Crime prevention and community policing

In short

Crime prevention is any deliberate action - by the State, the police, planners or citizens - intended to reduce the risk of a crime occurring or the harm it causes, before and apart from punishment. Community policing is the organisational strategy that makes the police work *with* residents as co-producers of safety, through beats, liaison groups, joint patrols and problem-solving, rather than only reacting to calls. The topic is examined in criminology, police administration and social-work papers (UGC-CBCS BA/B.Sc Criminology, MA Criminology, UGC-NET Criminology), and in recruitment for State police (SI/DSP), CAPFs, UPSC and State PSCs. Employers also test it for roles in corporate security, retail loss prevention, city-safety and NGO/CSO victim-support work. Working knowledge matters because the Bharatiya Nagarik Suraksha Sanhita, 2023 (BNSS) retains and restates the police's preventive powers, and nearly every State now runs a named community-policing scheme.

Explained

Concept, scope and classifications

• Crime prevention - anticipating, recognising and appraising a crime risk and initiating action to remove or reduce it (the classic police-training definition). It is *forward-looking*; detection and prosecution are *backward-looking*.
• Brantingham and Faust (1976) public-health model:
  • Primary - stops crime before it happens, aimed at the whole population or the environment (street lighting, locks, civics education, employment).
  • Secondary - targets identifiable at-risk people, places or situations (hot spots, school drop-outs, repeat-victim households).
  • Tertiary - works on known offenders to prevent repetition (probation, correctional treatment, after-care, de-addiction, open prisons).
• By method (van Dijk / Tonry-Farrington): (i) developmental / social prevention - attacks root causes: poverty, parenting, schooling, substance abuse; (ii) situational - alters the immediate setting of the crime; (iii) community - builds collective efficacy and informal social control; (iv) criminal-justice prevention - deterrence, incapacitation, rehabilitation.
• Deterrence theory basis: Beccaria and Bentham - prevention depends on certainty, celerity (swiftness) and severity of punishment, with certainty mattering most. Visible policing raises perceived certainty.
• Victim-centred prevention: repeat victimisation is highly concentrated, so protecting the recently victimised ("the hottest of hot spots") prevents more crime per rupee than general patrol.

Theories behind situational prevention

• Routine Activity Theory (Cohen and Felson, 1979): a crime needs the convergence in time and space of a motivated offender, a suitable target and the absence of a capable guardian. Remove any leg of the triangle and the crime does not occur. Eck's problem analysis triangle adds handlers (for offenders), guardians (for targets) and place managers (for places).
• Rational Choice Perspective (Clarke and Cornish): offenders make bounded, situation-specific choices; raise effort and risk, cut reward, and the choice changes.
• Crime Pattern Theory (Brantingham and Brantingham): offences cluster along offenders' awareness space - nodes, paths and edges; this is why crime maps show hot spots along transport corridors and market edges.
• Broken Windows (Wilson and Kelling, 1982): visible disorder signals that nobody is in charge and invites serious crime; hence attention to litter, graffiti, abandoned vehicles, illegal parking. Criticised for drifting into aggressive "zero-tolerance" order-maintenance that harms poor and minority groups.
• Social Disorganisation / Collective Efficacy (Shaw and McKay; Sampson): crime is high where residential instability, heterogeneity and poverty weaken informal control; community policing tries to rebuild that control.
• Defensible Space (Oscar Newman, 1972) and CPTED (C. Ray Jeffery, 1971): design that creates territoriality, natural surveillance, access control, image/maintenance and activity support.

