Time is supposed to heal all wounds and to help people forgive and let go of past transgressions. But what if that’s not always the case? What if, instead of leading to forgiveness, the passage of more time before a crime is punished leads judges to levy more severe penalties?

For example, the COVID pandemic created a considerable backlog in court systems across the U.S. and pushed many trials and sentences back by weeks, months or even years. Instead of leading to forgiveness, those time delays created frustration for people impacted by crimes and their families and for prosecutors.

It occurred to us that people in a position to determine justice—whether they are judges or other evaluators—often expect swift consequences. When this process is disrupted, we reasoned, they may find it unfair. Did they seek to correct for a process that they believed had unfairly benefited the transgressor? In a series of studies, we discovered that is indeed the case. Delays in arrests or sentencing increased punishment severity.

[Read more about the psychology of disproportionate punishment]

We began by accessing more than 150,000 felony sentencing decisions from Cook County, Illinois. Cook County, which encompasses Chicago, is the second-most populous county in the U.S. The data, which were released to offer more transparency into the prosecution process, provided a detailed view of how delays may influence sentencing. Importantly, some of the crimes occurred back in the 1980s, meaning that justice may have been delayed for years or even decades after the crime was committed.

We uncovered a consistent pattern: the more time that passed before the judgment of a crime, the longer the sentence a transgressor received. This occurred regardless of whether we computed delays from the time span between the crime and the arrest or sentencing or between the arrest and the sentencing. We also controlled for the number of charges and the severity of crimes committed, which ruled out alternative explanations.

Still, we wanted to replicate these findings in another context. Instead of looking at civilian sentencing, we acquired a data set of police misconduct cases from the New York Police Department. These data included such examples as an officer’s use of excessive force and abuse of authority. As before, our results revealed a consistent effect: the more time that elapsed between the report of misconduct and the closure of a case, the more severe the recommended punishment was. These results held even after accounting for the number of charges officers faced, the number of officers associated with the accusation and the type of accusation.

Together these two studies showed robust support for the effect of delays on punishment. But we still wanted to understand why time delays seemed to increase punishment severity. We therefore designed a series of experiments with 6,029 adult participants recruited via online panels. In these studies, people learned about a hypothetical crime, such as shoplifting, and then decided how many months they would sentence the transgressor to prison for. In one set of experiments, participants were randomly assigned to a scenario where the transgressor was arrested within a day after the offense (i.e., a short time delay), whereas others read that the transgressor was not arrested for 30 days (i.e., a long time delay). Once again, we found that participants punished the transgressor significantly more severely in the long-time-delay condition.

We also tried to gauge what might influence these punishment decisions. For instance, we asked participants how fairly or unfairly the transgressor was treated, including how much they felt the perpetrator had benefited from delays. We also asked how outraged they were by what transpired because some people may have stronger emotional reactions than others. And we asked other questions that explored how competent they saw the perpetrator—because a delay in arrest might indicate an especially savvy criminal. Participants also shared the degree to which they saw punishment as an effective way to deter other transgressors, which might predict more severe sentencing. Of all these factors, we found that only the perceived unfairness of the delays—the notion that a criminal had unfairly benefited from that extra time—consistently explained why longer time delays resulted in harsher punishments.

Importantly, these effects held even when we presented scenarios where the transgressor was not responsible for the time delay. In one experiment, participants opted to provide an additional punishment when they found out a guilty transgressor would start their prison sentence six months later because of court backlogs. In other words, they felt that extra time was an unfair advantage that warranted additional punishment, even though the timing was in no way the transgressor’s fault.

A final experiment examined whether it is possible to mitigate the desire for greater punishment. When we told participants that a judge had already accounted for the time delay while deciding on someone’s sentence, they did not levy an additional punishment. In other words, participants were satisfied that justice had already been served, provided someone in the system had accounted for that time.

Our studies reveal an interesting and important pattern. We should not assume that the passage of time has healing properties. In fact, it can potentially exacerbate punishment. Moreover in cases where the time delay is not the fault of the transgressor—as with the backlog of court cases during the COVID pandemic—people need to recognize that time delays may lead to biased sentencing. That’s something that should concern all of us.

Are you a scientist who specializes in neuroscience, cognitive science or psychology? And have you read a recent peer-reviewed paper that you would like to write about for Mind Matters? Please send suggestions to Scientific American’s Mind Matters editor Daisy Yuhas at pitchmindmatters@gmail.com.

This is an opinion and analysis article, and the views expressed by the author or authors are not necessarily those of Scientific American.