The Cartographic School and Economic Theories of Crime

Before the 1820s, almost no government kept national, standardised records of crime. Prosecutions were logged locally, in inconsistent formats, and rarely aggregated into anything a researcher could compare across regions or years.

That changed in France in 1827, when the Ministry of Justice began publishing the compte general de l'administration de la justice criminelle, an annual national accounting of every person prosecuted before French criminal courts, broken down by department, age, sex, and offence type. Belgium built a comparable statistical apparatus around the same period under its young national statistical bureau.

This was not a neutral bureaucratic accident. The post-Napoleonic states of Western Europe were consolidating centralised administration, and a wave of officials and social reformers, many trained in the natural sciences, believed that society itself could be studied with the same tools used for astronomy, physics, and public health.

Crime, suicide, marriage, and literacy were treated as social facts with underlying regularities waiting to be discovered, an outlook sometimes called moral statistics, meaning the statistics of moral conduct rather than statistics with a moral agenda.

Two figures did more than anyone to turn this raw administrative data into a scientific movement. Andre-Michel Guerry, a French lawyer working inside the justice ministry, used the compte general to build the first true crime maps of a nation.

Adolphe Quetelet, a Belgian astronomer and mathematician, generalised the same statistical instincts into a broader theory of social regularity. Their combined output is remembered as the cartographic school, sometimes also called the geographic or statistical school of criminology, and it is the direct ancestor of every modern crime map, hotspot analysis, and official crime statistics program in use today.

Neither man was trained as a lawyer studying crime for its own sake, and that outsider position mattered. Guerry brought an administrator's habit of checking a claim against the register before repeating it, while Quetelet brought an astronomer's comfort with treating large, noisy datasets as evidence of an underlying regularity rather than a set of unrelated accidents.

Both habits were unusual for their era, when writing about crime still mostly meant moral commentary on the offender's character, and both left French and Belgian criminal justice administrations with a lasting practice of publishing annual crime statistics that continues into the present day.

In 1833, Guerry published Essai sur la statistique morale de la France, a study that took six years of the compte general and laid it out department by department on shaded maps of France, one of the earliest large-scale uses of choropleth mapping for social data anywhere in Europe. Guerry separated offences into two broad categories, crimes against persons and crimes against property, and mapped each independently rather than treating crime as a single undifferentiated total.

The result overturned a comfortable assumption of the period. Reformers had generally believed that poverty and low education directly bred crime, so the poorest and least literate departments should show the highest offending.

Guerry's maps showed something more complicated: the departments with the most property crime were often the wealthier, more urbanised, and more literate regions of northern France, including the area around Paris, while violent crime against persons clustered instead in poorer southern departments. Wealth and opportunity, not poverty and ignorance alone, tracked with property offending.

Guerry drew a methodological lesson that still holds up: a single national crime rate hides more than it reveals, and disaggregating by offence type and region can overturn intuitive causal stories. He also insisted that any factor correlated with crime, whether education, wealth, or population density, needed to be tested against the map rather than assumed.

This insistence on empirical mapping over armchair moralising is why historians treat Guerry, alongside Quetelet, as a founder of scientific criminology, decades before Cesare Lombroso's positivist school turned attention to the offender's body instead of the offender's geography.

Quetelet approached the same French and Belgian data as an astronomer trained to look for regularity in noisy measurements. In works published through the 1830s, most notably Sur l'homme et le developpement de ses facultes (1835), he argued that human traits, including the propensity to commit crime, cluster around a statistical norm the way physical measurements do, a concept he called l'homme moyen, or the average man.

If enough individual cases are pooled, Quetelet argued, the annual number of murders, thefts, and even the choice of murder weapon in a given society becomes almost eerily predictable from one year to the next, a claim he summarised as the constancy of crime.

Quetelet's most durable specific finding is what later writers named the thermic law of crime. Working through French and other European judicial statistics, he found that crimes against persons, meaning violent offences, rose in warmer months and in warmer southern regions, while crimes against property rose in colder months and in colder northern regions.

Quetelet did not claim heat itself caused aggression in any simple mechanical sense. His explanation leaned on seasonal patterns of social contact and opportunity: summer brought people outdoors, into taverns, markets, and festivals, increasing the friction and provocation behind violent confrontation, while winter drove people indoors and left homes and stores more predictably vulnerable to burglary and theft.

Quetelet's broader claim, that crime is a stable social product rather than a string of unrelated individual choices, was genuinely radical for its time. It shifted the unit of explanation away from the sinner or the criminal and toward society as a whole, since Quetelet argued that society itself prepares the crime and the individual is merely the instrument that carries it out.

That reframing is the direct conceptual ancestor of every later structural theory of crime, including the economic theory Bonger would build seven decades afterward, and it gave later reformers a statistical basis for arguing that crime prevention should target social conditions rather than only the punishment of individuals.

The cartographic school's core empirical claims have aged well. Modern criminology confirms that recorded property crime does show a real, if modest, seasonal pattern in many countries, and that aggregate crime rates in a stable society tend to move gradually rather than erratically, which is essentially Quetelet's constancy thesis restated in contemporary language.

Guerry's insight that offence types must be disaggregated and mapped separately is now standard practice in every police crime-mapping unit and every academic hotspot study, and GIS-based crime mapping used by police departments worldwide is a direct methodological descendant of his choropleth maps of France.

