DRAWING ON LESSONS he learned as a seminarian, social worker, and then probation officer, George Kelling became a major thinker on law enforcement, envisioning a new approach to policing that combined a strong regard for public safety with a keen sensitivity to community concerns and perspectives. When Kelling died this spring, at age 83, William Bratton, the former Boston and New York City police commissioner, told the New York Times he had been “the most profound influence on American policing in the last 40 or 50 years.”
But his death has revived a long-standing debate over the nature of Kelling’s outsized influence.
To some, Kelling was one of the visionary architects of community policing, a forward-thinking approach that fosters closer ties between police and residents. But that view has largely been overshadowed by critics who say he helped usher in a heavy-handed approach to law enforcement that has badly eroded community trust in the police. At its worst, say critics, Kelling’s model gave rise to an aggressive style of “zero tolerance” law enforcement that led to the rash of high-profile incidents in recent years that have claimed the lives of unarmed black men at the hands of police.
At the center of the controversy — and the seemingly irreconcilable views of Kelling’s legacy – is a lengthy article he co-authored nearly 40 years ago in The Atlantic. “Broken Windows,” an essay he cowrote in 1982 with conservative political scientist James Q. Wilson, argued that addressing community disorder and less serious quality-of-life issues was an important component of crime prevention and neighborhood safety.
“One unrepaired broken window is a signal that no one cares, and so breaking more windows costs nothing,” they wrote. Kelling and Wilson, who died in 2012, argued that “disorder and crime are usually inextricably linked.”
The article proved to be hugely influential, setting the stage for a wholesale rethinking of the role of police in American cities.
The “broken windows” approach is based on the belief that residents are not just worried about serious crime, but are often equally concerned about various types of “disorder” — from graffiti to unruly drinking on street corners — that can degrade their quality of life and sense of safety. Wilson and Kelling’s essay provided the framework for police departments across the country to shift their mission from solely responding to emergency calls to also trying to help communities deal with the kinds of disorder that can lead “to the breakdown of community controls.” But when their ideas were translated into strategies to deal with disorder during a time of soaring crime and violence in US cities, far from bringing residents and police together, broken-windows policing often became a wedge that worsened already strained relations between communities and police.
NOT JUST ‘ARREST AND PROSECUTION’
Ed Davis, the former Boston police commissioner who earlier led the Lowell Police Department, brought Kelling in as a consultant when he took the reins as police chief in Lowell in 1994. “He got us thinking about disorder and got us thinking about the role of the community and prevention,” says Davis. “It was about more than just arrest and prosecution,” he says of the approach he had officers bring to chronic problems in neighborhoods.
Davis pushed officers to work more closely with residents and focus not only on responding to emergency calls but also “on the quality of life of people in the neighborhoods.” Over the ensuring 12 years, Lowell experienced one of the biggest decreases in crime of any similar-sized city in the country. “I attribute that to George’s philosophy and what he gave me and my officers,” Davis says.
Davis calls Kelling — who was a fellow at the Manhattan Institute, a conservative New York think tank, at the time of his death, but spent many years in Boston at Northeastern University and Harvard’s Kennedy School — “a legend in the field.”
William Gross, Boston’s current police commissioner, is also a big fan of Kelling and his ideas. The broken windows approach has been a “definite positive for Boston,” says Gross, who got to know Kelling while serving as a deputy superintendent under Davis. “It’s all about working in partnerships with the community to problem solve. It’s the little things that count,” he says, citing police attention to problems caused by people congregating at a vacant house as one example.
Striking the right balance in dealing with disorder has been at the center of recent debate over how police have handled the large number of homeless people and those suffering from drug addiction in the area near Newmarket Square in Boston, where a lot of drug treatment facilities and other services are based. After a corrections officer heading to the nearby South Bay House of Correction was assaulted by several people congregating in the area, police made more than two dozen arrests, mostly for outstanding warrants on a variety of charges, including some violent offenses and drug distribution cases.
Residents in the adjacent South End and Roxbury neighborhoods say they’ve been inundated with problems ranging from assaults and discarded needles to people using their property as an outdoor bathroom. But some, including Suffolk County District Attorney Rachael Rollins, have been critical of the police action, saying arrests aren’t the solution to the complicated troubles caused by homelessness, drug addiction, and mental illness.
“Public health and public safety have to work together,” says state Rep. Liz Miranda, who represents the Roxbury neighborhoods that abut the problem area. She voices compassion for those dealing with addiction and other problems, but says poor and largely minority neighborhoods are unfairly bearing the burden of negative, sometimes dangerous, behavior from this population. “We need help,” she says.
