & Decision Making]** - **Definition** -- unwritten rule that police officers have the right to be selective in how they do their jobs if they stay within departmental guidelines (deciding WHICH rules and WHETHER to apply them) - **Lord Scarman (1981)** -- discretion lies at heart of successful policing function. - **Goldstein (1960)** -- officer's decision not to report adequately could lead to possible failure of service - Discretion is the defining feature of police work, **concentrated in lower ranks** of police organisation (street-level bureaucrats -- Lipsky 1980) - **Sources of discretion:** - **Nature of criminal law** - vague definitions, law used for social & medical issues - **Police work environment** -- limited supervision of officers, private encounters - **Limited police resources** -- myth of full enforcement - **Decisions:** - Recording crime - Stop & search - Collecting evidence - Arrests - Proactive investigation - Case finalisation - **Factors influencing discretion:** - Situational -- seriousness, strength, victim preference, relationship between victim & suspect - Immediate work environment (location, time) - Force characteristics (specific policy, professionalism (Wilson)) - Officer characteristics (race, gender, education) - **Problems:** - Denial of equal protection - Poor police-community relations - Poor personnel management and planning **[Research Police and Policing]** - Carried out by academics, governments, police, think tanks, campaign groups, independent agencies - Evidence-based policing - 'Three Rs' -- Random patrol, rapid response, reactive investigations - 'Triple-T' -- Targeting, testing, tracking [Policing DV examples] - **Minneapolis DV experiment (field)** - Which police response is most effective for DV - **IV** -- officers made arrest or separated parties - **DV** -- re-offending by perpetrators - **Control** -- officers offered advice or mediated situation - **Results** -- arrest (19%), separate (33%), advise/mediate (37%) - **Critiques** -- small sample, random assignment, failure to replicate - **DV arrest decision (cross-sectional)** - Demographic/attitudinal/situational influences on decision to arrest DV - Situational provide most explanation for arrest (witnesses present, victim/suspect cohabitating, last hour of shift) - **Risk-led policing and DASH risk tool (mixed methods)** - Analysis of 43 forces to describe approach and identify different ones - 61 interviews w/ police, 120 hours of observation, 1296 online survey responses - Findings & Critique: - quality & completeness is a concern, understanding of DV often informed by physical violence, focus on IPV rather than DVA more broadly **[Cop Culture]** - 'Rational-legal' dominated policing until 1970s -- hierarchal/disciplined - Wide discretion of frontline police - Police occupational culture - The way police view the social world and their place (Reiner 2010) - Set of beliefs, values shared across police - 'Canteen culture' -- values expressed in off-duty socialising - 'Cop culture' -- norms expressed in course of police work (Waddington 1999) [Ethnographic contribution -- immersive fieldwork, participant & non-participant observation] - Immersion in back/front stage worlds of policing - Deeper cultural understanding - Direct observation of policing - Ongoing trust w/ relationship - Ethical challenge w/ informed consent - Practicalities w/ access and time - Hawthorne Effect (knowledge of observation) - Researcher bias/going native - Qualitative interviews - In depth research of key values, follow ups, first-hand accounts - Researcher effects, retrospective framing - Quantitative surveys - Objective, standardised, statistical testing, generalisation to wider population - Some aspect not able to measure quantitatively, reduction to individual characteristics, validity of self-report [Key characteristics of cop culture ] - Sense of mission/way of life action - Cynicism and pessimism - Suspicion - Isolation & solidarity -- us vs them - Machismo -- aggression, bravery - Prejudice of sexism/racism/homophobia - Conservatism -- law & order, preserve status quo [Where does cop culture come from?] - **Selection hypothesis** -- culture reflect individual characteristics of people accepted into police - **Socialisation hypothesis** -- generated by fundamental nature of policework - **Null hypothesis** -- there is no distinctive set of characteristics, police are same as wider population [Individual personality traits] - UK research - Police recruit attitudes reflect wider population (Bowling 2019) - US research -- police recruits show distinctive personality characteristics from wider populations (TenEyk 2024) [Cop culture -- structurally shaped] - Nature of policework - Skolnick's 3 features of working personality, Loftus - Unique coercive authority - Risk/unpredictable situations - Pressure for results - Interactions w/ people - Hostile encounters - Certain groups 'police property' - Socialisation -- learning on the job [How is cop culture shaped by police function & structure of policework?] - Mission -- action-based/heroic