Biological and Biosocial Theories of Crime

Cesare Lombroso, an Italian physician working in the 1870s, is widely regarded as the founder of biological criminology. Examining soldiers and prison inmates, Lombroso concluded that a subset of criminals were atavists: evolutionary throwbacks whose bodies retained primitive features that marked them as biologically different from law-abiding citizens. He catalogued physical stigmata including a sloping forehead, pronounced brow ridges, large jaw, asymmetric face, and abnormal ear shape. His 1876 work L'Uomo Delinquente (The Criminal Man) presented these measurements as scientific evidence that criminality was an inherited, bodily condition.

Lombroso's framework had two significant problems beyond its obvious ethical implications. First, his comparison groups were inadequate: he compared prison populations with military units, not with representative samples of the general population. Second, he had no control for the fact that poverty, malnutrition, and poor healthcare, which disproportionately affected working-class and colonised populations, produce physical differences that have nothing to do with criminality. Charles Goring's 1913 study The English Convict addressed these flaws directly, comparing 3,000 prisoners with matched samples from the general population across 37 physical measurements. Goring found no reliable physical differences and concluded that Lombroso's stigmata did not exist as distinguishing features.

Other early biological approaches followed a similar arc of claim and refutation. Phrenology, which claimed that skull shape indicated brain development and character, was popular in the first half of the nineteenth century but had collapsed as a scientific enterprise by 1900. William Sheldon's somatotype theory, published in the 1940s, argued that mesomorphic (muscular) body types were more crime-prone than ectomorphs or endomorphs; subsequent controlled studies found no reliable relationship. The XYY chromosome hypothesis, prominent in the 1960s, suggested that men with an extra Y chromosome were more violent; longitudinal studies found that while XYY men were slightly overrepresented in prison populations, their offending was predominantly non-violent and their rates of criminal conviction were much lower than early reports suggested.

The most methodologically defensible evidence for a genetic contribution to antisocial behaviour comes from twin and adoption studies conducted over the past five decades. Twin studies compare concordance rates for delinquency or criminal conviction between monozygotic (MZ) pairs, who share virtually all their DNA, and dizygotic (DZ) pairs, who share on average half. If the trait has a heritable component, MZ concordance should be higher than DZ concordance.

The evidence across multiple studies is consistent: MZ twins are significantly more concordant for antisocial behaviour than DZ twins. A major meta-analysis by Rhee and Waldman (2002) pooled data from 51 twin and adoption studies, finding that genetic factors accounted for approximately 41 per cent of variance in antisocial behaviour. Shared environment (factors common to both twins, such as family income and neighbourhood) accounted for 16 per cent. Non-shared environment (factors unique to each twin) accounted for 43 per cent. The conclusion is that genes matter, but so does environment, and no single factor dominates.

Adoption studies offer a complementary design. If adopted children's offending correlates more strongly with their biological parents' criminal records than with those of their adoptive parents, this indicates a genetic transmission pathway. Mednick, Gabrielli, and Hutchings' 1984 Danish adoption study examined 14,427 adoptees and found exactly this pattern: biological parent conviction predicted adoptee conviction, while adoptive parent conviction did not, even controlling for socioeconomic status. The relationship was strongest for property offences and weaker for violent offences.

Molecular genetics research has moved beyond identifying single genes toward genome-wide association studies (GWAS) that examine thousands of genetic variants simultaneously. The picture that has emerged is one of polygenicity: no single gene accounts for more than a tiny fraction of variance in antisocial behaviour. The monoamine oxidase A (MAOA) gene received significant attention after Caspi and colleagues' 2002 study showed that the low-activity MAOA variant predicted adult violence, but only in men who had experienced childhood maltreatment. This gene-environment interaction became one of the most cited examples in biosocial criminology, though its replication record across different populations has been mixed.

Neuroscience has identified a set of brain-based correlates of antisocial behaviour that are distinct from genetic explanations but often related to them. The most replicated finding is that individuals with histories of violent offending show reduced grey matter volume and lower metabolic activity in the prefrontal cortex (PFC), the brain region responsible for impulse control, planning, and moral reasoning. Adrian Raine's neuroimaging research, including a 1994 PET scan study of 41 murderers, found significantly reduced glucose metabolism in the PFC compared to matched controls. This finding has been replicated across multiple studies and populations.

Psychophysiological research has produced a simpler and equally well-replicated finding: low resting heart rate is one of the strongest biological correlates of antisocial behaviour identified to date. Across longitudinal studies in the UK, Sweden, and the United States, low resting heart rate in childhood and adolescence predicts later criminal conviction, even after controlling for social background. One proposed mechanism is that low autonomic arousal generates sensation-seeking behaviour as individuals pursue stimulation to reach a more comfortable arousal level. A second mechanism is that low arousal reduces the fear response that normally functions as a deterrent.

Neurotransmitter systems, particularly serotonin and dopamine, have also been implicated. Low serotonin levels correlate with impulsive aggression in both animal and human studies. Dopamine dysregulation has been linked to reward-seeking and sensation-seeking behaviour patterns associated with delinquency. These neurochemical findings are difficult to translate directly into criminal justice applications because they are population-level correlations with large individual variation, and because the same neurochemical profiles are associated with many non-criminal traits.

