Two ubiquitous phrases in debates about policing and criminal-justice policy are “nonviolent crime” and “nonviolent offender.” Are these categorizations really useful? The best test of whether an offense or offender is truly nonviolent is to intervene and try to put a stop to a so-called low-level offense and see what reaction you get. It’s not hard to see how getting involved in many such scenarios could prove dangerous. Tragically, a recent case in San Francisco illustrates these dangers vividly.

Five days after sustaining serious injuries, a San Francisco store clerk named Yowhannes “John” Tewelde has died. He was beaten with a baseball bat by a man whom he tried to stop from stealing beer from his store. Tewelde’s death marks yet another homicide in the Bay Area, which, between San Francisco and Oakland, has seen an enormous amount of crime and mayhem in recent years. 

The nature of Tewelde’s death illustrates an important reality that progressives working to ease up on supposedly nonviolent crimes don’t seem to appreciate: even “minor” offenses like retail theft, open-air drug use, and smoking on subway platforms are frequently backed by a threat of violence. Had Tewelde not intervened, allowing the thief to take what he wished, progressive prosecutors like former San Francisco district attorneys George Gascón and Chesa Boudin probably would have refused to prosecute him, citing the nonviolent nature of the offense. All we’d have is a lowly shop owner forced to shoulder the cost of a few cheap beers. No big deal, right?

Wrong. The decision to ignore the theft would allow an individual who was in truth both willing and able to kill over $10 to remain on the street. It would be only a matter of time before he did something terrible.

Tewelde’s death is one among many tragic instances in which attempts to stop thieves from running roughshod over their victims are met with deadly violence. Earlier this year, a loss-prevention worker at a California Home Depot was shot and killed after confronting a female suspected of shoplifting. At another Home Depot this past January—this time in North Carolina—an 83-year-old worker was killed after being shoved to the ground when he confronted a man suspected of shoplifting. In Georgia, a Lowes employee was beaten by suspected shoplifters earlier this year when she intervened to prevent the theft (and for her trouble, Lowes fired her for violating store policy on dealing with shoplifters). In 2018, a man detained under suspicion of shoplifting at a Dallas Home Depot was accused of shooting and killing a police officer.

It’s not just thieves. Other supposedly low-level offenses are often carried out by individuals willing to resort to violence if they don’t get their way. In March of this year, a New York City subway rider was stabbed after asking another man to stop smoking marijuana inside the train car. In 2020, two others were stabbed aboard a New York City subway train after asking a man to put out his cigarette. In 2021, a man in Florida was beaten nearly to death, requiring him to be put in a medically induced coma, after he asked his neighbors to turn down their music.

In the era of criminal-justice reform, advocates of the tougher approach to crime and punishment that characterized the mid-1990s and early 2000s have grown increasingly perplexed. How could anyone believe that cutting the prison population in half won’t make us less safe? What possible justification could there be for adopting an official policy of not prosecuting retail theft? The people backing the policies that prompt these exasperating questions don’t just differ in their views about how to solve such problems. They fundamentally misapprehend the reality of crime.

What progressives seem not to understand is that “minor offenses” are often manifestations of the broadly antisocial dispositions of individuals who likely have a much greater propensity for violence than the law-abiding. How much more violence will it take for the reformers to realize this fundamental error?