narratives to often difficult and mundane job - Cynicism/pessimism -- dealing with raw end of society's problems - Isolation -- coercive powers, unsocial hours - Solidarity -- need to rely on backup - Suspicion - Machismo -- stereotyped male attributes of toughness - Prejudice -- collective experience of hostility with particular segments of society - Conservatism -- preserve established order [Problematising 'cop culture' -- Waddington 1999] - Cop culture & police behaviour? - Saying vs doing - Social psychological research -- good people can do bad things (vice versa) -- Milgram Prison Experiment - Cop culture as distinctive police phenomenon - Police characteristics shared w/ wider population but there's psychological research on characteristics of police - Cop culture as universal & homogenous - Variations in cop cultures -- by rank, force, characteristics (Bowling 2019) - Major societal changes since classic studies of 70s/80s [Changing cop cultures -- Standard responses] - Recruitment -- targeted campaigns, recruitment targets, reduction of complex cultural traits to problematic individuals - Training -- increased entry standards, formal training to address problematic attitudes, does not address fundamental shapers of cultural traits [Changing cop cultures -- wider approaches] - Micro level -- individual accountability - Frontline policing more visible via tech and supervision - Disciplinary codes & complaint mechanisms - Rule tightening to restrict discretions (Police & Criminal Evidence Act 1984) - Meso level -- policing policy - Community based approaches, organisation accountability (elected commissioners) - Macro level -- societal context - Tackles broader patterns of inequality **[The Signal Crime Perspective and Reassurance Policing]** [Community policing ] - 'Softer side of policing' -- public relationships, foot patrol, accessibility and familiarity key to public reassurance [Decline of 'local bobby'] - 1960s -- Home Office widen area coverage - Unit Beat policing -- introduction of patrol cars and personal radios [Criticism over policing] - Urban riots from 1980s - Insensitive policing between police and communities (Scarman 1981) - Big impact on public perception of security [Reassurance Gap] - Decline in recorded crime from mid-1990s, but crime surveys showed people thought it was rising - Ongoing falling public trust & confidence in police - CSEW data -- fear of crime correlated with signs of disorder - Incidents of disorder played crucial role in shaping risk perceptions (youths hanging around, graffiti, vandalism) [Fear of crime] - Public perceptions of crime crucial in symbolic construction of social space - Shape risk perceptions -- send message to residents that local area is 'out of control' - Physical & social signs of disorder concentrated in poorer areas - Forms basis of Signal Crime Perspective [Signal Crime Perspective (Innes 2004)] - **Communicative properties** of incidents & responses - Crime incidents have **disproportionate** impact on individuals experience of local area - **Signals** -- expression, content, effect - **Types of effect** -- emotional, cognitive, behavioural - **Control signals** -- visible authority figures or physical security - **Formal** control signals (police, private security) - **Informal** control signals (organic features of local areas, natural surveillance, mixed use of public space) - Citizen focus -- engage w/ communities to assess priorities - Diagnosis of why people fear certain issues - Identifies problems to target problem solving - Opportunity to engage w/ public via community intelligence gathering methodology [Reassurance Policing & Neighbourhood Policing] - **National Reassurance Policing Programme 2003-2006 pilots** - Expanded number of officers - Strategy to increase visibility - PCSOs - **Neighbourhood Policing Programme 2008** - Local policing teams in neighbourhoods - Local citizen panels - Impact of austerity and cuts in police numbers **[Zero-Tolerance Policing]** Broken Windows -- Wilson & Kelling 1982 - 70s research -- police patrol has little impact on crime rate - Disorder raises fear of crime - Law abiding people avoid public spaces - Erosion of informal community controls - Communities reach 'tipping points' when disorder and crime spiral and feed off each other - Community controls & order maintenance - Police should act to reclaim streets & clamp down on disorder - Trigger circle of reduced disorder & fear, more informal social control via greater use of public space and citizen confidence, reducing crime rates [Critiques of Broken Windows Theory] 1. **Empirical Concerns** - No proven causal link between disorder and crime (Harcourt, 1999). - Disorder/crime may stem from deeper issues (poverty, inequality). - Crime reductions may be due to: - Police presence (deterrence). - Arresting offenders (incapacitation). - Policing style matters (e.g., community vs. intensive enforcement). 2. **Conceptual Issues** - Overly simplistic labels: \"orderly/disorderly,\" \"respectable/problematic.