Watching all the terrible news in the wake of the police killing of George Floyd, it’s been hard not to think about Eric Garner. The cases have so many similarities. Once again, an unarmed African-American man in his forties has been asphyxiated in broad daylight by a police officer with a history of abuse complaints. He and his fellow officers ignore cries of “I can’t breathe,” and keep subduing their target even after he stops moving, unconcerned that he’s being filmed.

Five years ago, while sketching the outline for a book about the Garner case called I Can’t Breathe, my editor suggested I take on a larger question.

Why, he asked, do we even have police? After all, the history of policing in our country, especially as it pertains to minority neighborhoods, has always rested upon dubious justifications. The early American police forces evolved out of slave patrols in the South, and “progressed” to enforce the Black Codes from the Civil War period and beyond, on to Jim Crow through the late sixties if not longer.

In an explicit way, American policing has almost always been concerned on some level with enforcing racial separatism. Because Jim Crow police were upholding a way of life, the actual laws they were given to enforce were deliberately vague, designed to be easily used as pretexts for controlling the movements of black people. They were charged with punishing “idleness” or “impudence,” and encouraged to enforce a range of vagrancy laws, including such offenses as “rambling without a job” and “leading an idle, profligate, or immoral course of life.”

I ended up not taking on that question, focusing on the hard-enough question of what had led two young, amped-up policemen to choke the life out of a harmless father and street character like Garner. I was more interested in those police than all police, and part of me – the white part, probably – thought the answer to the question of why we need police at all was at least somewhat self-evident.

But the Garner story ended up graphically revealing the way modern “Broken Windows” policing had evolved to fit the tactics of those centuries of racial enforcement. I learned that “vagrancy” laws had been replaced in cities like New York with essentially identical offenses like “obstructing pedestrian traffic” and “obstructing government administration.”

In Staten Island, a borough that to this day remains very segregated – white and black residents alike refer to the Staten Island Expressway that bisects black neighborhoods to the north and white neighborhoods to the south as the “Mason-Dixon line” – the young black men who lived in and around the Tompkinsville area where Garner was killed told stories of being stopped and ticketed whenever they crossed into the wrong neighborhoods.

The new strategies rely upon extremely high numbers of contacts between police and subject populations, who are stopped for every conceivable minor offense – public intoxication, public urination, riding bicycles the wrong way down a sidewalk, refusing to obey police orders, jumping subway turnstiles, and, in Garner’s case, selling loose cigarettes.

This idea of high-engagement policing was born in the mind of a Midwestern academic/corrections official named George Kelling. Kelling conducted a number of studies for think tanks like the Police Foundation and eventually co-authored a hugely influential 1982 article in the Atlantic called Broken Windows.

Kelling in his research found that while people may not actually be safer, they feel safer when there is less visible “disorder” in their neighborhoods, e.g. panhandling, litter, graffiti, etc. Also, research suggested such disorder was incentive to further disorder: as Stanford researcher Philip Zimbardo put it, “If a window in a building is broken and left unrepaired, all of the rest of the windows will soon be broken.”

“Broken Windows” revolutionized policing, changing it from a business of fighting crime to doing what Kelling described as “order maintenance.” If earlier police theorists like Orlando “O.W.” Wilson hoped to defeat crime by putting officers in squad cars and giving them advanced tools to react more quickly to offenses, the new strategy stressed stopping crime before it got started, by building and maintaining something not defined in law books – “order.”

Once again, police were charged with enforcing not rules but a way of life, and were asked again to view the law as more of a tool than an end in itself. The famous “Broken Windows” article spoke approvingly of officers in Chicago who read between the lines of the law to chase gang members out of a project: “In the words of one officer, ‘We kick ass.’”

The Kelling revolution was credited with early successes, like the cleaning up of the New York City subway. Soon the “Broken Windows” strategy (sometimes euphemistically called “community policing”) was the norm in big cities. Mass stops and arrests led to amazing numbers, like Baltimore under Mayor Martin O’Malley arresting 100,000 people in 2004 alone, or the city of Chicago stopping 250,000 people in 2014, a stop rate four times higher than New York in the peak years of “stop-and-frisk.”

