On the night of Friday, June 12, a police officer shot and killed Rayshard Brooks, a 27-year-old Black man, outside an Atlanta Wendy’s drive-through. Atlanta police officers were called to the scene after receiving a complaint that Brooks was sleeping in his vehicle, which was blocking the drive-through and forcing other cars to drive around it.
Video evidence shows the interaction starts out calm. Brooks repeatedly asks to leave his car parked and walk to his sister’s home, which he says is nearby. But the officer insists he take a field sobriety test, revealing that Brooks had a blood alcohol level slightly above the legal limit. The officer attempts to handcuff Brooks, Brooks resists, and a physical struggle ensues. Brooks grabs the officer’s Taser, begins running away, and turns to fire it. Seconds later he is lying on the ground motionless with three bullets inside him.
This was not the first time a Black man was killed in a police interaction that began with falling asleep in a parked car. It wasn’t even the first time in recent weeks. At 5:30 am on Memorial Day — the same day of George Floyd’s killing — Dion Johnson was sleeping in his car on the side of a north Phoenix highway when he was approached by an Arizona state trooper who planned to arrest Johnson for “suspicion of driving impaired.” According to the officer’s account, Johnson resisted arrest and reached for the officer’s gun. The officer shot and killed Johnson in self-defense.
Both cases raise the same basic question: Why were weapons-carrying agents of the state the chosen response to men sleeping in their cars? The same can be asked about a dispute over a (possibly) counterfeit $20 bill. A mistaken drug bust. A traffic citation. A man selling untaxed cigarettes. None of these infractions began with violence; yet, each of them ended in a Black man or woman killed at the hands of armed police. And stories abound of white, Native American, Latino, and Asian American civilians being killed by police in similar situations.
This dynamic reflects the structure of policing in the US. Here, the same officers who write accident reports and respond to noise complaints also have the capacity to shoot and kill. That means a single police officer has a monopoly over the entire force continuum, from casually talking to aggressively handcuffing to shooting and killing. A situation can escalate from calm conversation to the use of deadly force in a matter of seconds, entirely at their discretion. If whoever had responded to the call had not been carrying tasers and firearms, Brooks would be alive today.
Many European countries view the use of lethal force as a narrow specialization and structure their police forces accordingly. “If this happened in the UK, the first person to approach Brooks would have been a community support officer,” says Colin Rogers, a former UK police inspector turned criminologist at the University of South Wales. “They certainly wouldn’t have been armed.” And even if that interaction went poorly, the officer’s backup would have been an officer armed with only a baton and handcuffs, not a gun.
In 2015, a Guardian investigation found that British police shot dead fewer people (55) in 24 years in England and Wales than American police killed in the first 24 days of 2015 in the US. That disparity can only partly be explained by differences in armed encounters: American police still shot and killed 161 unarmed people in 2015 alone. That’s partly because of uniquely high levels of gun ownership in America, which leaves police officers in a constant state of hypervigilance.
It’s also because in the UK, the state officials who hold a vast majority of public safety responsibilities — from patrolling the streets to responding to nonviolent crimes — do not carry firearms. Only about 10 percent of British police carry guns, and they mostly operate on teams of highly trained specialists whose full-time responsibility is to answer calls of the highest possible threat level, like an active shooter or terrorist attack.
What if we decided to do the same? What if we decided to make traditional policing — defined by the capacity to deploy deadly force — a narrow specialization, as many calling to defund the police advocate? What would that world look like? What would police no longer do and who would take their place?
“It is easy for me to imagine a world in which Rayshard Brooks gets driven home that night instead of shot to death,” says Georgetown Law professor Christy E. Lopez. “The question I have is whether we have the will and commitment to create a public safety system that makes that world a reality.”
Over the past few weeks, I’ve spoken with more than a dozen sociologists, criminologists, policing experts, nonprofit leaders, and legal scholars to better understand the range of alternatives that exist to our current one-size-fits-all model of police response — and how we could design those alternatives to meet unique challenges like the overwhelming presence of firearms, intertwined histories of racism and policing, and relatively high rates of violent crime in the US. Here are four ideas they offered.
1) Create specialized traffic patrol officers
The vast majority of police-civilian interactions happen on the road. According to a 2015 Department of Justice report, of the 50 million Americans who came into contact with the police that year, 25 million were pulled over in a car they were driving or were a passenger in (Black Americans were the most likely to be pulled over). Another 8 million were involved in a car accident. And many of the 9 million who called the police to report non-crimes were reporting traffic accidents.
There is no justifiable reason why armed police officers should be in charge of road safety. Police officers are not hired for a particular talent in highway navigation, accident report taking, or citation writing. And deploying armed officers to perform such routine tasks introduces the risk of unnecessary lethal force into many millions of encounters every year. The police killing of Philando Castile in 2016 was one instance (there are plenty of others) of a routine traffic stop going horribly wrong — and it simply wouldn’t have happened if the officer hadn’t been carrying a gun. Sandra Bland’s arrest and subsequent suicide following a failure to signal a lane change was another.
