Shutterstock; Paul Spella / The Atlantic

Updated at 1:41 p.m. ET on June 22, 2020.

The story I can’t get out of my head is the death of Philando Castile.

In the summer of 2016, in the suburbs of St. Paul, Minnesota, Castile, a 32-year-old black man, was pulled over in a car he was driving, along with his partner and her 4-year-old daughter.

“How are you?” Castile asked the approaching officer, according to a published transcript.

“Good,” said the cop, a 28-year-old Hispanic American named Jeronimo Yanez. At the driver-side window, he asked for a license and proof of insurance.

“Sir, I have to tell you,” Castile said, “I have a firearm on me.” As his mother would later tell The New York Times, she had instructed her son to “comply, comply, comply” with law enforcement. So he did.

The statement made the officer nervous. “Don’t reach for it,” Yanez said. The gun, he meant.

“I’m,” Castile replied, “I was reaching for—” The wallet, he meant.

“Don’t pull it out!”

“I’m not pulling it out.”

“He’s not pulling it out,” Castile’s partner affirmed.

“Don’t pull it out!” Yanez yelled again. Then the officer unholstered his own gun and fired seven shots at point-blank range. Five bullets hit Castile. Two pierced his heart. Within minutes, he was dead. A licensed firearm sat untouched in the dead man’s pocket. Philando Castile was shot and killed reaching for his wallet.

The following summer, Yanez was found not guilty of second-degree manslaughter by a jury of 12 men and women. His lawyers had argued, persuasively it seems, that the mere presence of a gun—even untouched, in the trousers of a man with a legal permit to carry it—was enough to exonerate the point-blank execution of an innocent black driver with a busted brake light.

It’s been four years now. The list of men and women killed by law enforcement seems to grow every month. So why can’t I stop thinking about Philando Castile?

Because I think few other chapters in the long epic of American police brutality so capture the hell into which we have all chosen to walk, arm in arm, under the banner of the Constitution. Castile was set up to die by a country that proclaimed his inviolable right to a gun, legally approved his right to carry, and then excused his killing by virtue of the fact that the object he’d been permitted to keep in his pocket could also be used as a precondition for his slaughter.

Castile was killed by a cop in a country where it is more dangerous for a black man to exercise his Second Amendment right than it is for a white man; that is undeniable. But he also died at the hands of a culture that, in celebrating widespread gun ownership, makes it all but inevitable that the United States has more armed police than similarly rich countries, more panicky officers, more adversarial police encounters, more officer shootings, and more civilian killings.

The morbid exceptionalism of American police violence cannot be explained by the amount of money the U.S. spends on police, or by the number of cops it employs. The U.S. spends less on police than the European Union does, as a share of GDP. Italy has more officers per capita than any state in the U.S., according to a comparison of FBI and Eurostat databases. Greece has more officers per person than Newark, New Jersey; Baltimore; and Chicago.

But none of those places shares our epidemic of police violence. American police kill about 1,000 people every year. Adjusted for population, that body count is five times higher than that in Sweden, 30 times higher than that in Germany, and 100 times higher than that in the United Kingdom.

Many differences between the U.S. and the European Union can partly explain these gaps, including our history of systemic racism and our porous social safety net. But without the mention of guns, no explanation for America’s record of police violence is complete.

Let’s begin with the simplest fact: Life is more dangerous in the presence of firearms—period. A 2013 study of U.S. states in the American Journal of Public Health found that for each percentage-point increase in gun ownership, the overall firearm homicide rate increased by 0.9 percent, controlling for other factors. The correlation between gun availability and violent crime is statistically significant at every level of income. More money can spare Americans from the material and psychological ravages of poverty, but it does not buy an exception from the deadly social physics of guns.

