I started reporting this article with a simple question: What would it look like to build a better police department from the ground up?
Police in the US, after all, are more likely to shoot and kill someone than their peers around the developed world, and disproportionately the victims are Black Americans. Meanwhile, serious crimes are often unsolved — with almost half of murders in 2020 going uncleared.
So I asked a dozen experts, focused on criminal justice, what could be done about this to build better police departments. They gave me a lot of different answers, with a consensus on more accountability, a greater focus on crime prevention and more serious offenses over minor ones, and support for non-police efforts to address root causes of crime, among other ideas.
But they consistently gave the same caveat: America’s gun problem. The US has the most civilian-owned firearms in the world, with more than one gun in circulation for every person. A bevy of research has linked greater gun ownership to more deadly violence in the US — and, America, relatedly, has the highest murder rate out of the world’s developed countries.
For police, the huge number of guns in America also means that every single call is treated as if someone involved could be armed — and that an otherwise nonviolent wellness check, mental health call, or traffic stop could turn into a deadly encounter. US law generally allows police to use force because they merely perceive a threat, and the many firearms in civilian hands give police officers a reason to believe they’re in danger.
“It’s Schrödinger’s gun: It’s always there, but it’s not there until you see it,” Michael Sierra-Arévalo, a sociologist at the University of Texas Austin, told me. “That cost is borne by two parties: It’s borne by the public, when police make mistakes, and it’s borne by police themselves, when they’re attacked by firearms.”
Of course, other factors play a role in how US police behave. Racism, at the individual and systemic level, is a real force throughout much of American society. Racial disparities in all aspects of American life, from health to the economy, can translate to higher crime rates in minority communities, where police are subsequently deployed in greater force. And since the 1970s and ’80s, US policymaking has trended toward a “tough on crime” approach that encourages police to act very aggressively.
But guns act as a ratchet in policing. Firearms make every call to the police more risky, but also make officers and the public perceive every situation as inherently more risky. This helps explain not just how cops themselves behave but why police are involved in so many different calls to begin with, from murders to wellness checks. Armed officials ended up in charge of so many areas of society in part because the US has more guns and sees more deadly violence than its peers.
This complicates any effort to reduce the role of the police in American society. One of the more popular proposals today is to get the police out of mental health crises, replacing the cops called about people in crisis with special teams that take a softer, more public health–minded approach.
But the vast number of firearms makes it more likely these calls could escalate, endangering a member of the response team and potentially requiring armed backup. Eugene, Oregon’s vaunted CAHOOTS program, for example, has reportedly diverted 5 to 8 percent of dispatch calls away from the police by deploying unarmed, health-oriented staff to crisis situations. But as the Eugene Police Department explains, sometimes officers have to be deployed along with CAHOOTS, or even beforehand, to secure a possibly dangerous scene.
Reducing the footprint of police isn’t impossible. But the abundance of guns places limits on how far these reforms can go. To put it another way, there’s a choice that America, as a whole, and its leaders have to make: Do something about all of the guns in circulation, or limit the scope of police reform.
The US has more civilian-owned firearms than any country on Earth. There are about 120 guns for every 100 people, according to 2018 data from the Small Arms Survey. Yemen, in second place, has about 53 guns per 100 people. Canada has about 35 per 100, England and Wales — where police are often unarmed — have nearly five per 100, and Japan has fewer than one per 100.
A long line of research has connected more guns to more gun violence, including police shootings. The issue is not that America has more crime or violence than other developed countries, but that guns make it much easier for an event to escalate from a merely criminal offense to a deadly encounter. For police, this reality makes them more guarded, and, potentially, more likely to shoot unnecessarily.
“Police officers are being asked to make these often very subtle decisions in situations in which they legitimately feel their life is really threatened,” Emily Owens, a University of California Irvine economist focused on crime and policing, told me. “The prevalence of firearms in the United States doesn’t help that situation, certainly.”
To be sure, other factors besides guns, from personal views to systemic issues, contribute to those subtle decisions officers make as well. There are reforms that could be tried even within the context of Americans’ massive stockpile of firearms. But guns act as a constant force in the background, drawing boundaries around how far reforms can go and how well they can work.
As one example, the abundance of guns complicates a key concept in many police reform proposals: a higher bar for getting officers involved at all.
American law enforcement respond to a lot of calls that don’t involve violence or even conflict between people. One recent study in Police Quarterly found the top three calls across nine departments were about traffic, public disturbances (like noise violations, graffiti, fireworks, and public urination), or suspicious people and activities; just 7.2 percent were about violence or involved some kind of conflict between different people. The hope is that police, as armed and possibly violent state actors who can escalate a situation themselves, could be removed from the many lower-level calls.
“If police are going to be the armed emergency first responder, what do you want these people with guns to do?” Tracey Meares, the founding director of the Justice Collaboratory at Yale Law School, said. “There are people whose dogs poop in my front yard, and there’s a law against that. Do I think it’s a good idea to call a person with a gun to deal with that? No, I don’t. Just like I don’t think it’s a good idea for a person with a gun to deal with a noise complaint. I can come up with a whole bunch of other examples.”
But the number of guns among the civilian population raises the chances any given call in America will turn into violence, either by a police officer or by a civilian on the scene. In the UK or Japan, anyone responding to a mental health call — police or otherwise — can safely assume a gun won’t be present; in the US, that’s far from a sure bet.