In Lawrence, police arrested 80 people over several days in June during a series of sweeps focused on minor offenses like aggressive panhandling as well as several prostitution-related charges. Police Chief Roy Vasque says violent crime is always a priority, “but oftentimes when we go to a community meeting and meet with business owners or church leaders, we hear they’re more concerned with prostitution and noise and graffiti and aggressive panhandling and things like that.”
“We can’t turn a blind eye to complaints from citizens,” he says. Though the recent action resulted in a lot of arrests, he says that’s not the department’s goal. “That’s not the focus — making arrests,” says Vasque. “It’s about changing behavior.”
If law enforcement leaders in Massachusetts credit Kelling’s ideas with getting police departments here to work more closely with residents, the broken windows focus on curbing disorder has been blamed for fueling aggressive tactics in places like New York City, where the approach morphed into the controversial “stop-and-frisk” policy that saw thousands of innocent black and Hispanic men stopped in an effort to combat crime and gun violence. In 2013, a federal judge ruled that the approach violated the constitutional rights of minority residents and ordered reforms in department procedures.
Meanwhile, the dangers of a “zero tolerance” approach to minor nuisance crimes and disorder came into horrific view in 2014, when Eric Garner, who was illegally selling loose cigarettes on a Staten Island street corner, was put in a chokehold by a New York police officer and died despite repeatedly crying out, “I can’t breathe,” a gruesome scene captured on cellphone video.
“You can see the abuses of broken windows in cities all across America. Eric Garner is a perfect example,” says Rev. Jeffrey Brown, who was part of Boston’s clergy effort against gang violence in the 1990s and now heads a national faith-based anti-violence organization.
As for the connection between broken-windows policing and reduced crime rates, the evidence is “very mixed,” says Anthony Braga, director of the School of Criminology and Criminal Justice at Northeastern University, who worked closely with Kelling. A 2016 analysis Braga led of the 30 rigorously conducted studies on the question found an overall “modest crime reduction effect” associated with the approach. But digging deeper into the data, the researchers made an important observation: Much stronger benefits were seen in the subset of 20 studies of “community problem-solving programs,” in which police worked with local residents and merchants to identify problems. The 10 studies of departments using “aggressive order maintenance strategies,” which primarily relied on arrests and issuing summonses for various violations, did “not generate significant crime reductions,” the study concluded.
“If you go back to the original article,” Braga says of Kelling and Wilson’s 1982 essay in The Atlantic, “what they are making an argument for is doing that sort of community problem-solving.”
But that is not how the approach has often been applied.
BROKEN WINDOWS-POLICING ‘RUN AMOK’?
Kelling maintained that a key element of using the broken windows theory was having buy-in from residents, who he said police should work with to set norms for the kind of behavior to target in a neighborhood. Instead, the idea of clamping down on disorder and more minor crimes got folded into aggressive policing in New York and other cities that inflamed tensions between police and minority communities.
In the late 1970s, Kelling helped direct a field research project in Newark that demonstrated the benefit of police foot patrols on residents’ sense of safety, one of several studies that shaped his thinking in developing the “Broken Windows” article.
More than three decades later, in 2016, Newark was the focus of a PBS “Frontline” episode that cast Kelling’s approach in a very different light. “The Problem with Broken Windows Policing” documented how the theory became the foundation for “aggressive over-policing of minority communities” like Newark.
Kelling told “Frontline” that his ideas had been increasingly misapplied. As he heard about more and more departments adopting the approach without the sort of training and careful use of discretion he felt was crucial, Kelling said he had a blunt, two-word reaction: “Oh, shit.”
A year earlier, in 2015, Kelling wrote a piece for Politico responding to the growing criticism of the approach. “A lot of sins have been committed in the name of ‘broken windows,’” Kelling wrote. He decried the fact that the Ferguson, Missouri, police stop of Michael Brown for jaywalking and police confronting Eric Garner in New York City for selling loose cigarettes were said to be examples of “broken-windows policing run amok.” Kelling said his policing theory “has been largely misunderstood,” and emphasized that “broken windows was never intended to be a high-arrest program.” [Kelling cited the police stops of Brown and Garner in his article, not their killing, as originally reported.]
The idea has become so toxic that Democratic presidential candidate Pete Buttigieg highlighted “broken windows-style policing” as a “dangerous and discriminatory” practice that should be ended as part of his “Douglass Plan,” billed as his agenda for the “empowerment of black America.”
Although the idea behind broken windows is to pay attention to things that matter to residents in communities where various types of disorder take a toll on daily life, criminal justice reform advocates have turned the focus to ways the approach has harmed those committing minor offenses, saddling them with criminal records and onerous fines.
