More policing, less law enforcement

If we're really going to reimagine policing, let's stop acting like police are nothing more than an extension of the criminal justice system.

When I was a rookie, I got a call from a veteran officer who I’ll call Steve. Steve told me he needed backup to meet him at a residence, but didn’t say why. I looked up the address in question and found out a registered sex offender lived there. I figured the guy probably had a warrant and Steve wanted me to help arrest him, but when I checked the guy’s name, he wasn’t wanted.

Baffled, I met up with Steve outside the sex offender’s house. I asked him: Are we arresting this guy? What are we doing here? No, Steve told me, we aren’t making any arrests. It turned out this particular sex offender had been posting ads on the internet, asking women to meet with him. He had been falsely claiming to be a recruiter of some sort. Several women had met with him, then discovered he was sex offender and made police reports. The women making reports were (rightfully) very concerned by his behavior, but he hadn’t done anything criminal - yet.

Steve told me: “We’re just going to knock on this guy’s door, and tell him that we know what he’s doing and that we’re are watching him.”

If you’re a defense attorney, this kind of thing might make you cringe. Here’s the police “harassing” a man who has not committed a crime. After all, he had served his time. What if he had attacked us and we had to use force? Then we’d have “escalated” the situation for no lawful purpose.

And yet I would maintain that this was good police work, and I think most normal people would agree. Nothing we did was illegal - we just wanted to talk to this guy. We had no intention of beating him him up or making any threats. Isn’t it better to use our social authority as cops to solve this obvious problem even if there is no arrest-and-prosecute solution? I think it’s better than waiting to see if this guy commits a rape, and then going through the process of prosecuting and imprisoning him.

I use this example because I think it highlights a certain problem when it comes to police and the criminal justice system. Many lawyers, whether they are prosecutors, public defenders, judges, or law school professors, have come to view the police solely as an extension of the criminal justice system. This is also true of many academics and criminologists. To them, police action is merely the first step in the prosecution process as spelled out in the criminal procedure textbook: Police apprehend Defendant, then Defendant goes to court. But this view fundamentally misunderstands the actual role of police in our society.

Policing should solve social problems

Some of the calls I am most proud of as a police officer ended with no arrest, no jail time, and/or no prosecution. Of course I was glad when I arrested offenders who had committed serious violent felonies and wanted those cases resolved with a conviction and prison time. But those cases are relative rarities compared to other social problems, which are overwhelmingly related to public disturbances, alcohol-related behavior, and other disputes between mostly normal people

Here’s a couple examples of how we solved smaller problems in my precinct:

  1. A young woman, heavily intoxicated, had been caught drawing graffiti with a sharpie on a local business. The business owner was irate because of the pervasive graffiti problem in the area, but the woman had no criminal record at all. The graffiti in question amounted to about $20 worth of damage. I spoke to the woman and the business owner, and we came to the following arrangement: We would all trade contact information, and she would return tomorrow when sober and clean the graffiti. If she did not, the business owner would contact me and I would file charges against her with the city attorney.

  2. A taxi driver called to complain that his (very drunk) fare had stiffed him and was trying walk away. By the time we found them, the taxi driver had gotten into a scuffle with the drunk guy, knocking him over. I pulled the two of them aside and told the drunk guy he needed to pay the fare or get arrested for theft of services. I told him he needed to make a complaint with the taxi commission if he had a problem with the fare. I told the taxi driver he was an idiot for getting into a physical confrontation over a $20 fare and that he easily could have ended up in jail had the drunk guy gotten hurt. Both parties were then sent on their way.

  3. A very drunk guy threw a punch at someone outside a bar where we often had large crowds of intoxicated people fighting. A group of cops rushed in, arrested the guy, pushed his friends back, and took him to the precinct. We told his friends (who were now much less interested in fighting) that they could come get him from the precinct, which they did. We let him go home with them about an hour later. He was never booked into jail or charged with any kind of crime.

Nothing in “the law” or my policy manual explicitly allowed us to arrest someone and then release them with a warning. Nor did it allow me to invent a graffiti diversion program. But nothing prohibited it either, and I would maintain that these solutions were the best way to solve these particular problems. Could I have done less? Yes, but then I wouldn’t be doing my job. Could I have booked all these people into to jail and asked for charges? Yes, but what would be the point? For the most part, these situations required a brief coercive intervention to resolve a public disturbance or prevent harm. Our power to arrest was not primarily a way to start the machinery of the criminal justice process. It was a means by which we solved social problems.

The historical role of the policeman

The grandfather of today’s beat cop was the night watchman of Victorian England, a system which dated back to at least the 1200s. The watchmen were a sort of informal nighttime police service. At least in some places, being a watchman was a part-time duty imposed on householders as a form of public service. It was not considered an especially prestigious position, and people apparently tried to get out of it.

