C: Community Crime Prevention to Crime Mapping

Core idea

This is the operational chapter — the entries describe the tools and categories that contemporary criminal justice actually uses. The cluster reads as a journey across three layers. The community layer (community crime prevention, community sentences) describes how the system extends itself into neighbourhoods. The theoretical layer (constitutive criminology, the crime/deviance distinction) describes the categories the system uses and the postmodern objections to them. And the infrastructural layer (corporate crime, crime data, crime mapping) describes the things the system measures and, just as importantly, the things it overlooks.

Read together, the entries surface the topic's central tension: the criminal-justice apparatus has become extraordinarily sophisticated at measuring, mapping, and managing the everyday crimes of poor people — even as the much larger financial and physical harms perpetrated by corporations are, by design, kept outside the system's view.

Why it matters

The data infrastructure is the part of criminology most students underestimate. Once you understand what police statistics, victimisation surveys, and self-report studies do and do not capture — and why corporate-crime measurement is so weak — you can read most contemporary "crime is up / crime is down" arguments correctly. The community-and-prevention entries are equally important: they describe how the state has displaced part of its policing function onto residents themselves.

Mental model

The seven entries divide into three layers — community/operational, theoretical, and infrastructural — with crime data sitting underneath them all.

Copy sourceZoom

Community crime prevention and community safety

The 1990s saw the British state try to delegate part of its policing function. Where formal policing had failed to reach certain neighbourhoods — usually impoverished, often ethnic minority — and where the relationship with the police had broken down, the answer was to enlist residents themselves. Neighbourhood Watch schemes, CCTV, citizen reporting hotlines, partnership boards: a whole vocabulary of "community engagement" emerged. The criticisms have become standard. Most programmes are imposed from outside rather than emerging from local demand. They focus exclusively on low-level visible street crime, ignoring less visible offences. They constitute a form of net-widening, classifying more behaviours as nuisance and bringing them under control. And they leave the underlying conditions — economic exclusion, broken local services — entirely untouched.

Community sentences and community punishments

The whole category of non-custodial sanctions — fines, compensation orders, unpaid work, mandatory rehabilitation, curfews, electronic tagging — has expanded enormously since the 1980s. The driver is partly fiscal (prison is expensive) and partly evidential (reoffending rates from custody are high). The political resistance is permanent and predictable: the popular framing treats anything short of imprisonment as a "soft option", which limits how confidently any government can expand the category.

Constitutive criminology

A self-consciously postmodern intervention by Dragan Milanovic and Stuart Henry (1996). Their move is to redefine crime as harm — specifically, "the harm resulting from humans investing energy in harm-producing relations of power", with two sub-categories: harms of reduction (immediate loss to the victim, e.g. assault, theft) and harms of repression (restriction of the victim's future potential, e.g. intimidation, exploitation). The conceptual scaffolding is heavy — chaos theory, structuration, poststructuralism, psychoanalysis all wheeled in simultaneously — and the practical payoff is "replacement discourses" that subvert mainstream definitions of crime and give voice to the marginalised.

The objection O'Brien and Yar press is genuine: every question that constitutive criminology asks about "crime" (who defines it, whose interests does it serve, why is it variable across time and place?) can be asked of "harm" with equal force. Substituting one contestable category for another is not, by itself, progress.

Corporate crime

Author's argument: corporate crime causes financial harms in the hundreds of billions and physical harms running to thousands of deaths and serious injuries per major incident, yet it receives a tiny fraction of criminal-justice resources and is mostly handled through civil or administrative law rather than criminal sanction. This gap is a structural feature of the system, not an oversight.

Corporate crime is the offence committed by a legitimate business in the course of its ordinary activities — distinguished from "occupational crime" (offences committed by individuals in their employment) by being organisational in character. Edwin Sutherland was already calling for the discipline to take it seriously in 1947. The literature breaks the category into financial offences (fraud, tax evasion, mis-selling, price-fixing) and harms to bodies and environments (unsafe working conditions, pollution, complicity in state violence — Bhopal, Shell in Nigeria, private military contractors in Iraq).

The structural explanation for the under-attention is that corporate power exercises systematic influence over what counts as crime, what counts as enforcement, and what resources criminal-justice agencies allocate to which offences. Market liberals counter that this is rather the proper boundary of state intervention into business. Either way, the asymmetry is real and the discipline has yet to close it.

Crime and deviance

A small entry that does heavy conceptual work. Crime is the formal-legal category — behaviour prohibited and punishable under criminal law. Deviance is the informal-cultural category — behaviour that breaches social norms whether or not it is criminal. The two cross in interesting ways. Some behaviours are deviant but legal (consensual sadomasochistic sex). Some are criminal but not socially considered deviant (most motorists view speeding this way). And the historical traffic between the two categories is constant: criminalisation tends to follow shifts in cultural sensibility, and decriminalisation follows the reverse.

Holding the distinction in mind is what lets you ask the productive questions: why is this behaviour criminalised and that one not? What changed when this was decriminalised? Whose moral sensibility was tracked when the law was rewritten?

Crime data

The infrastructure on which most empirical criminology runs. Three principal sources, each with characteristic biases.

Police-recorded statistics (originating in 1830s France) are derived from the incidents reported to and then recorded by police. They miss everything not reported (most sexual offences, most domestic violence, much fraud) and they distort everything recorded (police discretion, force priorities, political pressure on what to count). They consistently show that property crime dominates the totals and that "the typical offender" is young and male.

Victimisation surveys (since the 1970s) ask a sample of households about their experiences of crime. They picked up huge dark-figure categories — rape, domestic violence — that police statistics had hidden, but they rely on respondents' recall and willingness to participate, and some social groups participate less than others.

Self-report surveys ask a sample of individuals to anonymously confess their own offending. They get at the offender side, including the social characteristics. They suffer from recall problems, fear of consequences despite anonymity, and a small but real population of respondents who exaggerate.

Qualitative methods — in-depth interviews, participant observation — sit alongside the quantitative ones, and most serious research now combines both.

Crime mapping

GIS-based overlays of police data, demographic data, school locations, business locations, and so on. Operationally useful: it lets a police force concentrate resources on hot-spots. Strategically loaded: it accelerates the drift toward a surveillance society in which any administrative dataset is potentially indexed against any other. The Chicago School was already mapping crime in the 1920s; what is new is the granularity, the integration with personal data, and the absence of any public conversation about which datasets are appropriate to overlay.

Practical application

When you read a "crime is rising" or "crime is falling" claim in the press, run it through the topic's filter.

Example

A police force announces with some fanfare that recorded burglaries in the city centre have fallen 22% year-on-year, citing improved CCTV coverage and a community partnership scheme that distributed property-marking kits. The mayor declares the city "safer than ever".

Apply the topic's filter. The headline number is police-recorded — it doesn't capture unreported burglaries, and the very same year the force published it, the local victimisation survey showed a flat rate. The metric reported is property crime; nothing in the press release addresses domestic abuse, online fraud, or workplace injury rates in the local logistics warehouses. The intervention is classic community crime prevention — Neighbourhood Watch enrolment, CCTV — but it is targeting precisely the crime type with the largest pre-existing visibility bias in the data. The fall therefore could easily be partly compositional (offenders shifting to harder-to-record crimes), partly displacement (offending moved to neighbouring areas), and partly real. The mayoral claim of "safer than ever" rests on whichever interpretation the audience is least equipped to challenge.

A criminologist reading the same press release is not opposed to falling burglary rates. They are opposed to the announcement being treated as evidence for any claim larger than "in this category, in this dataset, the number went down". The seven concepts in this topic give them the vocabulary to say so.

Related lessons