George Kelling knew he had the opportunity of a criminologist's lifetime - the chance to test his controversial theory in one of the most crime-ridden venues in America.
As a newly hired consultant to the New York City subway's police department, the tweedy professor set out to prove the seemingly heretical idea that paying attention to small crimes - such as panhandling and pickpocketing - can help reduce violent offenses.
His seminal essay, "Broken Windows," had struck a chord with crime-weary citizens. Yet his ideas had stayed tucked away in the ivory tower. No police chief was willing to give them a try. Until 1990. New York City police began using Dr. Kelling's theory in an attempt to turn the dirty and dangerous subway into a transit system where passengers would feel comfortable and safe.
Many said it couldn't be done. But within just a few months, minor crimes such as graffiti-spraying and public urination were sharply reduced - and along with them more serious offenses. Overall crime dropped 75 percent.
Today Kelling's theory is sparking a revolution in American policing.
His idea of focusing on so-called nuisance offenses, now in use throughout New York, is considered one reason for the city's dramatic drop in crime. It is also changing how police operate in neighborhoods from Boston to Seattle.
Kelling himself, a professor at Rutgers University in Newark, N.J., is popular on the lecture circuit at law-enforcement seminars and in precinct houses.
"We're going to have our ups and downs, but the old ideology is dead," says a matter-of-fact Kelling, speaking in his spare office at the School of Criminal Justice here. "People are saying all over, 'This is all common sense.' It's kind of why I think we haven't seen the best yet. More and more cities are looking for ways to implement these policies."
At the heart of it all is Kelling's "broken windows" theory. According to Kelling, the theory goes like this: "Just as a broken window left untended is a sign that nobody cares and leads to more damage and vandalism, so disorderly behavior left untended is a sign that nobody cares and leads to more serious crime."
THE theory derives its name from a test conducted by a behavioral psychologist in the late 1960s. The psychologist parked a car, with no license plate and with its hood up, on a quiet street in Palo Alto, Calif. The car sat untouched for more than a week. Then he smashed one of its windows. Within a few hours, the car was overturned and destroyed. This car later became Kelling's "handy metaphor" for illustrating his theory.
In the stations and underground tunnels of New York's subway, Kelling's theory took shape in policies such as ejecting loiterers, arresting aggressive panhandlers, and continuing a campaign against graffiti. These pervasive problems showed no one cared, just as the broken window did, Kelling reasoned. They intensified fears of passengers and gave confidence to lawbreakers.
Working hand-in-hand with Kelling was William Bratton, an up-and-coming police sergeant who shared views similar to Kelling's. The criminologist credits Mr. Bratton with leading the rank and file to implement the new ideas.
"One couldn't ask for a better laboratory than the subway," Kelling says. "First of all, it was so bad and outrageous. And secondly, it was large. Thirdly, it's a less complex system than the streets."
Kelling vividly remembers the moment when the merits of his theory became clear to all involved in the subway fight against crime.
A crackdown on fare-beaters, individuals who jump the turnstiles without paying, netted some unexpected results. Transit police "discovered in some stations that 1 out of 7 fare-beaters they were arresting either had a warrant out for a violent felony or had a weapon," Kelling says. "Wow, that was the link."
That breakthrough proved that an anticrime strategy focused on low-level offenses would nab some high-level offenders. More important, it showed that this approach helped to prevent more serious crimes - a finding that overturned the dominant police thinking of the day.
Until then, leading theories held that crime stemmed from social ills such as homelessness, joblessness, and poverty, problems that police had no power to rectify.
"As serious crime was increasing, more and more deviance was being discounted," Kelling says of the policing trends of the 1970s and '80s. "It was to the point in New York City that burglary and car theft were being almost decriminalized because there was more and more violent crime, more and more robberies, and the focus was there.
"In many respects that made a lot of intuitive sense," Kelling says. "It was hard to take on, as long as you maintained the idea that victims are individuals and that communities can't be victimized." But if you look at communities as a whole, and if you examine the effects of a cluster of incidents rather than individual incidents, another view emerges, Kelling says. "The aggregate of behavior of 20 drunks in the neighborhood can be much more destructive to a neighborhood than one incident of robbery."
Certainly, Kelling's theory has its detractors. Cities where his ideas have been put into practice have fended off lawsuits from civil libertarians who say prosecutions of vandals and panhandlers are often arbitrary and can infringe on the rights of the accused. Other criminologists pin the reduction in crime in New York and other cities to demographics or to prison policies - not a new way of policing.
But there is no doubt that, in large part because of the high-profile drop in crime in New York City, order-maintenance theories of Kelling and others are getting more attention than ever.