The watchmen were supposed to confront suspicious characters and deal with night-time disorder and disturbances, which was important in an era when cities had no street lighting. But they also did all kinds of things that had nothing to do with catching criminals. They kept an eye out for fires, checked businesses to make sure the doors were locked, and called out the time and the weather every hour.

Semi-professional constables and night watchmen were apparently not especially interested in actually enforcing the law. The literature of the time makes fun of them for being lazy goof-offs, and common law courts had such a hard time getting them to serve arrest warrants that according to Orin Kerr, they actually developed a substantial body of law imposing liability on constables who allowed prisoners to escape:

The part-time officials such as constables (and I'll just call them all constables for the sake of brevity) didn't have much interest in making arrests and detaining people after the arrest.  It was dangerous and time-consuming work, and they in general weren't paid for it.  Who wants to risk getting hurt arresting someone and forcibly bringing him to the local judge?   There's nothing in it for the constable.  So part of the law regulating constables at common law was about forcing the constables to do their jobs—to make arrests and to detain prisoners—or else face civil suits or criminal punishment.

In England, the night watch was eventually abolished in 1829 and replaced with the London Metropolitan Police. American states began to establish municipal police forces (particularly in Boston and New York) around the same time. But American courts and legislatures continue to recognize that modern day police retain an inherent, non-criminal community caretaking authority to address public hazards and perform services that have nothing to do with enforcing the criminal law. Many state statutes continue to refer to police as “peace officers” and almost all states authorize police to detain dangerously mentally ill people who are not committing any crime.

As late as 1970, American sociologist Egon Bittner observed that most of what police do in modern society still has very little to do with law enforcement, and everything to do with solving social problems reported by citizens who are “calling the cops”:

Many puzzling aspects of police work fall into place when one ceases to look at it as principally concerned with law enforcement and crime control, and only incidentally and often incongruously concerned with an infinite variety of other matters. It makes much more sense to say that the police are nothing else than a mechanism for the distribution of situationally justified force in society.

“Situationally justified force” sounds bad. But it makes more sense if you understand that what Bittner meant by “force” was not necessarily the use of a baton or the gun, but rather using the implied threat of the state’s coercive authority to solve a social problem.1 Bittner goes on to give examples of how police solved problems that were reported by citizens who called the cops:

Two patrolmen were directed to report to an address located in a fashionable district of a large city. On the scene they were greeted by the lady of the house who complained that the maid had been stealing and receiving male visitors in her quarters. She wanted the maid's belongings searched and the man removed. The patrolmen refused the first request, promising to forward the complaint to the bureau of detectives, but agreed to see what they could do about the man. After gaining entrance to the maid's room they compelled a male visitor to leave, drove him several blocks away from the house, and released him with the warningnever to return.

In a tenement, patrolmen were met by a public health nurse who took them through an abysmally deteriorated apartment inhabited by four young children in the care of an elderly woman. The babysitter resisted the nurse's earlier attempts to remove the children. The patrolmen packed the children in the squad car and took them to Juvenile Hall, over the continuing protests of the elderly woman…

In a middIe-class neighborhood, patrolmen found a partly disassembled car, tools, a loudly blaring radio, and five beer-drinking youths at the curb in front of a single-family home. The homeowner complained that this had been going on for several days and the men had refused to take their activities elsewhere. The patrolmen ordered the youths to pack up and leave. When one sassed them they threw him into the squad car, drove him to the precinct station, from where he was released after receiving a severe tongue lashing from the desk sergeant.

Bittner observed that this kind of policing tended to make lawyers, judges, and civil libertarians uncomfortable. If you accept that this is how police operate, you accept that police are, to a large degree, unconstrained by the law. The exclusionary rule is no deterrent to a cop uninterested in making a criminal case. And this kind of thing leaves a lot of room for discrimination and police abuse, as the last example illustrates. Bittner noted that as a result, there was a substantial gap between how judges and lawyers viewed the role of the police and how the police actually behaved.

The growth of the “law enforcement” mindset

Over the last half-century, the gap between what average people expect police to do (solve my problem) and the institutional view of what police should do (enforce laws) has only grown. The legal community widely rejects the idea of police as a community resource which uses coercion to solve disorder, and now views them almost entirely as serving a law enforcement role. Making matters even worse, police have actually started to see themselves this way too. The average police officer, if asked what profession he is in, will now say he’s in “law enforcement.”2 This trend extends to police executives too, who are now obsessed with crime rates and “stats” and the latest “evidence based” strategies for reducing crime.