The institutional habit Guerry and Quetelet started, a state publishing its own crime data broken down by place and offence type, now spans very different legal systems: New York City's police department built its CompStat model on precinct-level crime mapping from 1994, England and Wales began publishing street-level crime maps to the public through the police.uk website in 2011, and India's National Crime Records Bureau has run the Crime and Criminal Tracking Network and Systems since 2009 to link crime records across police stations nationally.

The school's central weakness was that correlation on a map is not the same as a causal mechanism inside an individual. Guerry and Quetelet could show that wealthy, literate departments had more property crime and poorer, less literate departments had more violent crime, but neither could specify why a particular person in a particular department actually offended.

Later critics labelled this the ecological fallacy: inferring facts about individuals from patterns that hold only at the level of aggregated groups. A department can have a high theft rate for reasons, such as more valuable property being available to steal, that say nothing about the character or motives of any single resident living there.

There was also a data problem the cartographic school could not fully escape. The compte general and its Belgian equivalent recorded only crimes that reached prosecution, not crimes that occurred, so the maps reflected policing intensity and reporting behaviour as much as underlying offending. This is the same limitation that modern criminology addresses through victimisation surveys and the concept of the dark figure of crime.

Even with these limits, the cartographic school proved that crime could be studied as a measurable social pattern, which set the stage for every structural theory that followed, Bonger's economic theory included, and pushed nineteenth-century criminal justice reform debates toward evidence rather than only rhetoric.

Willem Bonger, a Dutch criminologist and sociologist working within a Marxist framework, published Criminalite et conditions economiques in French in 1905 as his doctoral dissertation, translated into English in 1916 as Criminality and Economic Conditions.

Bonger's argument extended the structural instinct of the cartographic school into a fully causal theory: it was not enough to say crime correlated with poverty or wealth on a map, he wanted to explain the psychological mechanism that connected a society's economic system to the choices of the people inside it.

Bonger's central claim was that capitalism, organised around competition, private accumulation, and the pursuit of individual profit, systematically cultivates egoism, a self-centred orientation that weakens the altruistic social sentiment that would otherwise restrain people from harming others for personal gain.

In a capitalist economy, Bonger argued, the wealthy learn egoism through the everyday practice of exploiting labour and competing for markets, while the poor learn it through the everyday experience of deprivation and exclusion from that same economy. Crime, in his account, is what egoism looks like once it runs into either desperate want or unchecked opportunity.

A frequently overlooked feature of Bonger's theory is that he explicitly rejected the idea that crime is a problem confined to the poor. He devoted substantial attention to what he called crimes of the well-to-do, arguing that fraud, embezzlement, and commercial offending were the wealthy's version of exactly the same egoistic pressure that produced theft among the poor, just channelled through different opportunities.

This made his theory a critique of capitalism as a whole rather than a theory of lower-class deviance alone, and it anticipated the twentieth-century literature on property crime and economic offending among elites long before that literature had a name.

Bonger drew a distinction between absolute poverty, meaning deprivation measured against a fixed subsistence floor such as inadequate food or shelter, and relative poverty, meaning deprivation measured against the standard of living and expectations that a person's own society has taught them to want.

He treated relative poverty as the sharper driver of crime, since a person can be materially secure by an earlier era's standard and still feel acutely deprived next to the consumption and comfort visible around them under capitalism. This distinction anticipated the logic of twentieth-century strain and relative-deprivation theories by several decades, and it remains one of Bonger's most cited contributions even among scholars who reject the rest of his Marxist framework.

Bonger's theory drew the same core methodological criticism that had already been levelled at Guerry and Quetelet: correlation between an economic system and a crime rate does not establish that the system causes the crime through the specific psychological mechanism claimed.

Critics also pointed out that Bonger's own comparative evidence, largely European crime statistics from capitalist economies of his period, offered no genuine control case, since he had no large-scale socialist economy's crime data to compare against at the time he wrote.

Later attempts to test his claims against state-socialist crime records in the twentieth century produced mixed results at best, undermining the strong version of his thesis that capitalism specifically, rather than inequality or economic instability more generally, is the necessary cause of crime.

Even so, the combined legacy of the cartographic school and Bonger's economics runs directly into contemporary criminology. Modern hotspot policing, GIS-based crime mapping, and the entire discipline that studies official crime statistics descend from Guerry's method of plotting offending by place and type.

Contemporary inequality-and-crime research, which consistently finds that income inequality predicts violent crime more reliably than absolute poverty does across many national datasets, is a direct descendant of Bonger's relative-poverty argument, even where present-day researchers no longer share his specifically Marxist framing of capitalism as the singular cause of crime.

A widely cited cross-national test of this link, by the economists Pablo Fajnzylber, Daniel Lederman, and Norman Loayza in 2002, found that income inequality predicted higher homicide and robbery rates across dozens of countries even after accounting for each country's own fixed characteristics.

The United Nations Office on Drugs and Crime has reported that Brazil and Nigeria together, with only about six per cent of the world's population between them, account for roughly a fifth of global homicides, a concrete illustration of how the inequality Bonger treated as central to crime plays out well beyond the European cases available to him.