What a 1970s Report on Recidivism Reveals About Modern-Day Misinformation The story of 20th century sociologist Robert Martinson offers an object lesson on the power, and peril, of science communication, writes Project Fellow Ashley Belanger.

Editor’s Note: This is part of a series of dispatches from the Knight Science Journalism Program’s 2020-21 Project Fellows.

It’s rare that one sociologist releases a single report that inspires a complete overhaul of the U.S. criminal justice system. But in 1974, that’s exactly what happened.

Following a chilling 15 years in which the U.S. homicide rate doubled, American sociologist Robert Martinson published, without the consent of his collaborators, his own interpretation of the findings of their landmark study on violent crime. Martinson published his piece in the neoconservative journal The Public Interest; the official report, published in the same journal, wouldn’t appear until the next year. In Martinson’s preview, he described how they’d reviewed 231 studies on correctional rehabilitation of people imprisoned for violent offenses and concluded that no treatment was effective in preventing repeat offenses, or recidivism. His report’s primary conclusion: Nothing works.

It didn’t matter that the conclusion was wrong, or that many respected researchers easily and immediately disproved Martinson, noting that his sample targeted treatment programs that were underfunded, with untrained staff, and that he’d ignored treatments that were proven to be efficacious. The damage was done, and criminology experts say it’s because, to the American public, the idea that “nothing works” just sounded right. People, even top psychologists, are fundamentally bad at assessing risk, and this was one moment where we, as a society, made the risky choice to ignore evidence-based safety guidance.

For both liberals and conservatives who were eager to see officials get “tough on crime,” the Martinson report was evidence enough that investing in prison reform was a waste of time. If rehabilitation didn’t work, a war on crime could be waged instead, by crafting new policies like habitual offender laws that eschewed recidivism research in favor of simply locking up more people for life.

This led to the prison overcrowding and mass incarceration we see today.

Criminology researchers have said the Martinson Report provided the foundation for one of the most significant shifts in modern American corrections, not because it was groundbreaking — plenty of researchers had drawn “nothing works” conclusions before — but because of how remarkably well the report’s results were communicated to a general public that was eagerly awaiting confirmation of their belief that the current justice system wasn’t working. A painful irony is that Martinson, a once-jailed Freedom Rider, was moved to betray his co-authors and write his own report so that he could use their results to convince society that prison is futile.

As a KSJ Project Fellow, I’ve spent much of the past year researching methods for assessing sexual violence risk assessment, and I’ve spoken to dozens of researchers, state employees, and policy advocates about the unique challenges of communicating that risk. They say that what frequently happens is the opposite of what happened with the Martinson Report: Sound evidence emerges of a new safety guidance that can help communities more effectively prevent sexual violence, but amid confusion over the actual risks involved, the evidence fails to sway popular opinion and the guidance doesn’t impact policy.

Like politicians wanting to appear tough on crime without actively working to reduce violence, some states are more invested in maintaining laws that purport to prevent sexual violence than in educating the public on the reality of what is and isn’t working.

Under new federal guidance that ignores recidivism research, more kinds of sex crimes have become registerable offenses. And as states deal with the resulting growth in the number of registered sex offenders, which nearly doubled between 2006 and 2018, experts say communities could benefit from more flexible laws that support continually updated, empirically tested risk guidance. But historically, updating registration laws has been cost-prohibitive and politically unpopular.

The problem has become even more acute as sexual violence has spiked during the pandemic. It’s made me wonder if there could ever be a report on sexual violence risk that has the kind of impact that the Martinson Report had. What would it take for a review of sexual violence risk research to convince communities that the violence is not stopping, and that they should demand that officials “get tough on sexual assault”?

It’s easy to think what’s needed is the right authority figure. Because the Martinson Report was well-received by the policymakers and journalists who avidly read The Public Interest, Martinson soon became a talking head, echoing the misinformation that “nothing works” on popular news shows like 60 Minutes and CBS News. The rapid spread of Martinson’s bad faith message wasn’t just about the platforms he chose, though. More important, say criminology researchers, was how clearly Martinson wrote for general audiences. He was a gifted science communicator. In his report, he delivered his results in a way that the average Joe could follow, presenting the evidence in a question-and-answer format that made the case seem open-and-shut.

During uncertain times, people like to hear that they’re right much more than what’s right.

In essence, the Martinson Report was a case of the wrong message delivered at the right moment — a moment when his audience was primed to find misinformation more attractive than accurate safety guidance. We’ve also seen that play out with the botched risk communication during the pandemic, with experts appearing on popular shows to push ineffectual Covid-19 treatments, make misleading claims about herd immunity, and otherwise promote politicized, improper public health guidance. During uncertain times, people like to hear that they’re right much more than what’s right.

We now know that many people convicted of violent crimes can and should be helped to safely reemerge into communities. But our prisons are still designed around the dogma that “nothing works,” so we’re stuck with ineffective rehabilitation and overcrowded prisons until the next criminal justice revolution, which many of today’s activists hope will be a move to abolish prisons entirely. So far, however, no single study like the Martinson Report has given politicians reason to amplify the grassroots rally cry to explode the current prison system.

One thing Martinson did prove with his infamous report was something anyone who reposts a meme will recognize today: Once an idea is repeated enough, most people don’t consider the original source. And this can be dangerous for everyone, especially when it comes to monitoring violence. For women fearing sexual violence, uncertain times are always.

Within five years of his report, Martinson began to question his own results. No longer sought as a talking head, he died after jumping from his Manhattan apartment in 1980. The year before, Martinson publicly retracted his “nothing works” doctrine, saying in the Hofstra Law Review, “Contrary to my previous position, some treatment programs do have an appreciable effect on recidivism. Some programs are indeed beneficial.”

---

Ashley Belanger is an investigative science journalist whose health and politics reporting has been funded by the National Geographic Society and has appeared in Teen Vogue, Ars Technica, and other outlets.