\" - Disorder as a political/social construct. [Zero Tolerance Policing (ZTP) in New York City] - **Introduced by Giuliani & Bratton** - Charismatic leadership. - Media-focused strategies. - \"Quality of life\" policing targeting minor crimes. - Use of COMPSTAT (crime mapping and accountability). - Increased police numbers (1990-1995: 36K → 47K). - **Impact on Crime Rates (1990-1997)** - Homicide: -66% (2,262 → 767). - Car theft: -70%. - Burglary & robbery: -61%. - Rape: -35%. - **Broader Factors Influencing Crime Decline** - Economic growth, demographic shifts, reduced crack cocaine markets, increased imprisonment. - General US crime decline began in early 1990s. [Zero Tolerance Policing in Britain] - **Limited Applications** - Trials in some forces (1990s) targeting minor crimes. - Influenced political rhetoric (e.g., Tony Blair's 'New Labour'). - **Legislation Inspired by 'Broken Windows'** - **Crime and Disorder Act 1998**: Anti-social behaviour orders (ASBOs). - **Anti-Social Behaviour Act 2003**: Focused on minor disorders/incivilities. [Debates on Zero Tolerance Policing] 1. **United Kingdom** - 2011 riots → Cameron: Need for more \"zero tolerance\" approaches. - Bill Bratton advised UK government on gang strategies. 2. **USA** - Bratton reappointed NYC Police Commissioner (2013). - Controversies: Police killings (e.g., Eric Garner, Mike Brown) → \#BlackLivesMatter. - George Floyd's murder (2020) → Global protests against police brutality. [Evaluating Zero Tolerance Policing] 1. **Attractions** - Clear message against disorder and crime. - Potential for crime reduction. 2. **Limitations** - Limited long-term impact. - Focuses on symptoms, not root causes of crime. - Risks: Police brutality, strained community relations, over-labelling. **[Policing & Ethnic Minorities]** **Charging/Prosecution Decisions** - **Crown Prosecution Service (CPS):** Decisions guided by two tests -- *evidential* and *public interest*. - **Ethnic disparities in charges:** Lammy Review (2017) found: - For every 100 white male suspects charged, only 98 Black and 92 Asian suspects were charged. - **Deaths in custody (2008/09--2018/19):** - White: 85% deaths (86% population) - Black: 8% deaths (3% population) - Asian: 3% deaths (8% population) - Mixed: 2% deaths (2% population) **Explaining Over-Policing** 1. **Explanation A: Racism and Discrimination** - **A1: Direct Discrimination** - Evidence of racial disparities in stop & search, arrests, and CPS discontinuity rates. - Higher rates of stop & search in discretionary offenses like public order and drug possession. - Direct discrimination indicated but not the sole factor shaping disparities. - **A2: Institutional/Indirect Discrimination** - MacPherson Report (1999): Processes, attitudes, and behaviours disadvantage minority groups. - Deployment decisions and internal policies can perpetuate bias. - Indirect discrimination often linked to outdated recruitment rules or deployment priorities. 2. **Explanation B: Differential Offending Rates** - Structural factors (poverty, socio-economic status) disproportionately affect ethnic groups. - Self-report studies show no significant differences in drug use/offending rates between ethnic groups. - Official statistics reflect systemic bias in policing and justice systems. 3. **Explanation C: Structural and Demographic Factors** - Ethnic minorities often have younger population profiles and live in intensively policed inner-city areas. - **Availability:** Ethnic minorities more visible in public spaces, partly due to structural exclusion -- Waddington 2004 **Tackling Racial Disparities in Policing** 1. **Transforming Police Culture** - **Recruitment:** Targeted drives, stricter vetting, professionalization (e.g., policing degrees). - **Training:** Cultural awareness and anti-racism programs. - Problems: Slow progress, focus on individual issues rather than systemic inequalities. 2. **Restricting Police Discretion** - **Policy changes:** Body-worn cameras, better record-keeping, stricter disciplinary actions. - Supervision to prevent prejudiced actions. 3. **Changing Policing Patterns** - Reduce stop & search and make it intelligence led. - Shift focus from enforcement to community-based problem-solving. 4. **'Defunding' the Police** - Redirect funds to crime prevention, mediation, and welfare services. - Limit police power and scope; enhance local accountability. - Challenges: Centralised funding in England & Wales, lack of robust community alternatives. **Key Conclusions** - **Bias and direct discrimination:** Evident in stop & search and frontline practices. - **Indirect and structural factors:** Contribute significantly to racial disparities. - **Wider societal inequalities:** Employment, housing, and education inequities heighten risks of adversarial policing for ethnic minorities. **[Facial Recognition Technology]** 1. [UK Policing context (2010-17)] - Budget cuts for police -- less resources available, more pressure - Retirement of experienced officers - Shift from neighbourhood policing - Technological developments start 2. [What is facial recognition?] - Biometric processing