Why did this happen? I don’t actually know for sure. But I have a few guesses:

  • The massive crime wave that peaked in the early 1990s put immense pressure on police leaders to focus on crime, leading to things like Compstat.

  • Our popular culture is obssessed with detective programs. Think about your average police procedural show - is it about police solving problems, or is it about police catching criminals?

  • The rise of “procedural justice” theory among police reformers, which in order to promote transparency and fairness, tends to demand that police approach all situations in a standardized and legalistic manner.

  • Expanding judicial regulation of police, to the point where violating the Fourth Amendment now carries a serious risk of civil penalties as civil rights litigation against governments has exploded in recent decades.

The problem is that when we look at what people actually ask the police to do, the average citizen doesn’t seem to think police are here just for law enforcement. This means there is now a substantial and increasing disconnect between what the public is asking police to do every day, and what judges, lawyers, police executives, and policymakers are actually telling police they should do.

For an example, check out this chart from criminologist Jerry Ratcliffe, which shows all the types of calls Philadelphia Police handled in 2019:

It turns out only about half of what people call the police for in Philadelphia is really related to crime and even some of that (“person screaming”) could probably go either way. Just for kicks, I did the same thing with call data from Seattle - a much wealthier and whiter city. After removing proactive calls generated by officers, I found basically the same pattern, albeit with less violent crime:

My data isn’t cleaned up like Prof. Ratcliffe’s, but I think the general trend is still the same: At least when it comes to calling the police, people still expect a service that is very similar to that performed by the night watchman from a bygone era. They expect the cops to deal with thefts and assaults. But they also expect them to handle public disturbances, road hazards, trespassers, suspicious people, and alarms.3 They expect the police to assist the public, keep the peace, and maintain order.

What happens when police fail to do that? Imagine a person who flags down a police officer, asks for directions, and is told “Sorry, that’s not my job.” Or imagine a person who is having a dispute with his neighbor, calls the police for help, and when they come is told that no crime was committed so the police won’t do anything. How will these citizens regard the police? I think it undermines police legitimacy, and leaves a huge number of social problems unsolved.


The incarceration problem

I also think that in trying to limit police authority, civil libertarians have shot themselves in the foot when it comes to the incarceration problem. Do we really want an army of police who never violate the Fourth Amendment, but robotically make arrests or issue tickets whenever they have probable cause to do so? Does the mere act of following constitutional procedures make an otherwise pointless arrest worthwhile? Too many cops I worked with - including myself when I was new - came straight from the police academy having been taught that following and enforcing the criminal law was an end in itself, not a tool we could use to make our city a better place.

Take, as an example, the new Illinois bail reform law. Among other things, this law will prohibit police in Illinois from making a physical arrest of a person who is trespassing or engaging in disorderly conduct. Instead, they can only issue the violator a criminal citation with a notice to appear in court, and let them go.

To me, this seems exactly backwards. Isn’t it better for a police officer to tell a trespasser or someone disturbing the peace “leave, or I’ll arrest you” rather than charging them with a crime and hauling them into court? And if a warning didn’t work, isn’t it better for police to arrest the person, drive them somewhere safe, and let them go rather than book them into jail? What’s more likely to be harmful to the arrested person? I would think being formally charged with a crime and forced to appear in court repeatedly over a petty trespassing charge would be worse than being told to leave the area or detained for a few hours. But in Illinois, a formal criminal charge is now the only option left in the police toolkit.

The “law enforcement” attitude has led to deleterious outlooks among police as well. A lot of cops I knew bitched mightily about having to deal with calls about the mentally ill, runaway kids, or neighbor disputes. They had gone to the police academy thinking they were going to be cool crime fighter guys like on TV, and instead ended up refereeing an argument between a landlord and tenant. But this is the essential nature of police work. The best police officers I knew understood this and took it to heart. They were excellent mediators. They were outgoing and social. But too many cops have been persuaded that their only function is to make a criminal case, and that if they can’t do that, they shouldn’t be doing anything at all.

It also leads to bad prosecutorial decisions. Prosecutors should not prosecute every arrest the police make. Many arrests and jail bookings have solved the immediate problem, and no other action is needed. Maybe the drunk guy who was fighting at the bar needed to sit in a jail cell for 12 hours because he was totally out of control. But now that’s over, and he’s not likely to do it again. What’s the point of prosecuting him? I would say none, and some prosecutors decline cases like this for exactly this reason. Yet it’s not uncommon for academics to argue that a high volume of dismissed misdemeanor cases means the police are doing something wrong.