It isn’t difficult to imagine handing over most traffic patrol duties to specialized employees — we already do the same for plenty of other public safety roles, like restaurant and food inspection. Highways England in the UK employs unarmed traffic officers who drive around in distinct vehicles, and many other of the country’s traffic duties are left up to “community support officers” who can give out citations but are both unarmed and lack arrest power.
Some US cities are even beginning to take steps in this direction, largely because armed police officers are a uniquely expensive way to handle traffic patrol. In 2017, the city of New Orleans endorsed NOPD hiring third-party report-takers for accidents in which there is no injury and no concern about a driver under the influence.
2) Deploy community mediators to handle minor disputes
A huge number of calls to the police involve relatively minor interpersonal disputes: disputes over noise levels, trespassing, misbehaving pets, or rowdiness; disputes between spouses, family members, roommates, or neighbors.
Without a mediator present, it is possible that what starts out as a minor dispute can escalate to violence. But there is no particular reason the job of mediation has to be assigned to armed police officers; if anything, traditional police tend to unnecessarily escalate these situations, resulting in arrests or worse.
That’s why a number of countries such as the UK, Belgium, France, the Netherlands, and South Africa have created a distinct class of what can be broadly called “community safety professionals.” They are unarmed, lack most formal policing powers, and perform responsibilities like youth outreach, conflict mediation, community patrol, and addressing low-level crime and disorder. Preliminary results of their impact on crime and community well-being have been promising.
“The idea was for community support officers [the UK’s version of this role] to act as a bridge between communities and police officers,” says Rogers. “Because we are unarmed, we police with and through communities, not at them.”
A similar approach has been pioneered by many “street outreach” programs in the US like Cure Violence and Advance Peace, which employ “violence interrupters” and “peacemakers” from within local communities to mediate conflicts before they escalate to the level of violence. Scholarly evaluations of these efforts have shown that non-police mediation can be quite successful when executed properly.
“If someone is upset or thinking about shooting, the violence interrupters are almost always able to cool that person down and stop them from acting,” says Cure Violence founder Gary Slutkin. “The goal is to contain things before they get to the police. If nothing has happened yet, it’s none of the police’s business.”
I asked A.T. Mitchell, a former “violence interrupter” who now runs Cure Violence’s ManUp! program in New York, what he thought of the dispute between George Floyd and the store clerk who claimed he was using a counterfeit $20 bill.
“What that situation needed was someone to resolve a conflict,” he told me. “Did [Floyd] owe an apology? Did he even know what was happening? We don’t know. But I’ll tell you something: If we got that call, we would’ve been able to step in between, and that brother would still be here today.”
One could imagine cities hiring cadres of “community mediators” as employees of the local public health department who are trained in conflict resolution, applied psychology, and relationship management. Like their European counterparts, these mediators would be completely unarmed, lack formal policing powers, and wear uniforms distinct from traditional officers. They could spend their quiet hours building relationships with local community members and maintaining a presence in schools, neighborhoods, and highly trafficked public spaces.
Cities could even develop a special 211 or 311 number for concerned neighbors, spouses, or citizens to call when they are witnessing heated exchanges, and could redirect relevant 911 calls to the community mediation unit. If any of these exchanges begin to escalate toward violence, the community mediators could have special silent alert systems (similar to the ones senior citizens use for medical assistance) to call in armed police for backup.
“Imagine a world where first responders are really good at getting situations calmed down so that people can go on with their lives and nobody ends up being arrested — or worse,” says Barry Friedman, the director of the Policing Project at New York University. “In that world, we don’t need nearly as many police walking around with guns.”
3) Create a mobile crisis response unit
Oftentimes, a police officer’s role bleeds over from mediation into something that resembles social work, usually involving populations like those who are homeless, intoxicated, substance abusers, or suffering from mental illness.
The results can be disastrous. About half of prison inmates were diagnosed with a mental illness. Around a quarter of fatal encounters with law enforcement involve someone with a mental health condition (and those numbers are possibly severe undercounts). A massively disproportionate number of police calls and arrests in cities across the country involve homeless populations. In Portland, Oregon, the city’s homeless population made up 52 percent of the city’s arrests in 2017 even though they comprise less than 3 percent of Portland’s population.
“You wouldn’t try to build a house with just a jackhammer,” says Zachary Norris, director of the Ella Baker Center for Human Rights and author of We Keep Us Safe. “But that’s what we’re doing when we task police officers with dealing with public health issues like substance abuse, homelessness, and mental illness.”