of captured images, matches to database and identifies individuals - Analyses key features, generated mathematical representation, compares them in database - Police use: - **Live Facial Recognition (LFR)** -- live cam feed of faces against predetermined watchlist, alerts match - **Retrospective Facial Recognition (RFR)** -- post event use, compares still images - **Operator Initiated Facial Recognition (OIFR)** -- mobile phone app, compares photo of face on phone to watchlist 3. [Factors affecting FR] - **Organisational** -- policing routines, operating procedures, when & where to deploy, whether to tell public, positioning of cameras - **System** - algorithm (Pasquale 2015), threshold score (how certain it is positive match), number of faces the system can process, image quality, cameras steaming up, zone of recognition - **Operator** - discretion of officers, trust in credibility of alerts, length of time working in LFR vans 4. [Trials] - **2018 controlled FR trials** - test system accuracy and technological bias, which decreases for: - Older individuals - Females - People of colour - **National Physical Laboratory (NPL) Equitability study** -- examined true (TPIR) & false positive (FPIR) identification rate across groups - Threshold dependency: - Equitability depends on face-match threshold settings, influenced by size & demography of database - Equitability across gender/ethnicity: - Face-match threshold of 0.6 (TPIR is equitable) - Age group variations: - Differences in TPIR across age, higher for older, lower for younger - Environmental factors - Crowds & subject height contributed to variations in TPIR across age 5. [How facial recognition impacts suspicion] - **Normative Suspicion:** - Reiner 1992 -- surveillance focuses on populations labelled 'police property' - Officers gain 'cop cultural capital' by recognising gang members & behaviour (Innes 2020) - Unequal distribution of police gaze - **Matza's Suspicion Framework (1969)** - **Incidental Suspicion** - crimefighter myth, skilled investigator determines whodunit by linking incident to suspect with means & motives. - **Bureaucratic Suspicion** - based on prior knowledge of individuals with criminal behaviour or socio-demographic traits, rounds up usual suspects - **FR & Suspicion (Fussey 2021)** - Reinforces bureaucratic suspicion - Watchlist contains custody images, target known individuals - Higher scores = more likely match, frame suspicions - **Officer decision-making (Skolnick 1966)** - High discretion in decisions, influenced by cop cultures values, leads to inconsistency 6. [How facial recognition impacts discretion] - Highlights disparities between law in books & in action - **Influence of police on FR** -- decisions about deploying, watchlist formulation, setting threshold scores for matches - **Influence of FR on police discretion** -- adjudication (decide if match is correct), in RFR -- operators review list of potentials and make decision 7. [FR, Existing Bias & Disproportion] - Bureaucratic suspicion highlights usual suspects but frames persons of interest as different to others regardless of evidence (Matza 1969) - FR surveillant gaze oriented towards watchlist - Discretion & Discrimination: - Most likely when there's no clear guidelines for decision-making - Where decisions depend on subjective judgement - Where there is considerable scope for individual discretion 8. [FR & Underlying Bias ] - **Existing Social Biases** - Disproportionate focus on young people and African Caribbean/ethnic minorities (The Lammy Review, 2017). - Police \"gaze\" unevenly distributed, targeting \"police property\" - Ethnic minorities, especially Black individuals, disproportionately stopped, searched, subjected to force, and arrested. - **Technological Biases in FR** - Lower accuracy for older individuals, women, and people of colour - Recent findings: equitable for ethnicity and gender, but age bias persists - **Remaining Disparities** - FR targets individuals on custody databases/watchlists, reinforcing existing bias. - Deployment in heavily policed areas overlaps with structural and demographic inequalities. 9. [Regulation of FR] - Deployments must be targeted, intelligence led, time-bound and geographically limited - Criticisms: - Too broad -- forces retain discretion in deployments & watchlist - No set offence severity threshold - Other jurisdictions have stricter rules **[Policing Domestic Violence]** **Overview of Policy Evolution** - Historically, police adopted a **non-interventionist approach**: - Focus on keeping the family unit together. - Domestic violence considered a private matter. - Viewed as \"garbage work\" by police culture. - Officers trained to mediate or separate rather than arrest. **Key Policy Milestones** - **Home Office Circular (60/1990):** - Interventionist approach encouraged; arrest assailants where offences committed. - **National Policing Improvement Agency (2008):** - Positive obligations under the Human Rights Act 1998: - Article 2 (Right to Life). - Article 3 (Freedom from Torture/Inhumane Treatment). - Article 8 (Respect for Private and Family Life). - Failure to arrest could risk victim safety and expose police to legal challenges. - **College of Policing (2015):** - Updated to reflect policy/law changes and HMIC (2014) recommendations. - Reinforces \"positive action\" in domestic abuse cases. **Key Legal Developments** - **Section 76, Serious Crime Act 2015:** - Criminalized controlling/coercive behaviour in intimate/family relationships. - **Domestic Abuse Act 2021:** - Established a statutory definition of domestic abuse: - Any controlling, coercive, threatening behaviour, violence, or abuse between individuals aged 16+ in intimate or family relationships. **Reasons for Policy Change** - Women's liberation movement and media attention (e.g., *The Burning Bed*). - Rise of refuges for battered women. - Legal challenges (e.g., **Thurman v. City of Torrington -- victims not provided with equal protection under law, police sued for failure to protect**). - Research evidence: - **Minneapolis Domestic Violence Experiment** (Sherman & Berk, 1984): Highlighted the deterrent effect of arrests. **Domestic Abuse Statistics (2023)** - **2.1 million adults (16-59)** experienced domestic abuse. - **889,918 crimes** recorded as domestic abuse related. - **69,634 arrests** made for domestic abuse-related crimes. - **47,361 prosecutions** by CPS; **39,198 convictions** secured. - **566 convictions** specifically for coercive and controlling behaviour. **[Factors Influencing Arrest Decisions]** **Decreases Likelihood:** - Suspect fled. - Victim uncooperative/unlikely to prosecute. **Increases Likelihood:** - Severity of incident. - Use of weapon. - Presence of witnesses. **Research on Arrest (Key Studies)** - **Minneapolis DV Experiment (Sherman & Berk, 1984):** Arrest reduced recidivism by **50%** within 6 months. - **Replication Studies:** - **Omaha, NB:** No difference between arrest, separation, or mediation. - **Milwaukee, WI:** Arrest deterred employed but not unemployed offenders. - **Charlotte, NC:** Arrest not more effective. - **Colorado Springs, CO:** Mixed results (official data vs. victim reports). - **Miami, FL:** Arrest + follow-up services reduced revictimization. - **Pooled analyses:** Arrest reduced revictimization by up to **25%**. **Benefits of Arrest** - Creates victim safety and disrupts abuse patterns. - Enables evidence collection (e.g., DNA, fingerprints). - Allows bail conditions to protect victims. - Signals seriousness to both victim and perpetrator. **Victims' Expectations and Experiences** - **Key Factors (Robinson & Stroshine, 2005):** 1. **Voice:** Feeling heard. 2. **Neutrality:** Fair, unbiased decisions. 3. **Ethicality:** Respectful treatment. 4. **Accuracy:** Competent responses. 5. **Correctability:** Ability to fix mistakes. - **Positive Interaction:** Collecting evidence, making arrests, courteous and empathetic demeanour. - Victims prioritize respectful behaviour over technical actions. **Challenges to Justice** - Lack of victim voice can deter sharing critical details. - Dismissive police responses undermine trust and satisfaction. **Case Examples (Home Affairs Select Committee, 2008):** 1. **Victim strangled:** Police asked if arrest was wanted; victim, fearing for safety, couldn't decide. 2. **Unaddressed threats:** Police failed to act despite ongoing danger; victim felt unsupported. **Ecological Model of Domestic Violence** - **Four Levels:** Individual, Interpersonal, Community, Societal. - Illustrates complexity and limitations of \"one-size-fits-all\" approaches. **Key Points from Robinson & Stroshine (2005): Victim Satisfaction** - Victim satisfaction depends on whether police actions meet their expectations. - **Types of Disconfirmation:** - **Positive disconfirmation:** Police exceed expectations. - **Zero disconfirmation:** Police meet expectations. - **Negative disconfirmation:** Police perform worse than expected. - **Police Behaviour:** Actions like collecting evidence, writing reports, making arrests. - **Police Demeanour:** Courteousness, respectfulness, taking time to listen. - Satisfaction is higher when victims' expectations about police behaviour and demeanour are met. - **Demeanour is more critical** to victim satisfaction than police actions (behaviour). **Risk-led approach to policing DV** - IDVAs -- linking tools to interventions - MARACs -- facilitating partnerships work amongst key agencies - DASH: - develop risk tools based on analysis of problem - 27 questions identify risk of further DV - **Research: National Mapping** - 43 forces describe approach - Identifies approaches or models of risk-led policing - 3 forces represent 3 diff models chosen for fieldwork - **Research: Fieldwork in 3 forces** - 61 interviews w/ police - 120 hours of observing, 1296 survey responses, analysis of 2000 DV incidents - **Research: US & UK police perception of DV** - Online survey of 773 officers -- 20 risk factors 'a small constellation article' & violent vs non-violent scenario - Within-country & between-country differences - Risk factors: weapon, strangulation, injury, escalation of