The potential risk of a hypothetical gun is further complicated by the unpredictable nature of policing. Temple University criminologist Jerry Ratcliffe analyzed 911 calls in Philadelphia for a study in Crime Science. He found that calls for one thing can often turn into an entirely different matter — those about crime often turn out to be mental health cases or “sick assists” (such as helping a person who’s physically ill), and wellness checks sometimes turn out to be violent crimes or missing persons situations.
Even if someone thinks that they might be going into a relatively safe call, it could turn out that’s not the case. Add in the risk presented by America’s guns, and you may have a very volatile, potentially dangerous situation. “You don’t know what you’re getting,” Ratcliffe told me. “You don’t know for sure it’s a nonviolent call when you turn up.”
Most police calls are resolved safely without any serious incident. As New York City Police Department analyst John Hall noted, “just one in every 6,959 [traffic] stops results in an assault on an officer … an officer sustaining serious injury or death from a traffic stop is even rarer.” Still, each cop can respond to multiple calls while on duty — and each call carries a roll of the dice that ends in a dangerous encounter. As Hall put it, “Over the course of a career, these stops add up.”
The officers responding to these calls are also planning for the worst, not the ideal. If there’s a decent chance that someone will encounter a gun at a call — especially if something has already happened to a colleague — officers will tend to be more guarded.
This doesn’t excuse criminal acts or horrifying, avoidable mistakes by police officers. Other factors can drive up the risk of violence at any given call, from racial profiling to insufficient housing to poor mental health systems.
But guns are the one uniquely American factor that can escalate a police call.
Ideally, policing in the US would look very different. Several experts pointed to the principles laid out by Sir Robert Peel, who established the London Metropolitan Police Force in 1829, emphasizing crime prevention, rather than reaction to crime, and efforts to build public support. They called for evidence-based police training, stronger accountability measures, more use of research-backed crime prevention strategies, and greater focus on violence and interpersonal conflicts, leaving lower-level offenses and incidents to unarmed officials when possible.
Some activists have gone further, with calls to “defund the police” and redirect savings to other programs that address root causes of crime, such as poverty, mental health care, and housing.
But guns are also a root cause of violence, and not addressing it makes police reform approaches less likely to succeed as intended. What happens, for instance, when staff members of an unarmed team tasked with responding to nonviolent calls get shot? Do they ask for police escorts or backup — diminishing the purpose of the reform? Do they ask to be armed — also defeating the purpose of the reform?
University of Missouri St. Louis criminologist Richard Rosenfeld said that the latter has happened before: Probation and parole officers frequently started out unarmed but over time have armed themselves because, in their view, “they were endangered by their armed clients.”
That doesn’t mean other reforms aren’t worth trying, experts said. But they are likely to be limited in scope and reach by the reality of guns in America.
In some cases, police reform may even conflict with the task of addressing root causes — making it less likely the reform can succeed on all fronts. For example, a lot of attention has gone to police’s involvement in routine traffic stops, with Philadelphia recently banning officers from stopping drivers for low-level offenses.
But it turns out traffic stops are also a big source of the guns police take off the streets. Hall’s analysis for the Manhattan Institute found 42.3 percent of the NYPD’s gun arrests in 2020 were during vehicle stops. Many of these calls can start over a broken taillight or reckless driving, only for the officer to discover an illegal firearm. And, unfortunately, it’s really hard for officers to know which stops will go in this direction; you can’t tell who’s carrying a gun simply by looking at the vehicle or driver.
It also may not be that police’s footprint in US society — and all the costs that brings — are taking up resources from better solutions, but that police are necessary because US society has failed to address root causes of crime and violence first. As University of Pennsylvania criminologist Aaron Chalfin told me, “The police are the residual claimants on all the stuff that no one else is willing or able to deal with. We put them in that position.”
In the case of guns, police are frequently needed because a country awash with firearms requires some sort of armed presence to keep people safe. Only once that abundance of guns is reduced can the police safely retreat.
Stricter gun laws could help. A 2016 review of 130 studies in 10 countries, published in Epidemiologic Reviews, found legal restrictions on owning and purchasing guns tended to be followed by a drop in gun violence — a strong indicator that reducing access to guns can save lives. In the US, there’s particular evidence for requiring a license to purchase and own a firearm. But for political and cultural reasons, America has resisted new, serious national measures for decades, letting firearm purchases continue with few if any checks.
This has contributed to the dynamic of police acting as American society’s backup solution, which is what has saddled officers with so much responsibility to begin with. It’s not that cops wanted more duties. In my years of reporting on this issue, many officers have told me the opposite: that they were called to fill in — by lawmakers and the public — when society had already failed.
To describe these extra duties, police officers “use different terms — nonsense, bullshit, whatever they want to call it,” Sierra-Arévalo, the sociologist, said. “That’s a consistent thing: They don’t think they should be going to a lot of these things.”
America’s tremendous number of guns is at the center of all of this, exacerbating many of the country’s problems by adding a higher risk that any situation can escalate into deadly violence. Once this problem is seen, it’s hard to unsee; it makes it clear why police are responding to so much of the “nonsense” and “bullshit” in the first place.
Doing something about the guns may be the only hope of truly altering that reality — and allowing more police reform.