One of the most promising alternatives to a police-centric model of social work is a program called Cahoots, a collaboration between local police and a community service called the White Bird Clinic that operates in Eugene and Springfield, Oregon. In these cities, police officers aren’t dispatched to handle every single 911 call. Instead, about 20 percent of calls — often those involving the homeless, addicted, intoxicated, or mentally ill — are routed to a separate team of specialists extensively trained in mental health counseling, social work, and crisis deescalation.
Cahoots responders don’t brandish weapons of any kind. They dress in black sweatshirts, listen to their police radios via earbuds, and purposefully speak in calm tones with inviting body language. Their role is closer to that of an EMT for social issues than a traditional police officer: They assess the situation, assist the individual as best they can, and then direct that individual to a higher level of care or service if needed. If the situation escalates, they can also call police in for backup, but that’s rare. In 2019, Cahoots received around 24,000 calls and had to call in police backup less than 1 percent of the time.
“In 30 years, we’ve never had a serious injury or a death that our team was responsible for,” Ebony Morgan, a Cahoots crisis worker, told NPR. “I think that’s important.”
To top it off, Cahoots saves the Eugene and Springfield police departments around $15 million a year, according to clinic coordinator Ben Brubaker, by taking care of incidents that would otherwise have to be handled by law enforcement or emergency rooms, both of which are far more costly solutions.
Cities could also think about building on and improving that model. The existing program’s biggest limit is the fact that its jurisdiction is only in decidedly “non-criminal” calls. That means ordinary police officers may very well be sent to deal with situations that Cahoots crisis workers are far better trained to handle.
There are two possible remedies here. One is the decriminalization of issues like addiction and homelessness. “Right now, police response to homelessness is driven by city laws criminalizing it,” says Maria Foscarinis, executive director of the National Law Center on Homelessness & Poverty. “Typically, calls about homeless people are about issues like sleeping or begging that should not be addressed by law enforcement.” Changing laws that criminalize such behavior can broaden the range of activities that a mobile crisis response unit like Cahoots is able to address.
Another idea is to deploy hybrid response units consisting of both police officers and mobile crisis responders to situations that would normally fall outside of the Cahoots purview. For instance, police might be called to the scene to break up a violent fight. But it is easy to imagine a Cahoots team entering the situation first and attempting to defuse it while police officers wait around the block, out of sight, to be called in only if deemed necessary.
“I’m really excited about imagining an entirely different model of first responders,” says Friedman. “What that means is training people in a very different way, dispatching them in a different way, and giving them a different reward system than we’ve given cops.”
4) Experiment with community self-policing
Those first three ideas involve solutions that local government officials could incorporate fairly easily into their existing policing models. But what if we changed the model completely? What if, instead of policing at communities, we gave them the resources to police themselves?
A little over 20 years ago, the Australian government did just that.
The history of the indigenous community in Australia is thick with repression, brutality, and violence at the hands of the state. Descriptions of Indigenous-police relationships read as though they could be pulled straight from the contemporary African American experience in the United States (not to mention the US native communities). As Harry Blagg, a criminologist at Charles Darwin University in Australia, writes:
Historically, policing was an instrument for controlling, limiting, denying or supervising Indigenous egress into the white domain. According to criminologists, this has left a legacy of over-policing of Indigenous people in the public realm – where they may constitute a threat to public order – and under-policing (underservicing might be a better term) of Indigenous people within their own communities.
This began to change in the 1990s when a government commission found that Indigenous people were highly overrepresented in prisons and jails as a result of systemic bias. The authors concluded that the only way to end this injustice was to entirely reimagine the way Australians interact with the criminal justice system.
One recommendation they made was for the government to fund local forms of community self-policing, like the Julalikari Night Patrol in northern Australia. The idea behind the night patrols was simple: to enhance public safety by establishing a buffer between Indigenous people and police forces. This is how Princeton sociologist Patrick Sharkey described his visit to the Nyoongar patrol in Perth, Australia, in his 2018 book Uneasy Peace: The Great Crime Decline, the Renewal of City Life, and the Next War on Violence:
I joined a team led by Annie and Rachel, two extraordinary women who were remarkable to watch in action. I looked on as they tried to calm a shirtless man who was drunk and belligerent in front of a crowded bar. I saw them talk to a man who looked unwell, lying on a bench in the middle of a city plaza, and stay with him as the emergency medical technician asked him questions and eventually took him to get treatment. …
The challenges that emerge over the course of a shift change on a nightly basis, but the overarching goal of the patrol teams is to maintain a presence in the public spaces where young people hang out, to search for Aboriginal people who look as if they could use some help, and to give anyone who is causing trouble the chance to cool off or to go home before the police get involved. At times the patrol team’s intervention comes with a stern warning, but usually it comes with a warm smile.