Situational crime prevention and CPTED in practice

• Clarke and Eck's 25 techniques group under five mechanisms:
  1. Increase the effort - target hardening (grilles, shutters, steering locks), control access, screen exits, deflect offenders, control tools/weapons.
  2. Increase the risk - extend guardianship, assist natural surveillance (lighting, trimmed hedges), reduce anonymity, use place managers, strengthen formal surveillance (CCTV, beat constable).
  3. Reduce the rewards - conceal and remove targets, identify property (engraving, IMEI, chassis/engine numbers, CEIR blocking of stolen handsets), disrupt markets, deny benefits (ink tags, PIN-locked devices).
  4. Reduce provocations - cut frustration and stress (queue management), avoid disputes (segregating rival crowds), reduce emotional arousal, neutralise peer pressure, discourage imitation.
  5. Remove excuses - set rules, post instructions, alert conscience, assist compliance (litter bins, public urinals), control drugs and alcohol.
• CPTED strategies in Indian settings: lit and continuous footpaths; "eyes on the street" through mixed land use; no dead walls along long compound boundaries; safe audits of bus stops, subways and campus routes; women's safety audits used in Safe City projects.
• Displacement and diffusion: prevention may push crime elsewhere (spatial, temporal, target, tactical, offence or perpetrator displacement) or may spread benefits beyond the treated area (diffusion of benefits, "free rider" effect). Evidence shows displacement is usually partial, not total - so a scheme can still be a net gain. Measured with the Weighted Displacement Quotient (WDQ).
• Limitations: situational measures can be costly, can create fortress-like exclusionary spaces, raise privacy concerns (mass CCTV), and do nothing about causes - hence they are combined with social prevention.

Problem-oriented policing and the SARA model

• Problem-Oriented Policing (Herman Goldstein, 1979): the unit of police work should be the recurring *problem*, not the individual incident; the police should diagnose causes and craft tailored responses, using arrest as one option among many.
• SARA cycle:
  1. Scanning - identify and prioritise recurring problems from FIR/CCTNS data, Dial-112 calls, beat registers and residents' complaints.
  2. Analysis - study the problem with the crime triangle: who, where, when, how, what facilitates it; map hot spots, hot times, hot products, hot victims.
  3. Response - design a response, usually partly non-police (municipal lighting, excise action on illicit liquor, transport department on unauthorised stands, school counselling).
  4. Assessment - measure whether the problem fell, allowing for displacement and for trends in a control area; feed back.
• Hot spot policing: concentrating patrol on the small share of addresses producing most crime; the "Koper curve" suggests roughly 10-15 minute visible stops at a hot spot, at irregular intervals, give the best residual deterrence - long static postings waste resources.
• Intelligence-led and predictive policing: uses CCTNS/ICJS data, call records and mapping to forecast risk; raises bias, profiling and data-protection concerns. In India, data handling currently sits under the IT Act, 2000 and the SPDI Rules; the Digital Personal Data Protection Rules, 2025 impose Data Fiduciary duties only from 13 May 2027, and the DPDP Act in any case exempts specified State processing.
• Evidence hierarchy: evaluations are ranked by design - the Maryland Scientific Methods Scale (Level 1 correlation to Level 5 randomised trial). "What works" reviews (Campbell Collaboration, EMMIE framework) rate hot spot policing and focused deterrence as well supported, and generic foot patrol or scared-straight programmes as weak or harmful.

