Marked by Time: How Social Change Has Transformed Crime and the Life Trajectories of Young Americans Robert J. Sampson Harvard Univ. Press (2026)
‘Greed is the iron cage of our times’ — why nationalism is here to stay
In Stanley Kubrick’s film adaptation of the 1962 novel A Clockwork Orange, gang leader Alex DeLarge is portrayed as an ultraviolent miscreant. Once imprisoned, he is subjected to aversion therapy, which serves only to reinforce the deeply rooted nature of his criminal disposition.
DeLarge is one of cinema’s most memorable villains, and his behaviour aligns with a common explanation for why some people become chronic criminal offenders: ‘bad character’.
In Marked by Time, criminologist Robert Sampson sets out the degree to which academic theory, policing and court practice depend on this idea. He also lays out evidence that it is a fundamentally wrong-headed assumption: it neglects the role of changing historical circumstances in influencing a person’s chances of first being arrested, which subsequently affects whether they become a repeat offender.
Disquieting but effective, Sampson’s book makes a compelling case for rooting out character-based assumptions — and for factoring in historical context — at all stages of the criminal-justice system.
A matter of time
The idea that a person’s character is central to their risk of committing crime is prominent in the field of crime theory. As Sampson concedes, some evidence does indicate that children who are impulsive or lack self-control are more likely to engage in criminal activity as teenagers and adults. A classic study, for example, showed that antisocial behaviour in childhood was a strong predictor of antisocial behaviour in adulthood (L. N. Robins Psychol. Med. 8, 611–622; 1978).
How wealthy tech entrepreneurs seek to shape politics, culture and the future — and why we must resist
Yet Sampson notes that many, if not all, of these long-term studies are based on samples from single historical periods. If character were the key factor in determining who becomes a criminal, its relevance should remain stable across history, even during periods of rapid social change.
To analyse this idea, Sampson taps into a rich trove of data from a longitudinal study called the Project on Human Development in Chicago Neighborhoods. It ran from 1995 to 2024 and was designed to help researchers to understand the historically high crime rates in many US cities in the early 1990s. Researchers collected survey data on the individual, family and neighbourhood circumstances of more than 1,000 young people who were born in Chicago, Illinois, between 1980 and 1995. Sampson was one of its first scientific directors.
What no one knew when the project began was that the mid-1990s marked the beginning of a high decline in crime rates that continued into the mid-2010s — a phenomenon observed in many countries and for which there is no universally accepted explanation. Because the cohort lived through eras with drastically different crime levels, Sampson could explore how the time in which a person lives affects their risk of committing crime and of arrest.
Young people growing up in Chicago took part in a 29-year study of human development.Credit: Scott Olson/Getty
He cites two cases for comparison. Andre and Darnell were both born in the same disadvantaged Chicago neighbourhood, which was racially segregated and had high levels of poverty. But whereas Andre was born in 1980 — and grew up exposed to historically high levels of violence and policing enforcement — Darnell was born in 1995, and his adolescence occurred during a more tranquil period. Andre was arrested by age 19, but Darnell avoided entanglement with the criminal-justice system.
Arresting trends
Sampson argues that differences in the likelihood of arrest across time are a major factor in the boys’ differing life trajectories. The experience of arrest can itself alter a person’s life prospects by disrupting their education, reducing employment opportunities and leading to stigmatization in their community. And the difference in arrest likelihood over time, he shows, cannot be fully explained by demographic, family or neighbourhood conditions, or by self-control: children with high self-control born in the 1980s were just as likely to be arrested as were those with low self-control born in the next decade.