abuse - **Management of risk** - Activities undertaken by agencies to reduce likelihood of further victimisation - "Doing something" based on risk, but few options for victims not considered high risk - MARACs struggling with demand - Multi-agency partnerships enshrined in law and practice - **Challenges (Hoyle 2008)** - Ambiguity in risk concepts - Uneven training - Implications for delivering services across risk spectrum **Coercive Control** - **Coercion** -- physical & sexual violence, intimidation. Threats - **Control** -- isolation, exploitation, micro-regulation of behaviour - **Characterised as a liberty crime -- disproportionately men against women** - **Criminalisation:** - **England & Wales -- Section 76 of Serious Crime Act 2015** - **Scotland -- Section 1 of Domestic Abuse Act 2018** - **Ireland -- Section 39 of DV Act 2018** - **Northern Ireland -- Domestic Abuse & Civil Proceedings Act 2021** - **Brennan 2020 -- half as many coercive control crimes result in arrest vs DVA (25% vs 50%) & a far higher proportion are discontinued** - **ONS 2023 -- Only 566 convictions for coercive control in 2023** - **US vs UK** - **UK officers more likely than USA to believe controlling behaviour was abuse and took more actions in non-violent scenario** - **Challenges for effective agency responses** - **Negative impact of police occupational culture** - **Lack of understanding by officers on DVA dynamics** - **Police frustration with being called to 'minor issues'** - **Challenges with interpreting broad DVA definition** - **Struggles with identifying subtle behaviours vs physical assault (dog bowl)** - **Improving agency responses** - **Implementation of training programmes to increase knowledge** - **Tools which prompt practitioners to evaluate patterns rather than incidents of abuse** **[Policing, technology & hate crime]** **Cybercrime - definitions** - **Traditional crimes** -- statics, socially & politically rationalised, offenders often socio-economically marginalised - **Cybercrimes** -- distanciated, lack of social & political recognition, offenders socio-economically privileged - **3 categories of cybercrime (Wall 1998)** - Facilitates existing criminal activity - New crimes recognised by existing laws - New 'harms' unrecognised by laws - **Counting cybercrime** - Cyber-enabled (existing crimes transformed e.g. hate) - Cyber-dependents (new crimes e.g. viruses) - Personal cybercrime -- Eurobarometer Cybersecurity Survey (ID theft) - Increased victimisation -- selling on auction sites, using public computers, younger & poorer **Cybercrime law** - **Computer Misuse Act 1990** - Hacking, malware, denial of service (inc terrorism) - **Offences Against the Person Act 1861, Malicious Comms Act 1988, Crime and Disorder Act 1998** - Threats of violence or menacing messages - Hate speech/hateful media posts (grossly offensive, threatening, racial/religion/sexual orientation hatred - Prosecutor operated a high threshold -- consider public interest and impact upon targeted victim - Must be satisfied that communication is not protected under free speech principle - **Protection from Harassment Act 1997** - Online stalking - **Criminal Justice & Courts Act 2015** - Disclosing private sexual images without consent - **Sexual Offences Act 2003** - Online Grooming **Cybercrime Regulation** - Regulatory mechanisms - Mix of public/private, state/non-state, national/inter-national institutions - Current policy initiatives in UK: - **UK National Security Strategy (HM Govt 2015)** - Cybercrime a tier 1 threat - **UK Cyber Security Strategy (Cabinet Office, 2022-2030)** - Invest £2.6 billion -- graduate jobs - Manage cyber risks - Protect against attack - Detect cyber events - **GCHQ National Cyber Security Centre** - Supports gov, industry - Education/training - UK first response to national cyber threat (e.g. WannaCry) - **National Crime Agency** - Coordinates child protection online/offline - Money laundering, fraud - Target hardening **Public/private partnerships** **Nhan & Huey 2008** -- 'nodal clusters' that form cybercrime reduction network - Government - Law enforcement - Private industry - General public **Dupont 2004** -- 5 forms of capital that shape nodal networks - Social capital -- maintain relations w/ other nodes - Cultural capital -- knowledge possessed by a node - Political -- theoretical knowledge of structures - Economics -- knowledge of markets and power - Symbolic capital -- organisation legitimacy **eCrime Partnership Mapping Study (Levi & Williams 2011)** - Perceptions of eCrime prevalence largely symmetrical - Significant gaps in cooperation frequency & quality between govt and finance sector - Third sector organisations and local govt on periphery of UKIA network **Explanations for poor cooperation** - Over-crowded cybersecurity space - CJS poor record in prosecution - Low levels of network capital **Technology as a Regulator** -- Williams (2015) Guardians Upon High - Applied Routine Activities Theory (RAT) to online identity theft. - Key components: Offender, Opportunity, Lack of Guardianship. - Insights from the Eurobarometer Special Cybersecurity Survey. **Guardianship Findings (Williams 2015):** - **Positive effect of Internet penetration**: - Countries with **high Internet penetration**: - Effective personal and passive physical guardianship reduces identity theft incidents. - Countries with **low Internet penetration**: - Increased identity theft incidents despite adoption of similar guardianship methods. - **Cybersecurity strategy maturity**: - Countries with mature strategies: - Passive guardianship results in decreased identity theft incidents. **Technology as a Regulator** - **Tri-modal regulation**: 1. Law - retroactive. 