Community policing: meaning, elements and models

• Definition: a philosophy and organisational strategy of police-community partnership for joint identification and solution of local problems of crime, disorder and fear, with decentralised, permanently assigned, accountable officers.
• Roots: Sir Robert Peel's principles (1829) - "the police are the public and the public are the police"; prevention of crime is the basic mission; police power depends on public approval and consent, not force.
• Three core components: (i) community partnership; (ii) problem solving (SARA); (iii) organisational transformation - flattened hierarchy, geographic accountability, beat officers with discretion, changed training, appraisal and reward criteria.
• Key operating tools: permanent beat system and beat books; foot and cycle patrol; community liaison / police station management committees; mohalla and peace committees; neighbourhood watch and resident-welfare-association tie-ups; school and college cadet programmes; women's and senior-citizen help desks; grievance camps; police-public meetings at the station; "friendly" FIR reception and victim assistance.
• Indian statutory and policy base:
  • Police Act, 1861 - the colonial parent Act, still in force in several States; built for regime control, not service delivery.
  • Model Police Act, 2006 (drafted by the Soli Sorabjee committee) - expressly provides for community policing: police-community liaison, village/ward level committees, a service orientation, Police Establishment Boards, State Security Commission, Police Complaints Authority. Several States (Kerala Police Act 2011, Bihar, Rajasthan, Gujarat, Punjab, among others) enacted new Acts drawing on it; it is a model, not binding law.
  • Prakash Singh v. Union of India (2006) - Supreme Court's seven directives: State Security Commission, merit-based DGP selection with a minimum two-year tenure, two-year tenure for key field officers, separation of investigation from law and order in towns above a stated population, Police Establishment Board, Police Complaints Authority at State and district level, and a National Security Commission. Compliance across States remains partial.
  • Committees: Gore Committee on Police Training (1971-73) - training for a service role; National Police Commission (1977-81), 8 reports - the foundation document, recommended a Police Act replacement and a District Police Complaints Authority; Ribeiro (1998-99), Padmanabhaiah (2000), Malimath Committee (2003) on criminal-justice reform, and the Second Administrative Reforms Commission, 5th report, "Public Order" (2007).
  • BNSS, 2023 (replacing the CrPC, 1973) retains the preventive machinery: security for keeping the peace and for good behaviour, dispersal of unlawful assemblies, removal of public nuisances, orders in urgent cases of nuisance or apprehended danger (section 163 BNSS, the former CrPC s.144), the duty of every police officer to interpose to prevent a cognizable offence, preventive arrest where a cognizable offence is about to be committed, and information to be given by village officers and residents about certain offenders. Substantive offences are now in the Bharatiya Nyaya Sanhita, 2023 and evidence in the Bharatiya Sakshya Adhiniyam, 2023.
• Indian schemes (named, and commonly asked): Janamaithri Suraksha Project, Kerala (2008) with beat officers and Janamaithri Samithis; Friends of Police, Tamil Nadu (Ramanathapuram, 1993); Mohalla committees, Bhiwandi and Mumbai (from 1988), credited with communal-peace maintenance; Maithri and Hyderabad "Nenu Saitham"/She Teams, Telangana; Suraksha Setu, Gujarat; Samadhan, Madhya Pradesh; Community Liaison Groups and Yuva/Youth outreach, Delhi Police; Student Police Cadet (SPC) scheme - Kerala 2010, taken nationwide from 2018 by MHA and the Ministry of Education; Mahila Police Volunteers; Village Defence Committees/Guards in Jammu and Kashmir; Gram Rakshak/Gram Suraksha Dal in several States; Police Mitra/Prahari variants.
• Supporting national infrastructure: CCTNS (crime and criminal records across police stations) and ICJS linking police, courts, prisons, prosecution and forensics; ERSS Dial 112; Nirbhaya Fund Safe City projects in major cities (CCTV, safety islands, women's help desks); NAFIS fingerprint database; CEIR for blocking stolen mobile handsets; NCRB *Crime in India* and BPR&D *Data on Police Organisations* as the standard data sources.
• SMART policing (stated by the Prime Minister in 2014): Strict and Sensitive, Modern and Mobile, Alert and Accountable, Reliable and Responsive, Techno-savvy and Trained.

Advantages, limitations and evaluation

• Advantages: better information flow and intelligence; reduced fear of crime even where recorded crime is unchanged; faster, locally tailored responses; improved police legitimacy and willingness to report; fewer complaints and less use of force; cheap compared with more personnel; helps communal and crowd-conflict management.
• Limitations and criticisms: vague, elastic concept measured loosely; capture by dominant caste, class or political interests in the committee; exclusion of women, migrants, slum dwellers, minorities; "cosmetic" implementation as photo-opportunity meetings; risk of vigilantism and of outsourcing coercion to untrained volunteers; frequent transfers destroy the beat officer's local knowledge; resource squeeze - severe personnel shortfalls (actual civil police strength is roughly 150-160 per lakh population against a sanctioned level near 195, far below the often-quoted UN reference of about 222; BPR&D publishes the exact figure each year, so quote the current edition); heavy VIP-security and law-and-order duties leave little patrol time; appraisal systems that still reward arrests and detections, not prevention.
• Evaluation design: compare treated area with a matched control area before and after; report crime rate, clearance rate, repeat-victimisation rate, fear-of-crime and satisfaction scores from a victimisation survey, and WDQ for displacement; beware regression to the mean when hot spots are picked on one unusually bad year, and recording-change artefacts (a successful outreach drive raises reported crime).