"It's part of the common police parlance now," says Wesley Skogen, a political scientist at Northwestern University in Chicago, who has researched the links between disorder and crime.
After Mr. Bratton was named New York City police commissioner in late 1993, Kelling's theory was put to test citywide.
Bratton's approach was to devolve authority from headquarters to the precincts; build a new, better-articulated community-policing effort; and make quality-of-life crimes the top priority, as had been done in the subway. Three years later, the number of reported felonies has dropped by half.
What the statistics show, Bratton says, is that "police can impact crime. While society tries to figure out what to do about poverty, et cetera, police can effectively give society the breathing room to deal with long-standing problems."
George Kelling is a bear of a man, tall and broad with a hearty moustache. He's spent a good many of his years in the upper Midwest, and his demeanor mirrors the ruggedness of the region. He has also spent much of his life on the road, parachuting in as a consultant to one police department or another, gathering research at residents' crime-watch meetings, lecturing at police seminars, or promoting his latest book. His office at Rutgers wears the telltale signs of a man on the go - it's nearly empty. A bulging briefcase sits at Kelling's feet.
Kelling didn't start out to study cops. His first taste of police work came while he was attending a Lutheran seminary and volunteered to help troubled youths. Attracted more to his volunteer work than to his books, Kelling dropped out and became a probation officer in Minneapolis. Eventually, he went back to school for his master's degree in social work, began his work as a criminologist, and after earning a PhD from the University of Wisconsin was snapped up by the fledgling Police Foundation in Washington.
The Police Foundation was starting to evaluate policing tactics, to find what worked in one city and to replicate it in others. One of Kelling's first projects was to study police officers walking the beat here in Newark.
"That was where I got the idea [for the broken-windows theory]," Kelling says. One thing he noticed is that residents felt safer when cops walked the streets. "We had this remarkable response in terms of fear reduction," he says. "If you added foot-patrol officers, fear dropped and appreciation of police soared. And if you took it away, appreciation dropped and fear increased.
"The question is: Why should fear drop so much? The more I looked at my own field notes, the more I recalled that I was watching this low-level order maintenance," Kelling says. Police, by being on foot, got to know the community where they were working. Because of this rapport, they knew who the troublemakers were and could arrest them for the smallest infractions, to get them off the streets. They also found ways to respond to community concerns about minor, but aggravating, crimes, he says.
The irony is that the Police Foundation had sent Kelling to Newark to prove that foot patrols were an antiquated idea. His marching orders had been to show that foot patrols were a waste of time, a finding that would buttress police department's determination to get back to their "real" job of tracking violent offenses. Instead, it became Kelling's life work to understand what he had seen in Newark's foot patrol.
In 1982, Kelling published the "Broken Windows" essay with colleague James Q. Wilson. He thought he had his big break soon after. The National Institute of Justice in Washington wanted to sponsor test sites for the theory in Newark and Houston. The US attorney general planned to announce the initiatives.
"His staff took one look at the idea of order maintenance and canceled the experiment," Kelling says. "It was just too politically volatile even for a conservative Republican administration to deal with."
Kelling moved to Massachusetts and Harvard University, and continued low-level research on the idea of order maintenance. While waiting for an opportunity to test his theory, he started working with Bratton, then a Boston police sergeant. His break came with New York's subway system, which brought in Bratton to head it.
The rest, as they say, is history.
"A lot of things happened in the '80s to make it very difficult for police to think that anything they were doing was having an effect," says Professor Wilson, now a political scientist at the University of California, Los Angeles. There were the drug wars, the rise of violent youth gangs, and the growth of police corruption. Creating specialized units and controlling officers from headquarters were seen as the answers.
"It was an interesting solution," Wilson says. "In my view, we literally depoliced America's streets."
Kelling says he expects the broken-windows theory to be put into practice in more cities, as the public and professional organizations apply pressure to control crime. He also anticipates that continuing court cases will more clearly distinguish between what constitutes order maintenance and what amounts to abuse of authority.
He sees two tasks ahead for America's police departments. One is to better explain community policing to their own ranks - and to portray it as aggressive, preventative policing that is not soft on crime.
The other is to establish guidelines for professional police behavior. Police headquarters often seek to regulate officers by spelling out rules for each police action. Now, Kelling says, departments need to define acceptable behavior but at same time allow officers enough discretion to be effective problem-solvers.
"Fortunately, now we've had some real wins," Kelling says. The experience in "New York, in both the subways and the streets, has really brought the debate into the public in ways that it hasn't been in a long time."