Constraining discretion

Emphasizing problem solving over law enforcement means finding a way to appropriately constrain police authority to keep cops from making arrests for “contempt of cop” or on the basis of prejudice. As a starting point though, I would say that people must always extend at least some trust to police to exercise discretion appropriately. After all, if an officer truly wants to be malicious, nothing really stops him from doing that right now. An officer with no moral bearing, no regard for the truth, and no interest in following the law can easily evade any constitutional constraint, at least for a while. This is the terrifying power of policing, and it is why we try so hard to hire good people to be police officers and keep bad people out.

A second safeguard has to be apprpriate supervision and training. The emphasis for a police officer in his training, his supervision, and in his work should be on doing the right thing from a moral perspective. He should be taught not to view himself as a mere arm of the criminal law, but as a mediator of disputes and promoter of social norms in the community where he works. Criminal and constitutional law should serve as an outer boundary on his discretion, but not an end in and of themselves. He should be out there giving directions to tourists, playing football with kids in the park, and generally serving as a community resource. In deciding whether to make an arrest or issue a citation, he should ask: Will this solve the problem I’m here to solve?

There must also be a substantial shift in order to tighten the connection between the officer and the neighborhood he works. Cops should spend almost all their time out of their stations and cars and on the beat. Police reports needed solely for insurance purposes should be made online or via an appointment at the local police station. Agencies should offer police officers substantial incentives to live in or near the place they work, like take-home police cars or discounted housing. And finally, this would require hiring many more police officers and putting a lot of them out on foot in densely populated urban neighborhoods and commercial districts. Officers should be assigned to a small, specific patrol area for a fixed period of time and be held accountable for knowing the people in it and the problems they need solved.

This idea is actually what drove the original “Broken Windows” theory back in 1982. Over the decades, the concept behind Broken Windows been twisted into “police should arrest and charge a lot of people for minor offenses because that prevents larger crimes” as the “law enforcement” mentality took over policing. But the original idea espoused by Kelling and Wilson envisioned something a lot closer to true community policing, with officers working regular beats on foot and enforcing social norms in the manner the neighborhood wanted them enforced. Kelling and Wilson did believe that addressing disorder would reduce crime, but they also recognized that reducing social disorder was a good thing even if it did not affect overall crime rates.

In response to the issue of police accountability, Kelling and Wilson noted the inherent dynamic of foot patrol held a policeman physically accountable to the community in a way that the cop in a car is not:

An officer on foot cannot separate himself from the street people; if he is approached, only his uniform and his personality can help him manage whatever is about to happen. And he can never be certain what that will be—a request for directions, a plea for help, an angry denunciation, a teasing remark, a confused babble, a threatening gesture.

In a car, an officer is more likely to deal with street people by rolling down the window and looking at them. The door and the window exclude the approaching citizen; they are a barrier. Some officers take advantage of this barrier, perhaps unconsciously, by acting differently if in the car than they would on foot.

In short: an officer on foot is more physically accessible and accountable to people in the neighborhood where he works. And there is empirical evidence suggesting that 1) officers on foot patrols have a more positive view of the community in which they work and 2) foot patrol reduces fear of crime and improves public perceptions of police officers.

The future of policing

My greatest worry is that the current wave of police reform risks forcing cops even further into a narrow “law enforcement” silo. The skills that make a good cop - talking to people, solving problems, serving as a mediator - these are the kind of things that have be embedded at a cultural level, taught by veteran officers, and learned via experience. Once those skills have faded from the institution - and I’m afraid they have already faded quite a bit - it will be very hard to bring them back. I saw this happening in my department by the time I left.

Police executives and urban policymakers deserve a lot of blame here. They’ve had twenty years of relatively low crime to reorient their police departments away from the law-enforcement based model of thinking and towards a more truly community-driven model of problem solving. But despite giving problem oriented policing a ton of lip service, most of them are still obsessed with 911 response times and foot patrol remains a rarity. Now that violent crime is surging and cities are cutting police budgets, most urban police departments will have no choice but to focus their limited resources on violent crime.

If we are serious about reimagining policing and reducing incarceration, we don’t have to invent anything new. Police just need to return to their historical roots: Stop thinking of police as a mere extension of the criminal justice system, and get back to thinking of police as people who use the power of law to solve society’s problems.


Some people might argue that state coercion outside of criminal law is always inappropriate, but given our vast body of civil law and administrative law, I think this is a hard argument to sustain. The only real disagreement is about when coercion should be applied and in what manner (monetary fines, etc).


There are of course agencies which really only do law enforcement, like the FBI. And some police roles, like that of a detective, are exclusively about law enforcement. But that’s not the focus of contemporary debate.


Seattle Police categorize trespassers as “prowlers” for some reason.