Key points

• Crime prevention - action to remove or reduce the risk of crime before it occurs, distinct from detection and punishment.
• Primary / secondary / tertiary - whole population and environment / at-risk people and places / known offenders.
• Routine Activity Theory - motivated offender + suitable target + absent capable guardian; break any leg.
• Problem analysis triangle - handler, guardian, place manager are the controllers of the three legs.
• Rational choice - raise effort and risk, cut reward, provocation and excuses: the logic of situational crime prevention.
• 25 techniques of SCP - Clarke and Eck's matrix of five mechanisms × five techniques.
• CPTED - natural surveillance, access control, territoriality, maintenance, activity support; defensible space is Newman's version.
• Broken windows - disorder breeds crime; the basis, and the weakness, of zero-tolerance policing.
• Displacement vs diffusion - crime moving away vs benefits spreading; usually partial, measured by WDQ.
• SARA - Scanning, Analysis, Response, Assessment; the operating cycle of problem-oriented policing.
• Hot spot policing - concentrate on the few addresses generating most crime; short, repeated, irregular visits (Koper curve).
• Community policing - partnership + problem solving + organisational transformation; consent-based, from Peel's principles.
• Model Police Act, 2006 - model law with community-policing, liaison bodies and accountability mechanisms; States adopt it voluntarily.
• Prakash Singh (2006) - seven directives including DGP two-year tenure, Police Establishment Board, Police Complaints Authority, separation of investigation from law and order.
• BNSS, 2023 - carries forward preventive powers: peace bonds, dispersal of unlawful assembly, s.163 orders, preventive arrest, duty to interpose.
• Indian schemes - Janamaithri Suraksha (Kerala), Friends of Police (Tamil Nadu), mohalla committees (Bhiwandi/Mumbai), Suraksha Setu (Gujarat), Student Police Cadet (national since 2018).
• Data sources - NCRB *Crime in India* for crime, BPR&D *Data on Police Organisations* for strength and ratios.
• Repeat victimisation - prevention concentrated on recent victims yields the highest return per rupee.

Formulas and examples

Example 1 - crime rate and clearance rate. A city of 24,00,000 people records 9,600 cognizable offences in a year; investigation is completed in 9,000 of them and 6,240 are charge-sheeted.

• Crime rate = 9,600 ÷ 24,00,000 × 1,00,000 = 400 per lakh population.
• Charge-sheeting rate = 6,240 ÷ 9,000 × 100 = 69.3%.
• If the next year's offences fall to 8,400 with population 24,50,000: rate = 8,400 ÷ 24,50,000 × 1,00,000 = 342.9 per lakh, a fall of 14.3% on the rate (and 12.5% on the raw count) - always compare *rates*, not counts, when population moves.

Example 2 - police-population ratio and shortfall. A district of 18,00,000 people has 2,520 civil police in position against a sanctioned 3,420.

• Actual ratio = 2,520 ÷ 18 = 140 per lakh; sanctioned ratio = 3,420 ÷ 18 = 190 per lakh.
• Vacancy = 3,420 − 2,520 = 900, i.e. 26.3% of sanctioned strength.
• To reach 190 per lakh in position, 900 more are needed; to reach the often-quoted reference of 222 per lakh, 222 × 18 = 3,996, i.e. 1,476 more.

Example 3 - displacement and diffusion (WDQ). A CCTV-and-lighting scheme covers a market (target). Street thefts, before → after, over equal 6-month periods:

• Cr = 190 ÷ 200 = 0.95.
• Expected target after = 120 × 0.95 = 114; actual 60 → denominator = 60 − 114 = −54 (54 crimes prevented).
• Expected buffer after = 40 × 0.95 = 38; actual 50 → numerator = 50 − 38 = +12 (12 crimes displaced).
• WDQ = 12 ÷ (−54) = −0.22: displacement occurred, but it absorbed only 22% of the gain. Net prevented = 54 − 12 = 42 thefts. A naive before-after figure of 60 would have overstated the effect by 43%.