2. Market/Social - retroactive. 3. Technical - proactive. - **Situational crime prevention**: 1. Increase perceived effort of committing a crime. 2. Increase perceived risks to offenders. 3. Reduce anticipated rewards of crime. - **Lessig\'s \'Digital Realism\'**: 1. Technology disrupts human action. 2. Malleable, pervasive, rapidly adaptive, preventative, and less contentious. **Criticism of Technology as Regulator** - Hosein, Tsiavos, and Whitley (2003): - Technology is a biased cultural artifact. - Code-writers function as alternative sovereigns. **New Tech as 'Digital Forensic Data'** **Social Media as a Mirror** - Each social media use is a sensor of offline phenomena which detects social and environmental disorder via signs e.g. graffiti, vandalism & posts a related tweet - Users can publish info about disorder in 4 ways - As victims - As first-hand witnesses - As second-hand observers (media reports or rumour) - As perpetrators **Social Media Driver for Action** - Opposing online communities emerge due to algorithms - Cyclical reaction to divisive offline & online events increases polarisation - Engagement algorithms push content containing extreme views **Predicting Crime Patterns** - **Bendler 2014** -- relationship between activity on twitter in SF and location/likeliness of crime types. Absence of twitter posts predictive in relation to burglary & theft - **Malleson & Andresen 2014** -- measure mobile populations at risk from violent crime, found hotspots, twitter data represents mobile populations at higher resolutions than other data. - **Gerber 2014** -- statistical analysis of tweets correlates with crime in Chicago - First two studies use twitter as a proxy for population density -- ignores rich content of tweets themselves - Third study combines analysis of text but fails to show how topics relate to crime types - More nuanced approach -- classification of crime related content in tweets and correlate with recorded crime - **Williams 2017** - Studied posts referencing neighbourhood degradation in London. - **'Broken windows' tweets**: Positively correlated with criminal damage, theft from motor vehicles, drug possession, and violence in low-crime areas. - **Tweeting patterns**: More tweets about neighbourhood decline in low-crime areas than high-crime areas. - **Model instability**: Predictions unstable across London due to uneven tweet distribution. **HateLab** - Monitors and counters online harms - Focus areas -- anti-black, anti-Asian, anti-Muslim, anti-women, anti-LGBT - Platforms analysed -- Twitter, Reddit, 4Chan, Telegram - Machine learning classifiers trained on expert-annotated data for rapid detection - Partnerships with trusted flaggers for data sharing - Pilot use in law enforcement, government **Predicting Hate Crime Patterns -- Muller & Schwarz 2020** - Trump's anti-Muslim tweets were almost 2x more likely to be retweeted by followers. - 58% spike in hashtags like \#BanIslam and \#StopIslam. - A one standard deviation increases in Twitter usage correlated with a 32% rise in anti-Muslim offline hate crimes since 2016 primaries. - The rise was specific to anti-Muslim assault and vandalism, not other crimes, ruling out reporting/recording bias. - **In other research:** 1,000 additional tweets → 4% rise in racially/religiously aggravated harassment in a given month within a London LSOA. - Significant interaction between hate speech count and BAME population. - In LSOAs with 70% BAME population and 300 hate tweets/month, predicted incidence rate of racially/religiously aggravated violence: 1.75--2. **Applications** - Regional real-time monitoring detects tensions before street-level events - Relationship between online & offline tensions is bi-directional and dynamic - Hate crime is a process not a discrete event & can move online to offline or vice versa - Statistical studies miss personal & complex interactions Hate Crime Definitions & Laws **Hatred Motivation Model** - Common in Europe; used in England & Wales. - Requires identity-based hostility in law (verbal or historical evidence). - Excludes crimes targeting vulnerability (e.g., theft from disabled victims). **Group Selection Model** - Used in the United States. - No need for prejudice or hostility in law. - Focuses on intentional victim selection based on group identity. - Captures more cases but harder to prove intent. **NPCC & CPS Definitions** - **Hate Motivation**: Crimes/incidents where