Rollins, the Suffolk County DA, has drawn attention — and controversy — by declaring the default position of her office will be to not prosecute 15 lower-level offenses. Even though Kelling said broken windows is not meant to be a “high-arrest program,” supporters of order-maintenance policing say Rollins’s policy sends a dangerous message of indifference to minor offenses and quality of life concerns.
Rollins declined multiple requests to speak with her.
“Some of the anti-broken windows criticism is rooted in the problematic application of the approach, and is well-deserved,” says Braga, the Northeastern University criminal justice professor.
But attacks on Kelling and the broken windows approach, as he conceived it, have been “profoundly unfair,” says David Kennedy, a professor at John Jay College of Criminal Justice in New York.
“What George says in ‘Broken Windows’ is that fear and disorder matter,” says Kennedy. But he said “expressly that arrests should be a last resort. One of George’s deepest convictions was that the role of police should be negotiated between the communities and the police. Zero tolerance is an abhorrent and toxic idea and it did tremendous damage. That was not George, and that was not George’s thinking.”
Much of the dispute ends up centering on how Kelling and Wilson’s idea gets defined. To Brown, the Boston minister and anti-violence activist, the broken windows approach was always too narrow.
“I was never a broken windows fan,” says Brown. “I’m a community policing fan, and I see community policing as different than the broken windows theory. I see broken windows leading to things like stop-and-frisk and zero tolerance because the approach was really about taking data and applying it to a community without the community’s input. It’s not so much how he put it out there as it is how people perceived it and put it in place in their own communities,” he says of Kelling’s idea.
Mark Moore, a professor at Harvard’s Kennedy School who hired Kelling to direct the criminal justice program there in the 1980s, traces much of the controversy over broken windows-policing to the fact that the idea became separated from other components of community policing that emphasize residents’ role in public safety. Moore helped lead a set of “executive sessions” on policing at the Kennedy School over a span of several years in the late 1980s and early 90s. The confabs, which Kelling took part in, brought together police leaders from across the country and researchers and served as the laboratory where many of today’s ideas underpinning community policing took form.
“What emerged from that was three strands that were woven together in a rather complicated story that we were trying to tell,” Moore says. The three elements, he says, were the broken windows idea of clamping down on disorder; a police focus on devising longer-term solutions to the problems, especially without resorting to arrests; and strong partnerships that include residents in the identification of problems, something that is key to securing community support for police efforts.
The “broken windows strand got way out in front of the others and started dominating, and that was bad news,” says Moore. When the order-maintenance approach was applied without the problem-solving or community relations strands, “broken windows became aggressive prevention policing, or kicking ass and taking names,” says Moore. “What started off as an idea designed to strengthen the relationship between the police and local communities by acting on problems that the local communities disliked” ended up with communities having “fear of the police rather than seeing them as a force that could help them.”
Frank Hartmann, a research fellow at the Kennedy School, says Kelling embraced all three elements and was driven by a strong belief that “police can’t be an occupying army, just showing up when we’ve got a serious problem.” Kelling’s idea was to “use police in such a way that the community has a sense of itself and its own ability to sustain and protect itself,” says Hartmann, who worked with him at Harvard.
If Kelling is to be faulted, says Hartmann, it’s because “he let some of the distortions go on too long before speaking out.”
Some of the potential downsides of the broken windows approach, however, were anticipated by Kelling and Wilson from the outset.
“We might agree that certain behavior makes one person more undesirable than another but how do we ensure that age or skin color or national origin or harmless mannerisms will not also become the basis for distinguishing the undesirable from the desirable?” they wrote in their original 1982 article. “How do we ensure, in short, that the police do not become the agents of neighborhood bigotry?” They conclude: “We can offer no wholly satisfactory answer to this important question.” The key, write Kelling and Wilson, is in the “selection, training, and supervision” of the police.
Kennedy, the John Jay College professor, directs the National Network for Safe Communities, an initiative that works to promote community-based efforts that bring residents and police together to tackle crime and public safety issues. He says that type of collaboration is a much more accurate reflection of Kelling’s vision than the heavy-handed policing his ideas have become identified with.
The focus on disorder and quality-of-life problems in neighborhoods requires careful application of discretion by officers and strong police leadership to avoid the risks Kelling and Wilson wrote about. But Kennedy says the broader community policing philosophy that Kelling insisted broken windows should be part of remains the best hope for American cities, not a return to the “rapid response” model in which police stay at arm’s length from residents and see their job as only responding to 911 emergency calls.
“The labels and the framing of broken windows has not survived the historical moment that we’re in,” he says. “But that broader project that George was so committed to, I think, is actually very healthy.”