Example 4 - benefit-cost of a prevention project. The same scheme costs ₹1.20 crore (capital plus 5 years of maintenance). It prevents, net, 42 thefts per half-year = 84 per year; the average loss plus investigation and court cost per theft is taken as ₹35,000.

• Annual benefit = 84 × 35,000 = ₹29,40,000; over 5 years = ₹1.47 crore.
• BCR = 1.47 ÷ 1.20 = 1.23 - positive but thin; it turns on the cost-per-crime assumption, which should be stated, not hidden.
• If fear-of-crime reduction brought 10% more evening footfall and the civic body valued that separately, the BCR would rise - but such benefits must be measured, not assumed.

Example 5 - victimisation survey and the dark figure. A survey of 4,000 households finds 480 experienced at least one property crime in the year; 216 of them reported to the police.

• Victimisation rate = 480 ÷ 4,000 × 100 = 12%.
• Reporting rate = 216 ÷ 480 = 45%; dark figure = 55%.
• Police records would show 216 ÷ 4,000 × 1,00,000 = 5,400 per lakh households, against a true 12,000 per lakh - the recorded rate captures 45% of reality. A community-policing drive that lifts reporting to 70% would *raise* recorded crime by 56% while actual crime was flat: an evaluation must separate the two.

Example 6 - patrol coverage (Koper logic). A beat has 15 km of street. A constable on a motorcycle patrols effectively for 6 hours of an 8-hour shift at an average 20 km/h.

• Distance covered = 6 × 20 = 120 km → 120 ÷ 15 = 8 passes per shift, one about every 45 minutes of patrol time.
• Switching to five 15-minute foot stops at the beat's five worst addresses uses 75 minutes and, on hot-spot evidence, buys roughly 2-3 times that duration of residual deterrence at those addresses - a better allocation than uniform cruising.

Where it is used

• State police forces: beat and community-policing wings, crime-prevention cells, women's help desks, Dial-112 ERSS control rooms, hot-spot deployment planning from CCTNS data - roles from Constable/Beat Officer to SP (Law and Order).
• City government and urban planning: CPTED inputs to street lighting, subways, bus terminals and park design; women's safety audits under Nirbhaya Fund Safe City projects in major metros; smart-city command and control centres.
• Corporate and retail security: loss-prevention and shrinkage control in organised retail, electronic article surveillance, access control and CCTV design in IT parks, banks and warehouses - PSARA-licensed private security agencies, and risk roles in banks, insurers and e-commerce.
• Banking, telecom and payments: fraud-prevention teams applying "reduce the rewards" logic - PIN/OTP controls, transaction limits, IMEI and SIM-swap checks, CEIR blocking of stolen handsets.
• Correctional and NGO sector: probation and after-care officers, prison welfare, juvenile-justice boards and child-welfare committees under the Juvenile Justice Act, 2015, de-addiction and victim-support programmes - tertiary prevention.
• Research and policy: NCRB, BPR&D, State police research bureaus, Institute for Criminology and Forensic Science, university criminology departments and think tanks evaluating schemes and conducting victimisation surveys.

