A band marches down a Baton Rouge street in a "second line" vigil for Alton Sterling, who was killed by police officers TuesdayGerald Herbert / AP

I began writing this on Wednesday morning. Just hours before, late on Tuesday night another black person passed from life to death to the strange immortalization of a Twitter hashtag. Alton Sterling, a 37-year-old black man was shot and killed by police officers while selling CDs in front of a store in Baton Rouge, Louisiana. He was alleged to be armed. A widely-circulated video of the incident, which appears to show an officer aiming a pistol at Sterling and firing while Sterling was prone and restrained by another officer, kicked off protests in the streets of Baton Rouge.

A day later, I was still writing. Late on Wednesday night another black person passed from life to the immortalization of a hashtag. Philando Castile, a 32-year-old black man, was shot and killed by a police officer during a traffic stop in Falcon Heights, Minnesota. He was alleged to be armed. A widely-circulated video of the incident, in which Castile’s girlfriend Diamond Reynolds reacted after Castile had been shot—the officer still training his gun on him—kicked off protests in the streets of the Twin Cities.

The next shooting may not come tonight or tomorrow. By the math, though, every two days a black person of some age—14 or 18 or 43 or 37—armed or unarmed, sober or under the influence, resisting arrest or providing officers with identification will be shot and killed by an officer or officers. Video of the incident will likely be circulated. Protests will likely follow. But any sort of end to this violence remains truly unlikely.

According to The Guardian’s “The Counted” project, Sterling was the 560th person killed by police this year. Castile was the 561st. Despite a year of mass protests, despite the proliferation of videos of police violence and brutality, and despite a shifting national mood that acknowledges police brutality as both racially unjust and part of a broad pattern of behavior, the number of people killed by police is roughly on pace to be similar to last year’s number of people killed by police. Vigilance, complaints, scrutiny, and even rare prosecutions have done little to curb routine violence.

There is a sense of normalcy to what should be absolutely extraordinary. In Baton Rouge, swelling crowds prompted an immediate press conference from District Attorney Hillar Moore to announce that the Department of Justice would be investigating Sterling’s death right away. Just a year ago that might have seemed an entirely unlikely victory for those who pushed for scrutiny in officer-involved shootings. Now, it is unclear if even these extraordinary proceedings will result in justice or an end to the onslaught.

Moore’s press conference also suggested why policing is so difficult to change:

Well again, you know, it’s another person that’s dead, killed by law enforcement officers who have the authority by the state and the people— because we get our power, really, from the people, not just a piece of paper—that authorizes law enforcement to take a life in certain situations. Which is always—that’s why you guys are here in this situation and not on the streets of Baton Rouge where we have other killings, because this is potentially a state-authorized killing. [The law] gives law-enforcement officers the authority and mandates them to kill when in defense of themselves or others. So I think whenever there’s that situation and law enforcement officers [are] involved, it’s a completely different case than a person in the streets being killed.

As my colleague Brentin Mock notes, Moore’s comments seem unusually blunt. Police officers are in the business of state-authorized killing, to defend their own lives or those of others. And when they exercise that authority, Moore implies, their actions deserve greater scrutiny than other homicides.

The right to kill in self-defense is in keeping with the generally expansive and bloody American doctrine of self-defense. But officers are not ordinary Americans. They likely carry with them the same sense of intimidation and mistrust of people of color that many people across the country carry, but are simultaneously trained to aggressively interact with them on a daily basis. That fear and hypervigilance may be supplemented by what appears to be common racism. Officers are also allowed a far more generous interpretation of self-defense in disciplinary and court proceedings. In practice, simply claiming that they feared for their lives often proves sufficient grounds to secure acquittal. In essence, police officers are given lethal weapons, taught and authorized to use them rather liberally, and then deployed in a manner as to create situations to use those weapons.

At some point, a system constructed in such a way that killings are inevitable asymptotically approaches a system in which killings are deliberate so as to make the difference difficult to discern. Moore’s comments encapsulate what I tried to express in an analysis last week about the failings of criminal justice around the Freddie Gray case and the need for a new way to assess police violence. Some have defended those trials and others like them as exercises in justice working the way it should: Defendants are offered the fullest and most charitable version of due process, supported by strong legal teams, and prosecutors who cannot meet the evidentiary burden to convict them are left with nothing.

But that defense is orthogonal to the actual issues at hand and the implications of Moore’s statement. The expansive authority of police to kill forestalls accountability for all but the best-documented and most-egregious acts of violence by police officers and reduces the ability of the criminal-justice system to meet its burden of safeguarding citizens. Because police officers are agents of criminal justice, they benefit from a presumption that their acts are just. But if widespread police killings are in fact an injustice, then there is a pressing need to look beyond individual trials for some systemic remedy.  

There is little public will to do so, however. As Moore put it, the agents of the criminal-justice system “get our power, really, from the people, not just a piece of paper.” Those comments laid bare exactly why incidents of police brutality—even killings—seem like routine elements of American life. It’s because they are. They are not aberrations, but the predictable and inevitable consequence of common encounters enabled by policy and sustained by the will of society. If there actually is any resolve to keep history from repeating itself and to end the parade of death, Americans will have to challenge the state’s authorization of violence beyond individual police acts, and investigate the purposes of policing that drive its use. Until then, people will continue to die.

This article is part of our Next America: Criminal Justice project, which is supported by a grant from the John D. and Catherine T. MacArthur Foundation.

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