When such policing became hot in the nineties, as advocates like Bill Bratton became national celebrities (here he is on the cover of Time in 1996 under the headline, “Finally, we’re winning the war against crime. Here’s why”), police departments became infected by a corporate-like mania for “goal-setting” and “deliverables.” There was no numerical way to impress politicians if police just worked cases as they came: to show progress, Bratton believed, one had to order police to produce concrete quantities of stops, searches, arrests.

Commissioners demanded captains deliver numbers and captains began browbeating lesser officers, who in turn pushed quotas on patrol cops, for reasons that often had nothing to do with crime. As depicted in the The Wire, in the stats revolution, “shit always rolls downhill.” The point was to get lieutenants promoted to captain, to get mayors re-elected, and help provide the rationale for the prison jobs state legislators were bringing home to suburban districts. All of this was greased by the lobbying money of construction firms, prison vendors, even private prison corporations – a great business for all, and all that was needed to keep it going was an endless stream of jailable people.

This is why, even as rates of both violent crime and property crime have been decreasing steadily since the early nineties, rates of incarceration have been exploding in the other direction. For most of the 20th century the rate of incarceration in America was roughly 110 per 100,000 people. As of last year, the number was 655 per 100,000. Although the numbers have dipped slightly in recent years, down from a high of about 760 per 100,000 in 2013, the quantity of prisoners in America remains absurdly high.

Such aggressive, military-style policing would be not be tolerated by voters if it were taking place everywhere. It’s popular, and continues to be embraced by politicians in both parties, because it’s only happening in “those” neighborhoods (or, as Mike Bloomberg once put it, “where the crime is”). Even during the Covid-19 crisis, 80% of the summonses for social distancing violations are given out to blacks and Hispanics. Does anyone really think that minorities account for that massive a percentage of those violations? Do they think black people really commit 3.73 times as many marijuana offenses as white people?

Basically we have two systems of enforcement in America, a minimalist one for people with political clout, and an intrusive one for everyone else. In the same way our army in Vietnam got in trouble when it started searching for ways to quantify the success of its occupation, choosing sociopathic metrics like “body counts” and “truck kills,” modern big-city policing has been corrupted by its lust for summonses, stops, and arrests. It’s made monsters where none needed to exist.

Because they’re constantly throwing those people against walls, writing them nuisance tickets, and violating their space with humiliating searches (New York in 2010 paid $33 million to a staggering 100,000 people strip-searched after misdemeanor charges), modern cops correctly perceive that they’re hated. As a result, many embrace a “warrior” ethos that teaches them to view themselves as under constant threat.

This is why you see so many knees on heads and necks, guns drawn on unarmed motorists, chokeholds by the thousand, and patterns of massive overkill everywhere – 41 shots fired at Amadou Diallo, 50 at Sean Bell, 137 at Timothy Russell and Malissa Williams in Cleveland, and homicides over twenty bucks or a loose cigarette.

Police are trained to behave like occupiers, which is why they increasingly dress like they’ve been sent to clear houses in Mosul and treat random motorists like potential car-bombers – think of poor Philando Castile, shot seven times by a police officer who leaped back firing in panic like he was being attacked by Freddy Krueger, instead of a calm, compliant, educated young man. Officers with histories of abuse complaints like Daniel Pantaleo and Derek Chauvin are kept on the force because senior officers value police who make numbers more than they fear outrage from residents in their districts. The incentives in this system are wrong in every direction.

The current protests are likely to inspire politicians to think the other way, but it’s probably time to reconsider what we’re trying to accomplish with this kind of policing. In upscale white America drug use is effectively decriminalized, and Terry stops, strip searches, and “quality of life” arrests are unknowns. The country isn’t going to heal as long as everyone else gets a knee in the neck.

Where did policing go wrong?