Interview questions

• Q: How does crime prevention differ from crime detection? A: Prevention acts before the offence to remove opportunity, motive or vulnerability; detection acts after it to identify and prosecute the offender. Prevention is measured by crimes that did not happen, detection by charge-sheeting and conviction rates.
• Q: State Routine Activity Theory and use it on vehicle theft at a railway station. A: Crime needs a motivated offender, a suitable target and no capable guardian in the same place and time. At the station the targets are unattended two-wheelers; so install a fenced paid stand with an attendant (guardian/place manager), lighting and CCTV, and require handlebar locks - removing two legs of the triangle.
• Q: What are the five mechanisms of situational crime prevention? A: Increase the effort, increase the risk, reduce the rewards, reduce provocations and remove excuses - Clarke and Eck's 25 techniques are five under each.
• Q: Is displacement a reason to abandon situational prevention? A: No. Evidence shows displacement is usually partial, and diffusion of benefits to nearby areas is as common; the test is net effect, measured with a buffer and a control area through the WDQ, not before-after counts in the target alone.
• Q: What is CPTED and name its principles. A: Crime Prevention Through Environmental Design - shaping the built environment to reduce crime through natural surveillance, access control, territorial reinforcement, image and maintenance, and activity support.
• Q: Explain the SARA model with an Indian example. A: Scanning, Analysis, Response, Assessment. Repeated night chain-snatchings on one road: analyse FIR and Dial-112 data for the exact stretch and hours, find unlit sections and an illegal liquor outlet, respond with municipal lighting, excise action and 15-minute patrol stops at peak hours, then assess against a comparable road.
• Q: What are the three components of community policing? A: Community partnership, problem solving, and organisational transformation - decentralised, permanently posted, locally accountable officers with discretion, and appraisal that rewards prevention.
• Q: Which committees and judgments drive police reform in India? A: The National Police Commission (1977-81, eight reports), Gore Committee on Police Training, Ribeiro, Padmanabhaiah and Malimath Committees, the Second ARC's "Public Order" report, the Model Police Act, 2006, and the Supreme Court's seven directives in *Prakash Singh v. Union of India* (2006).
• Q: Where do the police get preventive legal powers now that the CrPC is repealed? A: From the BNSS, 2023, which carries forward security for keeping the peace and good behaviour, dispersal of unlawful assemblies, removal of nuisances, orders in urgent cases under section 163 (the former CrPC s.144), the duty to interpose to prevent a cognizable offence, and preventive arrest.
• Q: Name two Indian community-policing schemes and what each does. A: Kerala's Janamaithri Suraksha Project posts a dedicated beat officer who visits every household in the beat and works with a Janamaithri Samithi; Tamil Nadu's Friends of Police enrols trained civilian volunteers to assist with information, traffic and local problem solving.
• Q: Community policing is criticised as elite capture. How would you guard against it? A: Fix the composition of the liaison committee with reserved places for women, SC/ST, minority and slum or migrant representatives, rotate membership, publish minutes and the action taken, and track whose complaints are acted on - not merely whether meetings were held.
• Q: Recorded crime rose after you started a community-policing drive. Has it failed? A: Not necessarily - outreach typically raises *reporting*, shrinking the dark figure, so recorded crime can rise while actual crime falls. Judge it with a victimisation survey, repeat-victimisation data, fear-of-crime scores and a control area, not with FIR counts alone.
• Q: Why do Indian police struggle to do preventive work despite policy support? A: Large vacancies against sanctioned strength, heavy law-and-order, court, escort and VIP-security duties, frequent transfers that erode beat knowledge, weak separation of investigation from law and order, and performance appraisal that still counts detections rather than prevented crime.

Common mistakes

• Treating community policing as a public-relations activity - meetings, melas and photographs - instead of a change in how the force is organised and appraised.
• Using raw crime counts instead of rates per lakh, or comparing two years without a control area, so a general trend or a population change is credited to the scheme.
• Assuming displacement is total and therefore writing off situational prevention; or ignoring buffer areas and claiming the full before-after fall as the benefit.
• Confusing the classifications - mixing up primary/secondary/tertiary with social/situational/community, or calling imprisonment "primary prevention".
• Attributing CPTED to Newman and defensible space to Jeffery; the order is Jeffery (CPTED, 1971) and Newman (defensible space, 1972).
• Citing the CrPC, the IPC or the Evidence Act as current law - they are the former laws, replaced by the BNSS, the BNS and the Bharatiya Sakshya Adhiniyam, 2023 - or quoting a repealed section number from memory.
• Calling the Model Police Act, 2006 binding on States, or presenting the *Prakash Singh* directives as fully implemented; police is a State subject and compliance is partial.
• Quoting police-population ratios or crime rates from memory without naming the year and source (NCRB *Crime in India*, BPR&D *Data on Police Organisations*), when these figures change every edition.

Related topics

Police organisation and administration in India · Crime investigation: FIR to charge-sheet · Victimology and victim compensation · Penology and correctional administration · Juvenile delinquency and the Juvenile Justice Act, 2015 · Theories of crime causation