Alton Sterling and Philando Castile: Two black men who were killed by police officers this week, their deaths captured on camera and protested by demonstrators and online activists who say that the men weren’t posing a threat, let alone one that justified deadly force. Their deaths have driven renewed attention to the more than 1,000 people killed each year by police officers.
Outrage about police killings has gone national since Darren Wilson killed Michael Brown in Ferguson, Missouri, nearly two years ago, creating a movement to count them and to prevent them, propelling activists to national attention and putting violence on the agendas of the two leading Democratic presidential candidates. Some police chiefs and the director of the Federal Bureau of Investigation have said rising violent crime rates were the result of increased scrutiny of police: a supposed “Ferguson effect.”
Yet the best available data suggests that if police officers are being watched more closely, that hasn’t reduced the frequency with which they kill people. In fact, they might be killing people more often. And the people dying still are disproportionately black.
These conclusions are somewhat fuzzy because there is no single, universally accepted data source for police killings. The federal government acknowledges that its agencies’ various counts are incomplete. Fatal Encounters and Killed By Police are two crowdsourced efforts that try to fill in the gaps, using media and other reports. Mapping Police Violence collates and fact-checks these along with an older source, the U.S. Police Shootings Database, and does its own original research. The Guardian added its resources to the effort last year. And The Washington Post and the Gun Violence Archive count fatal police shootings, which are the majority of police killings. A web of amateurs, volunteers and journalists have filled the void left by the government.
These sources don’t all agree on raw numbers, and even the highest counts probably are missing some cases. But the sources all agree that police killings have held steady since they started trying to count comprehensively. That holds whether you look at full-year totals, compare numbers through June 30, or look at totals between the date of Michael Brown’s death and the end of the year, in the years before and after he died.
And blacks continue to make up about 30 percent of the people dying from police violence, though they make up 13 percent of the nation’s population.
“Would this have happened if the passengers were white? I don’t think it would have,” said Minnesota Gov. Mark Dayton about Castile’s death in Falcon Heights, Minnesota. President Obama linked the Castile and Sterling shootings to “the racial disparities that appear across the system year after year,” and Hillary Clinton, who is running to succeed Obama, tweeted, “Alton Sterling Matters. Philando Castile Matters. Black Lives Matter.”
The counts don’t all track whether the people killed are unarmed, and that can be disputed or hard to categorize. For instance, the police said Sterling and Castile had guns, but witnesses said the two men weren’t holding their weapons when police officers shot them. Those organizations that do keep data on this agree that most people killed by police officers are armed, but black people who are killed are and have been less likely to be armed than those who aren’t black.
Because these counts aren’t comprehensive, it’s possible that something else is driving the increase. For instance, news organizations may have missed earlier police killings, and may be counting more now that they are paying closer attention.
“My gut tells me that media is picking up on it more, and that so many folks are videoing from smartphones, there is a level of forced transparency,” said Mark Bryant, executive director of the Gun Violence Archive. D. Brian Burghart of Fatal Encounters agreed that news outlets might be missing fewer cases, but also proposed an alternate explanation: “I believe the increased reporting actually does create fear among the police and the people who are policed,” he said. “It makes everyone more nervous, more likely to do something out of panic, when people and police interact.”
Burghart urged the federal government to expand its effort to collect better stats on the problem. “If a tiny group of people with no budget can do it,” he said, “the feds with their millions and trained data scientists could have a complete database in a week, and begin examining policies across the country to find the training that produces the best outcomes.”