Gun prevalence—roughly 400 million firearms are in circulation in the U.S.— is a danger for cops, too.  As the Vox reporter German Lopez writes, police officers are especially likely to be shot dead in states with more guns. A 2015 study published in the American Journal of Public Health examined the relationship between state firearm-ownership rates and police killings, controlling for factors that relate with homicide rates, such as income, poverty, property crime, and alcohol consumption. The researchers concluded that “a 10% increase in firearm ownership correlated to ten additional officer homicides” from 1996 to 2010.

Where guns are abundant, civilians are more likely to kill civilians and cops, and cops are, in turn, more likely to kill civilians. A 2018 study from Northeastern University and the Harvard Injury Control Research Center found that “rates of police shooting deaths are significantly and positively correlated with levels of household gun ownership,” even after accounting for other variables, such as poverty.

One of the most common criticisms of modern policing is that officers dealing with traffic violations and homeless people shouldn’t be lugging around military-style weaponry or exhibiting a “warrior” mentality on the street. Seth W. Stoughton, a professor at the University of South Carolina School of Law, calls that mentality “the most problematic aspect of modern [police] policy.”

Both police militarization and this pernicious “warrior mentality” are a direct response to gun violence and mass shootings. After the 1965 Watts riots in Los Angeles, where police officers faced sniper fire over several bloody days, the LAPD responded by creating the nation’s first SWAT team. After the University of Texas clock-tower shootings the following year, other police departments added their own quasi-military units. In 1997, the LAPD faced off against two bank robbers in North Hollywood, who held off the officers with automatic rifles and body armor. This highly publicized event spurred police departments to arm patrol officers with heavy weapons, and “the AR-15 rifle, a semiautomatic, civilian version of the military’s M-16 rifle, became the weapon of choice for patrol officers,” Stoughton wrote in the Wake Forest Law Review in 2016.

Today’s warrior-cop training is inappropriate, as violent crime has declined by more than 70 percent since 1993. But perhaps some police paranoia comes from the fact that the U.S. has added twice as many guns as people since the mid-1990s. “Generally, when a police officer pulls up to a car in Australia, they don’t expect someone to be armed,” Terry Goldsworthy, a criminologist at Bond University in Queensland, Australia, has said. Indeed, just 6 percent of Australian households own a gun. In the U.S., that figure is 43 percent.

We should stop thinking about guns as just an acute threat, and start thinking about them as something more like lead poisoning—an environmental toxicity that builds over decades and leads to a host of social and cognitive problems. Gun prevalence increases civilian violence and officer shootings, which makes cops more concerned about getting killed, which in turn leads officers to bedeck themselves in quasi-military gear, escalate conflicts that don’t deserve escalation, and, too often, shoot and kill. “When police are involved in an encounter where guns are more prevalent in general, I wouldn’t be surprised that their level of anxiety were heightened,” says Matt Miller, a health-sciences professor at Northeastern and a co-director of the Harvard Injury Control Research Center.

The solutions are socially and politically improbable—but perhaps no more improbable than abolishing the police. They could begin with banning guns of war, requiring universal background checks, and instituting national “red flag” laws to keep guns from potentially dangerous people. These measures might sound small-bore compared with the challenge before us, but they could still make a difference. Deborah Azrael, also of the Harvard Injury Control Research Center, found in 2017 that 22 percent of gun owners who had acquired their most recent firearm in the previous two years did so without a background check.* There are not many issues that easily unite police reformers and police unions, but reducing the social toxin of gun prevalence, especially in high-crime areas, ought to be at the top of that short list.

In a country where guns are protected by the Constitution and cherished by tens of millions of Americans, meaningful change will require a major social movement. Perhaps we’re seeing the seeds of that groundswell right now. As long as reformers are imagining the future of an unbundled police force, they have to be clear-eyed about the roots of civilian and police violence. Which means they have to talk about guns.


* This article originally cited a 1994 survey on the prevalence of firearm purchases without background checks.

We want to hear what you think about this article. Submit a letter to the editor or write to letters@theatlantic.com.