When reading that description, it’s hard not to think about how differently things would have gone for Rayshard Brooks or Dion Johnson if members of this local night patrol had been on duty. Maybe they drive Brooks to his sister’s house to spend the night. Maybe they take Johnson to a local shelter to sober up with a warm breakfast. The police are never called.
Today, hundreds of these night patrols have been established in Indigenous communities across Australia, many of them government-funded. The patrols lack formal policing powers, but their legitimacy comes from the fact that they are established by community councils, endorsed by elders, utilize local knowledge, and work within the boundaries of Indigenous law and culture.
By many indicators, the patrols have been extremely successful. “Relationships between night patrols and police [are] generally excellent these days,” Blagg says. “The police can’t manage without them.” Police don’t have a full-time presence in most communities with night patrols. They will step in to calm things down or make an arrest, but typically only when contacted by the local patrol. One study found that patrols in three areas were able to reduce arrests by around 30 percent.
“That was one of the most inspiring nights of my life,” Sharkey told me of his visit to the Nyoongar patrol. “It gave me a vision of what public safety can look like if it is driven by people acting out of genuine concern for their communities.”
A community-based approach to public safety has also been pioneered in some of America’s most violent neighborhoods by many of the “street outreach” programs that I mentioned above, the largest and most rigorously evaluated of which is Cure Violence Global.
Like the Indigenous Australian night patrollers, Cure Violence’s “violence interrupters” are locals with strong community ties, many of whom have done prison time themselves. Their job is to build relationships throughout the community such that they are aware of ongoing disputes, interpersonal tensions, and potential fights before they escalate to the level of either civilian violence or police intervention.
“We have a level of trust with the community that the police will never have,” says Mitchell. “That’s because we only hire those who are indigenous to the neighborhood. Information gets around to us long before it ever gets to the police.”
The role of violence interrupters goes beyond just on-the-spot mediation. They provide mentorship and economic opportunities to individuals who are considered “at risk” of committing violence. In the aftermath of violent shootings, they mobilize the family and friends of victims and respected community leaders to prevent retaliation. And in quiet times, they work to spread social norms of nonviolence.
Cure Violence programs have been implemented in 25 US cities, often in neighborhoods that are experiencing high levels of gun violence. And multiple independent analyses of programs in places like New York, Chicago, and Baltimore have shown that the model has the potential to bring about major reductions in violent crime and gun violence at a fraction of what it would cost police forces.
One option, then, is for local lawmakers to simply scale up the Cure Violence model from one or two neighborhoods to an entire borough or city. Mayor Bill de Blasio recently announced he would be investing an extra $10 million to expand the program to the 20 New York City precincts with the most gun violence.
“The time is more than right for a large investment in Cure Violence,” says Caterina Roman, a sociologist at Temple University who has conducted research on the organization’s approach. She points out that while the organization has never been tested at the scale that is now demanded, it is also one of the few models that has been shown to successfully make highly violent communities a lot less violent without using the tools of arrest, force, and incarceration.
Another option is for local lawmakers in the US to experiment with community approaches to policing. Sharkey believes city officials should bring together local civic organizations, community leaders, and residents to form a new community entity tasked with planning a new model for public safety in a predefined number of neighborhoods.
The group would receive funding equivalent to whatever the police department in that jurisdiction would receive. They would be allowed to use the funds however they choose. They would draw up plans for their community’s relationship to the local police department, which will likely serve in some sort of backup capacity in case things escalate. Then, they would be given a minimum of 10 years to run the experiment, with rigorous monitoring and evaluation along the way.
“Communities themselves should be the ones deciding on these issues,” says Tracie L. Keesee, a former police officer and the co-founder of the Center for Policing Equity. “Who do you think should provide services? Who should be in charge of public safety? These questions should be put to the community.”
These ideas might fail — but the current system is already failing
There’s no guarantee that any of these suggestions will succeed across the board. When it comes to policing alternatives, even the best existing models haven’t been attempted at scale, and there’s no telling how different communities will respond to them. To implement any idea on this list would mean venturing into relatively uncharted territory.
That means there will be failures. Things will go wrong. Systems will break down. Programs will fall apart. Violence may temporarily increase in some places. Occasionally, a violence interrupter or mobile crisis worker will be seriously injured or killed.
But our current system already represents a kind of profound failure. We live in a country that has built the largest system of human incarceration on earth, where agents of the state kill unarmed members of the communities they are supposed to protect and terrorize those who are still alive. Where peaceful protesters are beaten in the streets.
The question, then, isn’t whether we are willing to live with failure; communities across the country already live with failure every single day. That failure, at least in part, stems from the fact that police officers in the United States are tasked with responsibilities — from traffic patrol to mediation to crisis response — that amplify the risk of unnecessary violence.
There are plenty of models out there of how we could transfer these responsibilities to non-police personnel and, in doing so, make the use of lethal force far rarer. The question is: Are we willing to give them a try?