Dr. George Kelling

September 10, 2008

Thank you for the opportunity to meet with you.

During the past five years, I have worked on the ground in six cities: Newark, Los Angeles, Denver, Boston, Milwaukee, and Allentown (PA). In Newark, homicide is down in comparison to 2007 by 40%; in Los Angeles, 9% (a two year decline of 23%); Milwaukee, 30%; Boston, 13%; and Denver, 22%. Allentown's homicide rate has held steady, but our efforts have just begun there. In these cities I have worked with political and community leaders, citizens in neighborhoods, public and private agencies, and police officials ranging from chiefs to line police officers. Two common threads run through my experiences in each of these cities: first, the need for leadership, and, second, a shift in approach on the part of all concerned from reacting to crime after it occurs to "stopping the next crime."

The sources of leadership in addressing crime problems vary from community to community: in some locations it is political; in others, police; in others, both; in yet others, a mix of private and public agencies. Regardless, these leaders, almost all of whom saw violence surge in their cities, have been appalled by the carnage on their streets and deeply committed to ending it. As important, they had learned from colleagues, the literature, and their own experiences what had to be done to reduce violence. Specifically, they understood that the reactive model of crime control had failed miserably and that they had to take political and organizational risks to field effective violence prevention.

In the following, I will describe briefly the basic methods of crime prevention. I can provide more detail if there are questions. Second, I will revisit the experience of New York City, a city enjoying crime declines that arguably are unparalleled in history and from which I believe there is much to learn. There is much I will not discuss that relates to crime prevention and reduction: the need for social, spiritual, recreational, and educational services; employment; family assistance and support; and others. My focus, instead, will be on five proximate measures that most communities could move to immediately. None are very "sexy" or even new, but conceiving, implementing, and sustaining the programmatic forms they take can be complicated, depending on the agency, its resources, and the shape of the problems. Some measures will have impact on their own, others, depending on the neighborhood and the problems, will need to be implemented in tandem. In brief, the five are:

1. Increase the "Felt" Presence of Capable Guardians: Starting with police but moving on to prosecution, probation and parole, other governmental agencies, and even the courts, we must increase the real presence of each in neighborhoods. For police this means getting out of their cars, walking, riding bicycles, meeting with citizens, and in other ways becoming an active neighborhood presence. In prosecution, it would mean having community prosecutors meet regularly with citizens in neighborhoods to understand their problems and devise solutions. I could give examples for other disciplines as well.

2. Persuade People, Especially the Young, to Behave: Law enforcement agencies and others involved in crime reduction efforts need to think beyond formal measures. Among the most fundamental and successful tactics is persuasion. Persuading people can range from simply "talking to them" to complicated programs that link active law enforcement with persuasive ways of communicating with young people on the verge of serious trouble. Both John Jay College Professor David Kennedy and University of Illinois, Chicago Campus Professor Gary Slutkin have developed model programs to persuade people, especially violence-prone youths, to "back off." Kennedy focuses on persuasive efforts by law enforcement officials themselves while Slutkin's program uses reformed young people.

3. Restore Order: I am, of course, referring here to an idea that I helped develop: "broken windows." Put simply, broken windows argues that for a community to be safe and prosperous minimal levels of order must be established and maintained. It is no secret that broken windows has come under considerable academic criticism. Certainly, a broken windows approach - that is, aggressive "paying attention" to minor offenses and disorderly behavior - can be done inappropriately. Yet every city in which I have worked that has achieved substantial crime reduction has also paid careful attention to maintaining order--and with considerable success. I will discuss broken windows in more detail in the second half of my presentation.

4. Solve Problems: Until recently, police and other criminal justice agencies have treated violent acts as independent incidents rather than symptoms of problems with both history and future. Right now a relatively small number of youths are carrying guns, dealing drugs, draping themselves in gang colors, and shooting each other to settle what are often trivial disputes. The effects on communities are disastrous and degrading. While we certainly want police and other agencies, especially prosecutors, to be concerned about individual cases and offenders, they need to be equally concerned about the community problems that such cases represent and create. Both the Department of Justice and the Police Executive Research Forum have developed inventories of successful problem-solving efforts that address both the prevention of crimes themselves and contextual aspects that facilitate or discourage crime and related conditions in a community. These strategies can inspire similar innovations in other cities.