Prenatal exposure to alcohol, tobacco smoke, lead, and other neurotoxins has been linked to cognitive deficits, impulsivity, and attention problems in children, all of which are established risk factors for later delinquency. Fetal alcohol spectrum disorders (FASD), caused by maternal alcohol consumption during pregnancy, are associated with substantially elevated rates of criminal behaviour in affected individuals, primarily through impaired executive function and reduced understanding of consequences. Studies in Denmark, Sweden, and Canada have documented this association across nationally representative birth cohorts.

The Dutch Hunger Winter of 1944 to 1945, during which German occupation caused severe famine in the Netherlands, provided a natural experiment. Research by Neugebauer, Hoek, and Susser (1999) found that men prenatally exposed to famine had significantly higher rates of antisocial personality disorder in adulthood. This effect was strongest for exposure during the first and second trimesters, consistent with the hypothesis that critical periods of brain development are particularly vulnerable to nutritional deprivation.

Prenatal lead exposure has received substantial research attention following the observation that the rise and fall of violent crime rates in the United States, UK, and other countries tracked closely with the introduction and phasing out of leaded petrol, with a roughly 20-year lag corresponding to the age at which lead-exposed children reached peak offending years. Nevin (2000) and Reyes (2007) modelled this relationship quantitatively and found that lead exposure explained a substantial share of violent crime variance after controlling for other factors. This remains an active area of debate, but the convergence of evidence across multiple countries has persuaded many researchers that prenatal and early childhood lead exposure is a genuine crime risk factor, one that is entirely environmental in origin but operates through neurological pathways.

Perinatal complications, including birth trauma, obstetric complications, and low birth weight, have also been associated with later antisocial behaviour. Raine and colleagues' 1994 study found that the combination of birth complications and early maternal rejection predicted violent crime at age 18, while neither factor alone showed the same effect. This interaction pattern is typical of biosocial findings: the biological risk factor matters most in the presence of an adverse social condition.

Biosocial criminology's central contribution is the shift from asking whether biology or environment causes crime to asking how biological vulnerabilities and social environments interact. Gene-environment interaction (GxE) research shows that many genetic predispositions express differently depending on the social context in which an individual develops. A person with a neurological profile associated with impulsivity raised in a structured, supportive environment with consistent adult supervision may never come to the attention of the criminal justice system. The same profile in an environment characterised by abuse, poverty, and peer delinquency produces very different outcomes.

Terrie Moffitt's influential developmental taxonomy, published in 1993, distinguished two pathways to antisocial behaviour: life-course-persistent offenders, who begin antisocial behaviour in childhood and continue into adulthood, and adolescence-limited offenders, who offend during teenage years and desist by adulthood. Moffitt argued that life-course-persistent offending reflects a neuropsychological deficit, often rooted in prenatal risk factors, that interacts with adverse family environments. Adolescence-limited offending, by contrast, is primarily social, driven by peer influence and the social gap between biological maturity and social adulthood. This taxonomy has been tested extensively and broadly supported, though its two-category structure has been criticised as oversimplified.

Gene-environment correlation (rGE) adds another layer: individuals with certain genetic profiles may also be more likely to encounter particular environments because of their own behaviour or because of the environments their parents create. A child with a genetic predisposition toward impulsivity may evoke harsher parenting responses, select into delinquent peer groups, and live in more chaotic family settings, all of which are environmental risk factors. Separating the genetic and environmental contributions in this situation requires careful statistical design, which is why the twin and adoption study methodologies, despite their limitations, remain methodologically important.

Biosocial criminology faces three categories of criticism. The first is scientific: heritability estimates from twin studies assume equal environments between MZ and DZ pairs, an assumption that research shows is not fully met. MZ twins are treated more similarly than DZ twins, share more prenatal environment, and are more likely to spend time together. These factors inflate apparent heritability. Molecular genetics has not yet identified specific genetic variants that explain the heritability estimates suggested by twin studies, a problem sometimes called the missing heritability problem.

The second criticism is ideological: biosocial explanations risk directing attention toward individual-level risk factors and away from structural causes of crime such as inequality, unemployment, and housing deprivation. Critics in the tradition of critical criminology argue that identifying neurological or genetic risk factors in offenders, who are disproportionately drawn from disadvantaged populations, risks naturalising social disadvantage as biological difference. This concern is not hypothetical: the history of eugenics shows that biological framings of social problems have been catastrophically misused.

The third criticism is practical: biosocial risk factors cannot be translated into criminal justice policy in any straightforward way. Knowing that low resting heart rate predicts later offending does not tell a court what to do with that information. Predictive instruments based on biological markers raise serious due process concerns if used for pre-emptive detention or differential sentencing. Contemporary biosocial researchers generally argue that the appropriate applications of their findings are preventive, targeting biological risk factors through public health interventions such as lead remediation, prenatal care, and nutritional support, rather than through criminal justice mechanisms.