hostility or prejudice toward a group influences victim selection. - **Hate Incident**: Non-crime incident perceived by anyone as motivated by hostility/prejudice based on disability, race, religion, sexual orientation, or transgender identity. - **Hate Crime**: Criminal offense perceived as motivated by hostility/prejudice toward the same characteristics. - **Incitement to Hatred**: Criminal offenses include incitement to racial, religious, or sexual orientation-based hatred. - Overlaps with gang crimes, terrorism, political violence, and war. **Hate Crime Legislation** **Hate Crimes**: - **Crime and Disorder Act 1998** (amended by *Anti-terrorism, Crime & Security Act 2001*): - Covers racially/religiously aggravated assault, criminal damage, public order offenses, and harassment. - *Excludes*: sexual offenses, burglary, robbery, fraud, forgery, homicide, etc. **Enhanced Sentencing**: - **Criminal Justice Act 2003**: Aggravating factors include hostility toward race, religion, sexual orientation, and disability. - **Sentencing and Punishing Offenders Act 2012**: Added hostility toward transgender identity. **Stirring Up Hatred**: - **Public Order Act 1986**: Incitement to racial hatred (*threatening, abusive, or insulting*). - **Racial and Religious Hatred Act 2006**: Incitement to religious hatred (*threatening only*). - **Immigration and Criminal Justice Act 2008**: Incitement to hatred based on sexual orientation (*threatening only*). **CPS Guidance on Hate Crime** - **Hostility**: Ill-will, prejudice, unfriendliness; not always the main motivation. - **Proving Hostility**: Demonstration (e.g., slurs) is sufficient, though some may argue it's not always indicative of hate. - **Motivation**: Includes social media, hate symbols, or past hate crimes; hard to prove consistent hate. - **Importance of Demonstration**: Shows intentional hostility, impacting victims and communities more than motivation. - **Trial** **Procedures**: Guilty pleas for lesser charges not accepted in hate crime cases. **Charging & Sentencing (CJA s146)** - **Sentence Disparity**: CDA allows up to 400% higher sentences for racially or religiously aggravated offences compared to basic offences (e.g., common vs. aggravated assault). - **CJA s146 Issue**: Max sentence for basic offence, leading to fewer convictions for aggravation. - **Evidence Gathering**: Aggravation under CJA s146 often considered only at sentencing, leading to lost evidence and weaker investigations. - **Discrimination in Investigation**: Hate crimes against LGBT, disabled, and transgender victims less likely to be investigated as such compared to racial/religious offences. - **Lack of Flagging**: No tracking of repeat hate offenders for s146 crimes in PNC or prison/probation services. **Policing Hate Crime -- CSEW Data** **6 Hate crime strands:** - Race - Religion - Sexual orientation - Disability - Transgender identity **Hate crimes can take forms including:** - Physical assault, damage to property, offensive graffiti - Threat of attack, letters, calls, intimidation - Verbal abuse - Cyberhate **Police Statistics:** 23/24 -- 140,000 hate crimes, decrease of 5% from year before. - Majority race (70%) - Sexual orientation (16%) - Disability (8%) - Trans (3%) - Religious (8%) - 2% of all hate crimes were flagged as online - 40% increase of previous year -- also a rise in prosecutions for online hate - 47% of hate crimes came to attention of police - 27% of victims of household hate crime had been victimised more than once in previous year -- more likely perpetrated by someone you know **Police Encounter & Policy** **Victim Satisfaction** - 55% of hate crime victims satisfied - Hate crime victims more likely to be dissatisfied than CSEW victims - 70% of hate crime victims though police treated them fairly vs 79% of general CSEW victims **Police Strategy & Operational Guidance** **2005 National Policing Hate Crime Strategy (CoP)** - Positively respond to hate crime - Increase trust in police and investigation - Reduce underreporting **2014 Hate Crime Operational Guidance (CoP)** - Detailed definitions of hate crime - Legislative frameworks - Data sharing **Positive Practice:** - Gwent Police: use of hate crime 'champions' for victim contact and follow-up support, effective audit arrangements for hate crime flag application, creation of a 'cyber community support officer' for online hate crime. - Greater Manchester Police: training on responding to transgender victims. - Nottinghamshire Police: comprehensive hate crime risk assessments. - West Yorkshire Police: effective partner work to protect victims. - West Yorkshire Police: involving communities in scrutinizing police approach to hate crime. **Poor Practice:** - Inconsistent advice from control room staff due to lack of training. - Officers hesitant to probe underlying hate crime factors. - Inaccurate recording of hate crimes. - Inadequate data gathering on victims and incidents. - Ill-equipped to handle online hate crime. - Poor risk assessments for repeat victims. - Not all victims automatically referred to victim support.