5. When Formal Measures are Appropriate, Enforce the Law Swiftly and Fairly: Finally, law enforcement. I will not say much about this here. Let me summarize, however, by noting that a small population of offenders is busily nominating itself for incarceration by repeatedly committing both minor and serious offenses. For this group, we should have no reluctance to imprison them for extended periods of time. Unfortunately, however, there are at least two problems: first, in the absence of other preventive measures, incarceration has been overused and, second, law enforcement has been applied so capriciously that it often fails to serve as a deterrent to the "wanna bes" and other youths at the edge.

In sum, there are five proximate means of preventing and reducing violence: presence, persuasion, broken windows, solving problems, and law enforcement. I have no doubt that any community, given dedication, leadership, and sufficient resources can reduce violence. As Los Angeles Chief of Police William Bratton has said, "Preventing crime is not rocket science."

The primary question facing us now is: once we have initially reduced violence in a neighborhood, how do we sustain those gains? I think that close examination of what happened in New York City will help us answer this question. Let me summarize what I believe really happened in New York City.

During the late 1970s, the 1980s, and early 1990s a demand for order developed out of a disorder and crime crisis - things were simply so bad that citizens and institutions wanted change in, or wanted out of, New York. An idea surfaced as early as the 1970s - call it quality of life, broken windows, or what have you - that order had broken down and threatened the viability of the city itself. A theory of action developed across a wide swath of public and private organizations and sectors that specified goals and tactics (the goal was restoration of order, the methods ranged from law enforcement in the subway to environmental design in Bryant Park). And, this theory of action was implemented aggressively and persistently by a diverse set of organizations ranging from public transportation to Business Improvement Districts (BIDS). By the mid-1990s, police became seriously involved. Under Mayor Giuliani and Commissioner Bratton, police not only adopted a congruent theory of action, they brought with them their under-developed and under-utilized crime prevention capacities.

I would explain both the steepness and persistence of the crime decline in NYC as resulting from the fervent pursuit of self-interest by a critical mass of public and private agencies operating out of a congruent understanding of the nature of the problems and their solutions. Their self-interests included economic, neighborhood safety, the ability to provide services, and others. When joined by the NYPD, with its common understanding and its additional and unique capacities, this critical mass reached a tipping point.

If my interpretation is correct - a set of important agencies consistently pursued an idea and theory of action that resulted in NYC's steep and persistent crime declines - this also helps explain why, as Milwaukee Police Chief Edward Flynn comments, "other cities that attempt to emulate NYPD by focusing on 'comp-stat' identifying the links between order in public spaces, fear, and crime do not generate the same dramatic outcomes." He goes on: "My experience leading five police agencies is that 'critical mass' is by far the exception rather than the rule." This does not mean that emulating aspects of NY, especially Compstat, is without profit; it means that the achievements are limited - the tipping point that literally changes a city's culture cannot be reached.

Summarizing, we now have a lot of knowledge about ways to prevent crime that if assiduously applied reduce violence. For violence reduction to be sustained, however, a common theory of action must activate a critical mass of community agencies and resources. Without such a common theory of action, cities and communities will pick away here and there, never reaching the tipping point that New York City has.

Two final comments that might seem unrelated to what I have said above, but in fact are not. Both have to do with the fact that the war on terror and related assumptions that terror and common crimes are essentially different problems have resulted in the virtual gutting of the National Institute of Justice (NIJ), Bureau of Justice Assistance (BJA), and Community Oriented Policing Services (COPS), all in the name of terror prevention. These assumptions are not only faulty, they have been a disaster for localities. In fact, terrorists commit common crimes on their way towards terrorist acts and, in doing so, are vulnerable to action by local police.

Second, ongoing support for local law enforcement efforts is crucial to their future success. Their accomplishments in reducing crime during the last 10-12 years is a direct result of the research conducted during the last 40 years. This is not just a pitch for resources for research and other types of support from an academic--every chief with whom I have worked over the past years would make the same claim. If we are to maintain, and improve on, our gains of the recent past, the federal government must view ongoing crime control research and support as equally essential to that needed for medical problems. Both crime problems and terrorism in many senses are ultimately local problems, and must and will be resolved locally. Locals, however, are in need of support.

Thank you for your attention.